Surface defect small target detection method based on multi-scale feature interaction

By introducing SMK module, SELA-S attention mechanism and Fafusion module into the YOLOv11 network, combined with the Focaler-PIoUv2 loss function, the problem of missed detection and missed detection in traditional steel defect detection methods is solved, and high-precision and high-efficiency steel surface defect detection is achieved.

CN120374613AActive Publication Date: 2025-07-25湖南工商大学
5 Cites 8 Cited by

Patent Information

Application Number
CN202510855348.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-25
Publication Date
2025-07-25
Estimated Expiration
2045-06-25

AI Technical Summary

Technical Problem

Traditional steel defect detection methods are inefficient, susceptible to human factors and poor environmental adaptability, making it difficult to meet the needs of high precision and high efficiency, especially in the identification of complex or subtle defects, which are prone to missed or missed inspections.

Method used

The multi-scale feature interactive surface defect detection method based on YOLOv11 network is adopted, and the multi-scale feature extraction SMK module, SELA-S attention mechanism and feature fusion Fafusion module are introduced, and the Focaler-PIoUv2 bounding box loss function is combined to perform model training and detection.

Benefits of technology

It improves the accuracy and generalization ability of steel surface defect detection, can more accurately identify defect areas in complex backgrounds, reduces calculation complexity, and enhances the adaptability and detection efficiency of the model.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a surface defect small target detection method based on multi-scale feature interaction. Comprising the steps that a surface image, collected in real time, of a to-be-detected product is input into a detection model, and after the detection model processes the surface image, an image marked with possible defect classification categories, bounding box coordinates and defect target existence confidence is output; wherein the detection model is a model taking a YOLOv11 network as a basic structure and comprises a backbone network, a neck network and a detection head which are sequentially connected in series; a multi-scale feature extraction SMK module and an attention mechanism SELA-S module are introduced into the backbone network; two feature fusion Fault modules are introduced into the neck network. According to the method, the multi-scale target features under the complex background can be effectively extracted and fused, and the method has good processing performance for distinguishing defects in the picture from the complex background and multi-scale target detection. And in a surface defect detection task, the method can be accurately focused on a defect area.
Need to check novelty before this filing date? Find Prior Art