No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Sensing the occurrence and dynamics of Cercospora leaf spot disease using UAV-supported image data and deep learning
Abstract and Figures
The most damaging foliar disease in sugar beet is Cercospora leaf spot (CLS), caused by Cercospora beticola Sacc. The pathogen is expanding its territory due to climate conditions, generating the need for early and accurate detection to avoid yield losses. In Germany, monitoring and control strategies are based on visual field assessments, with the parameter disease incidence (DI). This parameter triggers warning systems when a threshold is achieved, and decision-making takes place for fungicide application. However, visual scoring is a time-consuming activity that requires well-trained personnel and is the principal bottleneck for CLS control. Digital technologies can support this process. Thus, the present work is based on two trial fields conducted and monitored in 2020 using an unmanned aerial vehicle (UAV) equipped with a multispectral camera. Image data were collected in time series during the vegetation period. Trials were sown with different sugar beet varieties; for field management, there was employed diverse fungicide strategies, and artificial inoculation took place in a spot manner. Parallel to the flight mission and additional assessment of DI, disease severity (DS) via KWS scale was collected by experts as so-called ground truth (GT). Combined with image-processing, it was possible to catalogize plants in field trials, identify them over time, and use them for training and testing models. A convolutional neural network (CNN) supported by cataloged data was trained to perform classification of the disease presence in time-series, and performance was evaluated. As the last image processing step, maps were generated showing site-specific distribution of the diseased plants in the field. Generated maps can serve as a basis for application maps in practical cultivation or the evaluation of variety performance in variety trials. The presented methodological approach provides high precision and sensitivity in CLS detection and offers the potential to automate processes of CLS monitoring for different application areas.
Figures - uploaded by Abel Barreto
Author content
All figure content in this area was uploaded by Abel Barreto
Content may be subject to copyright.
... The accurate detection and monitoring of CLS through phenotyping is critical for informed decision-making in managing crop diseases. Some successes have been observed using individual sensor types, such as RGB [26,27,21], multispectral Barreto Alcántara et al. [6], Günder et al. [24,23], Jay et al. [27], and hyperspectral sensors [47,37,12,36,4], respectively. However, only two studies explore LiDAR [61] and thermal imaging technology [15] for CLS detection, contrasting with the numerous examples found for other crops and scenarios, as highlighted by [39,63,9,11,35]. ...
... The spectral signature of CLS has been studied across different wavelength ranges, including the UV (250 nm to 500 nm) [12], visible light (VIS)-near-infrared light (NIR) (400 nm to 1050 nm) [36], and upper NIR (975 nm to 2500 nm) [4]. Vegetative indices and machine learning techniques, such as support vector machines, classical ML, and CNNs, have been widely applied [36,6,26]. Soil-plant segmentation has also been explored using vegetative indices and thresholds, though with less accurate results due to the presence of dry material. ML has improved these results [6], and segmentation of individual leaves has also been investigated [7]. ...
... ML has improved these results [6], and segmentation of individual leaves has also been investigated [7]. Regarding the sensor mounting platform, multiple reports have used UAVs [26,27,21,7,24,23]. Only Jay et al. [27] demonstrates the use of a UGV and compares it with UAVs. ...
... Additionally, they play a crucial role in aiding the development of curative site-specific disease control strategies (Maes and Steppe, 2019;Mahlein et al., 2018). Modern Imaging UAV systems utilize high-resolution multispectral technology, which provides detailed canopy information in the visual and near-infrared spectral range, allowing field mapping under natural light conditions Ispizua Yamati et al., 2022;Jay et al., 2020). By capturing images and employing appropriate analysis routines, UAV systems offer a non-destructive assessment tool that delivers valuable disease intensity parameters, including disease severity (DS) and disease incidence (DI) , which are obtained through automated post-processing steps. ...
... In this thesis, it was mainly acquired ground truth at the plot level. However, the imbalanced distribution, especially low frequency of diseased specimens, and environmental light conditions lead to corroborate the disease status of leaf scoring units when first symptoms appear Ispizua Yamati et al., 2022). An acquisition of unit-to-unit ground truth data will be necessary in future studies for sensitivity determination in future approaches. ...
... The classification of remaining KWS-scales (2, 3, 4, 5, 6, and 7) is inaccurate with F1-scores lower than 40%, which suggest limited properties of trained classifier for class separability, and this can suggest the reduction of KWS-categories(Sokolova and Lapalme, 2009). Besides accuracy of expert estimation, the factors for this inaccurate prediction within KWS-categories should be related to similar intra-class features for class discrimination(Sokolova and Lapalme, 2009), and sugar beet plants distribution on the field, where neighbor plants with low or higher severity affected the performance of the prediction(Ispizua Yamati et al., 2022). Reducing the number of classes may indeed enhance the robustness of automatic predictions using UAV-based ...
