FIGURE 4 - uploaded by Pieter Blok
Content may be subject to copyright.
| One single hyperspectral line image, horizontally showing the spatial information of one line, vertically showing the spectral reflection between 400 and 1000 nm.

| One single hyperspectral line image, horizontally showing the spatial information of one line, vertically showing the spectral reflection between 400 and 1000 nm.

Source publication
Article
Full-text available
Virus diseases are of high concern in the cultivation of seed potatoes. Once found in the field, virus diseased plants lead to declassification or even rejection of the seed lots resulting in a financial loss. Farmers put in a lot of effort to detect diseased plants and remove virus-diseased plants from the field. Nevertheless, dependent on the cul...

Context in source publication

Context 1
... full sensor resolution of the FX10 is 1024 pixels in the spatial by 224 bands spectral. In order to improve light sensitivity and speed, the images were binned by a factor 2 in the spatial direction and a factor 4 in the spectral direction, resulting in line images of 512 pixels × 56 pixels (Figure 4). As the bands at the start and the end of the spectrum were noisy, only the central 35 bands were kept for further processing. ...

Citations

... Table 5 shows some relevant studies from recent years. For the detection of potato viruses, Polder et al. [101] designed a fully convolutional neural network that was adapted for hyperspectral images and trained on two experimental rows in the field. The trained network was validated on two other rows, with different potato cultivars. ...
Article
Full-text available
Pathogen infection has greatly reduced crop production. As the symptoms of diseases usually appear when the plants are infected severely, rapid identification approaches are required to monitor plant diseases at early the infection stage and optimize control strategies. Hyperspectral imaging, as a fast and nondestructive sensing technology, has achieved remarkable results in plant disease identification. Various models have been developed for disease identification in different plants such as arable crops, vegetables, fruit trees, etc. In these models, important algorithms, such as the vegetation index and machine learning classification and methods have played significant roles in the detection and early warning of disease. In this paper, the principle of hyperspectral imaging technology and common spectral characteristics of plant disease symptoms are discussed. We reviewed the impact mechanism of pathogen infection on the photo response and spectrum features of the plants, the data processing tools and algorithms of the hyperspectral information of pathogen-infected plants, and the application prospect of hyperspectral imaging technology for the identification of plant diseases.
... In potato research, detection of diseases through remote sensing has been advanced for late blight (Gold et al. 2020;Ray et al. 2011;Duarte-Carvajalino et al. 2018) and viruses (Chávez et al. 2009;Chávez et al. 2010;Griffel et al. 2018;Polder et al. 2019). Whereas there are clear challenges to implementing such technologies routinely at a landscape or regional level, obvious direct application could be available through automated screening for resistance in breeding trials and rapid screening for diseases in seed multiplication plots. ...
... In the field of plant disease identification, the hyperspectral imaging technology is usually used for object detection because the difference in reflectance of plant disease features is slight (Yue et al., 2015;Polder et al., 2019;Wang D. et al., 2019). The investigation of Nagasubramanian et al. (2019) demonstrated that soybeans infected the charcoal rot are more sensitive than healthy soybeans in the wavelengths of visible spectra (400-700 nm). ...
Article
Full-text available
Maize leaf diseases significantly reduce maize yield; therefore, monitoring and identifying the diseases during the growing season are crucial. Some of the current studies are based on images with simple backgrounds, and the realistic field settings are full of background noise, making this task challenging. We collected low-cost red, green, and blue (RGB) images from our experimental fields and public dataset, and they contain a total of four categories, namely, southern corn leaf blight (SCLB), gray leaf spot (GLS), southern corn rust (SR), and healthy (H). This article proposes a model different from convolutional neural networks (CNNs) based on transformer and self-attention. It represents visual information of local regions of images by tokens, calculates the correlation (called attention) of information between local regions with an attention mechanism, and finally integrates global information to make the classification. The results show that our model achieves the best performance compared to five mainstream CNNs at a meager computational cost, and the attention mechanism plays an extremely important role. The disease lesions information was effectively emphasized, and the background noise was suppressed. The proposed model is more suitable for fine-grained maize leaf disease identification in a complex background, and we demonstrated this idea from three perspectives, namely, theoretical, experimental, and visualization.
... 2.2 on the final model. In the absence of potato meristem segmentation studies, precision scores were benchmarked against the Potato Virus Y (Polder et al., 2019), whose primary symptom is chlorotic foliage akin to the signal being detected by the image analysis approach to label stems. Polder et al. (2019) found precision scores between 0.23 and 0.54 when a fully convolutional network was used to achieve semantic segmentation of Potato Virus Y. ...
... In the absence of potato meristem segmentation studies, precision scores were benchmarked against the Potato Virus Y (Polder et al., 2019), whose primary symptom is chlorotic foliage akin to the signal being detected by the image analysis approach to label stems. Polder et al. (2019) found precision scores between 0.23 and 0.54 when a fully convolutional network was used to achieve semantic segmentation of Potato Virus Y. This is comparable with the performance of the faster R-CNN approach but outperforms the image analysis method. ...
