Novel CropdocNet Model for Automated Potato Late Blight Disease Detection from Unmanned Aerial Vehicle-Based Hyperspectral Imagery

The accurate and automated diagnosis of potato late blight disease, one of the most destructive potato diseases, is critical for precision agricultural control and management. Recent advances in remote sensing and deep learning offer the opportunity to address this challenge. This study proposes a novel end-to-end deep learning model (CropdocNet) for accurate and automated late blight disease diagnosis from UAV-based hyperspectral imagery. The proposed method considers the potential disease-specific reflectance radiation variance caused by the canopy’s structural diversity and introduces multiple capsule layers to model the part-to-whole relationship between spectral–spatial features and the target classes to represent the rotation invariance of the target classes in the feature space. We evaluate the proposed method with real UAV-based HSI data under controlled and natural field conditions. The effectiveness of the hierarchical features is quantitatively assessed and compared with the existing representative machine learning/deep learning methods on both testing and independent datasets. The experimental results show that the proposed model significantly improves accuracy when considering the hierarchical structure of spectral–spatial features, with average accuracies of 98.09% for the testing dataset and 95.75% for the independent dataset, respectively.

The Omics Hunt for Novel Molecular Markers of Resistance to Phytophthora infestans

Wild Solanum accessions are a treasured source of resistance against pathogens, including oomycete Phytophthora infestans, causing late blight disease. Here, Solanum pinnatisectum, Solanum tuberosum, and the somatic hybrid between these two lines were analyzed, representing resistant, susceptible, and moderately resistant genotypes, respectively. Proteome and metabolome analyses showed that the infection had the highest impact on leaves of the resistant plant and indicated, among others, an extensive remodeling of the leaf lipidome. The lipidome profiling confirmed an accumulation of glycerolipids, a depletion in the total pool of glycerophospholipids, and showed considerable differences between the lipidome composition of resistant and susceptible genotypes. The analysis of putative resistance markers pinpointed more than 100 molecules that positively correlated with resistance including phenolics and cysteamine, a compound with known antimicrobial activity. Putative resistance protein markers were targeted in an additional 12 genotypes with contrasting resistance to P. infestans. At least 27 proteins showed a negative correlation with the susceptibility including HSP70-2, endochitinase B, WPP domain-containing protein, and cyclase 3. In summary, these findings provide insights into molecular mechanisms of resistance against P. infestans and present novel targets for selective breeding.

SE-CapsNet: Automated evaluation of plant disease severity based on feature extraction through Squeeze and Excitation (SE) networks and Capsule networks

Diseases in plants harm the quantity of the overall food production as well as the quality of the yield. Early detection, diagnosis and treatment can greatly reduce losses, both economic and ecological. Intuitively, reduction in the use of agrochemicals due to timely detection of the disease, would greatly help in mitigating the environmental impact. In this paper, the authors have proposed an improved feature computation approach based on Squeeze and Excitation (SE) Networks, before processing by the original Capsule networks (CapsNet) for classification, for estimating the disease severity in plants. Two SE networks, one based on AlexNet and another on ResNet have been combined with Capsule networks. Leaf images for the devastating Late Blight disease occurring in the Tomato crop have been utilized from the PlantVillage dataset. The images, divided into four severity stages i.e. healthy, early, middle and end, are downscaled, enhanced and given as input to the SE networks. The feature maps generated from the two networks are separately given as input to the Capsule Network for classification and their performances are compared with the original CapsNet, on two image sizes 32X32 and 64X64. SE-Alex-CapsNet achieves the highest accuracy of 92.1% and SE-Res CapsNet achieves the highest accuracy of 93.75% with 64X64 image size, as compared to CapsNet that results in 85.53% accuracy. The classification accuracies of six state-of-the-art CNN models namely AlexNet, SqueezeNet, ResNet50, VGG16, VGG19 and Inception V3 are also presented for comparison purposes. Accuracy as well as precision, recall, F1-score, validation loss etc. measures have been recorded and compared. The findings have been validated by implementing the proposed approaches with another dataset, achieving similar resultant accuracy measures. The implementation was also accomplished with datasets after noise addition in six different variations, to verify the robustness of the proposed model. Based on the performances, the proposed techniques can be exploited for disease severity assessment in other crops as well and can be extended to other areas of applications such as plant species classification, weed identification etc. In addition to improved performance, with reduced image size, the proposed methodology can be utilized to create a mobile application requiring low processing capabilities, to be installed on reasonably priced smartphones for practical usage by farmers.

