A steel structure surface defect detection method based on scanning technology

By combining directional line scanning and brightness standardization with specular reflection suppression and local trajectory coordinate remapping, a method for detecting surface defects in steel structures was established, which solved the problems of missed detection and inaccurate benchmarks, and achieved efficient defect identification and classification.

CN122367976APending Publication Date: 2026-07-10天津亿鸣钢结构有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
天津亿鸣钢结构有限公司
Filing Date
2026-04-15
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing defect detection methods suffer from missed or false detections in the detection of surface defects in steel structures, and it is difficult to accurately establish a baseline profile, especially in the case of high-gloss areas and curved welds.

Method used

By combining directional line scanning with brightness normalization and specular reflection suppression, the center trajectory of the weld is extracted and coordinate remapping is performed to establish a weld straightening diagram. Multidimensional defect response quantities are calculated, a defect evidence field is constructed and continuously enhanced, and finally the defect type and severity are identified.

Benefits of technology

It effectively eliminates specular reflection interference, accurately straightens bent welds, improves the detection rate and classification accuracy of small, gradual and discontinuous defects under complex working conditions, and solves the problems of missed detection and inaccurate reference.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122367976A_ABST
    Figure CN122367976A_ABST
Patent Text Reader

Abstract

This invention discloses a method for detecting surface defects in steel structures based on scanning technology, belonging to the field of defect detection technology. The method includes: normalizing and weighting multidimensional defect response quantities to obtain a defect evidence sequence; constructing a defect evidence field based on the defect evidence sequence; continuously enhancing the defect evidence field along the scanning direction to obtain an enhanced defect response map; extracting defect regions, defect boundaries, and defect center trajectories from the enhanced defect response map; calculating the geometric parameters of each defect region; and identifying the type and severity of surface defects in the steel structure based on the geometric parameters of each defect region and the multidimensional defect response quantities to obtain the surface defect detection results. This invention improves the detection rate, classification accuracy, and anti-interference ability of small, gradual, and discontinuous surface defects in steel structures under complex working conditions by generating surface defect detection results.
Need to check novelty before this filing date? Find Prior Art