Cercospora leaf spot (CLS) in sugar beet is a damaging leaf disease caused by the fungal pathogen Cercospora beticola Sacc. This disease leads to substantial yield diminishment, and its management poses a challenge owing to rapid sporulation and high genetic variability. Integrated pest management strategies, including cultural practices, cultivar resistance, and fungicide management, are used to mitigate the disease. Disease intensity evaluation plays a crucial role in plant breeding for resistance screening and in agricultural practice for guiding control measures. The use of optical sensor technology and unmanned aerial vehicles (UAVs) with multispectral or hyperspectral cameras provides a novel alternative to human-based disease assessment. These sensors capture reflected light in multiple wavelength bands, allowing high spatial resolution imaging with spectral information. Machine and deep learning techniques are utilized to analyze multispectral UAV images and extract relevant disease assessment information. The combination of multispectral UAV data and machine learning approaches holds great promise for assessing parameters such as disease incidence (DI) and disease severity (DS) as a basis for decision-making. This thesis focuses on using RGB and multispectral imaging sensor technologies, UAVs, and machine learning to monitor and assess CLS in sugar beet. Two main application scenarios were investigated: evaluating tolerance and resistance in variety trials, and assessing parameters for decision-making in integrated CLS control in agricultural practice. The results of this dissertation recommended utilizing multispectral UAV systems for evaluating CLS resistance, particularly through an image-based and pixel-wise quantification of healthy foliage and soil regions. The close association between healthy foliage and yield outcomes emphasizes the importance of the proposed pixel-wise methods in breeding procedures. Furthermore, the identification and standardization of image-based scoring units are crucial for crop protection. Accurate detection of diseased specimens is essential for efficient site-specific disease management. In the present work, machine learning models were adapted and developed to detect DI and DS parameters with high accuracy. Procedures considering plant, circle, and leaf scoring units were incorporated to optimize decision-making. However, limitations in spatial resolution and nadir UAV-perspective, as well as challenges in discriminating diseased tissue from bare soil under certain light conditions, may impact the sensitivity for detecting first symptoms of disease. Curative site-specific fungicide application and generation of multidisease application maps are potential future developments. Overall, the dissertation demonstrates the potential of multispectral UAV-based methodologies for advancing disease resistance breeding and precise disease control, offering valuable applications in practical agriculture for integrated control of CLS. The knowledge gained from studying Cercospora beticola Sacc. and sugar beet can be transferred to other relevant sugar beet diseases such as Powdery mildew, Rust, and virus yellows using the established UAV-based assessment pipeline.
... Weekly severity scoring was carried out to evaluate disease progression. The scoring scale was adopted from earlier work [10]. The scale ranged from 0 (i.e., stage before canopy closure), over 9 (i.e., plants died off) to 10 (i.e., complete recovery from the disease). ...
... The accuracies were comparable to using original satellite imagery. Limitations and potentials of such models have already been described [10], [17], [18] have shown the potential of using multispectral images for detection of CLS. A problem is the quality of the scoring of the experiments [10], [19]. ...
... Limitations and potentials of such models have already been described [10], [17], [18] have shown the potential of using multispectral images for detection of CLS. A problem is the quality of the scoring of the experiments [10], [19]. Expert scores may not be necerssarily be considered as ground "truth" and may carry similar errors as the remotely sensed data. ...
In the field of vegetation remote sensing, the concept and conversion across scales has been addressed—but not concluded—in recent years. A large array of different sensors is deployed using various platforms such as uncrewed aerial vehicles (UAVs) and satellites. In this context, while multiple concepts of scale exist, the influence of spatial scale plays a major role when comparing and validating optical data derived from either of these platforms. Until to date, no satisfying method to allow cross sensor and scale studies exists. The presented work proposes a custom U-Net architecture to convert multispectral UAV imagery to multispectral satellite imagery. U-Nets are convolutional networks that have shown to be capable of learning mapping patterns between an input image and an output image. Therefore, artificially created images resembles original images at a high level of local detail and global semantic coherence. By using a multi-year, multispectral image UAV dataset, capturing a biologically and structural complex pathosystem (Cercosopora beticola + Beta vulgaris L.), it is shown that the developed custom U-Net can convert original UAV images with low structural and spectral error (MSE of 0.0001, SSIM test score of 0.0003, DWT test score of 0.0104, and a PSNR value of 40.3551) into artificial satellite images. Also, it is compared whether band center and bandwidth are accurately converted. To showcase an application, artificial images have been used to classify levels of Cercospora Leaf Spot in sugar beet. Classification accuracies are similar to those derived from the classification using original images. Therefore, it can be concluded that image to image translation using U-Nets can be a powerful tool for scaling studies in remote sensing. But they also offer to address the issue of data sparsity in machine learning as the conversion allows to generate images across different scales, therefore potentially adding valuable data to time-series studies.
