Abstract:Purpose In the field of steel defect detection, the existing detection algorithms struggle to achieve a satisfactory balance between detection accuracy, computational cost and inference speed due to the interference from complex background information, the variety of defect types and significant variations in defect morphology. To solve this problem, this paper aims to propose an efficient detector based on multi-scale information extraction (MSI-YOLO), which uses YOLOv8s as the baseline model. Design/methodology/approach First, the authors introduce an efficient multi-scale convolution with different-sized convolution kernels, which enables the feature extraction network to accommodate significant variations in defect morphology. Furthermore, the authors introduce the channel prior convolutional attention mechanism, which allows the network to focus on defect areas and ignore complex background interference. Considering the lightweight design and accuracy improvement, the authors introduce a more lightweight feature fusion network (Slim-neck) to improve the fusion effect of feature maps. Findings MSI-YOLO achieves 79.9% mean average precision on the public data set Northeastern University (NEU)-DET, with a model size of only 19.0 MB and an frames per second of 62.5. Compared with other state-of-the-art detectors, MSI-YOLO greatly improves the recognition accuracy and has significant advantages in computational cost and inference speed. Additionally, the strong generalization ability of MSI-YOLO is verified on the collected industrial site steel data set. Originality/value This paper proposes an efficient steel defect detector with high accuracy, low computational cost, excellent detection speed and strong generalization ability, which is more valuable for practical applications in resource-limited industrial production.

A light weight multi-scale feature fusion steel surface defect detection model based on YOLOv8

Efficient Detection Model of Steel Strip Surface Defects Based on YOLO-V7

An efficient steel defect detection model based on multi-scale information extraction

GDM-YOLO: A Model for Steel Surface Defect Detection Based on YOLOv8s

WFRE-YOLOv8s: A New Type of Defect Detector for Steel Surfaces

A high-speed YOLO detection model for steel surface defects with the channel residual convolution and fusion-distribution

SteelGuard-yolo: Steel Surface Defect Detection Network Based on Improved YOLOv5s

Steel Surface Defect Detection Method Based on Improved YOLOv8

CRGF-YOLO: An Optimized Multi-Scale Feature Fusion Model Based on YOLOv5 for Detection of Steel Surface Defects

LSD-YOLOv5: A Steel Strip Surface Defect Detection Algorithm Based on Lightweight Network and Enhanced Feature Fusion Mode

A Highly Efficient and Lightweight Detection Method for Steel Surface Defect

Real-Time Steel Surface Defect Detection with Improved Multi-Scale YOLO-v5

Development of a cross-scale weighted feature fusion network for hot-rolled steel surface defect detection

A deep learning model for steel surface defect detection

An Improved YOLOv5 Algorithm for Steel Surface Defect Detection

FEB-YOLOv8: A Steel Surface Defect Detection Algorithm Based on Improved YOLOv8s

Research on Steel Surface Defect Detection Algorithm Based on Improved YOLOv8n

Research on surface defect detection model of steel strip based on MFFA‐YOLOv5

Multiscale Local and Global Feature Fusion for the Detection of Steel Surface Defects

DF-YOLOv7: steel surface defect detection based on focal module and deformable convolution

A Novel Real-Time Steel Surface Defect Detection Method with Enhanced Feature Extraction and Adaptive Fusion