Article
To read the full-text of this research, you can request a copy directly from the authors.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The majority of remote sensing methods use passive optical sensors with ratios of different spectral reflectances to calculate LAI [1,3,5]. Classically, the calculations are based on vegetation indices such as the normalized difference vegetation index (NDVI), which is sensitive to chlorophyll absorption [6]. This often becomes a measurement of GAI which relies on the plant pigment [5]. ...
... This often becomes a measurement of GAI which relies on the plant pigment [5]. These techniques may be impacted by the illumination conditions during collection [7] along with background elements [6,8]. Cereal crops, such as winter wheat, begin browning with their spectral reflectance becoming similar to that of soil as they approach senescence, where the greenness no longer reflects the whole surface area of the aboveground biomass, i.e., PAI, and passive optical methods are challenged [5]. ...
... Considering that LiDAR provides a better characterization of the crop canopy structure, combining the LiDAR height and PAILiDAR values may give better estimates of biomass than previously seen with multispectral and photogrammetry methods [2]. GAI of winter wheat could also be calculated using the intensity of the LiDAR returns since the signal is emitted close to NIR [6]. The more products that this sensor can calculate without supporting data from other sensors and as a standalone sensor, the more efficient it becomes. ...
Article
Full-text available
Monitoring of canopy density with related metrics such as leaf area index (LAI) makes a significant contribution to understanding and predicting processes in the soil–plant–atmosphere system and to indicating crop health and potential yield for farm management. Remote sensing methods using optical sensors that rely on spectral reflectance to calculate LAI have become more mainstream due to easy entry and availability. Methods with vegetation indices (VI) based on multispectral reflectance data essentially measure the green area index (GAI) or response to chlorophyll content of the canopy surface and not the entire aboveground biomass that may be present from non-green elements that are key to fully assessing the carbon budget. Methods with light detection and ranging (LiDAR) have started to emerge using gap fraction (GF) to estimate the plant area index (PAI) based on canopy density. These LiDAR methods have the main advantage of being sensitive to both green and non-green plant elements. They have primarily been applied to forest cover with manned airborne LiDAR systems (ALS) and have yet to be used extensively with crops such as winter wheat using LiDAR on unmanned aircraft systems (UAS). This study contributes to a better understanding of the potential of LiDAR as a tool to estimate canopy structure in precision farming. The LiDAR method proved to have a high to moderate correlation in spatial variation to the multispectral method. The LiDAR-derived PAI values closely resemble the SunScan Ceptometer GAI ground measurements taken early in the growing season before major stages of senescence. Later in the growing season, when the canopy density was at its highest, a possible overestimation may have occurred. This was most likely due to the chosen flight parameters not providing the best depictions of canopy density with consideration of the LiDAR’s perspective, as the ground-based destructive measurements provided lower values of PAI. Additionally, a distinction between total LiDAR-derived PAI, multispectral-derived GAI, and brown area index (BAI) is made to show how the active and passive optical sensor methods used in this study can complement each other throughout the growing season.
... Morphological parameters, such as plant height, stem diameter, leaf area or leaf area index (LAI), leaf angle, stalk length, and in-plant space [1], can be determined with LiDAR (light detection and ranging). Research on phenotyping using LiDAR often focusses on one specific crop, for example, wheat [2,3] or cotton [4]. ...
... For high-throughput phenotyping, traits such as plant-height, LAI, and leaf cover fraction are determined directly in the field, using a LiDAR-based system mounted on a vehicle or RGB cameras mounted on a UAV (unmanned aerial vehicle) [3,4,7]. Tractor based LiDAR systems data have shown good correlation with in situ field measurements of plant height. ...
Article
Full-text available
Phenotyping of crops is important due to increasing pressure on food production. Therefore, an accurate estimation of biomass during the growing season can be important to optimize the yield. The potential of data acquisition by UAV-LiDAR to estimate fresh biomass and crop height was investigated for three different crops (potato, sugar beet, and winter wheat) grown in Wageningen (The Netherlands) from June to August 2018. Biomass was estimated using the 3DPI algorithm, while crop height was estimated using the mean height of a variable number of highest points for each m2. The 3DPI algorithm proved to estimate biomass well for sugar beet (R2 = 0.68, RMSE = 17.47 g/m2) and winter wheat (R2 = 0.82, RMSE = 13.94 g/m2). Also, the height estimates worked well for sugar beet (R2 = 0.70, RMSE = 7.4 cm) and wheat (R2 = 0.78, RMSE = 3.4 cm). However, for potato both plant height (R2 = 0.50, RMSE = 12 cm) and biomass estimation (R2 = 0.24, RMSE = 22.09 g/m2), it proved to be less reliable due to the complex canopy structure and the ridges on which potatoes are grown. In general, for accurate biomass and crop height estimates using those algorithms, the flight conditions (altitude, speed, location of flight lines) should be comparable to the settings for which the models are calibrated since changing conditions do influence the estimated biomass and crop height strongly.
... Lidar includes a single-beam narrowband laser and a receiving system. The laser generates and emits a light pulse, which hits the target object and reflects back to a receiver [199]. ...
... The lidar point cloud from UAVs can be processed by CloudCompare or by writing code as needed [112], [199]. Currently, UAVs can carry different sensors (RGB cameras, multi/hyperspectral cameras, thermal cameras, and lidar) to estimate different crop traits at varying spatial scales. ...
Article
Crop yields need to be improved in a sustainable manner to meet the expected requirements of the worldwide increase in population in the coming decades and with anticipated climate change. Genomics-assisted breeding has become popular in recent years as an approach to contribute to food security, and in this regard, the crop breeding community needs to better link the relationships between phenotype and genotype. While high-throughput genotyping is feasible at low cost, high-throughput crop phenotyping methods and the data analytical capacity need to be improved. High-throughput phenotyping offers powerful methods to assess particular phenotypes in large-scale experiments to monitor and quantify plants in breeding nurseries and field experiments at multiple scales, using high-tech sensors, advanced robotics and image processing systems. In addition, new bioinformatics platforms are able to embrace large-scale multi-dimensional phenotypic datasets. Environmental responses and gene functions may now be dissected at unprecendent resolution with the combined analysis of phenotyping and genotyping data. This will aid in finding better solutions to the current limitations of incremental improvements in crop yields. This review will show the origin and definition of crop phenotyping. Secondly, it will introduce the developments of current sensors for crop phenotyping traits in the field, discussing the advantages and limitations of different technologies available and their potential applications. Thirdly, it will highlight the development of ground and aerial phenotyping platforms and their applications for crop phenotyping traits in the field. Finally, new opportunities and developments for crop phenotyping studies in the future will be discussed.
... Images can be also taken from the top of the canopy to get the green fraction (GF, the fraction of green pixels in an image) by identifying the green pixels from which the GAI will be estimated. 3D techniques have been also developed more recently using terrestrial laser scanners (Liu et al. 2017;Yan et al. 2019) or stereovision (Biskup et al. 2007). The resulting 3D point clouds are exploited to get the directional canopy transmittance and derive the corresponding PAI if no distinction is made between the green and non-green elements, or to get the directional green fraction leading to GAI estimates when the green points are identified. ...
... For both species, 810 canopies were generated according the variables listed in Table 11. The range of values considered for the input variables were derived from previous experiments Abichou et al. 2013;Liu et al. 2017). Each input variables was varied between the minimum and maximum values by a number of equal steps ( Table 1). ...
Thesis
Measuring leaf and canopy characteristics from remote sensing acquisitions is an effective and non destructive way to monitor crops both for decision making within the smart agriculture practices or for phenotyping under field conditions to improve the selection efficiency. With the advancement of computer computing power and the increasing availability of high spatial resolution images, retrieval methods can now benefit from more accurate simulations of the Radiative Transfer (RT) models within the vegetation. The objective of this work is to propose and evaluate efficient ways to retrieve leaf and canopy characteristics from close and remote sensing observations by using RT models based on a realistic description of the leaf and canopy structures. At the leaf level, we first evaluated the ability of the different versions of the PROSPECT model to estimate biochemical variables like chlorophyll (Cab), water and dry matter content. We then proposed the FASPECT model to describe the optical properties differences between the upper and lower leaf faces by considering a four-layer system. After calibrating the specific absorption coefficients of the main absorbing material, we validated FASPECT against eight measured ground datasets. We showed that FASPECT simulates accurately the reflectance and transmittance spectra of the two faces and overperforms PROSPECT for the upper face measurements. Moreover, in the inverse mode, the dry matter content estimation is significantly improved with FASPECT as compared to PROSPECT. At the canopy level, we used the physically based and unbiased rendering engine, LuxCoreRender to compute the radiative transfer from a realistic 3D description of the crop structure. We checked its good performances by comparison with the state of the art 3D RT models using the RAMI online model checker. Then, we designed a speed-up method to simulate canopy reflectance from a limited number of soil and leaf optical properties. Based on crop specific databases simulated from LuxCoreRender for wheat and maize and crop generic databases simulated from a 1D RT model, we trained some machine learning inversion algorithms to retrieve canopy state variables like Green Area Index GAI, Cab and Canopy Chlorophyll Content (CCC). Results on both simulations and in situ data combined with SENTINEL2 images showed that crop specific algorithms outperform the generic one for the three variables, especially when the canopy structure breaks the 1D turbid medium assumption such as in maize where rows are dominant during a significant part of the growing season.
... Our work provides data that could be used in structural, data-driven 3D models such as Adel-Wheat (Fournier et al., 2003), which allow calculation of plant-environment interactions based on realistic 3D dynamic structures. This would help in defining ideotypes for light capture, and to investigate other consequences of the changes in sowing design and plant architecture, such as the perception of light quality (Chelle et al., 2007), propagation of pathogens (Robert et al., 2018), interception of pesticides (Dorr et al., 2008), or the formation of phenotyping signals (Liu et al., 2017). ...
... These observations correspond to a single year-site experiment and involved intrusive methods to characterize the 3D architecture. High throughput phenotyping with 2D and LiDAR imagery (Liu et al., 2017) allows investigating in field conditions and without any contact with the plants, a large number of cultivar traits such as position of individual ears and leaf orientation, which in the present study were measured by digitization. In this context, manipulating interrow distance is a convenient way to create reproducible representations of spatial heterogeneity. ...
Article
Full-text available
Achieving novel improvements in crop management may require changing interrow distance in cultivated fields. Such changes would benefit from a better understanding of plant responses to the spatial heterogeneity in their environment. Our work investigates the architectural plasticity of wheat plants in response to increasing row spacing and evaluates the hypothesis of a foraging behavior in response to neighboring plants. A field experiment was conducted with five commercial winter wheat cultivars possessing unique architectures, grown under narrow (NI, 17.5 cm) or wide interrows (WI, 35 cm) at the same population density (170 seeds/m2). We characterized the development (leaf emergence, tillering), the morphology (dimension of organs, leaf area index), and the geometry (ground cover, leaf angle, organ spreading, and orientation). All cultivars showed a lower number of emerged tillers in WI compared to NI, which was later compensated by lower tiller mortality and by shoots producing larger blades. The rate of leaf emergence and the final leaf number were higher in WI compared to NI, except for one cultivar. Around the start of stem elongation, pseudo-stems were more erect in WI, while around the time of flowering, stems were more inclined and leaves were more planophile. Cultivars differed in their degrees of responses, with one appearing to prospect more specifically within the interrow space in WI treatment. Altogether, our results suggest that altering interrow distance leads to changes in the perceived extent of competition by plants, with responses first mimicking the effect of a higher plant density and later the effect of a lower plant density. Only one cultivar showed responses that suggested a perception of the heterogeneity of the environment. These findings improve our understanding of plant responses to spatial heterogeneity and provide novel information to simulate light capture in plant 3D models, depending on cultivar behavior.
... This characteristic allows for accurate canopy height measurements (Harkel et al., 2020) and can also provide canopy density metrics in respect to the rate of signal penetration (Bates et al., 2021). The intensity of the LiDAR signal which is often within the near infrared (NIR) bandwidth can also be an indicator for green area index (GAI) (Liu et al., 2017). In this study we evaluate the combination of LiDAR height, intensity, and multilayer density products within an ANN model when monitoring winter wheat over the growing season. ...
... In our case, this was one hidden layer consisting of 4 nodes. (Liu et al., 2017). The model training and implementation with raster data was done in RStudio with the neuralnet package. ...
Article
Full-text available
Biomass is an important indicator in the ecological and management process that can now be estimated at higher temporal and spatial resolutions because of unmanned aircraft systems (UAS). LiDAR sensor technology has advanced enabling more compact sizes that can be integrated with UAS platforms. Its signals are capable of penetrating through vegetation canopies enabling the capture of more information along the plant structure. Separate studies have used LiDAR for crop height, rate of canopy penetrations as related to leaf area index (LAI), and signal intensity as an indicator of plant chlorophyll status or green area index (GAI). These LiDAR products are combined within a machine learning method such as an artificial neural network (ANN) to assess the potential in making accurate biomass estimations for winter wheat.
... More importantly, the shortwavelength laser can penetrate the vegetation canopy to characterize vegetation inner structures, which compensates for the deficiencies of previously-mentioned optical image technologies (Berk et al., 2016). During the past three decades, lidar has been widely used to quantify plant structure (Coops et al., 2007;Davies and Asner, 2014;Dubayah and Drake, 2000;Hyyppä et al., 2020b;Lefsky et al., 2002;Liang et al., 2016;Liu et al., 2017b;Nelson et al., 1988;Omasa et al., 2006;Popescu and Zhao, 2008;Saarinen et al., 2017;Zhao et al., 2018) and has been actively studied to extract phenotypes for boosting plant breeding and improving plant management in the last decade (Asner et al., 2017;Calders et al., 2020;Cao et al., 2019;Hosoi and Omasa, 2012;Lim et al., 2003b;Popescu and Wynne, 2004;Simonson et al., 2012;Wulder et al., 2012). ...