... First, we have to define a DS annotation scale that serves as a guideline for all human expert annotations and finally as "unit" of the model input. In this work, the rating scale developed in [30] will be used with an extension for non-infested and newly sprouted plants. Fig 2 shows the numerical scale with exemplary plant images. ...
... Finally, a rating of 9 is given when the foliage experiences complete death [31]. In order to complement the scale, we added the 0 for non-infested sugar beets before canopy closure, and the 10 for newly sprouted plants as in [30]. In order to apply the model also on regions, where only soil is visible, we further added a -1 as "no plant" or "soil only" label by still maintaining the continuous fashion of the severity scale. ...
Remote sensing and artificial intelligence are pivotal technologies of precision agriculture nowadays. The efficient retrieval of large-scale field imagery combined with machine learning techniques shows success in various tasks like phenotyping, weeding, cropping, and disease control. This work will introduce a machine learning framework for automatized large-scale plant-specific trait annotation for the use case of disease severity scoring for CLS in sugar beet. With concepts of DLDL, special loss functions, and a tailored model architecture, we develop an efficient Vision Transformer based model for disease severity scoring called SugarViT. One novelty in this work is the combination of remote sensing data with environmental parameters of the experimental sites for disease severity prediction. Although the model is evaluated on this special use case, it is held as generic as possible to also be applicable to various image-based classification and regression tasks. With our framework, it is even possible to learn models on multi-objective problems, as we show by a pretraining on environmental metadata. Furthermore, we perform several comparison experiments with state-of-the-art methods and models to constitute our modeling and preprocessing choices.
... Similarly, other factors negatively influence the variability and repeatability of observations, including the effect of noise, heat, exhaustion or time allocated for an assessment [5]. All those factors emphasize the limitations of visual scoring methods and motivate the development of innovative and automated UAV-based imaging approaches for quantification of plant diseases [8][9][10][11]. ...
... Jay et al. [8] and Görlich et al. [9] segment limits of plot regions from RGB and multispectral orthomosaic images to calculate DS in variety trials. Similarly, a pipeline by Günder et al. [12] to segment individual plants, was extended for an application in disease quantification of Cercospora leaf spot (CLS) to classify infested plants according to their disease categories [10]. Circular regions within a plot are considered for an automated analysis of multispectral images to calculate DI, DS, and additional parameters such as area of foliage, area of healthy foliage, number of lesions and mean area of lesion by unit of foliage [11]. ...
In crop protection, disease quantification parameters such as disease incidence (DI) and disease severity (DS) are the principal indicators for decision making, aimed at ensuring the safety and productivity of crop yield. The quantification is standardized with leaf organs, defined as individual scoring units. This study focuses on identifying and segmenting individual leaves in agricultural fields using unmanned aerial vehicle (UAV), multispectral imagery of sugar beet fields, and deep instance segmentation networks (Mask R-CNN). Five strategies for achieving network robustness with limited labeled images are tested and compared, employing simple and copy-paste image augmentation techniques. The study also evaluates the impact of environmental conditions on network performance. Metrics of performance show that multispectral UAV images recorded under sunny conditions lead to a performance drop. Focusing on the practical application, we employ Mask R-CNN models in an image-processing pipeline to calculate leaf-based parameters including DS and DI. The pipeline was applied in time-series in an experimental trial with five varieties and two fungicide strategies to illustrate epidemiological development. Disease severity calculated with the model with highest Average Precision (AP) shows the strongest correlation with the same parameter assessed by experts. The time-series development of disease severity and disease incidence demonstrates the advantages of multispectral UAV-imagery in contrasting varieties for resistance, as well as the limits for disease control measurements. This study identifies key components for automatic leaf segmentation of diseased plants using UAV imagery, such as illumination and disease condition. It also provides a tool for delivering leaf-based parameters relevant to optimize crop production through automated disease quantification by imaging tools.
... First, we have to define a DS annotation scale that serves as a guideline for all human expert annotations and finally as "unit" of the model input. In this work, the rating scale developed in [19] will be used with an extension for noninfested and newly sprouted plants. Figure 3 shows the numerical scale with exemplary plant images. ...
... The scale is based on the usual CLS rating scale. We added the 0 for non-infested sugar beets before canopy closure, and the 10 for newly sprouted plants as in [19]. different labels. ...
