改进YOLOv5的路面坑洞检测算法OA

Detection and recognition of road potholes are crucial for autonomous driving and highway maintenance.In order to address the challenges of road pothole recognition under different lighting conditions and water accumulation,and to apply it in resource-constrained field environments,this article proposes an improved YOLOv5 algorithm for detecting road potholes.The C3 module was improved using PConv to achieve faster detection speed,and different perception modules using dynamic detection heads were used to adapt to different features of road potholes.At the same time,the loss function CIoU of the model is improved to SIoU,improving the accuracy and convergence ability of the model.The experimental results show that the improved YOLOv5 algorithm has an average accuracy increase of 5.19%in road pothole detection,and the model volume has also been reduced by 13%.