An Intelligent Optimization Based Yolov5 Framework to Detect the Rice Leaf Disease

Rice leaf disease detection and classification is critical and plays a significant role. The present work focuses classification procedure in rice leaf disease. Traditional categorization systems, on the other hand, have a high proportion of false rates. In this paper, a new hybrid strategy is proposed to classify the disease present in the rice leaves. An optimized Ant Lion based YOLOv-5 framework was developed that can creates images of the disease affected image and without affecting images. Moreover, the proposed work provides vital information for treating rice leaf diseases and causes. Generally, the techniques are used for detecting the disease. The proposed method is a hybrid approach, which includes the corresponding key steps such as enhancement, segmentation, extraction of features and training images with YOLOv5 options. Moreover, MATLAB was used for implementation and simulation outcomes are compared with existing techniques. From the comparison, the proposed work has attained better classification accuracy as 99.9%.