... Mobile lidar systems have been developed for a variety of environments with the advancements in mobile measurement technologies (Yuan et al., 2019), which have lifted observation angles and reduced labor costs along with high-throughput capability of outdoor lidar systems . Unmanned ground vehicle (UGV) platforms are used primarily in simple working environments , such as orchards, but have been also developed for phenotyping large collections of genotypes (Liu et al., 2017b). By contrast, manned ground vehicle systems, such as mounting on a sprayer , are more flexible and have wider applications (Bai et al., 2016;Jimenez-Berni et al., 2018). ...
Article
Full-text available
Plant phenomics is a new avenue for linking plant genomics and environmental studies, thereby improving plant breeding and management. Remote sensing techniques have improved high-throughput plant phenotyping. However, the accuracy, efficiency, and applicability of three-dimensional (3D) phenotyping are still challenging, especially in field environments. Light detection and ranging (lidar) provides a powerful new tool for 3D phe-notyping with the rapid development of facilities and algorithms. Numerous efforts have been devoted to studying static and dynamic changes of structural and functional phenotypes using lidar in agriculture. These progresses also improve 3D plant modeling across different spatial-temporal scales and disciplines, providing easier and less expensive association with genes and analysis of environmental practices and affords new insights into breeding and management. Beyond agriculture phenotyping, lidar shows great potential in forestry, horticultural , and grass phenotyping. Although lidar has resulted in remarkable improvements in plant phenotyping and modeling, the synthetization of lidar-based phenotyping for breeding and management has not been fully explored. We identify three main challenges in lidar-based phenotyping development: 1) developing low cost, high spatial-temporal, and hyperspectral lidar facilities, 2) moving into multi-dimensional phenotyping with an endeavor to generate new algorithms and models, and 3) embracing open source and big data.
... The traditional plant phenotype measurement method is inefficient, unable to meet the needs of high-throughput and automated phenotypic analysis, and cannot achieve real-time monitoring. At present, the sensors used in plant phenotyping research mainly include red-green-blue (RGB) cameras [4,5], thermal cameras [6,7] fluorescence sensors [8][9][10], multispectral cameras [6], hyperspectral cameras [11,12] and lidar devices [13,14], etc. Compared to other sensors, RGB cameras have the advantage of being affordable and providing high-resolution images. ...
Article
Full-text available
In crop growth management, phenotypic traits are an important basis for judging growth status. Manual measurements are labor-intensive, unstable and time-consuming. We propose an image processing pipeline to estimate multiple lettuce traits (fresh weight, dry weight, plant height, diameter, leaf area) based on the lightweight DeepLabV3 + network. Accurate and rapid segmentation of crops from backgrounds is the basis for phenotypic research. First, we propose to combine DeepLabV3 + and MobilenetV2 to realize a high-precision and fast segmentation of lettuce in complex backgrounds and illuminations. Based on the segmentation results, we extracted the morphological factors and vegetation indices. Random forest (RF), partial least squares regression (PLSR) and support vector machine were applied to predict the multiple lettuce traits and compared for optimal model selection. Results showed that DeepLabv3 + (with Mobilenetv2) has the best segmentation performance with pixel accuracy of 97.520% and 99.821%, mIoU of 88.661% and 98.517%, and segmentation speeds with 0.094 and 0.049 ms per image in dataset D3 and dataset D4. PLSR had the highest accuracy in predicting fresh weight, dry weight, diameter and leaf area, with \({R}^{2}\) of 0.898, 0.899, 0.931 and 0.904, respectively. RF yielded the highest accuracy in predicting plant height, with \({R}^{2}\) of 0.858. We proposed method for estimating phenotypic characteristics of lettuce based on deep learning has excellent performance and important application value for lettuce growth monitoring and yield estimation.
... The conventional method of phenotype acquisition is mainly urate-based. Phenotype traits obtained using images can be used for studies such as biomass estimation [6], nutritional diagnosis [7], growth and development monitoring [8], plant structural analysis [9], quantitative description of phenotypic traits [10], and pest and disease identification [11]. In recent years, significant breakthroughs have been achieved in digital image processing using deep learning; the performance of deep learning-based image processing has been much better than traditional methods in applications such as object recognition and segmentation [12][13][14]. ...
Article
Full-text available
Deep learning can enable image-based high-throughput phenotype analysis of plants. However, deep learning methods require large amounts of artificially annotated data. For application in plant phenotyping, the available data sets are usually small; it is expensive to generate new data and challenging to improve model accuracy with limited data. In this study, the L-system was used to generate virtual image data for training deep learning models. The precision (P), recall (R), and F-score (F) of the image segmentation model using a combination of virtual data and real data reached 0.95, 0.91, and 0.93, respectively; Mean Average Precision (mAP) and Intersection over Union (IoU) of the target detection model reached 0.96 and 0.92, respectively; the coefficient of determination (R2) and the standardized root mean square error evaluation of the leaf count model reached 0.94 and 0.93, respectively; all the results outperformed the results of training with only real data. Thus, we demonstrated that virtual data improves the effectiveness of the prediction accuracy of deep neural network models, and the findings of this study can provide technical support for high-throughput phenotype analysis.
... Among all the indirect methods available for estimating LAI, the technique that uses hemispherical images taken with a fisheye-type lens is the most used due to its robustness [30,31]. This method is based on the estimated position, size, density, and distribution of canopy gaps, which characterize the canopy geometry through which the intercepted solar radiation is measured [32]. The gap fraction is calculated using thresholding in order to distinguish pixels that are occupied by leaves from pixels that are occupied by the sky or ground [33]. ...
Article
Full-text available
Remote and non-destructive estimation of leaf area index (LAI) has been a challenge in the last few decades as the direct and indirect methods available are laborious and time-consuming. The recent emergence of high-throughput plant phenotyping platforms has increased the need to develop new phenotyping tools for better decision-making by breeders. In this paper, a novel model based on artificial intelligence algorithms and nadir-view red green blue (RGB) images taken from a terrestrial high throughput phenotyping platform is presented. The model mixes numerical data collected in a wheat breeding field and visual features extracted from the images to make rapid and accurate LAI estimations. Model-based LAI estimations were validated against LAI measurements determined non-destructively using an allometric relationship obtained in this study. The model performance was also compared with LAI estimates obtained by other classical indirect methods based on bottom-up hemispherical images and gaps fraction theory. Model-based LAI estimations were highly correlated with ground-truth LAI. The model performance was slightly better than that of the hemispherical image-based method, which tended to underestimate LAI. These results show the great potential of the developed model for near real-time LAI estimation, which can be further improved in the future by increasing the dataset used to train the model.
... Non-contact measurement methods are mainly divided into active and passive ranging methods. These methods provide efficient way to estimate tree height and crown diameter [14][15]. The remote sensing technology supports for large-scale forest resource survey and management, which can help to timely understand the dynamic changes of vegetation [16]. ...
Article
Full-text available
The tree height and crown diameter are important measurement attributes in forest resource survey and management. Hence, we propose a passive measurement method of tree height and crown diameter based on monocular camera of a smartphone. First, we use an feature-adaptive Mean-Shift algorithm to segment the image and extract tree’s contour. Furthermore, an adaptive feature coordinate system is established to help study the conversion relationship of the coordinate systems. It has been proved that for the image points with the same abscissa pixels, their ordinate pixels have a linear relationship with its actual imaging angles. A depth extraction model is built according to this principle. Then, we obtain the rotation and translation matrix and established tree height and crown diameter models according to the mapping transformation relationship of coordinates. Experimental results reveal significant correlation between calculated and truth values. The RMSE is 0.267 m ( rRMS=2.482%) for tree height and 0.209 m ( rRMS=5.631%) for crown diameter. The relative errors of tree heights are less than 5.76 (MRE=2.159%); for crown diameter, the relative errors are less than 9.73% (MRE=4.95%). Overall, the accuracy of this method falls within the requirements of the continuous inventory of Chinese national forest resources.
... These models aim to produce a complete description of the dynamics of the 3-D architecture, based on a limited number of measurements. They have shown a number of applications related to the calculation of plant-environment interactions (Barillot et al., 2014;Liu et al., 2017;Robert et al., 2018;Vidal et al., 2018). If complemented with a 3-D description of leaves such as used in Fournier and Andrieu (1998) or Guo et al. (2006), the present model could be the basis for a structural model for maize which would be generic enough to represent a wide range of phenotypes. ...
Article
Background and aims: The dynamics of plant architecture is a central aspect of plant and crop models. Most models assume that whole shoot development is orchestrated by the leaf appearance rate, which follows a thermal time schedule. However, leaf appearance actually results from leaf extension and taking it as an input hampers our ability to understand shoot construction. The objective of this work was to assess a modelling framework for grasses, in which the emergence of leaves and other organs is explicitly calculated as a result of their extension. Methods: The approach builds on the model of Zhu et al. (2014), which uses a set of rules co-ordinating the timing of development within and between phytomers. We first assessed rule validity for four experimental data sets, including different cultivars, planting densities and environments and accordingly revised the equations driving the extension of the upper leaves and of internodes. We then fitted model parameters for each data set and evaluated the ability to simulate the measured phenotypes across time. Finally, we carried out a sensitivity analysis to identify the most impacting parameters and investigate model behaviour. Key results: The modified version of the model simulated correctly the contrasting maize phenotypes. Co-ordination rules accounted for the observations in all studied cultivars. Factors with major impact on model output included the extension rates, the time of tassel initiation and initial conditions. A large diversity of phenotypes could be simulated. Conclusions: This work brought direct experimental evidence for co-ordination rules and illustrated the capacity of the model to represent contrasting phenotypes. These rules play an important role in patterning shoot architecture and some of them should be further assessed, considering contrasting growth conditions. To make the model more predictive, several parameters could be considered in the future as internal variables driven by plant status.
... More developments in 3D modeling and inversion are thus required to develop operational pipelines over given species . This can be supported by field phenotyping experiments that allows detailed description of crop architecture thanks to 3D point clouds generated by Structure from Motion (SfM) or LIDAR techniques ( Liu et al., 2017). ...
Article
Full-text available
Agriculture provides humanity with food, fibers, fuel, and raw materials that are paramount for human livelihood. Today, this role must be satisfied within a context of environmental sustainability and climate change, combined with an unprecedented and still-expanding human population size, while maintaining the viability of agricultural activities to ensure both subsistence and livelihoods. Remote sensing has the capacity to assist the adaptive evolution of agricultural practices in order to face this major challenge, by providing repetitive information on crop status throughout the season at different scales and for different actors. We start this review by making an overview of the current remote sensing techniques relevant for the agricultural context. We present the agronomical variables and plant traits that can be estimated by remote sensing, and we describe the empirical and deterministic approaches to retrieve them. A second part of this review illustrates recent research developments that permit to strengthen applicative capabilities in remote sensing according to specific requirements for different types of stakeholders. Such agricultural applications include crop breeding, agricultural land use monitoring, crop yield forecasting, as well as ecosystem services in relation to soil and water resources or biodiversity loss. Finally, we provide a synthesis of the emerging opportunities that should strengthen the role of remote sensing in providing operational, efficient and long-term services for agricultural applications.
... A few recent studies used synthetic DHP for evaluating LAI retrieval (Leblanc and Fournier, 2014;Zou et al., 2018). Other studies utilized simulated LiDAR point clouds to evaluate the estimation accuracy of tree height (Hämmerle et al., 2017), light interception (Perez et al., 2018), LAI (Liu et al., 2017b;Chen et al., 2018), and leaf inclination angle Vicari et al., 2019). To the best of our knowledge, no studies assessing TLS and DHP estimation of LIDF have been carried out. ...
Article
Leaf inclination plays a crucial role in regulating the radiation, carbon and water fluxes in plant canopies. Accurate measurement of the probability density function of leaf inclination (i.e., the leaf angle distribution or leaf inclination distribution function (LIDF)), is very important for modelling photosynthesis as well as measuring leaf area index. In spite of its importance, in situ measurement of LIDF is very challenging. Both digital hemispherical photography (DHP) and terrestrial LiDAR (TLS) have been used to measure LIDF. However, the consistency and relative accuracy of these two techniques has never been evaluated. In this research, we aimed to evaluate which in situ technique, either DHP or TLS, could measure LIDF more accurately, with respect to both field-based and synthetic datasets. The field-based datasets were collected from 36 natural European beech stands covering a range of forest structures. The synthetic datasets were generated from 44 virtual forest scenes using TLS and DHP simulators. Due to the inability of differing leaf and woody materials in DHP, the average plant inclination angle from DHP and TLS was selected for LIDF comparison. The average inclination angle was retrieved from TLS point clouds using a geometric method, and from DHP using three gap fraction inversion methods including the NC method (Norman and Campbell, 1989), as well as the CAN-EYE and Hemisfer software. Results from the field-based datasets showed a significant difference and inconsistency between the average inclination angle retrieved from TLS and DHP (respectively (40°, 58°) from TLS, (15°, 86°) from DHP NC, (36°, 78°) from DHP CAN-EYE, (0°, 67°) from DHP Hemisfer). Results from the synthetic datasets demonstrated that the accuracy from TLS was considerably higher than that obtained from DHP (R2: 0.90 > 0.74; RMSE: 5.38° < 13.30°). This study demonstrated that the LIDF estimated from TLS and DHP were not coherent. Based on the experimental results as well as deduction from theoretical arguments, we recommended using TLS when measuring leaf inclination in broadleaf forests, especially for stands with a heterogeneous structure.
