article · IEEE Access
Early detection and treatment of plant diseases are critical to maintaining agricultural yields, yet conventional diagnostic methods remain difficult and time-consuming. To improve multi-class plant disease classification, pre-trained deep convolutional neural networks were assessed against an open dataset encompassing 52 categories of healthy leaves and various diseases. Several architectures, including Xception, InceptionResNetV2, InceptionV3, and ResNet50, were tested alongside an EfficientNetB3 network combined with adaptive augmented deep learning. Model performance was evaluated using measures such as accuracy, precision, recall, and F1 score across variations in batch size, dropout, and epoch counts. The EfficientNetB3-adaptive augmented deep learning approach achieved an accuracy of 98.71 percent, outperforming the other pre-trained networks and standard feature-based techniques. This demonstrates that pairing transfer learning with adaptive augmentation can deliver high-accuracy classification for agricultural diagnostics.
Unchecked plant diseases pose a substantial threat to agricultural productivity and food security. Because manual disease identification is slow and labour-intensive, high-performing automated classification models can support faster detection. Achieving reliable identification across dozens of crop conditions offers a route towards earlier treatment, protecting harvest yields and reducing farming losses.
The model indicates potential for integration into real-time disease diagnostic tools within agricultural systems, serving farmers, agronomists, or crop monitoring service providers. Based on the abstract, the research represents an applied and tested model evaluated on an open image dataset. Moving toward commercial software or field deployment would require transition from dataset benchmarks to real-world operational environments.
AI-generated from the published abstract. Always read the original work before citing.
Plant diseases can significantly impact agricultural productivity if not promptly identified and treated. Traditional plant disease classification methods are often challenging and time-consuming, making the identification of diseases a challenging task. This paper aims to bridge research gaps and address challenges in existing methodologies by proposing an efficient, effective multi-class plant disease classification approach. The research explores the application of pre-trained deep convolutional neural networks (CNNs) in this classification task, utilizing an open dataset comprising 52 categories of various diseases and healthy plant leaves. This study evaluated the performance of pre-trained deep CNN models, including Xception, InceptionResNetV2, InceptionV3, and ResNet50, paired with EfficientNetB3-adaptive augmented deep learning (AADL) for precise disease identification. Performance assessment was conducted using parameters such as batch size, dropout, and epoch counts, determining their accuracy, precision, recall, and F1 score. The EfficientNetB3-AADL model outperformed the other models and conventional feature-based methods, achieving a remarkable accuracy of 98.71%. This investigation highlights the potential of the EfficientNetB3-AADL model in offering accurate, real-time disease diagnostics in agricultural systems. The findings suggest that transfer learning and augmented deep learning techniques enhance the accuracy and performance of the model.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1109/access.2023.3303131
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.
