<![CDATA[Chili plants are a strategic agricultural commodity with high economic value, yet their production is often disrupted by disease attacks causing significant yield reduction. Early and accurate detection of chili leaf diseases is crucial for implementing precision agriculture practices. This research implements the YOLO11n (You Only Look Once version 11 nano) model for automated chili leaf disease detection using the "Chili Plant Leaf Disease and Growth Stage Dataset from Bangladesh" containing 1,856 high-resolution images across six categories: Bacterial Spot, Cercospora Leaf Spot, Curl Virus, Healthy Leaf, Nutrition Deficiency, and White Spot. The model was trained for 100 epochs on Google Colab with Tesla T4 GPU using 640×640 pixel input resolution. Evaluation results demonstrate excellent detection performance with precision of 83.7%, recall of 84.3%, mAP@0.5 of 92.3%, and mAP@0.5:0.95 of 74.5%. Per-class analysis reveals that Nutrition Deficiency achieved the highest performance (mAP@0.5 = 99.2%), while Curl Virus presented the greatest detection challenge (recall = 55.6%). The lightweight YOLO11n architecture with only 2.58 million parameters and 6.3 GFLOPs, making it highly suitable for deployment on edge devices such as agricultural drones, mobile applications, and IoT monitoring systems. This research contributes to smart agriculture applications by providing an efficient and accurate solution for automated chili leaf disease detection under real field conditions.]]>
Copyrights © 2026
