Steel Surface Defect Detection Algorithm based on Improved YOLOv8

Abstract

Aiming at the problem of steel surface defects, a defect detection algorithm based on YOLOv8 is constructed. Firstly, SimAM is added to the head to improve the expression ability of the features to enhance the detection ability of the model for tiny defects or small targets. Then the CIoU loss function is replaced with Wise-IoU to enhance the detection accuracy of the model. The experimental results show that the constructed improved models P, mAP0.5 and mAP0.5:0.95 reach 74.6%, 75.6% and 43.4%, respectively, which are improved by 5.9%, 0.5% and 0.6%, respectively, compared with the original YOLOv8n model. The detection accuracy was effectively improved.