The Automation of Hyperspectral Training Library Construction: A Case Study for Wheat and Potato Crops

The potential of hyperspectral measurements for early disease detection has been investigated by many experts over the last 5 years. One of the difficulties is obtaining enough data for training and building a hyperspectral training library. When the goal is to detect disease at a previsible stage, before the pathogen has manifested either its first symptoms or in the area surrounding the existing symptoms, it is impossible to objectively delineate the regions of interest containing the previsible pathogen growth from the areas without the pathogen growth. To overcome this, we propose an image labelling and segmentation algorithm that is able to (a) more objectively label the visible symptoms for the construction of a training library and (b) extend this labelling to the pre-visible symptoms. This algorithm is used to create hyperspectral training libraries for late blight disease (Phytophthora infestans) in potatoes and two types of leaf rust (Puccinia triticina and Puccinia striiformis) in wheat. The model training accuracies were compared between the automatic labelling algorithm and the classic visual delineation of regions of interest using a logistic regression machine learning approach. The modelling accuracies of the automatically labelled datasets were higher than those of the manually labelled ones for both potatoes and wheat, at 98.80% for P. infestans in potato, 97.69% for P. striiformis in soft wheat, and 96.66% for P. triticina in durum wheat.

Potato late blight disease prediction using meteorological parameters in Northern Himalayas of India

Weather parameters play an important role in the spread of potato late blight of caused by Phytophthora infestans (Mont.) de Bary has historically been serious disease of potatoes through worldwide, including India. Due to spatial variation in prevailing weather conditions, its severity varies from region to region. Disease development process and the weather parameters are well understood and have been utilized for disease developing forecasting models and decision support system. Therefore, an experiment was conducted for two consecutive cropping seasons (2017 & 2018) to develop a forecasting model against late blight of potato using stepwise regression analysis for Northern Himalayas in India. Maximum and minimum temperature, relative humidity, rainfall and wind speed appeared to be most significant factors in the potato late blight disease development. The meteorological conditions conducive for the development of potato late blight disease were characterized. Maximum and minimum temperatures in the range of 15.0 – 28.0°C and 2.0 – 12.0°C were found favorable for potato blight disease. Similarly, relative humidity, rainfall and wind speed in the range of 85 - 95 per cent, 15.5 - 20.75 mm and 1.0 - 5.5 Km h-1, respectively, were conducive for potato late blight disease which are helpful in disease development.

Comparative Phenotypic and Agronomic Assessment of Transgenic Potato with 3R-Gene Stack with Complete Resistance to Late Blight Disease

Transgenic potato event Vic.172, expressing three naturally occurring resistance genes (R genes) conferring complete protection against late blight disease, was evaluated for resistance to late blight, phenotypic characterization, and agronomic performance in field conditions at three locations during three seasons in Uganda. These trials were conducted by comparison to the variety Victoria from which Vic.172 derives, using identical fungicide treatment, except when evaluating disease resistance. During all seasons, the transgenic event Vic.172 was confirmed to have complete resistance to late blight disease, whereas Victoria plants were completely dead by 60–80 days after planting. Tubers from Vic.172 were completely resistant to LB after artificial inoculation. The phenotypic characterization included observations of the characteristics and development of the stems, leaves, flowers, and tubers. Differences in phenotypic parameters between Vic.172 and Victoria were not statistically significant across locations and seasons. The agronomic performance observations covered sprouting, emergence, vigor, foliage growth, and yield. Differences in agronomic performance were not statistically significant except for marketable yield in one location under high productivity conditions. However, yield variation across locations and seasons was not statistically significant, but was influenced by the environment. Hence, the results of the comparative assessment of the phenotype and agronomic performance revealed that transgenic event Vic.172 did not present biologically significant differences in comparison to the variety Victoria it derives from.