Detection of Corn Leaf Diseases Using Image Processing and YOLOv7 Convolutional Neural Network (CNN)

Given the specific challenges confronting the agricultural sector in the Philippines, this thesis introduces an approach in corn leaf disease detection utilizing a YOLOv7-based CNN model. Concentrating on diseases like Common Rust, Northern Corn Leaf Blight, and Gray Leaf Spot, this study focuses on developing an optimized YOLOv7 model for precise disease detection in corn leaf plants. The optimized model demonstrates remarkable training results, achieving a precision of 82%, recall of 85%, and an mAP50 of 89%. In the subsequent testing phase, the model maintains its high performance, displaying an overall precision of 87%, a recall of 85%, and an mAP50 of 88%. This research is of paramount significance in the context of the Philippines, where agricultural productivity is indispensable for the nation's economy and food security. Addressing the critical need for accurate disease detection, this study has the potential to revolutionize local farming practices, ensuring crop health and sustainability in the region. Furthermore, the insights gained from this research contribute significantly to the broader field of agricultural technology, marking it as a pivotal endeavor with profound and far-reaching implications.