Browsing by Subject "Early detection"

Now showing 1 - 2 of 2

Classifying early-stage soybean fungal diseases on hyperspectral images using convolutional neural networks
(2025) Hsiao, Chieh Fu; Feyrer, Georg; Stein, Anthony
Using convolutional neural networks (CNNs) to detect plant diseases has proven to reach high accuracy in the classification of infected and non-infected plant images. However, most of the existing researches are based on RGB images due to the availability and the comparably low cost of image collection. The limited spectral information restricts the detectability of plant diseases, especially in the early stage where often symptoms of pathogen infection have not yet become visible. To this end, in this study, hyperspectral imaging (HSI) data are combined with deep learning models to test the classification ability of two soybean fungal diseases: Asian soybean rust (Phakopsora pachyhizi) and soybean stem rust (Sclerotinia scleroriorum). Different CNNs employing 2D, 3D convolution, and hybrid approaches are compared. The influences of the depth of the convolutional layer and the regularization techniques are also discussed. Besides, image augmentation methods are investigated to overcome the problem of data scarcity. The results indicate the 6-convolutional-layer depth hybrid model to have the best capacity in classifying Asian soybean rust in the early-mid to mid-late stage when there are over 2 % visible symptoms but a limited detectability in the early stages when there are below 2 % visible symptoms on leaves. On the other hand, the optimized CNN model shows a limited capability to detect both diseases when there are no visible symptoms observable. Overall, this study suggests a hybrid 2D-3D convolutional model with augmentation and regularization methods has a high potential in the early detection of fungal diseases. This research is expected to contribute to a new cropping system that vastly reduces the chemical-synthesis plant protection products, where a continuous pathogen disease monitoring plays a key to manage the crop stands.
Multi-crop early detection of spider mite damage using hyperspectral data and XGBoost
(2026) Mandrapa, Boris; Spohrer, Klaus; Wuttke, Dominik; Ruttensperger, Ute; Dieckhoff, Christine; Müller, Joachim
The two-spotted spider mite is a globally significant pest affecting over 150 crop species, including cucumbers and strawberries. Its feeding activity leads to chlorophyll degradation and physiological changes in leaf tissue, which alter spectral reflectance properties and enable image-based detection. In this study, hyperspectral imaging (HSI) under controlled conditions was used to classify healthy and spider mite-infested leaves of cucumber and strawberry plants, including asymptomatic infested leaves. Spectral data were analyzed and classified with three supervised machine learning algorithms built on extreme gradient boosting (XGBoost) models. The study had three objectives: (1) to assess the ability of XGBoost to classify multiple infestation states, (2) to evaluate model performance with a reduced set of effective wavelengths, and (3) to determine whether infestation across both crops can be classified using a single, merged model. Using all wavelengths, results showed that classification accuracy was 93 % for cucumber leaves, 84 % for strawberry leaves, and 87 % when combined. With five most effective wavelengths, classification accuracy reached 70 % for cucumber leaves, 65 % for strawberry leaves, and 65 % for cucumber and strawberry leaves combined. The most effective wavelengths were consistently selected from the red-edge and near-infrared (NIR) spectral regions, which highlights their importance for early detection. To the best of our knowledge, this is the first known study to successfully apply a combined machine learning model for early spider mite detection across two different crop species using hyperspectral data under controlled conditions. The results show the potential of machine learning for multi-crop pest detection and could lay the groundwork for practical, sensor-based tools in precision agriculture.