... Despite the existing successes in applying the gap model of Equation 1 or its variants for estimation of LAI and LAD, many practical issues remain (Chianucci & Cutini, 2012;Li et al., 2016;Liu et al., 2017;Weiss, Baret, Smith, Jonckheere, & Coppin, 2004). Foremost, existing algorithms require first converting raw data, such as hemiphotos, into gap fractions. ...
Article
1.Probabilistic modeling of gaps for light‐canopy interactions has long served as a theoretical basis to estimate vegetation structural parameters—leaf area index (LAI) and leaf angle distribution (LAD)—from optical measurements such as hemispherical photos. Direct inversion of such probabilistic models provides a reliable statistical algorithm for parameter estimation, but this inferential paradigm has been seldom explored. Even worse, many classical LAI algorithms implicitly assume “wrong” statistical models inconsistent with the underlying probabilistic gap models—a subtle issue not articulated before but known to cause practical issues. 2. Here, we clarified how to improve LAI and LAD estimation by directly inverting binary gap/non‐gap data of hemispherical photos via binary nonlinear regression (BNR). We implemented the new BNR method and some classical algorithms in an R package “hemiphoto2LAI”, comprising a total of 135 models for LAI estimation. 3. Compared to classical algorithms, BNR features many theoretical advantages and allows estimating LAI and LAD simultaneously. BNR can address questions difficult to answer by classical algorithms (e.g., how better is one LAD than another?). We demonstrated the utility of the BNR paradigm based on both synthetic and real data. 4. Overall, BNR is statistically more justifiable but its potential has been under‐appreciated. We encourage the community to embrace this new paradigm for reliable analyses of hemispherical photos or other gap data for canopy research.
... The distance between the LiDAR and the targeted object is determined by a product of the speed of light and the time interval between when a laser signal is emitted and when the reflected laser signal is received. LiDARs were used to evaluate height, shape, structure, and contours of trees (Van der Zande et al., 2006;Côté et al., 2012;Shi et al., 2013Shi et al., , 2015McMahon et al., 2015), as well as to describe surface features of vegetation canopies (Saeys et al., 2009;Li et al., 2014;Liu et al., 2017). LiDAR has the advantage of being accurate, fast, and compatible for use in outdoor environments. ...
Article
Full-text available
Plant architecture characteristics contribute significantly to the microclimate within peanut canopies, affecting weed suppression as well as incidence and severity of foliar and soil-borne diseases. However, plant canopy architecture is difficult to measure and describe quantitatively. In this study, a ground-based LiDAR sensor was used to scan rows of peanut plants in the field, and a data processing and analysis algorithm was developed to extract feature indices to describe the peanut canopy architecture. A data acquisition platform was constructed to carry the ground-based LiDAR and an RGB camera during field tests. An experimental field was established with three peanut cultivars at Oklahoma State University's Caddo Research Station in Fort Cobb, OK in May and the data collections were conducted once each month from July to September 2015. The ground-based LiDAR used for this research was a line-scan laser scanner with a scan-angle of 100°, an angle resolution of 0.25°, and a scanning speed of 53 ms. The collected line-scanned data were processed using the developed image processing algorithm. The canopy height, width, and shape/density were evaluated. Euler number, entropy, cluster count, and mean number of connected objects were extracted from the image and used to describe the shape of the peanut canopies. The three peanut cultivars were then classified using the shape features and indices. A high correlation was also observed between the LiDAR and ground-truth measurements for plant height. This approach should be useful for phenotyping peanut germplasm for canopy architecture.
... LiDAR can rapidly estimate canopy height, width, volume and other structural parameters [45, 88,89]. The application of LiDAR covers a diverse range of crops from fruit trees and forests [90][91][92][93][94] to field crops and pasture grasses [95][96][97][98]. In perennial ryegrass, ground-based LiDAR was used to measure variation among 12 cultivars with high accuracy of fresh and dry biomass estimation (R 2 = 0.76-0.78) ...
Article
Full-text available
Increasing the yield of perennial forage crops remains a crucial factor underpinning the profitability of grazing industries, and therefore is a priority for breeding programs. Breeding for high dry matter yield (DMY) in forage crops is likely to be enhanced with the development of genomic selection (GS) strategies. However, realising the full potential of GS will require an increase in the amount of phenotypic data and the rate at which it is collected. Therefore, phenotyping remains a critical bottleneck in the implementation of GS in forage species. Assessments of DMY in forage crop breeding include visual scores, sample clipping and mowing of plots, which are often costly and time-consuming. New ground-and aerial-based platforms equipped with advanced sensors offer opportunities for fast, nondestructive and low-cost, high-throughput phenotyping (HTP) of plant growth, development and yield in a field environment. The workflow of image acquisition, processing and analysis are reviewed. The "big data" challenges, proposed storage and management techniques, development of advanced statistical tools and methods for incorporating the HTP into forage breeding systems are also reviewed. Initial results where these techniques have been applied to forages have been promising but further research and development is required to adapt them to forage breeding situations, particularly with respect to the management of large data sets and the integration of information from spaced plants to sward plots. However, realizing the potential of sensor technologies combined with GS leads to greater rates of genetic gain in forages.
... The distance between the LiDAR and the targeted object is determined by a product of the speed of light and the time interval between when a laser signal is emitted and when the reflected laser signal is received. LiDARs were used to evaluate height, shape, structure, and contours of trees (Van der Zande et al., 2006;Côté et al., 2012;Shi et al., 2013Shi et al., , 2015McMahon et al., 2015), as well as to describe surface features of vegetation canopies (Saeys et al., 2009;Li et al., 2014;Liu et al., 2017). LiDAR has the advantage of being accurate, fast, and compatible for use in outdoor environments. ...
Article
Full-text available
Phenotypic information of peanut canopy, including height, width, shape, and density are important in the selection of the best cultivars of peanuts. However, current methods to acquire these data are mainly by manual measurements or qualitative scorings. These methods are laborious, time-consuming, and subjective. In this study, a ground-based peanut canopy phenotypic system was developed to improve the efficiency and accuracy of the data collection on peanut canopy architecture. The system was on a ground-based, remote controlled cart with a sensor suite of two RGB cameras, a thermal camera, a laser scanner and an RTK GPS. Software programs was developed to control the system and collect, store, and analyze the data. This system was tested in the peanut growth season of 2017. The result showed that the system was able to complete the data collection at least four times faster than previous manual collection. The data collected were with a much higher resolution, thus could be used to acquire detailed features of peanut canopy.
... In addition, many agronomical experiments or breeding programmes involve small plots separated by paths, thereby introducing heterogeneous situations among plants for light interception. Progress in phenotyping now allows precise estimation of the individual leaf area and spatial arrangement of plants in phenotyping platforms (Alvarez Prado et al., 2018), or leaf area index and gap fraction in the field (Liu et al., 2017), thereby potentially allowing calculation of RIE and RUE at the scale of micro-plots in the field or of plants in phenotyping platforms. A methodological challenge is now to take into account the non-homogeneity of canopies and the plant-to-plant interactions to identify the genetic and environmental contributions to RIE and RUE. ...
Article
Full-text available
Multi-genotype canopies are frequent in phenotyping experiments and raise increasing interest in agriculture. Radiation interception efficiency (RIE) and radiation use efficiency (RUE) have low heritabilities in such canopies. We propose a revised Monteith equation that identifies environmental and genetic components of RIE and RUE. An environmental term, component of RIE, characterizes the effect of the presence/absence of neighbours on light interception. The ability of a given plant to compete with its neighbours is then identified, which accounts for the genetic variability of RIE of plants having similar leaf areas. This method was used in three experiments in a phenotyping platform with 765 plants of 255 maize hybrids. As expected, the heritability of the environmental term was near-zero, whereas that of the competitiveness term increased with phenological stages, resulting in the identification of quantitative trait loci. In the same way, RUE was dissected in an effect of intercepted light and a genetic term. This approach was used for predicting the behaviour of individual genotypes in virtual multi-genotype canopies. A large effect of competitiveness was observed in multi-genotype but not in single-genotype canopies, resulting in a bias for comparing genotypes in breeding fields.
... Two main technologies may be used to acquire a CSM: (i) LiDAR and (ii) photogrammetry. Terrestrial laser scanning (TLS) was first used to create CSMs (Hoffmeister et al. 2010;Liu et al. 2017;Sun et al. 2017). UAV photogrammetry is an inexpensive alternative to LiDAR, as it only requires a good quality RGB camera and the UAV (Hugenholtz et al. 2013). ...
Article
Full-text available
Estimating aboveground biomass is important for monitoring crop growth in agronomic field experiments. Often this estimation is done manually, destructively (mowing) or not (counting) on a relatively limited number of sub-plots within an experiment. In the presence of spatial heterogeneity in experiment fields, sensors developed for precision agriculture, have shown great potential to automate this estimation efficiently and provide a spatially continuous measurement over an entire plot. This study investigated the suitability of using an unmanned aerial vehicle (UAV) for biomass and yield estimations in an agronomic field experiment. The main objectives of this work were to compare the estimates made from manual field sampling with those made from UAV data and finally to calculate the improvement that can be expected from the use of UAVs. A 6-ha maize field was studied, with plot treatments for the study of the exogenous organic matter (EOM) amendment effect on crop development. 3D surface models were created from high resolution UAV RGB imagery, before crop emergence and during crop development. The difference between both surface models resulted in crop height which was evaluated against 38 reference points with an R2 of 0.9 and prediction error of 0.16 m. Regression models were used to predict above-ground biomass and grain yield (fresh or dry). Dried grain yield prediction with a generalized additive model gave an error of 0.8 t ha−1 calculated on 100 in-field validation measurements, corresponding to a relative error of 14.77%. UAV-based yield estimates from dry biomass were 15% more accurate than manual yield estimation.
... Thence, measurement of above-ground biomass is laborious and subject to large experimental error. Recent developments in remote and proximal sensing for high-throughput field phenotyping have led to proposed alternatives to destructive sampling, including the use of digital photography and NDVI sensors (Li et al., 2010;Pask et al., 2012), across multiple scales (Hawkesford and Lorence, 2017) using both aerial (see review by Yang et al., 2017) and ground platforms (Busemeyer et al., 2013;Andrade-Sanchez et al., 2014;Deery et al., 2014;Barker et al., 2016;Liu et al., 2016;Virlet et al., 2016;Kirchgessner et al., 2017). ...
Article
Full-text available
Crop improvement efforts are targeting increased above-ground biomass and radiation-use efficiency as drivers for greater yield. Early ground cover and canopy height contribute to biomass production, but manual measurements of these traits, and in particular above-ground biomass, are slow and labour-intensive, more so when made at multiple developmental stages. These constraints limit the ability to capture these data in a temporal fashion, hampering insights that could be gained from multi-dimensional data. Here we demonstrate the capacity of Light Detection And Ranging (LiDAR), mounted on a lightweight, mobile, ground-based platform, for rapid multi-temporal and non-destructive estimation of canopy height, ground cover and above-ground biomass. Field validation of LiDAR measurements is presented. For canopy height, strong relationships with LiDAR (r² of 0.99 and root mean square error of 0.017 m) were obtained. Ground cover was estimated from LiDAR using two methodologies: red reflectance image and canopy height. In contrast to NDVI, LiDAR was not affected by saturation at high ground cover, and the comparison of both LiDAR methodologies showed strong association (r²=0.92 and slope=1.02) at ground cover above 0.8. For above-ground biomass, a dedicated field experiment was performed with destructive biomass sampled eight times across different developmental stages. Two methodologies are presented for the estimation of biomass from LiDAR: 3D voxel index (3DVI) and 3D profile index (3DPI). The parameters involved in the calculation of 3DVI and 3DPI were optimised for each sample event from tillering to maturity, as well as generalized for any developmental stage. Individual sample point predictions were strong while predictions across all eight sample events, provided the strongest association with biomass (r²=0.93 and r²=0.92) for 3DPI and 3DVI, respectively. Given these results, we believe that application of this system will provide new opportunities to deliver improved genotypes and agronomic interventions via more efficient and reliable phenotyping of these important traits in large experiments.
... Thus, it will be able to generate high-resolution three-dimensional (3D) data of objects. The application of LiDAR sensors in estimating biophysical parameters (biomass, height, volume, etc.) has recently been reported in major crops such as in maize [60,61], cotton [62,63], corn [64,65], and wheat [66,67]. Until now, two studies using ground-based LiDAR data applied on perennial ryegrass have reported a good correlation between LiDAR volume and FM [13,68]. ...
