Rice leaf disease detection and classification using shallow trained YOLOv7 active deep learning approach

Rice is a staple food in more than a hundred nations world-wide (Oryza sativa). The cultivation of rice is vital to global economic growth. However, the main issue facing the agricultural industry is rice leaf disease detection and classification. The quality and quantity of the crops have declined due to the growth of diseases. As farmers in any country do not have enough knowledge about rice leaf disease, therefore, they are unable to diagnose it properly. Overall, it’s really a struggle for them to take proper care of rice leaves. As a result, the production is degrading day by day leading to the scarcity of food worldwide. With the continual advancement of object detection technology, Yolo family algorithms have extraordinarily high precision and better speed in various recognition tasks. In this work, 1,500 images of data sets are collected from the field and annotated for training purposes. After that, a shallow-trained YOLOv7 active deep learning approach is proposed for classifying and detecting rice leaf disease. The experimental results show improved object detection and classification results are found for the shallow-trained YOLOv7 network compared to the Yolov5 in terms of recall, mAP value, and F1 performance metrics.