Remote sensing and artificial intelligence are pivotal technologies of precision agriculture nowadays. The efficient retrieval of large-scale field imagery combined with machine learning techniques shows success in various tasks like phenotyping, weeding, cropping, and disease control. This work will introduce a machine learning framework for automatized large-scale plant-specific trait annotation for the use case disease severity scoring for Cercospora Leaf Spot (CLS) in sugar beet. With concepts of Deep Label Distribution Learning (DLDL), special loss functions, and a tailored model architecture, we develop an efficient Vision Transformer based model for disease severity scoring called SugarViT. One novelty in this work is the combination of remote sensing data with environmental parameters of the experimental sites for disease severity prediction. Although the model is evaluated on this special use case, it is held as generic as possible to also be applicable to various image-based classification and regression tasks. With our framework, it is even possible to learn models on multi-objective problems as we show by a pretraining on environmental metadata.
... During the last decade, the dominant tools for image analysis are different convolutional neural net (CNN) approaches. These approaches use the complete RGB image as an input to apply a model, which is able to be adapted to various application scenarios from segmentation (Jiang and Li 2020), plant (Barreto et al. 2021) and stress detection (Ispizua Yamati et al. 2022). This leads to the generation of high resolution and georeferenced field maps by merging various single images to a big map. ...
Maize ( Zea mays ) is one of the most important cereal crops globally, providing food, feed, and fuel for humans and animals. However, the production of maize is frequently challenged by various stresses that can severely impact crop yield and quality. Early detection and identification are critical for implementing timely and effective control measures to reduce their impact. Therefore, it is essential to establish effective disease and pest monitoring and management strategies to ensure sustainable maize production and maintain food security. Conventional detection methods relying on visual assessments by human experts are time-consuming, labour-intensive, and subjective. The emergence of imaging sensors, computer vision technologies, and high throughput platforms has revolutionised the detection and differentiation of crop diseases, offering a non-invasive and rapid approach. In this publication, we present a review on imaging sensors for the detection and differentiation of symptoms on maize caused by diseases and pests. The main advantages and limitations of each imaging sensor, along with its applications and case studies for maize disease detection, are introduced and discussed. Recent advances in the visible, near-infrared, and hyperspectral imaging for maize disease detection are highlighted, and the importance of different sensors is discussed. The goal is to provide a comprehensive overview of the current state-of-the-art in this field, highlighting the potential of imaging sensors for improving maize production and identifying future research directions in this area.
... Next, we simply define a frame around those pixels and get smaller "tiles." These can then be used for further steps, for instance, as a training set for neural network architectures as in Yamati et al. [33]. The fact that the images are linked to individual plants at multiple dates enables this dataset to be used not only for spatially related but also for time-series analyses. ...
Background
Unmanned aerial vehicle (UAV)–based image retrieval in modern agriculture enables gathering large amounts of spatially referenced crop image data. In large-scale experiments, however, UAV images suffer from containing a multitudinous amount of crops in a complex canopy architecture. Especially for the observation of temporal effects, this complicates the recognition of individual plants over several images and the extraction of relevant information tremendously.
Results
In this work, we present a hands-on workflow for the automatized temporal and spatial identification and individualization of crop images from UAVs abbreviated as “cataloging” based on comprehensible computer vision methods. We evaluate the workflow on 2 real-world datasets. One dataset is recorded for observation of Cercospora leaf spot—a fungal disease—in sugar beet over an entire growing cycle. The other one deals with harvest prediction of cauliflower plants. The plant catalog is utilized for the extraction of single plant images seen over multiple time points. This gathers a large-scale spatiotemporal image dataset that in turn can be applied to train further machine learning models including various data layers.
Conclusion
The presented approach improves analysis and interpretation of UAV data in agriculture significantly. By validation with some reference data, our method shows an accuracy that is similar to more complex deep learning–based recognition techniques. Our workflow is able to automatize plant cataloging and training image extraction, especially for large datasets.