Article
Full-text available
Perennial ryegrass biomass yield is an important driver of profitability for Australian dairy farmers, making it a primary goal for plant breeders. However, measuring and selecting cultivars for higher biomass yield is a major bottleneck in breeding, requiring conventional methods that may be imprecise, laborious, and/or destructive. For forage breeding programs to adopt phenomic technologies for biomass estimation, there exists the need to develop, integrate, and validate sensor-based data collection that is aligned with the growth characteristics of plants, plot design and size, and repeated measurements across the growing season to reduce the time and cost associated with the labor involved in data collection. A fully automated phenotyping platform (DairyBioBot) utilizing an unmanned ground vehicle (UGV) equipped with a ground-based Light Detection and Ranging (LiDAR) sensor and Real-Time Kinematic (RTK) positioning system was developed for the accurate and efficient measurement of plant volume as a proxy for biomass in large-scale perennial ryegrass field trials. The field data were collected from a perennial ryegrass row trial of 18 experimental varieties in 160 plots (three rows per plot). DairyBioBot utilized mission planning software to autonomously capture high-resolution LiDAR data and Global Positioning System (GPS) recordings. A custom developed data processing pipeline was used to generate a plant volume estimate from LiDAR data connected to GPS coordinates. A high correlation between LiDAR plant volume and biomass on a Fresh Mass (FM) basis was observed with the coefficient of determination of R2 = 0.71 at the row level and R2 = 0.73 at the plot level. This indicated that LiDAR plant volume is strongly correlated with biomass and therefore the DairyBioBot demonstrates the utility of an autonomous platform to estimate in-field biomass for perennial ryegrass. It is likely that no single platform will be optimal to measure plant biomass from landscape to plant scales; the development and application of autonomous ground-based platforms is of greatest benefit to forage breeding programs.
... In a field experiment, highthroughput phenotyping using unmanned aerial vehicles (UAVs) (Yang et al., 2017) and tractors (White et al., 2012) was used to measure plant growth. Among growth traits, the leaf area index (LAI) is often investigated because it is accessible from high-throughput phenotyping (Verger et al., 2014;Liu et al., 2017) while being sensitive to the environment, directly determining amount of light absorption, and thus affecting biomass production and yield. Until recently, however, these techniques were mainly used for crop management such as estimation of canopy state variables, soil properties and yield (Jin et al., 2018), and their applications to genetic dissection remain limited (Blancon et al., 2019). ...
Article
Full-text available
With the widespread use of high-throughput phenotyping systems, growth process data are expected to become more easily available. By applying genomic prediction to growth data, it will be possible to predict the growth of untested genotypes. Predicting the growth process will be useful for crop breeding, as variability in the growth process has a significant impact on the management of plant cultivation. However, the integration of growth modeling and genomic prediction has yet to be studied in depth. In this study, we implemented new prediction models to propose a novel growth prediction scheme. Phenotype data of 198 soybean germplasm genotypes were acquired for 3 years in experimental fields in Tottori, Japan. The longitudinal changes in the green fractions were measured using UAV remote sensing. Then, a dynamic model was fitted to the green fraction to extract the dynamic characteristics of the green fraction as five parameters. Using the estimated growth parameters, we developed models for genomic prediction of the growth process and tested whether the inclusion of the dynamic model contributed to better prediction of growth. Our proposed models consist of two steps: first, predicting the parameters of the dynamics model with genomic prediction, and then substituting the predicted values for the parameters of the dynamics model. By evaluating the heritability of the growth parameters, the dynamic model was able to effectively extract genetic diversity in the growth characteristics of the green fraction. In addition, the proposed prediction model showed higher prediction accuracy than conventional genomic prediction models, especially when the future growth of the test population is a prediction target given the observed values in the first half of growth as training data. This indicates that our model was able to successfully combine information from the early growth period with phenotypic data from the training population for prediction. This prediction method could be applied to selection at an early growth stage in crop breeding, and could reduce the cost and time of field trials.
... These tasks are time-consuming and laborious, and the measurement accuracy is closely related to worker proficiency. Ground LiDAR has also been applied to canopy measurements and has achieved a good level of accuracy [2]. However, it is a tedious method, and its high cost hinders its promotion and development. ...
Article
Full-text available
Citation: Wang, K.; Zhou, J.; Zhang, W.; Zhang, B. Mobile LiDAR Scanning System Combined with Canopy Morphology Extracting Methods for Tree Crown Parameters Evaluation in Orchards. Sensors 2021, 21, 339. https://doi. Publisher's Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Abstract: To meet the demand for canopy morphological parameter measurements in orchards, a mobile scanning system is designed based on the 3D Simultaneous Localization and Mapping (SLAM) algorithm. The system uses a lightweight LiDAR-Inertial Measurement Unit (LiDAR-IMU) state estimator and a rotation-constrained optimization algorithm to reconstruct a point cloud map of the orchard. Then, Statistical Outlier Removal (SOR) filtering and European clustering algorithms are used to segment the orchard point cloud from which the ground information has been separated, and the k-nearest neighbour (KNN) search algorithm is used to restore the filtered point cloud. Finally, the height of the fruit trees and the volume of the canopy are obtained by the point cloud statistical method and the 3D alpha-shape algorithm. To verify the algorithm, tracked robots equipped with LIDAR and an IMU are used in a standardized orchard. Experiments show that the system in this paper can reconstruct the orchard point cloud environment with high accuracy and can obtain the point cloud information of all fruit trees in the orchard environment. The accuracy of point cloud-based segmentation of fruit trees in the orchard is 95.4%. The R 2 and Root Mean Square Error (RMSE) values of crown height are 0.93682 and 0.04337, respectively, and the corresponding values of canopy volume are 0.8406 and 1.5738, respectively. In summary, this system achieves a good evaluation result of orchard crown information and has important application value in the intelligent measurement of fruit trees.
... LiDAR based LAI can be estimated by separating canopy woody and foliage components . In addition, LiDAR observations have the potential to characterize the vertical vegetation structure at different heights, and provide accurate three-dimensional (3D) point cloud data (Liu et al., 2017b). Such data provides new opportunities for detailed assessments of drought impact on canopy structure. ...
Article
Satellite based remote sensing offers one of the few approaches able to monitor the spatial and temporal development of regional to continental scale droughts. A unique element of remote sensing platforms is their multi-sensor capability, which enhances the capacity for characterizing drought from a variety of perspectives. Such aspects include monitoring drought influences on vegetation and hydrological responses, as well as assessing sectoral impacts (e.g., agriculture). With advances in remote sensing systems along with an increasing range of platforms available for analysis, this contribution provides a timely and systematic review of multi-sensor remote sensing drought studies, with a particular focus on drought related datasets, drought related phenomena and mechanisms, and drought modeling. To explore this topic, we first present a comprehensive summary of large-scale remote sensing datasets that can be used for multi-sensor drought studies. We then review the role of multi-sensor remote sensing for exploring key drought related phenomena and mechanisms, including vegetation responses to drought, land-atmospheric feedbacks during drought, drought-induced tree mortality, drought-related ecosystem fires, post-drought recovery and legacy effects, flash drought, as well as drought trends under climate change. A summary of recent modeling advances towards developing integrated multi-sensor remote sensing drought indices is also provided. We conclude that leveraging multi-sensor remote sensing provides unique benefits for regional to global drought studies, particularly in: 1) revealing the complex drought impact mechanisms on ecosystem components; 2) providing continuous long-term drought related information at large scales; 3) presenting real-time drought information with high spatiotemporal resolution; 4) providing multiple lines of evidence of drought monitoring to improve modeling and prediction robustness; and 5) improving the accuracy of drought monitoring and assessment efforts. We specifically highlight that more mechanism-oriented drought studies that leverage a combination of sensors and techniques (e.g., optical, microwave , hyperspectral, LiDAR, and constellations) across a range of spatiotemporal scales are needed in order to progress and advance our understanding, characterization and description of drought in the future.
... Phenomobiles, which consist of modified vehicles and sensors, form the basis of ground-based phenotyping (Qiu et al. 2019). In the last few years, different phenomobiles have been developed (Andrade-Sanchez et al. 2014;Busemeyer et al. 2013;Comar et al. 2012;Kicherer et al. 2017;Kirchgessner et al. 2017;Liu et al. 2017;Montes et al. 2011;Shafiekhani et al. 2017;Svensgaard et al. 2014;White et al. 2012). For example, a novel field-based robust phenotyping platform, Phenoliner, was developed to screen grapevine fields for biotic and abiotic stress. ...
Article
Full-text available
Development of high-throughput phenotyping technologies has made a considerable progress in the last 10 years. These technologies provide precise measurement of desired traits among thousands of fields grown crop plants under multiple environments which is a critical step towards selection of better performing lines with respect to yield, disease resistance, and stress tolerance for acceleration of crop improvement programs. High-throughput phenotyping techniques and platforms are helpful in unraveling the genetic basis of complex traits associated with plant growth and development and targeted traits. The review focuses on advancements in technologies involved in high-throughput phenotyping, field-based, aerial, and unmanned phenotyping platforms. Development of user-friendly data management tools and softwares for better understanding of phenotyping will increase the use of field-based high-throughput techniques which have potential to revolutionize breeding strategies to meet the future needs of all stakeholders.
... TLS data create a point cloud that allows extremely fine-scale (mm) measurement of microstructure and provides new, detailed features of forest crown structure in 3D (Atticus et al., 2017;Anderson et al.,2021). TLS provides accurate estimates of the 3D structure and geometric scaling properties of the forest and its dependence on environmental conditions (Disney, 2019;Liu et al., 2017;Momo et al., 2017). TLS has the potential to address the challenges associated with traditional forest monitoring techniques and has been used to directly measure or estimate variables including DBH, tree height, crown height, crown diameter, stand basal area, AGB, crown cover, tree density, and volume (Bienert et al., 2012;Brolly and Király, 2009;Hildebrandt and Iost, 2012;Király and Brolly, 2007). ...
Article
Estimating forest aboveground biomass (AGB) is a crucial step to better understand the carbon sequestration capacity of forest ecosystems and their interactions with climate change. The Light detection and ranging (Lidar) derived three dimensional (3-D) structural information makes it possible to accurately estimate forest AGB based on allometric growth relationships. In this study, we propose a novel physical-based parameter named “Lidar Biomass Index (LBI)” based on the lidar equation using point cloud data. Both terrestrial laser scanning (TLS) data and reconstructed point cloud data of analytical trees were used. By comparing lidar-based AGB with field-based deconstructed measurements of 57 trees (including 40 coniferous and 17 broadleaf trees) in Northeast China, our results showed that the LBI-HCmean-based tree-level AGB better explained variations in the field data obtained for coniferous species (Larix kaempferi) (R² = 0.948, RMSE = 23.301 kg) than that of broadleaf species (Fraxinus mandshurica) (R² = 0.881, RMSE = 19.428 kg). The LBI provides an effective solution for estimating tree-level AGB from a 3-D perspective.
... It is then possible to access a few phenological events such as heading [110] or flowering [97], and to describe the dynamics of canopy structure as a proxy of functional traits. The use of simple models or more sophisticated ones [101,107,111,112] offers great potential for providing breeders with new insights into the functioning of the crop. This will be the focus of future investigations where crop functioning models are combined with high-throughput phenotyping observations to tune model parameters that describe the reaction of the crop to environmental factors. ...
Article
Full-text available
There is currently a strong societal demand for sustainability, quality, and safety in bread wheat production. To address these challenges, new and innovative knowledge, resources, tools, and methods to facilitate breeding are needed. This starts with the development of high throughput genomic tools including single nucleotide polymorphism (SNP) arrays, high density molecular marker maps, and full genome sequences. Such powerful tools are essential to perform genome-wide association studies (GWAS), to implement genomic and phenomic selection, and to characterize the worldwide diversity. This is also useful to breeders to broaden the genetic basis of elite varieties through the introduction of novel sources of genetic diversity. Improvement in varieties particularly relies on the detection of genomic regions involved in agronomical traits including tolerance to biotic (diseases and pests) and abiotic (drought, nutrient deficiency, high temperature) stresses. When enough resolution is achieved, this can result in the identification of candidate genes that could further be characterized to identify relevant alleles. Breeding must also now be approached through in silico modeling to simulate plant development, investigate genotype × environment interactions, and introduce marker–trait linkage information in the models to better implement genomic selection. Breeders must be aware of new developments and the information must be made available to the world wheat community to develop new high-yielding varieties that can meet the challenge of higher wheat production in a sustainable and fluctuating agricultural context. In this review, we compiled all knowledge and tools produced during the BREEDWHEAT project to show how they may contribute to face this challenge in the coming years.
... The reconstruction speed using this method was improved by 79.9% compared with obtaining the complete 3D digitizing data of wheat plants one at a time (Table 7). However, this method takes longer for data acquisition than other data acquisition methods [21,22]. ...
Article
Full-text available
The three-dimensional (3D) morphological structure of wheat directly reflects the interrelationship among genetics, environments, and cropping systems. However, the morphological complexity of wheat limits its rapid and accurate 3D modelling. We have developed a 3D wheat modelling method that is based on the progression from skeletons to mesh models. Firstly, we identified five morphological parameters that describe the 3D leaf features of wheat from amounts of 3D leaf digitizing data at the grain filling stage. The template samples were selected based on the similarity between the input leaf skeleton and leaf templates in the constructed wheat leaf database. The leaf modelling was then performed using the as-rigid-as-possible (ARAP) mesh deformation method. We found that 3D wheat modelling at the individual leaf level, leaf group, and individual plant scales can be achieved. Compared with directly acquiring 3D digitizing data for 3D modelling, it saves 79.9% of the time. The minimum correlation R2 of the extracted morphological leaf parameters between using the measured data and 3D model by this method was 0.91 and the maximum RMSE was 0.03, implying that this method preserves the morphological leaf features. The proposed method provides a strong foundation for further morphological phenotype extraction, functional–structural analysis, and virtual reality applications in wheat plants. Overall, we provide a new 3D modelling method for complex plants.
