Faster Metallic Surface Defect Detection Using Deep Learning with Channel Shuffling

TL;DR

This paper introduces an improved real-time defect detection model for steel surfaces using YOLOv5 with channel shuffling and feature fusion, enhancing accuracy and speed for small and complex defect targets.

Contribution

The study proposes a novel defect detection model that integrates channel shuffling and feature fusion into YOLOv5 for better detection of small surface defects.

Findings

Achieved 77.5% mAP on NEU-DET dataset

Outperformed other models in accuracy and detection time

Enhanced detection of small and complex defects

Abstract

Deep learning has been constantly improving in recent years and a significant number of researchers have devoted themselves to the research of defect detection algorithms. Detection and recognition of small and complex targets is still a problem that needs to be solved. The authors of this research would like to present an improved defect detection model for detecting small and complex defect targets in steel surfaces. During steel strip production mechanical forces and environmental factors cause surface defects of the steel strip. Therefore the detection of such defects is key to the production of high-quality products. Moreover surface defects of the steel strip cause great economic losses to the high-tech industry. So far few studies have explored methods of identifying the defects and most of the currently available algorithms are not sufficiently effective. Therefore this study presents an improved real-time metallic surface defect detection model based on You Only Look Once (YOLOv5) specially designed for small networks. For the smaller features of the target the conventional part is replaced with a depth-wise convolution and channel shuffle mechanism. Then assigning weights to Feature Pyramid Networks (FPN) output features and fusing them increases feature propagation and the networks characterization ability. The experimental results reveal that the improved proposed model outperforms other comparable models in terms of accuracy and detection time. The precision of the proposed model achieved by @mAP is 77.5% on the Northeastern University Dataset NEU-DET and 70.18% on the GC10-DET datasets