Early diagnosis of diseases in agriculture is an important factor in reducing the negative environmental impacts by effectively and economically managing the losses caused by these diseases and reducing the use of chemicals. There are different options within the scope of remote sensing for the early detection of diseases. Among these, choosing a method that can detect diseases accurately without harming the plant and the environment is important. Today, positive developments have been made toward non-invasive and effective detection of diseases with thermal camera-based image processing techniques. In this context, there is potential for disease detection with data collection, image processing, and the determination of the characteristics of disease agents through thermal imaging. The research was based on Cercospora leaf spot (Cercospora beticola Sacc.) diseases which have significant economic loss potential in sugar beet. The effectiveness of the proposed method was evaluated in experiments involving Cercospora beticola, utilizing a climate station early warning system and UAV-based thermal images across three subjects and six replicate field trial plots. Analyses were made for the early detection of diseases by comparing thermal images taken from the field with multispectral images taken simultaneously. It was investigated whether it was possible to diagnose the disease early before physical symptoms were seen using image processing and machine learning methods. The variability of leaves was analyzed using field images, thermal images, and machine learning algorithms. Thermal imaging enables the rapid detection of potential disease development by measuring increases in leaf temperature in infrared wavelengths. However, a significant limitation of this method in practice is its sensitivity to climate factors such as air temperature and humidity, which can cause rapid fluctuations in the index. This study compared five machine learning algorithms based on four key metrics. MS imaging achieved about 25% higher accuracy in predicting early disease than TE imaging. This study indicates that thermal imaging provides valuable information but is not as effective as multispectral imaging in detecting early-stage stress factors related to diseases.
Rhizoctonia crown and root rot (RCRR), caused by Rhizoctonia solani, can cause severe yield and quality losses in sugar beet. The most common strategy to control the disease is the development of resistant varieties. In the breeding process, field experiments with artificial inoculation are carried out to evaluate the performance of genotypes and varieties. The phenotyping process in breeding trials requires constant monitoring and scoring by skilled experts. This work is time demanding and shows bias and heterogeneity according to the experience and capacity of each individual person. Optical sensors and artificial intelligence have demonstrated a great potential to achieve higher accuracy than human raters and the possibility to standardize phenotyping applications. A workflow combining red-green-blue (RGB) and multispectral imagery coupled to an unmanned aerial vehicle (UAV), and machine learning techniques was applied to score diseased plants and plots affected by RCRR. Georeferenced annotation of UAV orthorectified images. With the annotated images, five convolutional neural networks were trained to score individual plants. The training was carried out with different image analysis strategies and data augmentation, respectively. The custom convolutional neural network trained from scratch together with a pre-trained MobileNet showed the best precision in scoring RCRR (0.73 to 0.85). The average per plot of spectral information was used to score plots, and the benefit of adding the information obtained from the score of individual plants was compared. For this purpose, machine learning models were trained together with data management strategies, and the best-performing model was chosen. A combined pipeline of Random Forest and k-Nearest neighbors have shown the best weighted precision (0.67). This research provides a reliable workflow for detecting and scoring RCRR based on aerial imagery. RCRR is often distributed heterogeneously in trial plots, therefore, considering the information from individual plants of the plots showed a significant improvement of UAV based automated monitoring routines.
Remote sensing technology is vital for precision agriculture, aiding in early issue detection, resource management, and environmentally friendly practices. Recent advances in remote sensing technology and data processing have propelled unmanned aerial vehicles (UAVs) into valuable tools for obtaining detailed data on plant diseases with high spatial, temporal, and spectral resolution. Given the growing body of scholarly research centered on UAV-based disease detection, a comprehensive review and analysis of current studies becomes imperative to provide a panoramic view of evolving methodologies in plant disease monitoring and to strategically evaluate the potential and limitations of such strategies. This study undertakes a systematic quantitative literature review to summarize existing literature and discern current research trends in UAV-based applications for plant disease detection and monitoring. Results reveal a global disparity in research on the topic, with Asian countries being the top contributing countries (43 out of 103 papers). World regions such as Oceania and Africa exhibit comparatively lesser representation. To date, research has largely focused on diseases affecting wheat, sugar beet, potato, maize, and grapevine. Multispectral, reg-green-blue, and hyperspectral sensors were most often used to detect and identify disease symptoms, with current trends pointing to approaches integrating multiple sensors and the use of machine learning and deep learning techniques. Future research should prioritize (i) development of cost-effective and user-friendly UAVs, (ii) integration with emerging agricultural technologies, (iii) improved data acquisition and processing efficiency (iv) diverse testing scenarios, and (v) ethical considerations through proper regulations.