Chapter
The concentration of atmospheric carbon dioxide (CO2) has almost doubled since the preindustrial era due to global climate change and is expected to further increase if the current emission rates are not controlled. The impacts of elevated CO2 (e[CO2]) on growth, development, and yield of plant species, particularly crops, are very important concerns for the scientist. This is due to dynamic implications on global agricultural production and food security in the climate change scenario. Crops respond to the e[CO2] by stimulating the photosynthetic rate. which boosts crop yield. Higher levels of atmospheric carbon act like a carbon fertilizer for the plants and results in an increase in plant growth and productivity. Cereal crops grow larger in size and exhibit faster growth rates under e[CO2], and biomass production becomes higher. Crops have evolved strategies to enhance their physiological performance by increasing water use efficiency and reducing the transpirational water loss as well as lowering stomatal conductance under e[CO2]. C3 plants exhibit considerably higher increases in yield due to e[CO2] ranging from 20% and 35% as compared to C4 crops with only 10% to 15%. e[CO2] influences the qualitative attributes of crops, including the concentration of nutrients, which are fundamental food quality attributes having diverse implications on agricultural production, market value of crops as well as impacts on human health. Sharp declines are projected in the protein content and free amino acid of cereals under e[CO2] conditions. Under realistic field conditions experiments, free-air CO2 enrichment technology revealed significant increases in the photosynthesis activity, leaf carbohydrates, starch and sugars whereas the concentration of nitrogen per unit leaf mass has been found to decrease. The relative yield responses of grain crops under e[CO2] might increase under limiting nutrient and water conditions due to physiological adaptations. The major C3 cereals, including wheat and rice, undergo major shifts in physiological responses and C:N metabolism in response to e[CO2], However, a reduction in nutritional quality under e[CO2] appears to be a major challenge.
Chapter
In recent years, three-dimensional (3D) data acquisition and model reconstruction of plants have been developed as a hot topic of plant scientific researches. However, the morphological structure of plants is very complex and it is hard to describe the details. The data acquisition approaches are diverse for different parts of plants. This study introduces the data acquisition methods of different scales of maize. The grain, leaf and ear, individual plant and maize colony represents the target models for different scales. 3D data acquisition instruments are used to acquire the morphometric of each target. It is found that the organ scale is the simplest to obtain and process. The smallest grain needs high-resolution scanner to acquire the morphological details, while the plant canopy is the hardest one for point cloud process and modeling. The data and reconstructed models are oriented to digital plant, phenotyping analysis, FSPMs research, and popular science education application.
Article
Full-text available
High-throughput phenotyping technologies have become an increasingly important topic of crop science in recent years. Various sensors and data acquisition approaches have been applied to acquire the phenotyping traits. It is quite confusing for crop phenotyping researchers to determine an appropriate way for their application. In this study, three representative three-dimensional (3D) data acquisition approaches, including 3D laser scanning, multi-view stereo (MVS) reconstruction, and 3D digitizing, were evaluated for maize plant phenotyping in multi growth stages. Phenotyping traits accuracy, post-processing difficulty, device cost, data acquisition efficiency, and automation were considered during the evaluation process. 3D scanning provided satisfactory point clouds for medium and high maize plants with acceptable efficiency, while the results were not satisfactory for small maize plants. The equipment used in 3D scanning is expensive, but is highly automatic. MVS reconstruction provided satisfactory point clouds for small and medium plants, and point deviations were observed in upper parts of higher plants. MVS data acquisition, using low-cost cameras, exhibited the highest efficiency among the three evaluated approaches. The one-by-one pipeline data acquisition pattern allows the use of MVS high-throughput in further phenotyping platforms. Undoubtedly, enhancement of point cloud processing technologies is required to improve the extracted phenotyping traits accuracy for both 3D scanning and MVS reconstruction. Finally, 3D digitizing was time-consuming and labor intensive. However, it does not depend on any post-processing algorithms to extract phenotyping parameters and reliable phenotyping traits could be derived. The promising accuracy of 3D digitizing is a better verification choice for other 3D phenotyping approaches. Our study provides clear reference about phenotyping data acquisition of maize plants, especially for the affordable and portable field phenotyping platforms to be developed.
Chapter
Full-text available
Intensified drought stress threatens plant growth and productivity, while elevated CO2 (e[CO2]) alleviates the negative impact of drought stress on plants through alteration in water use and improvement in plant growth. In the terrestrial ecosystem, crops are particularly sensitive to drought and benefit from e[CO2]. To cope with the drier and CO2-enriched climate, plants have evolved various adaptive strategies. Water-dependent crops can benefit from e[CO2] but are species-dependent and depend on the intensities and durations of drought stress. In this chapter, we summarized drought impact on crops, crop performance under e[CO2], as well as their interactions in physiological, biochemical, and molecular levels.
Chapter
Full-text available
Maize is an important food crops across the world and provides at least 30% of the food calories to more than 4.5 billion people in 94 developing countries. Maize is also a basic constituent in animal feed and is used broadly in industrial products like biofuels production. Due to increasing demand and production, global maize supplies and prices have been badly affected. Further, climatic change and the consequences of changes raise the abiotic and biotic stresses. Climate change challenges reduced growth and yield which leads hunger and food insecurity for millions of poor consumers. In the context of climate change, this chapter summarizes the challenges faced by maize and how these challenges can cope to meet future maize demand. Consideration needs to be directed at the production of high yielding, stress-tolerant, and widely-adapted maize varieties through conventional and molecular breeding approaches. For long-term approaches, large public and private sector investment and sustained political commitment and policy support for new technology are needed to overcome hunger, raise the incomes of smallholder farmers and meet the challenges of growing demand for maize at the global level.
Article
Lidar (light detection and ranging) has been widely utilized for estimating the structural parameters of plants, such as tree height, leaf inclination angle, and biomass. However, individual trees have been primarily manually extracted from three-dimensional (3D) point cloud images. Automatically detecting each tree and analyzing its structural parameters is desirable. In this study, we propose a method to (1) detect each tree from 3D point cloud images obtained from ground-based lidar, (2) estimate the number of trees and diameter at breast height (DBH) from the detected 3D point cloud images of trees, and (3) segment each tree canopy. First, we focused on point clouds whose height ranged from 0.5 to 1.5 m and detected each cluster of tree trunks. Then, the clusters were expanded by classifying other points to the clusters that are located near the points and then repeating this process. The process assigns the points in the 3D point cloud image to each tree in the upward direction and separates not only tree trunks but also tree canopies. As a result, the trees in 3D point cloud images were detected with high accuracy, and the number of trees and DBH was estimated. Moreover, each tree canopy was segmented. © 2018, Society of Agricultural Meteorology of Japan. All rights reserved.
Article
[Objective] Morphological parameters acquisition of maize plant is an important part for maize breeding research. Now, the mainly acquisition method of plant type parameter is manual measurement, which has the problems of non-uniform standard, low accuracy and difficult to visualize, etc. Plant type parameter extraction from image or three dimensional (3D) point cloud depends on the algorithms of skeleton extraction, and it has a low accuracy of recent methods.[Method] In this paper, we proposed a maize plant type parameter extraction method by using 3D digitizing data. The maize plant skeleton data was obtained by using 3D digitizer, and an acquisition standard for maize stalk, leaf, tassel and ear was proposed to promise uniform morphological data. The 3D digitized data was regularized by moving to the origin of the space coordinate system and rotating parallel to the Z axis. Morphological parameters, including plant height, leaf insertion height, blade peak height, leaf length, leaf width, blade span length, leaf insertion inclination, and leaf azimuthal angle, were extracted according to the relationship of 3D digitizing point and the definition of each parameter. Exact formulas were given of each morphological parameter. Meanwhile, a novel method for calculating plant azimuth plane was proposed by iteratively for solving an L1 optimization problem, which is described by the minimum sum of every azimuth angle to the azimuthal plane. The algorithm could calculate the exact plant azimuthal range when the leaf number was even. An index called dev value was introduced to evaluate the deviation of maize leaves from the azimuthal plane. [Result] 3D digital data and manual measured parameters of 6 different cultivars of maize plant, with the abbreviations of JK665, JK968, MC812, ND108, XY335 and ZD958, were obtained to verify the extraction method. Experimental results showed that the error of leaf length, leaf insertion inclination and leaf azimuth were very small, the corresponding RMSE (root mean square error) were 3.44 cm, 3.41°, and 8.23°, respectively. The MAPE (mean absolute percent error) of leaf length and leaf insertion inclination were 4.06% and 4.72%, respectively. The error of leaf width with RMSE = 0.8 cm and MAPE = 7.21%, was a little larger than other parameters, because the curve shape vertical to the midrib on the leaf was different. The novel azimuthal plane estimation method gave a quantitative description and derived better results than the averaged azimuth angle approach. The dev value could be used for estimating the spatial expansion of maize plants. In theory, larger dev value cultivar maize plants could intercept more photosynthetically active radiation. [Conclusion] The research provided an accurate, convenient and visual way for extracting and analyzing maize plant type parameters and had an important role in the optimization of maize plant type, functional-structural plant modeling, and plant phenotyping research. © 2018 Editorial Department of Scientia Agricultura Sinica. All rights reserved.
Article
Recent changes to plant architectural traits that influence the canopy have produced high yielding cultivars in rice, wheat and maize. In breeding programs, rapid assessments of the crop canopy and other structural traits are needed to facilitate the development of advanced cultivars in other crops such as Canola. LiDAR has the potential to provide insights into plant structural traits such as canopy height, aboveground biomass, and light penetration. These parameters all rely heavily on classifying LiDAR returns as ground or vegetation as they rely on the number of ground returns and the number of vegetation returns. The aim of this study is to propose a point classification method for canola using machine learning approach. The training and testing datasets were clusters sampled from field plots for flower, plant and ground. The supervised learning algorithms chosen are Decision Tree, Random Forest, Support Vector Machine, and Naïve Bayes. K-means Clustering was also used as an unsupervised learning algorithm. The results show that Random Forest models (error rate = 0.006%) are the most accurate to use for canola point classification, followed by Support Vector Machine (0.028%) and Decision Tree (0.169%). Naïve Bayes (2.079%) and K-means Clustering (48.806%) are not suitable for this purpose. This method provides the true ground and canopy in point clouds rather than determining ground points via a fixed height rely on the accuracy of the point clouds, subsequently gives more representative measurements of the crop canopy.
Article
Full-text available
Manual assessments of plant phenotypes in the field can be labor-intensive and inefficient. The high-throughput field phenotyping systems and in particular robotic systems play an important role to automate data collection and to measure novel and fine-scale phenotypic traits that were previously unattainable by humans. The main goal of this paper is to review the state-of-the-art of high-throughput field phenotyping systems with a focus on autonomous ground robotic systems. This paper first provides a brief review of nonautonomous ground phenotyping systems including tractors, manually pushed or motorized carts, gantries, and cable-driven systems. Then, a detailed review of autonomous ground phenotyping robots is provided with regard to the robot’s main components, including mobile platforms, sensors, manipulators, computing units, and software. It also reviews the navigation algorithms and simulation tools developed for phenotyping robots and the applications of phenotyping robots in measuring plant phenotypic traits and collecting phenotyping datasets. At the end of the review, this paper discusses current major challenges and future research directions.
Article
Plant phenotyping forms the core of crop breeding, allowing breeders to build on physiological traits and mechanistic science to inform their selection of material for crossing and genetic gain. Recent rapid progress in high throughput techniques based on machine vision, robotics and computing (Plant Phenomics) enables crop physiologists and breeders to quantitatively measure complex and previously intractable traits. By combining these techniques with affordable genomic sequencing and genotyping, machine learning and genome selection approaches, breeders have an opportunity to make rapid genetic progress. This review focusses on how field based plant phenomics can enable next generation physiological breeding in cereal crops for traits related to radiation use efficiency, photosynthesis and crop biomass. These traits have previously been regarded as difficult and laborious to measure but have recently become a focus as cereal breeders find genetic progress from “Green Revolution” traits such as harvest index become exhausted. Application of LiDAR, thermal imaging, leaf and canopy spectral reflectance, chlorophyll fluorescence and machine learning are discussed using wheat and sorghum phenotyping as case studies. A vision of how crop genomics and high‐throughput phenotyping could enable the next generation of crop research and breeding is presented. This article is protected by copyright. All rights reserved.