Article
Full-text available
Potato (Solanum tuberosum) stem density variation in the field can be used to inform harvest timing to improve tuber size distribution. Current methods for quantifying stem density are manual with low throughput. This study examined the use of Unmanned Aerial Vehicle imagery as a high-throughput alternative. A colour-based feature extraction technique and a deep convolutional neural network (CNN) were compared for their effectiveness in enumerating apical meristems as a proxy to subtending stems. Two novel colour indices, named the cumulative blue differences index and blue difference normalized index, showed significant differences (P < 0.001) between meristematic leaves and mature leaves in comparison to other indices. The two indices were used to generate 500 pseudo-labelled human-corrected images as training data for the CNN. Benchmarked against a human labelled test dataset, the CNN performed better with a normalized Root Mean Square Error (nRMSE) of 0.09 than the sole use of the image analysis algorithm (nRMSE = 0.3) in predicting the number of meristems in a canopy at 52 days after planting. Furthermore, the CNN had better precision (Intersection over Union [IOU]: 0.49 and 0.56, respectively) than the image analysis algorithm (IOU: 0.33 and 0.13, respectively). Meristem counts in both approaches showed a linear relationship with actual subtending stem counts (P < 0.001). This study demonstrates the validity of using traditional image analysis and CNNs to generate meristem detectors with acceptable nRMSE. Transfer learning with CNN is proposed for developing meristem detectors for evaluating stem density variation from UAV images in the field.
... In the current state of the art, recent applications of HTPP suggest that automation can be performed for a wide variety of traits including drought tolerance [2,32], salt tolerance [33], and biotic stress [34,35] as well as for assessing the efficiency of plant protection agents [30]. Furthermore, facilities are not only used for phenotyping small model plants such as Arabidopsis thaliana [36,37] but also can be used for large plants such as Zea mays [2,32] and even, tree species [38]. ...
Article
Full-text available
High-throughput (HTP) plant phenotyping approaches are developing rapidly and are already helping to bridge the genotype–phenotype gap. However, technologies should be developed beyond current physico-spectral evaluations to extend our analytical capacities to the subcellular level. Metabolites define and determine many key physiological and agronomic features in plants and an ability to integrate a metabolomics approach within current HTP phenotyping platforms has huge potential for added value. While key challenges remain on several fronts, novel technological innovations are upcoming yet under-exploited in a phenotyping context. In this review, we present an overview of the state of the art and how current limitations might be overcome to enable full integration of metabolomics approaches into a generic phenotyping pipeline in the near future.
... LeNet, AlexNet, ResNet, GoogleNet, Visual Geometry Group (VGG) etc., are some of the examples of CNN based models. Few studies reported the use of CNN based plant disease classification [Chen, J et al., 2019, Wang G et al., 2017, Polder, G et al., 2019, Fuentes, A et al., 2018 and yielded promising results. [Afshar, P et al, 2018], disease classification [Sezer, A et al., 2019, Afshar, P et al., 2020, Verma, S et al., 2020 , drug detection , object detection [Kumar, A.D, 2019], hyper spectral image classification [Deng, F et al., ...
Preprint
Full-text available
Crop protection is the prime hindrance for food security. The plant diseases destroy the overall quality and quantity of the agricultural products. Grape is an important fruit and major source of vitamin C nutrients. The automatic decision making system plays a paramount role in agricultural informatics. This paper aims to detect the diseases in grape leaves using convolutional capsule networks. The capsule network is a promising neural network in the field of deep learning. This network uses a group of neurons as capsules and effectively represents spatial information of features. The novelty of the proposed work relies on the addition of convolutional layers before the primary caps layer which indirectly decrease the number of capsules and speed up the dynamic routing process. The proposed method is experimented with augmented and non-augmented datasets. It effectively detects the diseases of grape leaves with the accuracy of 99.12%. The performance of the method is compared with state-of-the art deep learning methods and produces reliable results.
... New plant breeding techniques (such as gene editing or CRISPR/Cas [50], based on experience with other crops [51]) are assessed as the question is how (fast) such systems can be developed and what these systems could eventually contribute towards the reduction of climate change. Special attention should be given to the negative feedback loop of viruses of which the detection is an important step forward to avoid progressing the negative side effects [52], and to the losses of nutrients that are emitted to air and soil, as these are more pronounced at the end of the response curves of any inputs, following the law of diminishing returns, until at the plateau where additional inputs eventually have no effect at all: they are completely lost to the environment [11,[53][54][55][56]. It is also important to know how the socioeconomic system reacts [57], if there are any ethical issues to address [58], and if NPBT is applied in a viable way [59,60]. ...
Article
Full-text available
The Butterfly framework of Wageningen University & Research (WUR) for assessing transitions towards a circular and climate-neutral society is presented. The Butterfly framework is built after analysis of existing frameworks that could only partly comply with the needs of the full set of stakeholders interlinked and operating in domains like society and well-being; food, feed, and biobased production; natural resources and living environment. It shows that for adequate action perspectives on and in these domains, the socio-ecological, socio-technical, and socio-institutional subsystems should be fully integrated, and stakeholders should be equally consulted and appreciated. In order to advance and integrate action perspectives of different stakeholders in the light of the transition to circularity with high-level ambitions like climate neutrality, stakeholders (groups) need to understand their position and links in a full systems perspective, which the Butterfly framework provides.
