A Hybrid Computer Vision enabled GAN Model for Pest and Disease Detection

Agriculture is among the most prominent occupation in India and among different parts of the globe, and plant diseases account for 35% of crop yield loss each year. Plants are affected by pests, environmental conditions, and nutritional deficits. Identifying and treating issues early is crucial for accomplishing desired yields. In this paper, an effective way for early detection of bugs in plants which combines Generative Adversarial Networks(GANs) with computer vision techniques is suggested. The proposed model utilizes advanced data augmentation, pre-processing, feature extraction, and classification to improve the precision of bug assessment in plants. Extensive set of simulations were performed to highlight the improved performance of the suggested approach. The experimental outcome has demonstrated promising results of the proposed model over the recent modern appraoches with maximum 95.2% accuracy, 92.6 % f-score, 93.8 % precision and 91.2% recall.