Chapter
Climatic factors affect various stages of plant growth, thereby affecting agricultural productivity and production stability. The growth and development of various tissues and organs during the complete growth period of plants are affected by changes in climate and environment. The genotype is the internal cause for the expression of the phenotype, while the environment is the external conditions for the appearance of various morphological characteristics. With the rapid progress of high-throughput plant phenotype measurement technology, combined with genomics, bioinformatics, and big data computing, plant phenotyping will greatly promote the process of functional genomics research and crop molecular breeding and efficient cultivation. Phenotypes can effectively track the links between genotypes, environmental factors, and phenotypes. Without detailed phenotype data, it is difficult to fully understand the complex effects of genomic and environmental factors on plant phenotypes. Therefore, difficulties in plant scientific research are gradually shifting from genetic analysis to phenotypic analysis. Plant phenotyping began at the end of the 20th century, and its core was to obtain high-quality, reproducible trait data, and then quantify the interactions of genotype and environment and their impact on key traits related to yield, quality, and stress resistance. As more and more plant and trait parameters must be measured quickly and accurately, many of the world's top scientific research institutions have shifted their research focus to solving practical problems such as experimental design, quantitative analysis, and interpretation of conclusions. Characterizing key traits through phenotypic analysis can provide big data-based decision support for breeding, cultivation, and agricultural practices. Plant genomes have undergone rapid development in recent years, but the lack of sufficient phenotypic data has limited human ability to parse quantitative trait genetics. This can be addressed by developing a plant phenotypic information collection platform and performing image analysis. High-throughput, automated, high resolution plant phenotypic information collection platforms and analytical technologies are critical to the acceleration of plant improvement and breeding, increasing yield, and resistance to pests and diseases. These systems are used to analyze genomic information and quantitatively study complex traits related to growth, yield, and adaptation to biotic or abiotic stress. It is an important pathway for integration, which can fulfill the gaps between genomic information and plant phenotypic plasticity. With the growth in the demands of scientific research and the development of imaging sensor technology, it has become possible to collect high-throughput, high-efficiency, high-precision, low-error, and low cost automatic phenotypic information. The data monitored by imaging sensors is objective and can monitor and analyze plants in real time. Therefore, automatic phenotype information collection technology has begun to be widely used in plant phenotype information collection platforms. With the development of modern and intelligent agricultural equipment, scholars hope to discover functionally diverse species and compare performance and plant response to the environment in a large number of plants. To generate a correlation between phenotypic traits, genes, environment, and expression, high-throughput plant phenotype information collection has been invented. High-throughput refers to the ability to measure more samples and/or more data points than manual phenotyping, not only with a high number of samples per unit time, but also with the synchronization and efficiency of data processing and parameter acquisition. The hardware platform scans the source data to match. High-throughput phenotyping relies on rapid transportation, automated sensing, data acquisition, data analysis methods and technical equipment, and is carried out by various sensors, such as CCD camera, near-infrared instrument, infrared instrument, thermal imager, spectral imager, fluorescence imager, etc., to monitor indoor and field plants, so as to obtain more phenotypic parameters in a short time. Due to the huge potential of plant phenotype information collection technology in the agricultural field, scientific research institutions and enterprises in various countries are actively developing and constructing high-throughput plant phenotype information collection platforms. Phenotypic data collection and analysis methods are the core part of plant phenotyping research, including indoor and outdoor parts. Modern indoor high-throughput plant phenotype detection platforms generally detect plants closely related to plant genetics and variation through automated transmission equipment and integrated sensors (such as visible light, near-infrared, far-infrared, fluorescence, multispectral, laser, hyperspectral, etc.). The acquisition of a plant dynamic growth and developmental phenotype dataset. Its advantages are high resolution, high controllability and high-throughput, which can provide high-quality multidimensional images and experimental data for subsequent plant phenotyping. Whether it is aboveground or underground, indoor phenotype monitoring usually takes the individual plant as the unit and describes the population characteristics by collecting the characteristics of the individual plant. Depending on the image sensor accuracy, the extracted phenotypic data can often be accurate down to the tissue cell level. Thus, the scale division of indoor phenotypic traits can range from populations to histiocytes. Field-oriented plant phenotyping platforms provide accurate and continuous collection of single leaves or plant organs, single plants, small plots, and entire farms from proximal to long distances, mainly including vehicle-mounted, self-propelled, gantry, and suspension as well as several types such as drones, aerial remote sensing and spectral satellite imaging. The field high-throughput platform mainly includes two types of field machinery and UAVs equipped with multi-sensors, which can achieve rapid and non-destructive acquisition of plant population phenotype information under field conditions. The combination of agricultural machinery and equipment with a multi-sensor platform can effectively reduce the variation in measurement results but is limited by the distribution of crops and soil conditions after irrigation. Therefore, it is difficult to quickly realize cross-regional applications, the operation efficiency is low, and it cannot be used over a large range. However, the rapid analysis of field crop phenotype information based on an UAV-equipped multi-sensor platform has high technical efficiency and low cost and is suitable for complex farmland environments. It has a wide range of applications in the analysis of crop height, chlorophyll content, LAI, disease susceptibility, drought stress sensitivity, nitrogen content and yield, making it an important means to obtain crop phenotype information. Plant roots are an important part of plants and have very important functions such as water and nutrient absorption and transport, organic matter storage, plant anchoring, and interaction with soil. Since roots grow underground, the core of collecting root phenotypes is how to visualize roots growing under natural conditions. Therefore, the collection and analysis of root phenotypic traits has become the focus and challenge of biological and phenotypic research. At present, the research progress of cultivation and improvement based on root traits is very slow, and the screening of root traits is still a very time-consuming task. Researchers must continue to develop systematic and large-scale high-throughput root phenotyping platforms suitable for different cultivation goals, so as to accelerate the screening of root traits and the understanding of root function, and to gain an in-depth understanding of how root traits are related to whole plant stress resistance strategies correlate to increase crop productivity and ultimately successfully identify key root traits for crop improvement. In recent years, with the continuous development of remote sensing and related sensor technologies, a variety of non-destructive plant detection technologies have emerged, providing effective means for crop phenotype monitoring, disease and insect pest monitoring, and crop identification. RGB cameras are a relatively common sensor device in crop phenotyping technology. In the experiment, a digital camera with visible light imaging on a UAV can be used to improve the efficiency of data collection and achieve rapid acquisition of high-definition images. It can be used to monitor crop emergence during agricultural production, rate, flowering dynamics, canopy coverage, and lodging. Thermal imaging sensors use an infrared detector and an optical imaging lens to receive infrared radiation energy in a photosensitive element infrared detector. Infrared imaging technology is used to measure plant canopy temperature to infer plant water use efficiency and photosynthesis efficiency. It is also used to measure the response of crops to osmotic stress such as salinity or drought. It can also measure the impact of other abiotic stresses on organisms and detect the presence of pests in the grain. Because the crop canopy temperature changes with time, conventional handheld infrared temperature measurement equipment is limited by its low measurement efficiency, and it is difficult to be widely used in large-scale breeding areas. However, UAVs are equipped with thermal imagers for breeding areas, where canopy temperature provides a new efficient and reliable method. At present, thermal infrared technology is increasingly used in the field of precision agriculture. To ensure stable agricultural production, increased production and food security, optimized agricultural production structure, and a reduction in the use of pesticides and fertilizers, it is necessary to accurately monitor and warn the occurrence of diseases during agricultural production. Experts at home and abroad have carried out a series of thermal infrared imaging technology to detect crop disease research. The importance of data interpretation in phenotyping research cannot be overstated. With the continuous development of phenotyping platforms and related technologies at home and abroad, the current plant high-throughput phenotyping platform can obtain massive digital images, point cloud data, spectral imaging and thermal imaging data, and undergo geometric correction, radiometric correction, and data modeling. And a series of processing processes, and finally realize the use of remote sensing means to analyze plant phenotype information. High-throughput phenotyping is expected to become the latest tool for sustainable production under global climate change. However, at present, there are few plant species as the research objects of plant phenotype information, and there is a lack of analysis and identification of plant phenotype information in complex natural environments. The breadth and depth of plant phenotype analysis research must be continuously expanded to establish a unified plant phenotype monitoring system and specifications, form a network resource sharing library, and strengthen mutual cooperation among developers of various phenotype platforms. Using modern mathematical analysis methods, the rapidity and effectiveness of the algorithm can be achieved in image processing and recognition software design to improve the analysis ability of the analysis module of the plant phenotype monitoring system. It is necessary to use multi-domain knowledge to carry out comprehensive control and establish background expert decision-making support systems to automatically monitor and analyze target plants in real time.
Thesis
La modélisation en biologie, plus particulièrement celle de la croissance et du fonctionnement des plantes, est un domaine actuellement en pleine expansion, utile pour appréhender les enjeux liés au changement climatique et à la sécurité alimentaire au niveau mondial. La modélisation et la simulation sont des outils incontournables pour la compréhension des relations complexes entre l'architecture des plantes et les processus qui influencent leur croissance dans un environnement changeant. Pour la modélisation des plantes, un grand nombre de formalismes ont été développés dans de nombreuses disciplines et à différentes échelles de représentation. L'objectif de cette thèse est de définir une architecture modulaire qui permette de simuler des systèmes structure-fonction en réutilisant et en assemblant différents modèles existants. Nous étudierons d'abord les différentes approches de la réutilisation logicielle, proposées par Krueger, les systèmes à tableau noir et les systèmes de workflows scientifiques. Ces différentes approches sont utilisées afin de faire coopérer, de réutiliser et d'assembler des artefacts logiciels de façon modulaire. A partir du constat que ces systèmes fournissent les abstractions nécessaires à l'intégration d'artefacts variés, notre hypothèse de travail est qu'une architecture hybride, basée sur les systèmes à tableau noir avec un contrôle procédural piloté par dataflow, permettrait à la fois d'obtenir la modularité tout en permettant au modélisateur de garder le contrôle sur l'exécution. Dans le chapitre 2, nous décrivons la plateforme OpenAlea, une plateforme à composants logiciels et offrant un système de workflow scientifique, permettant l'assemblage et la composition de modèles à travers une interface de programmation visuelle. Dans le chapitre 3, nous proposons une structure de données pour le tableau noir, associant une représentation topologique de l'architecture des plantes à différentes échelles, le Multiscale Tree Graph, et sa spatialisation géométrique à l'aide de la bibliothèque 3D PlantGL. Ensuite, dans le chapitre 4, nous présentons les $\lambda$-dataflows, une extension des dataflows permettant de coupler simulation et analyse.Puis, dans le chapitre 5, nous présentons une première application, qui illustre l'utilisation d'un modèle générique de feuilles de graminées dans différents modèles de plantes.Finalement, dans le chapitre 6, nous présentons l'ensemble des éléments de l'architecture utilisés pour élaborer un cadre générique de modélisation du développement des maladies foliaires dans un couvert architecturé.L'architecture présentée dans cette thèse et sa mise en œuvre dans OpenAlea sont un premier pas vers la réalisation de plateformes de modélisation intégratives ouvertes, permettant la coopération de modèles hétérogènes en biologie. L'utilisation du formalisme de workflows scientifiques en analyse et en simulation permet notamment d'envisager à court terme l'élaboration des plateformes de simulation collaboratives et distribuées à grande échelle.
Article
Full-text available
Plant phenomics (PP) has been recognized as a bottleneck in studying the interactions of genomics and environment on plants, limiting the progress of smart breeding and precise cultivation. High-throughput plant phenotyping is challenging owing to the spatio-temporal dynamics of traits. Proximal and remote sensing (PRS) techniques are increasingly used for plant phenotyping because of their advantages in multi-dimensional data acquisition and analysis. Substantial progress of PRS applications in PP has been observed over the last two decades and is analyzed here from an interdisciplinary perspective based on 2972 publications. This progress covers most aspects of PRS application in PP, including patterns of global spatial distribution and temporal dynamics, specific PRS technologies, phenotypic research fields, working environments, species, and traits. Subsequently, we demonstrate how to link PRS to multi-omics studies, including how to achieve multi-dimensional PRS data acquisition and processing, how to systematically integrate all kinds of phenotypic information and derive phenotypic knowledge with biological significance, and how to link PP to multi-omics association analysis. Finally, we identify three future perspectives for PRS-based PP: (1) strengthening the spatial and temporal consistency of PRS data, (2) exploring novel phenotypic traits, and (3) facilitating multi-omics communication.
Article
The definition of LAI (Leaf Area Index) is important when deriving it from reflectance observation for model application and validation. Canopy reflectance and the corresponding quantities of LAI, PAI (Plant Area Index), GAI (Green Area Index) and effective GAI (GAIeff) are first calculated using a 3D radiative transfer model (RTM) applied to 3D wheat and maize architecture models. A range of phenological stages, leaf optical properties, soil reflectance, canopy structure and sun directions is considered. Several retrieval methods are compared, including vegetation indices (VIs) combined with a semi-empirical model, and 1D or 3D RTM combined with a machine learning inversion approach. Results show that GAIeff is best estimated from remote sensing observations. The RTM inversion using a 3D model provides more accurate GAIeff estimates compared with VIs and the 1D PROSAIL model with RMSE = 0.33 for wheat and RMSE= 0.43 for maize. GAIeff offers the advantage to be easily accessible from ground measurements at the decametric resolution. It was therefore concluded that the most efficient retrieval approach would be to use machine learning algorithms trained over paired GAIeff and the corresponding canopy reflectance derived either from realistic 3D canopy models or from experimental measurements.
Preprint
Full-text available
In the era of high-throughput visual plant phenotyping, it is crucial to design fully automated and flexible workflows able to derive quantitative traits from plant images. Over the last years, several software supports the extraction of architectural features of shoot systems. Yet currently no end-to-end systems are able to extract both 3D shoot topology and geometry of plants automatically from images on large datasets and a large range of species. In particular, these software essentially deal with dicotyledons, whose architecture is comparatively easier to analyze than monocotyledons. To tackle these challenges, we designed the Phenomenal software featured with: (i) a completely automatic workflow system including data import, reconstruction of 3D plant architecture for a range of species and quantitative measurements on the reconstructed plants; (ii) an open source library for the development and comparison of new algorithms to perform 3D shoot reconstruction and (iii) an integration framework to couple workflow outputs with existing models towards model-assisted phenotyping. Phenomenal analyzes a large variety of data sets and species from images of high-throughput phenotyping platform experiments to published data obtained in different conditions and provided in a different format. Phenomenal has been validated both on manual measurements and synthetic data simulated by 3D models. It has been also tested on other published datasets to reproduce a published semi-automatic reconstruction workflow in an automatic way. Phenomenal is available as an open-source software on a public repository.
