基于YOLOv5s模型的端口块表面缺陷识别研究：优化与未来展望

In this paper, we utilized a model based on YOLOv5s to obtain both the location and category information of surface defects on port blocks. However, the process of optimizing the model was labor-intensive, involving data annotation and iterative training that consumed a significant amount of time. Moving forward, we will continue exploring more efficient and effective algorithms for target recognition.