article · EURASIP Journal on Image and Video Processing
Plant diseases cause visible damage to leaves, manifesting as spots with distinct shapes, colours, and positions. Accurately diagnosing these conditions requires identifying these spatial features, an area where standard convolutional neural networks often face constraints regarding orientation and relative positioning. An optimised capsule neural network offers an effective alternative by better capturing spatial positioning within leaf imagery. Evaluated on standard dataset images covering ten tomato leaf diseases, the model incorporates data augmentation and image preprocessing to reduce overfitting during training. Trained using an Adam optimiser set to a learning rate of 0.00001, the architecture achieved a classification accuracy of 96.39 percent alongside low loss values. Compared with existing state-of-the-art methods, this approach successfully demonstrates the capability of capsule networks to reliably identify and classify foliar tomato conditions according to lesion characteristics.
Foliar diseases can severely reduce crop yields, making precise disease detection essential for agricultural management. Standard deep learning models often struggle to evaluate how lesion spots relate spatially across a leaf. Demonstrating that capsule networks can achieve high diagnostic accuracy across ten tomato diseases provides plant pathology researchers with a more reliable computer vision technique for automated leaf monitoring.
This technology enables automated visual diagnostic tools for identifying tomato crop diseases from leaf imagery. Agricultural software developers, crop monitoring services, and greenhouse operators could integrate such models into diagnostic platforms. Because the model was trained and evaluated exclusively on standard image datasets rather than deployed in operational field conditions, the work is at an applied research stage that requires further testing before real-world commercial deployment.
AI-generated from the published abstract. Always read the original work before citing.
Abstract Plant diseases have a significant impact on leaves, with each disease exhibiting specific spots characterized by unique colors and locations. Therefore, it is crucial to develop a method for detecting these diseases based on spot shape, color, and location within the leaves. While Convolutional Neural Networks (CNNs) have been widely used in deep learning applications, they suffer from limitations in capturing relative spatial and orientation relationships. This paper presents a computer vision methodology that utilizes an optimized capsule neural network (CapsNet) to detect and classify ten tomato leaf diseases using standard dataset images. To mitigate overfitting, data augmentation, and preprocessing techniques were employed during the training phase. CapsNet was chosen over CNNs due to its superior ability to capture spatial positioning within the image. The proposed CapsNet approach achieved an accuracy of 96.39% with minimal loss, relying on a 0.00001 Adam optimizer. By comparing the results with existing state-of-the-art approaches, the study demonstrates the effectiveness of CapsNet in accurately identifying and classifying tomato leaf diseases based on spot shape, color, and location. The findings highlight the potential of CapsNet as an alternative to CNNs for improving disease detection and classification in plant pathology research.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1186/s13640-023-00618-9
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