Preprint
Full-text available
Background: Three-dimensional (3D) laser scanning technology could rapidly extract the surface geometric features of maize plants to achieve non-destructive monitoring of maize phenotypes. However, extracting the phenotypic parameters of maize plants based on laser point cloud data is challenging. Methods: In this paper, a rotational scanning method was used to collect the data of potted maize point cloud from different perspectives by using a laser scanner. Maize point cloud data were grid-reconstructed and aligned based on greedy projection triangulation algorithm and iterative closest point (ICP) algorithm, and the random sampling consistency algorithm was used to segment the stem and leaf point clouds of single maize plant to obtain the plant height and leaf parameters. Results: The results showed that the R² between the predicted plant height and the measured plant height was above 0.95, and the R² of the predicted leaf length, leaf width and leaf area were 0.938, 0878 and 0.956 respectively when compared with the measured values. Conclusions: The 3D reconstruction of maize plants using the laser scanner showed a good performance, and the phenotypic parameters obtained based on the reconstructed 3D model had high accuracy. The results were helpful to the practical application of plant 3D reconstruction and provided guidance for plant parameter acquisition and theoretical methods for intelligent agricultural research.
Article
Full-text available
The capacity of canopy light interception is a key functional trait to distinguish the phenotypic variation over genotypes. High-throughput phenotyping canopy light interception in the field, therefore, would be of high interests for breeders to increase the efficiency of crop improvement. In this research, the Digital Plant Phenotyping Platform(D3P) was used to conduct in-silico phenotyping experiment with LiDAR scans over a wheat field. In this experiment virtual 3D wheat canopies were generated over 100 wheat genotypes for 5 growth stages, representing wide range of canopy structural variation. Accordingly, the actual value of traits targeted were calculated including GAI (green area index), AIA (average inclination angle) and FIPARdif (the fraction of intercepted diffuse photosynthetically activate radiation). Then, virtual LiDAR scanning were accomplished over all the treatments and exported as 3D point cloud. Two types of features were extracted from point cloud, including height quantiles (H) and green fractions (GF). Finally, an artificial neural network was trained to predict the traits targeted from different combinations of LiDAR features. Results show that the prediction accuracy varies with the selection of input features, following the rank as GF + H > H > GF. Regarding the three traits, we achieved satisfactory accuracy for FIPARdif (R2=0.95) and GAI (R2=0.98) but not for AIA (R2=0.20). This highlights the importance of H feature with respect to the prediction accuracy. The results achieved here are based on in-silico experiments, further evaluation with field measurement would be necessary. Nontheless, as proof of concept, this work further demonstrates that D3P could greatly facilitate the algorithm development. Morever, it highlights the potential of LiDAR measurement in the high-throuhgput phenopyting of canopy light interpcetion and structural traits in the field.
Article
With continuous collaborative research in sensor technology, communication technology, plant science, computer science and engineering science, Internet of Things (IoT) in agriculture has made a qualitative leap through environmental sensor networks, non-destructive imaging, spectral analysis, robotics, machine vision and laser radar technology. Physical and chemical analysis can continuously obtain environmental data, experimental metadata (including text, image and spectral, 3D point cloud and real-time growth data) through integrated automation platform equipment and technical means. Based on data on multi-scale, multi-environmental and multi-mode plant traits that constitute big data on plant phenotypes, genotype–phenotype–envirotype relationship in the omics system can be explored deeply. Detailed information on the formation mechanism of specific biological traits can promote the process of functional genomics, plant molecular breeding and efficient cultivation. This study summarises the development background, research process and characteristics of high-throughput plant phenotypes. A systematic review of the research progress of IoT in agriculture and plant high-throughput phenotypes is conducted, including the acquisition and analysis of plant phenotype big data, phenotypic trait prediction and multi-recombination analysis based on plant phenomics. This study proposes key techniques for current plant phenotypes, and looks forward to the research on plant phenotype detection technology in the field environment, fusion and data mining of plant phenotype multivariate data, simultaneous observation of multi-scale phenotype platform and promotion of a comprehensive high-throughput phenotype technology.
Article
Full-text available
Leaf area index (LAI) is an important vegetation leaf structure parameter in forest and agricultural ecosystems. Remote sensing techniques can provide an effective alternative to field-based observation of LAI. Differences in canopy structure result in different sensor types (active or passive), platforms (terrestrial, airborne, or satellite), and models being appropriate for the LAI estimation of forest and agricultural systems. This study reviews the application of remote sensing-based approaches across different system configurations (passive, active, and multisource sensors on different collection platforms) that are used to estimate forest and crop LAI and explores uncertainty analysis in LAI estimation. A comparison of the difference in LAI estimation for forest and agricultural applications given the different structure of these ecosystems is presented, particularly as this relates to spatial scale. The ease of use of empirical models supports these as the preferred choice for forest and crop LAI estimation. However, performance variation among different empirical models for forest and crop LAI estimation limits the broad application of specific models. The development of models that facilitate the strategic incorporation of local physiology and biochemistry parameters for specific forests and crop growth stages from various temperature zones could improve the accuracy of LAI estimation models and help develop models that can be applied more broadly. In terms of scale issues, both spectral and spatial scales impact the estimation of LAI. Exploration of the quantitative relationship between scales of data from different sensors could help forest and crop managers more appropriately and effectively apply different data sources. Uncertainty coming from various sources results in reduced accuracy in estimating LAI. While Bayesian approaches have proven effective to quantify LAI estimation uncertainty based on the uncertainty of model inputs, there is still a need to quantify uncertainty from remote sensing data source, ground measurements and related environmental factors to mitigate the impacts of model uncertainty and improve LAI estimation.
Article
Full-text available
Background: Functional-structural plant models (FSPMs) explore and integrate relationships between a plant's structure and processes that underlie its growth and development. In the last twenty years, scientists interested in functional-structural plant modelling have expanded greatly the range of topics covered and now handle dynamical models of growth and development occurring from the microscopic scale, and involving cell division in plant meristems, to the macroscopic scales of whole plants and plant communities. Scope: The FSPM approach occupies a central position in plant science; it is at the crossroads of fundamental questions in system biology and predictive ecology. This special issue of Annals of Botany features selected papers on critical areas covered by FSPMs and examples of comprehensive models that are used to solve theoretical and applied questions, ranging from developmental biology to plant phenotyping and management of plants for agronomic purposes. Altogether, they offer an opportunity to assess the progress, gaps and bottlenecks along the research path originally foreseen for FSPMs two decades ago. This review also allows discussion of current challenges of FSP models regarding i) integration of multi-disciplinary knowledge, ii) methods for handling complex models, iii) standards to achieve interoperability and greater genericity and iv) understanding of plant functioning across scales. Conclusions: This approach has demonstrated considerable progress, but has yet to reach its full potential in terms of integration and heuristic knowledge production. The research agenda of FSP modelers in the coming years should place a greater emphasis on explaining robust emergent patterns, and on the causes of possible deviation from it. Modelling such patterns could indeed fuel both generic integration across scales and transdisciplinary transfer. In particular, it could be beneficil to emergent fields of research such as model-assisted phenotyping and predictive ecology in managed ecosystems.
Article
Full-text available
Leaf area index (LAI) is a key biophysical variable and an ecosystem condition indicator that is measured from multiple methods. In this study, LAI was measured by a terrestrial laser scanner (TLS) and Li-Cor LAI-2200 plant canopy analyzer for understanding differences. A total of six different methods that consider leaf clumping, leaf–wood separation, and orthographic and stereographic projection were compared. A reasonable agreement among all methods for LAI estimates was observed (i.e., correlations r > 0.50). The Bayesian linear regression (BLR) approach was used to scale up the six different LAI estimates and to produce continuous field surfaces for the Oak Openings Region in northwest Ohio using Landsat TM–derived spectral vegetation indices (weighted difference vegetation index [WDVI], difference vegetation index [DVI], normalized difference vegetation index [NDVI], soil-adjusted vegetation index [SAVI], and perpendicular vegetation index 3 [PVI3]). The BLR approach provides details about the parameter uncertainties that may arise from foliage and wooden biomass and can be used for model comparisons. In this study, the TLS estimates of foliage derived by orthographic projection had the lowest residual scatter and overall model uncertainty. The deviation departure from the mean BLR estimates revealed that sparse and open areas were associated with the highest error and spatial uncertainties.
Article
Full-text available
Leaf area index (LAI) or plant area index (PAI) are commonly used to represent canopy structure and dynamics, but daily estimation of these variables using traditional ground-based methods is impractical and prone to multiple errors during data acquisition and processing. Existing terrestrial laser scanners can provide accurate representation of forest canopy structure, but the sensors are expensive, data processing is complex, and measurements are typically confined to a single event, which severely limits their utility in the interpretation of canopy trends indicated by remotely sensed data. We tested a novel, low-cost terrestrial laser scanner for its capacity to provide reliable and successive assessments of canopy PAI in an evergreen eucalypt forest. Daily scans comprised of 920 range measurements were made by three scanners at one forest site over a two-year period, providing mostly consecutive estimates of PAI, and of vertical structure profiles (as Plant Area Volume Density, PAVD). Data filtering, involving objective statistical methods to identify outliers, indicated that scan quality was adversely affected by moist weather and moderate wind speeds (> 4 m s− 1), suggesting limited utility in some forest environments. Data cleaning (associated with sensor malfunctions) plus filtering removed 32 to 49% of scans, leaving on average 57% of data over the two-year period. Nonetheless, we found strong agreement between lidar-derived PAI estimates, and those from monthly hemispherical images (± 0.1 PAI); with both methods indicating mostly stable PAI over multiple seasons. The PAVD profiles from the laser scanner indicated that leaf flush in the upper canopy concomitantly balanced leaf loss from the middle canopy in summer, which was consistent with measured summer peaks in litter fall. This clearly illustrated the advantages of three-dimensional lidar data over traditional two-dimensional PAI estimates in monitoring tree phenology, and in interpreting changes in canopy reflectance as detected by air- and space-borne remotely sensed data.
Article
Full-text available
This article describes the methods and problems associated to the esti-mation of canopy characteristics from remote sensing observations. It is illustrated over the solar spectral domain, with emphasis on LAI estimation using currently available algorithms developed for moderate resolution sensors. The principles of algorithms are first presented, distinguishing between canopy biophysical and ra-diometric data driven approaches that may use either radiative transfer models or experimental observations. Advantages and drawback are discussed with due atten-tion to the operational character of the algorithms. Then the under-determination and ill-posedness nature of the inverse problem is described and illustrated. Finally, ways to improve the retrieval performances are presented, including the use of prior information, the exploitation of spatial and temporal constraints, and the interest in using holistic approaches based on the coupling of radiative transfer processes at several scales or levels. A conclusion is eventually proposed, discussing the three main components of retrieval approaches: retrieval techniques, radiative transfer models, and the exploitation of observations and ancillary information.
Article
Full-text available
The development of functional-structural plant models requires an increasing amount of computer modelling. All these models are developed by different teams in various contexts and with different goals. Efficient and flexible computationalframeworksarerequiredtoaugmenttheinteractionbetweenthesemodels,theirreusability,andthepossibility to compare them on identical datasets. In this paper, we present an open-source platform, OpenAlea, that provides a user-friendly environmentfor modellers,andadvanceddeploymentmethods.OpenAlea allowsresearchers tobuildmodels using a visual programming interface and provides a set of tools and models dedicated to plant modelling. Models and algorithms are embedded in OpenAlea 'components' with well defined input and output interfaces that can be easily interconnectedtoformmorecomplexmodelsanddefinemoremacroscopiccomponents.Thesystemarchitectureisbasedon the use of a general purpose, high-level, object-oriented script language, Python, widely used in other scientific areas. We present a brief rationale that underlies the architectural design of this system and we illustrate the use of the platform to assemble several heterogeneous model components and to rapidly prototype a complex modelling scenario.
Article
Full-text available
Terrestrial LiDAR scanners have been shown to hold great potential for estimating and mapping three dimensional (3-D) leaf area distribution in forested environments. This is made possible by the capacity of LiDAR scanners to record the 3-D position of every laser pulse intercepted by plant material. The laser pulses emitted by a LiDAR scanner can be regarded as light probes whose transmission and interception may be used to derive leaf area density at different spatial scales using the Beer–Lambert law or Warren Wilson's contact frequency method among others. Segmenting the canopy into cubic volumes –or voxels- provides a convenient means to compute light transmission statistics and describe the spatial distribution of foliage area in tree crowns. In this paper, we investigate the optimal voxel dimensions for estimating the spatial distribution of within crown leaf area density. We analyzed LiDAR measurements from two field sites, located in Mali and in California, with trees having different leaf sizes during periods with and without leaves. We found that there is a range of voxel sizes, which satisfy three important conditions. The first condition is related to clumping and requires voxels small enough to exclude large gaps between crowns and branches. The second condition requires a voxel size large enough for the conditions postulated by the Poisson law to be valid, i.e., a turbid medium with randomly positioned leaves. And, the third condition relates to the appropriate voxel size to pinpoint the location of those volumes within the canopy which were insufficiently sampled by the LiDAR instrument to derive reliable statistics (occlusion effects). Here, we show that these requirements are a function of leaf size, branching structure, and the predominance of occlusion effects. The results presented provide guiding principles for using voxel volumes in the retrieval of leaf area distributions from terrestrial LiDAR measurements.
Article
Full-text available
Estimation of grapevine vigour using mobile proximal sensors can provide an indirect method for determining grape yield and quality. Of the various indexes related to the characteristics of grapevine foliage, the leaf area index (LAI) is probably the most widely used in viticulture. To assess the feasibility of using light detection and ranging (LiDAR) sensors for predicting the LAI, several field trials were performed using a tractor-mounted LiDAR system. This system measured the crop in a transverse direction along the rows of vines and geometric and structural parameters were computed. The parameters evaluated were the height of the vines (H), the cross-sectional area (A), the canopy volume (V) and the tree area index (TAI). This last parameter was formulated as the ratio of the crop estimated area per unit ground area, using a local Poisson distribution to approximate the laser beam transmission probability within vines. In order to compare the calculated indexes with the actual values of LAI, the scanned vines were defoliated to obtain LAI values for different row sections. Linear regression analysis showed a good correlation (R 2 = 0.81) between canopy volume and the measured values of LAI for 1 m long sections. Nevertheless, the best estimation of the LAI was given by the TAI (R 2 = 0.92) for the same length, confirming LiDAR sensors as an interesting option for foliage characterization of grapevines. However, current limitations exist related to the complexity of data process and to the need to accumulate a sufficient number of scans to adequately estimate the LAI.
