Improving the YOLOv11 Model for Detecting Plant Diseases

In the agricultural sector, detecting plant diseases plays a crucial role in managing, controlling, and treating plant diseases, contributing to increased productivity and improved quality of farming products. With the swift progress in computer science and artificial intelligence technologies, numerous technological applications have been widely implemented in agriculture. This paper proposes a Parallel Decoupled Detection-You Only Look Once (PDD-YOLO) model is an enhancement of the YOLOv11m version, aimed at ensuring superior accuracy while maintaining the model’s complexity for the detection of various plant diseases across different crop types. The proposed approach involves improving the YOLOv11m model by replacing the Conv modules with the Conv_wavelet module. Experimental results indicate that the proposed model achieves higher accuracy and performs effectively, compared to the most advanced models, such as YOLOv11 and YOLOv10. The findings of this research suggest that the proposed model can be effectively deployed on mobile and embedded devices for real-time plant disease detection.