Surface quality inspection of H-beams based on laser detection and multi-dimensional information fusion
MA Zhiqiang
HU Hong
GUO Xinwen
ZHAN Weijin
ZHENG Dongyan
YAN Zhiwen
Abstract：This research establishes a high-precision measurement and visualization online contour and surface quality detection system for H-beams using a multi-color light source cross 3D imaging system device,360°multi camera point cloud stitching technology,and real-time dynamic compensation technology.Propose a surface defect image processing method based on 2D+3D multi-dimensional information fusion,attaching 3D depth information to 2D grayscale images,solving the problem of traditional machine vision methods that can only perform qualitative detection but not depth detection,achieving"qualitative"and"quantitative"detection with a defect detection rate of 100％.By using small sample data augmentation methods and classification priority network methods,the problems of insufficient feature extraction information and insufficient data samples are solved to ensure the recognition rate of key defects and general defects in H-beams.Effectively solved the problems of appearance size and surface defect detection for various specifications of steel products,ensuring inspection accuracy while reducing the labor intensity of workers.
Keywords：3D imagingH-shaped steeldetection systemsurface defects
Publication Date：2025-07-20
Online Publishing Date：2025-09-08(First online date of this platform, not the publication date of the document)
Pages：7( 14-20 )