Conference Paper
Full-text available
A parameterisation of wheat architecture was developed, having high flexibility to simulate contrasted genotypes and growth conditions with a reasonably low number of parameters. Field measurements at 4-5 dates allowed to simulate crops from emergence to maturity with a good agreement between simulated and measured ground cover and GAI. Dynamics of leaf angles were shown to impact strongly ground cover. See all FSPM2013 proceedings at http://www.metla.fi/fspm2013/proceedings.htm
Conference Paper
Full-text available
A 3D architectural model of winter wheat development is presented. The model simulates the kinetics of elongation of individual organs, together with their geometric shape. With the inclusion of two additional parameters, it was possible to model the development of tillers using the same parameter values as for the main stem. The duration of organ extension was found to be the same amongst phytomers, and the kinetics of extension of any organ could be simulated from its final length and two parameters. Finally, lengths of mature organs were found to follow simple functions of phytomer position. The resulting model simulates development of the whole plant with a small number of parameters, characterising the size and geometry of mature organs and the timing of collar appearance.
Article
Full-text available
Remote sensing of forest canopy cover has been widely studied recently, but little attention has been paid to the quality of field validation data. Ecological literature has two different coveragemetrics. Vertical canopy cover (VCC) is the vertical projection of tree crownsignoringwithin-crowngaps. Angular canopy closure (ACC) is the proportion of covered sky at some angular range around the zenith, and can be measured with a field-of-view instrument, such as a camera. We compared field-measured VCC and ACC at 15° and 75° from the zenith to different LiDAR (Light Detection and Ranging) metrics, using several LiDAR data sets and comprehensive field data. The VCC was estimated to a high precision using a simple proportion of canopy points in first-return data. Confining to a maximum 15° scan zenith angle, the absolute root mean squared error (RMSE) was 3.7-7.0%, with an overestimation of 3.1-4.6%. We showed that grid-based methods are capable of reducing the inherent overestimation of VCC. The low scan angles and low power settings that are typically applied in topographic LiDARs are not suitable for ACC estimation as they measure in wrong geometry and cannot easily detect small within-crown gaps. However, ACC at 0-15° zenith angles could be estimated from LiDAR data with sufficient precision, using also the last returns (RMSE 8.1-11.3%, bias -6.1-+4.6%). The dependency of LiDARmetrics andACC at 0-75° zenith angleswas nonlinear and wasmodeled fromlaser pulse proportionswith nonlinear regressionwith a best-case standard error of 4.1%. We also estimated leaf area index fromthe LiDARmetricswith linear regression with a standard error of 0.38. The results show that correlations between airborne laser metrics and different canopy field characteristics are very high if the field measurements are done with equivalent accuracy.
Article
Full-text available
An algorithm for minimizing a nonlinear function subject to nonlinear inequality constraints is described. It applies sequential quadratic programming techniques to a sequence of barrier problems, and uses trust regions to ensure the robustness of the iteration and to allow the direct use of second order derivatives. This framework permits primal and primal-dual steps, but the paper focuses on the primal version of the new algorithm. An analysis of the convergence properties of this method is presented.
Article
Full-text available
The ability to accurately and rapidly acquire leaf area index (LAI) is an indispensable component of process-based ecological research facilitating the understanding of gas-vegetation exchange phenomenon at an array of spatial scales from the leaf to the landscape. However, LAI is difficult to directly acquire for large spatial extents due to its time consuming and work intensive nature. Such efforts have been significantly improved by the emergence of optical and active remote sensing techniques. This paper reviews the definitions and theories of LAI measurement with respect to direct and indirect methods. Then, the methodologies for LAI retrieval with regard to the characteristics of a range of remotely sensed datasets are discussed. Remote sensing indirect methods are subdivided into two categories of passive and active remote sensing, which are further categorized as terrestrial, aerial and satellite-born platforms. Due to a wide variety in spatial resolution of remotely sensed data and the requirements of ecological modeling, the scaling issue of LAI is discussed and special consideration is given to extrapolation of measurement to landscape and regional levels.
Article
Full-text available
In even-aged, single species conifer plantations LiDAR height data can be modelled to provide accurate estimates of tree height and volume. However, it is apparent that growth models developed for single species stands are not directly transferable to a more general situation of mixed species plantations. This paper evaluates the ability of small footprint, dual-return, pulsed airborne LiDAR data to estimate the proportion of the productive species when mixed with a nurse crop in closed canopy plantations. A study area located in Galloway Forest District in Scotland is used as an example of Lodgepole pine and Sitka spruce mixed plantation; this area contains good examples of a wide range of pure and mixed species plantation types. Three species groups are studied: areas of pure Sitka spruce, areas of pure Lodgepole pine and areas where the two species have been planted together. Two approaches are assessed for detection of plantation mixtures: the first uses LiDAR intensity data to separate spruce and pine species and the second uses LiDAR-derived canopy density measures, coefficient of variation, skewness, percent of ground returns (which provides a measure of canopy openness) and the mean canopy height, which enables areas with height variations to be identified. From analysis of LiDAR data extracted from 54 study plots using logistic regression, the coefficient of variation and LiDAR intensity data provide the most useful predictors of the proportion of spruce in a pine/spruce mixture with coefficients of determination (R2) of 0.914 and 0.930 respectively. The method could be developed as a mapping tool, which in combination with existing inventory data should help to improve timber volume forecasting for mixed species even-aged plantations.
Article
Full-text available
The retrieval of biophysical variables using canopy reflectance models is hampered by the fact that the inverse problem is ill-posed. This leads to unstable and often inaccurate inversion results. In order to regularize the model inversion, a novel approach has been developed and tested on synthetic Landsat TM reflectance data. The method takes into account the neighbouring radiometric information of the pixel of interest, named object signature. The neighbourhood data can either be extracted from gliding windows, already segmented images, or using digitized field boundaries. The extracted radiometric data of the neighbourhood pixels are used to calculate 42 descriptive statistical properties that comprehensively characterize the spectral (co)variance of the image object (e.g. mean and standard deviation of the distributions, intercorrelations between spectral bands, etc.). Together with the habitual spectral signature of the pixel being inverted (6 variables), this object signature (42 variables) is used as input in an artificial neural net to estimate simultaneously three important biophysical variables (i.e. leaf area index, leaf chlorophyll, and leaf water content). The use of neural nets for the model inversion avoids time-consuming iterative optimizations and provides a computational effective way to consider simultaneously pixel and object signatures.
Article
Full-text available
The theoretical background of modeling the gap fraction and the leaf inclination distribution is presented and the different techniques used to derive leaf area index (LAI) and leaf inclination angle from gap fraction measurements are reviewed. Their associated assumptions and limitations are discussed, i.e., the clumping effect and the distinction between green and non-green elements within the canopy. Based on LAI measurements in various canopies (various crops and forests), sampling strategy is also discussed.
Article
Full-text available
Equations used to calculate extinction coefficients for radiation in plant canopies tend either to be too simple to describe canopy radiation accurately, or too complex for convenient computation. An equation has been derived using the assumption that the angular distribution of leaf area in a canopy is similar to the distribution of area on the surface of a prolate or oblate spheroid. This is therefore a generalization of the spherical leaf angle distribution which is frequently used for plant canopies. Simulated leaf angle distributions generated using this model closely approximate measured leaf angle distributions for plant canopies. Extinction coefficients calculated from the model give values which are virtually identical to those calculated using six leaf angle classes.
Article
Full-text available
Most vegetation indices (VI) combine information contained in two spectral bands: red and near-infrared. These indices are established in order to minimize the effect of external factors on spectral data and to derive canopy characteristics such as leaf area index (LAI) and fraction of absorbed photosynthetic active radiation (P). The potentials and limits of different vegetation indices are discussed in this paper using the normalized difference (NDVI), perpendicular vegetation index (PVI), soil adjusted vegetation index (SAVI), and transformed soil adjusted vegetation index (TSAVI). The discussion is based on a sensitivity analysis in which the effect of canopy geometry (LAI and leaf inclination) and soil background are analyzed. The calculation is performed on data derived from the SAIL reflectance model. General semiempirical models, describing the relations between VI and LAI or P, are elaborated and used to derive the relative equivalent noise (REN) for the determination of LAI and P. The performances of VIs are discussed on the basis of the REN concept.
Article
Full-text available
We explore an original strategy for building deep networks, based on stacking layers of denoising autoencoders which are trained locally to denoise corrupted versions of their inputs. The resulting algorithm is a straightforward variation on the stacking of ordinary autoencoders. It is however shown on a benchmark of classification problems to yield significantly lower classification error, thus bridging the performance gap with deep belief networks (DBN), and in several cases surpassing it. Higher level representations learnt in this purely unsupervised fashion also help boost the performance of subsequent SVM classifiers. Qualitative experiments show that, contrary to ordinary autoencoders, denoising autoencoders are able to learn Gabor-like edge detectors from natural image patches and larger stroke detectors from digit images. This work clearly establishes the value of using a denoising criterion as a tractable unsupervised objective to guide the learning of useful higher level representations.
Article
Full-text available
Leaf area index (LAI) is a key structural characteristic of forest ecosystems because of the role of green leaves in controlling many biological and physical processes in plant canopies. Accurate LAI estimates are required in studies of ecophysiology, atmosphere-ecosystem interactions, and global change. The objective of this paper is to evaluate LAI values obtained by several research teams using different methods for a broad spectrum of boreal forest types in support of the international Boreal EcosystemAtmosphere Study (BOREAS). These methods include destructive sampling and optical instruments: the tracing radiation and architecture of canopies (TRAC), the LAI-2000 plant canopy analyzer, hemispherical photography, and the Sunfleck Ceptometer. The latter three calculate LAI from measured radiation transmittance (gap fraction) using inversion models that assume a random spatial distribution of leaves. It is shown that these instruments underestimate LAI of boreal forest stands where the foliage is clumped. The TRAC quantifies the clumping effect by measuring the canopy gap size distribution. For deciduous stands the clumping index measured from TRAC includes the clumping effect at all scales, but for conifer stands it only resolves the clumping effect at scales larger than the shoot (the basic collection of needles). To determine foliage clumping within conifer shoots, a video camera and rotational light table system was used. The major difficulties in determining the surface area of small conifer needles have been largely overcome by the use of an accurate volume displacement method. Hemispherical photography has the advantage that it also provides a permanent image record of the canopies. Typically, LAI falls in the range from 1 to 4 for jack pine and aspen forests and from 1 to 6 for black spruce. Our comparative studies provide the most comprehensive set of LAI estimates available for boreal forests and demonstrate that optical techniques, combined with limited direct foliage sampling, can be used to obtain quick and accurate LAI measurements.
Article
Full-text available
This paper reviews LiDAR ground filtering algorithms used in the process of creating Digital Elevation Models. We discuss critical issues for the development and application of LiDAR ground filtering algorithms, including filtering procedures for different feature types, and criteria for study site selection, accuracy assessment, and algorithm classification. This review highlights three feature types for which current ground filtering algorithms are suboptimal, and which can be improved upon in future studies: surfaces with rough terrain or discontinuous slope, dense forest areas that laser beams cannot penetrate, and regions with low vegetation that is often ignored by ground filters.
Article
Full-text available
In this paper, we present PlantGL, an open-source graphic toolkit for the creation, simulation and analysis of 3D virtual plants. This C++ geometric library is embedded in the Python language which makes it a powerful user-interactive platform for plant modelling in various biological application domains. PlantGL makes it possible to build and manipulate geometric models of plants or plant parts, ranging from tissues and organs to plant populations. Based on a scene graph augmented with primitives dedicated to plant representation, several methods are provided to create plant architectures from either field measurements or procedural algorithms. Because they reveal particularly useful in plant design and analysis, special attention has been paid to the definition and use of branching system envelopes. Several examples from different modelling applications illustrate how PlantGL can be used to construct, analyse or manipulate geometric models at different scales.
Article
This work investigates the spatial distribution of wheat plants and its consequences on the canopy structure. A set of RGB images were taken from nadir on a total 14 plots showing a range of sowing densities, cultivars and environmental conditions. The coordinates of the plants were extracted from RGB images. Results show that the distance between-plants along the row follows a gamma distribution law, with no dependency between the distances. Conversely, the positions of the plants across rows follow a Gaussian distribution, with strongly interdependent. A statistical model was thus proposed to simulate the possible plant distribution pattern. Through coupling the statistical model with 3D Adel-Wheat model, the impact of the plant distribution pattern on canopy structure was evaluated using emerging properties such as the green fraction (GF) that drives the light interception efficiency. Simulations showed that the effects varied over different development stages but were generally small. For the intermediate development stages, large zenithal angles and directions parallel to the row, the deviations across the row of plant position increased the GF by more than 0.1. These results were obtained with a wheat functional-structural model that does not account for the capacity of plants to adapt to their local environment. Nevertheless, our work will extend the potential of functional-structural plant models to estimate the optimal distribution pattern for given conditions and subsequently guide the field management practices.