BUILDING A DEEP LEARNING MODEL FOR PLANT LEAF DISEASE DETECTION

Abstract

Building a deep learning model for plant leaf disease detection. This study develops a plant disease detection system using the YOLOv7 deep learning model for real-time and accurate identification. Trained on a dataset of plant leaf images, the model achieved a mean average precision (mAP) of approximately 0.80 at IoU 0.50 and 0.65 at IoU 0.50-0.95, demonstrating its robustness and reliability. Data augmentation and hyperparameter optimization were applied to enhance performance across diverse image conditions. The results highlight YOLOv7's potential for effective disease detection in agriculture. Current work focuses deployment on website and mobile platforms, and exploring advanced models for improved accuracy and usability in practical scenarios.