Spring dead spot (Ophiosphaerella herpotricha, O. korrae, O. narmari; SDS) is among the most destructive diseases of bermudagrass [Cynodon dactylon (L.) Pers.] and hybrid bermudagrass (C. dactylon x C. transvaalensis Burtt Davy) in the Transition Zone of the United States. Spring dead spot's primary causal agents in the United States, Ophiosphaerella korrae and O. herpotricha, infect bermudagrass in the fall with symptoms appearing in the spring when winter dormancy breaks. Patches of necrotic turfgrass often reoccur in the same location and expand into surrounding areas. Chemical control options are often expensive or provide inconsistent results. Our objectives were to develop SDS incidence maps, analyze these maps, and evaluate suppression efficacy of chemical applications guided by incidence‐based maps. Digital imagery captured with an unmanned aerial vehicle (UAV) was used to create SDS incidence maps in the spring of 2016, 2017, and 2018. In the fall of 2016 and 2017, a targeted, site‐specific penthiopyrad treatment was evaluated against blanket, full‐coverage applications of penthiopyrad and tebuconazole, and a nontreated control. Treatments were compared using digital image analysis of diseased area (DA) and post‐treatment SDS patch count (PC). Across both metrics, the penthiopyrad treatments had significantly less disease than both the tebuconazole and nontreated control in 2016–2017. Targeted penthiopyrad compared favorably to full coverage penthiopyrad for DA and PC in 2016–2017, but full‐coverage penthiopyrad was superior to targeted penthiopyrad and tebuconazole in 2017–2018 for both DA and PC. Targeted penthiopyrad using SDS incidence maps required 51% less fungicides in 2016–2017 and 65% less in 2017–2018 when compared to full‐coverage penthiopyrad.
Plant diseases can impact crop yield. Thus, the detection of plant diseases using sensors that can be mounted on aerial vehicles is in the interest of farmers to support decision-making in integrated pest management and to breeders for selecting tolerant or resistant genotypes. This paper investigated the detection of Cercospora leaf spot (CLS), caused by Cercospora beticola in sugar beet using RGB imagery. We proposed an approach to tackle the CLS detection problem using fully convolutional neural networks, which operate directly on RGB images captured by a UAV. This efficient approach does not require complex multi- or hyper-spectral sensors, but provides reliable results and high sensitivity. We provided a detection pipeline for pixel-wise semantic segmentation of CLS symptoms, healthy vegetation, and background so that our approach can automatically quantify the grade of infestation. We thoroughly evaluated our system using multiple UAV datasets recorded from different sugar beet trial fields. The dataset consisted of a training and a test dataset and originated from different fields. We used it to evaluate our approach under realistic conditions and analyzed its generalization capabilities to unseen environments. The obtained results correlated to visual estimation by human experts significantly. The presented study underlined the potential of high-resolution RGB imaging and convolutional neural networks for plant disease detection under field conditions. The demonstrated procedure is particularly interesting for applications under practical conditions, as no complex and cost-intensive measuring system is required.
Real-time application technologies based on the target crop, crop surface area and biomass using non-contact sensors for precise fungicide spraying in winter wheat have been developed in a joint research project. The decision support system proPlant expert.classic and the internet-version proPlant expert.com (proPlant GmbH) suggest appropriate fungicides and dosages for certain infection scenarios of eight important leaf and ear diseases of winter wheat. The Precision Farming Module "Fungicide", which runs on the on-board terminal in the tractor cabin, controls the spraying process. During the spraying process, the module defines the local target application amount using a local ultrasonic sensor value as an input parameter. Winter wheat field experiments were conducted in 2013 and 2014 (Agri Con Co., ATB) to analyse the relationship between the sensor values (ultrasonic and camera) and the leaf area index (LAI) and biomass crop parameters that are important for a locally adapted and variable fungicide application rate. Measurements were performed several times during the vegetation period at sampling points that were visually selected based on crop density. Regression analyses showed that after technical changes in 2014, a linear relationship was obtained between the sensor values and the two crop parameters.
Early and accurate detection and diagnosis of plant diseases are key factors in plant production and the reduction of both qualitative and quantitative losses in crop yield. Optical techniques, such as RGB imaging, multi- and hyperspectral sensors, thermography, or chlorophyll fluorescence, have proven their potential in automated, objective and reproducible detection systems for the identification and quantification of plant diseases at early time points in epidemics. Recently, 3D scanning has also been added as an optical analysis that supplies additional information on crop plant vitality. Different platforms from proximal to remote sensing are available for multi-scale monitoring of single crop organs or entire fields. Accurate and reliable detection of diseases is facilitated by highly sophisticated and innovative methods of data analysis that lead to new insights derived from sensor data for complex plant-pathogen systems. Non-destructive, sensor-based methods support and expand upon visual and/or molecular approaches to plant disease assessment. The most relevant areas of application of sensor-based analyses are precision agriculture and plant phenotyping.
The ability of learning networks to generalize can be greatly enhanced by providing constraints from the task domain. This paper demonstrates how such constraints can be integrated into a backpropagation network through the architecture of the network. This approach has been successfully applied to the recognition of handwritten zip code digits provided by the U.S. Postal Service. A single network learns the entire recognition operation, going from the normalized image of the character to the final classification.