... The reviewed works prove the possibility of detecting oil palm [36,[50][51][52][53][54][55][56][57][63][64][65][66], citrus [73][74][75][76][77][78], Solanaceae family crops [91][92][93][94][95][96][97][98][99][100][101][102][103] and wheat [24,[124][125][126][127][128][129][130][131][132][140][141][142][143][144][145] diseases using HRS. ...
... Additionally, as we have mentioned earlier, the methods of analyzing the data obtained (machine learning, neural networks, statistical analysis, manual analysis), in our opinion, are only methods of automation that do not make a significant contribution to solving the problem of the early detection of plant diseases with HRS [90][91][92][93][94]97,105,142,143]. ...
Article
Full-text available
The development of hyperspectral remote sensing equipment, in recent years, has provided plant protection professionals with a new mechanism for assessing the phytosanitary state of crops. Semantically rich data coming from hyperspectral sensors are a prerequisite for the timely and rational implementation of plant protection measures. This review presents modern advances in early plant disease detection based on hyperspectral remote sensing. The review identifies current gaps in the methodologies of experiments. A further direction for experimental methodological development is indicated. A comparative study of the existing results is performed and a systematic table of different plants’ disease detection by hyperspectral remote sensing is presented, including important wave bands and sensor model information.
... In potato research, detection of diseases through remote sensing has been advanced for late blight (Gold et al. 2020;Ray et al. 2011;Duarte-Carvajalino et al. 2018) and viruses (Chávez et al. 2009;Chávez et al. 2010;Griffel et al. 2018;Polder et al. 2019). Whereas there are clear challenges to implementing such technologies routinely at a landscape or regional level, obvious direct application could be available through automated screening for resistance in breeding trials and rapid screening for diseases in seed multiplication plots. ...
Chapter
Cassava is an important crop in sub-Saharan Africa for food security, income generation, and industrial development. Business-oriented production systems require reliable supplies of high-quality seed. Major initiatives in Nigeria and Tanzania have sought to establish sustainable cassava seed systems. These include the deployment of new technologies for early generation seed (EGS) production; the promotion of new high-yielding and disease-resistant varieties; the updating of government seed policy to facilitate enabling certification guidelines; the application of ICT tools, Seed Tracker and Nuru AI, to simplify seed system management; and the establishment of networks of cassava seed entrepreneurs (CSEs). CSEs have been able to make profits in both Nigeria (US$ 551–988/ha) and Tanzania (US$ 1,000 1,500/ha). In Nigeria, the critical demand driver for cassava seed businesses is the provision of new varieties. Contrastingly, in Tanzania, high incidences of cassava brown streak disease mean that there is a strong demand for the provision of healthy seed that has been certified by regulators. These models for sustainable cassava seed system development offer great promise for scaling to other cassava-producing countries in Africa where there is strong government support for the commercialization of the cassava sector.
... In potato research, detection of diseases through remote sensing has been advanced for late blight (Gold et al. 2020;Ray et al. 2011;Duarte-Carvajalino et al. 2018) and viruses (Chávez et al. 2009;Chávez et al. 2010;Griffel et al. 2018;Polder et al. 2019). Whereas there are clear challenges to implementing such technologies routinely at a landscape or regional level, obvious direct application could be available through automated screening for resistance in breeding trials and rapid screening for diseases in seed multiplication plots. ...
Chapter
Full-text available
Root, tuber, and banana (RT&B) crops are critical for global food security. They are vegetatively propagated crops (VPCs) sharing common features: low reproductive rates, bulky planting materials, and vulnerability to accumulating and spreading pathogens and pests through seed. These crops are difficult to breed, so new varieties may be released slowly relative to new emerging threats. VPC seed systems are complex and face several challenges: poor-quality seed of existing varieties, low adoption rates of improved varieties, and slow varietal turnover, limiting yield increases and farmers’ ability to adapt to new threats and opportunities. Addressing these challenges requires first identifying key knowledge gaps on seed systems to guide research for development in a holistic and coherent way. Working together across 10 crops and 26 countries in Africa, Asia, and Central and South America, the CGIAR seed systems research community has developed a “Toolbox for Working with Root, Tuber, and Banana Seed Systems,” which introduces 11 tools and a glossary to address four major gaps: (1) capturing the demand characteristics of different types of farmers; (2) identifying effective seed delivery pathways; (3) ensuring seed health and stopping the spread of disease; and (4) designing effective policies and regulations. We describe the toolbox and its creation and validation across 76 crop-and-country use cases, and illustrate how the tools, applied individually or in combination, are addressing the key knowledge gaps in RT&B seed systems. The tool developers are actively working to scale the toolbox, including identifying new partners and models for collaboration, developing new tools, and supporting new applications in VPCs, as well as for fruit, vegetable, grain, and pulse seed systems.