Enhancing Chili Disease Detection: A Hybrid Deep Learning Approach with YOLOv5 Algorithm

To minimize crop losses and maintain sustainable agricultural practices, identifying chili disease is imperative. Using the YOLOv5 algorithm, this work suggests a unique hybrid deep learning method to improve the identification of chili diseases. The methodology integrates advanced image preprocessing techniques with the YOLOv5 framework to achieve high accuracy in identifying various chili plant diseases. The proposed system influence a comprehensive dataset comprising labelled images of healthy and diseased chili plants, ensuring robust training and validation. Two significant developments are the use of transfer learning to maximize model performance and the integration of data augmentation procedures to overcome class imbalance. The experimental findings show that our hybrid technique achieves outstanding accuracy, recall, and F1-score metrics, greatly outperforming conventional approaches. The results show that the suggested approach can be a useful tool for early disease diagnosis, allowing prompt treatment and encouraging improved crop management techniques. This research highlights the potential of deep learning in revolutionizing agricultural disease diagnostics, offering a scalable and efficient solution for farmers and agricultural stakeholders.