Counting crop seedlings is a time-demanding activity involved in diverse agricultural practices like plant cultivating, experimental trials, plant breeding procedures, and weed control. Unmanned Aerial Vehicles (UAVs) carrying RGB cameras are novel tools for automatic field mapping, and the analysis of UAV images by deep learning methods can provide relevant agronomic information. UAV-based camera systems and a deep learning image analysis pipeline are implemented for a fully automated plant counting in sugar beet, maize, and strawberry fields in the present study. Five locations were monitored at different growth stages, and the crop number per plot was automatically predicted by using a fully convolutional network (FCN) pipeline. Our FCN-based approach is a single model for jointly determining both the exact stem location of crop and weed plants and a pixel-wise plant classification considering crop, weed, and soil. To determinate the approach performance, predicted crop counting was compared to visually assessed ground truth data. Results show that UAV-based counting of sugar-beet plants delivers forecast errors lower than 4.6%, and the main factors for performance are related to the intra-row distance and the growth stage. The pipeline’s extension to other crops is possible; the errors of the predictions are lower than 4% under practical field conditions for maize and strawberry fields. This work highlight the feasibility of automatic crop counting, which can reduce manual effort to the farmers.
The impact of Cercospora leaf spot, caused by Cercospora beticola, on the yield and quality of sugar beets (Beta vulgaris) was studied in one field trial in 1982 and two trials in 1983. Fungicides were applied to cultivars of various disease susceptibilities to obtain different levels of disease severity. Plants were rated for disease severity with a spot-percentage scale according to the number of lesions per leaf at intensities <3% and with standard area diagrams at intensities ?3%. A disease severity of 50% approximately 10 days before harvest corresponded to relative dollar losses of 43, 27, and 27% for trials 13, respectively. Reduced payment per hectare was sometimes evident for epidemics of >3% severity by mid-September. Increased concentration of impurities and decreased sugar concentration in beet roots and decreased root weight generally were correlated with disease severity. Beets with higher concentrations of impurities yielded less pure sugar and more unrefined sugar (molasses) in the extraction process. In diseased plots, sugar loss to molasses had a minor impact on dollar return compared with losses attributed to reductions in root weight and sugar content.
Preliminary observations and data on sugarbeet ( Beta vulgaris L.) yield and quality components suggested that a possible differential response to infection by the fungus Cercospora beticola exists among sugarbeet cultivars and genotypes with different inherent levels of resistance and possibly among cultivars with similar resistance. The objectives of this study were: 1) to determine, for cultivars known to vary in leaf spot resistance, the effects of leaf spot disease on sucrose yield components and on extract chemical components which affect sugarbeet juice purity and 2) to examine whether infected cultivars with similar resistance to leaf spot respond in similar or different ways. Eight sugarbeet cultivars ranging from leaf spot resistant to leaf spot susceptible were examined under an artificially induced leaf spot epidemic in a 2‐year field study. With increasing severity of C. beticola infection there was a general increase in nonsucrose chemical components, and decrease in gross sucrose yield, yield components, and purity. Of the chemical components, Na, nitrate, amino N, and total N consistently showed the greatest increases with increased infection by C. beticola . Numerous examples of differential response of the cultivars to C. beticola infection were found for nonsucrose chemical components as well as for sucrose yield and yield components.
Management of Cercospora leaf spot, caused by Cercospora beticola. is necessary for the economic production of sugar beet (Beta vulgaris). The objectives of this study were to evaluate the impact of two relative humidity thresholds (87 and 90%) on the daily infection values (DIVs) used to determine when fungicide applications were required, to determine whether current Cercospora management recommendations for northern areas of Minnesota and North Dakota could be used by rowers in the southern areas of these states, and to compare the utility of calendar-based fungicide applications with the Cercospora management model. Research was conducted in Breckenridge. MN and St. Thomas, ND in 2003 and 2004. Fungicide applications significantly (P = 0.05) reduced maximum disease severity (y(max)) and area under the disease progress curve (AUDPC) when compared with the nontreated control at both locations during 2003 and 2004.. Fungicides applied according to DIVs calculated at RH >= 87% or RH > 90% gave similar results. The mandatory second fungicide application 14 days after the first application for southern areas did not significantly decrease disease severity or AUDPC, or improve root yield or recoverable sucrose compared with treatments without the mandatory application. This research illustrates that a DIVs calculated at RH >= 87% would result in similar timing of fungicide applications compared with DIVs calculated at RH > 90%. The results further show that the recommendation of fungicide applications at initial symptom and subsequent applications based on DIV and disease severity should be used for both northern and southern growers. Finally, this research showed that fungicide applications based on the Cercospora management model provided similar. effective disease control with fewer fungicide applications compared with calendar-based applications.
