Recent Papers

Plant diseases substantially decrease crop yield and quality and are a source of significant threat to global food security. Conventional manual diagnosis is usually time-consuming, subjective, and expert-based and thus inappropriate for large-scale agriculture. Recent developments in artificial intelligence (AI) and computer vision have made it feasible to identify plant diseases automatically using image-based analysis. Deep learning models are used in this research to categorize leaf images as healthy or diseased. The PlantVillage dataset is used, and preprocessing involves resizing, normalization, and data augmentation to enhance model robustness. Different deep learning architectures like Convolutional Neural Networks (CNNs), ResNet, EfficientNet, MobileNet, and Vision Transformers are compared. The models have high classification accuracy between 9598%. Results show that transfer learning and light-weight architectures are especially well-suited for real-time deployment, allowing for deployment on mobile and edge devices. This work demonstrates the promise of deep learning for precision agriculture, providing an efficient and scalable solution for early disease detection and enhanced crop management.Future research can attempt larger, more heterogeneous data sets and deep learning hybrids for robust cross-crop disease detection.

Copyright © 2025 Anushree S. This is an open access article distributed under the Creative Commons Attribution License.