Enhanced Apple Quality Detection Algorithm Based on Improved YOLOv5 with Coordinate Attention Mechanism

Abstract

In recent years, the rapid advancement of artificial intelligence, particularly in the field of deep learning, has significantly improved the efficiency and accuracy of agricultural product quality inspection. This paper presents an improved apple quality detection algorithm based on the YOLOv5 model. The proposed approach incorporates a coordinate attention mechanism into YOLOv5, enhancing the model's feature extraction capabilities and improving its robustness across different scenarios. Experimental results demonstrate that the improved model achieves a recall rate of 88.8%, compared to the original 82.7%, and increases the mean average precision (mAP) from 83.7% to 85.9%. These enhancements not only improve the overall accuracy and recognition rate of apple quality detection but also contribute to reducing manual labor costs and enhancing automation in fruit quality inspection processes. This study provides an effective and efficient solution for intelligent apple quality assessment in modern agricultural applications.