Two disease severity scales for Cercospora leaf spot (CLS) assessment were compared. CLS was assessed in two experimental fields in the Netherlands in 1999 using two scales: a single leaf severity assessment, as used in the IPS-system (Integriertes Pflanzenschutzsystem) in Germany, referred to as DSIPS, and a whole plant assessment, Agronomica diagram from Italy, referred to as DSAGR. To obtain a range of disease severities, fungicides were applied at defined action thresholds based on disease incidence and severity. There was an exponential relationship between DSIPS and DSAGR (R2=86%) for pooled data with little change in the whole plant assessment above DSIPS=5%. An exponential curve best fitted DSIPS and root and sugar weight in fields severely infected with CLS, whereas a linear curve was found for mildly infected fields. A linear curve fitted DSAGR best with root and sugar weight in both severely and mildly infected fields. No relationship was found between both DSIPS and DSAGR and sugar content. The use of DSAGR was less time consuming in monitoring and was done with greater accuracy, efficiency and level of reproducibility than DSIPS. These results demonstrate that CLS assessment can be less time-consuming and more practical in application.
Unlabelled:
SUMMARY Leaf spot disease caused by Cercospora beticola Sacc. is the most destructive foliar pathogen of sugarbeet worldwide. In addition to reducing yield and quality of sugarbeet, the control of leaf spot disease by extensive fungicide application incurs added costs to producers and repeatedly has selected for fungicide-tolerant C. beticola strains. The genetics and biochemistry of virulence have been examined less for C. beticola as compared with the related fungi C. nicotianae, C. kikuchii and C. zeae-maydis, fungi to which the physiology of C. beticola is often compared. C. beticola populations generally are not characterized as having race structure, although a case of race-specific resistance in sugarbeet to C. beticola has been reported. Resistance currently implemented in the field is quantitatively inherited and exhibits low to medium heritability.
Taxonomy:
Cercospora beticola Sacc.; Kingdom Fungi, Subdivision Deuteromycetes, Class Hyphomycetes, Order Hyphales, Genus Cercospora.
Identification:
Circular, brown to red delimited spots with ashen-grey centre, 0.5-6 mm diameter; dark brown to black stromata against grey background; pale brown unbranched sparingly septate conidiophores, hyaline acicular conidia, multiseptate, from 2.5 to 4 microm wide and 50-200 microm long.
Host range:
Propagative on Beta vulgaris and most species of Beta. Reported on members of the Chenopodiaceae and on Amaranthus. Disease symptoms: Infected leaves and petioles of B. vulgaris exhibit numerous circular leaf spots that coalesce in severe cases causing complete leaf collapse. Dark specks within a grey spot centre are characteristic for the disease. Older leaves exhibit a greater number of lesions with larger spot diameter. During the latter stage of severe epiphytotics, new leaf growth can be seen emerging from the plant surrounded by prostrate, collapsed leaves.
Control:
Fungicides in the benzimidazole and triazole class as well as organotin derivatives and strobilurins have successfully been used to control Cercospora leaf spot. Elevated levels of tolerance in populations of C. beticola to some of the chemicals registered for control has been documented. Partial genetic resistance also is used to reduce leaf spot disease.
Agricultural Plant Cataloging and Establishment of a Data Framework from UAV-based Crop Images by Computer Vision
- M Günder
- F R Ispizua Yamati
- J Kierdorf
- R Roscher
- A.-K Mahlein
- C Bauckhage
- S To Gigascience Jay
- A Comar
- R Benicio
- J Beauvois
- D Dutartre
- G Daubige
- W Li
- J Labrosse
- S Thomas
- N Henry
- M Weiss
- F Baret
Günder M., Ispizua Yamati F.R., Kierdorf J., Roscher R., Mahlein A.-K.,
Bauckhage C. (2022): Agricultural Plant Cataloging and Establishment
of a Data Framework from UAV-based Crop Images by Computer Vision,
Preprint arXiv:2201.02885v1 [cs.CV] 8 Jan 2022, submitted to GigaScience
Jay S., Comar A., Benicio R., Beauvois J., Dutartre D., Daubige G., Li W.,
Labrosse J., Thomas S., Henry N., Weiss M., Baret F. (2020): Scoring Cercospora Leaf Spot on Sugar Beet: Comparison of UGV and UAV Phenotyping
Systems, Plant phenomics 2020, ID 9452123; DOI: 10.34133/2020/9452123
Empirical-deterministic prediction of disease and losses caused by Cercospora leaf spots in sugar beets
- P F J Wolf
- J A Verreet
Wolf P.F.J., Verreet J.A. (2009): Empirical-deterministic prediction of disease
and losses caused by Cercospora leaf spots in sugar beets, Journal für Kulturpflanzen 61(5), 168-177