Machine Vision-Based Detection Method and Implementation of Defects on Light Plate Paint Surface
This paper proposes a machine vision-based method for detecting defects on the paint surface of a light plate,which addresses the low efficiency of manual detection in the production process.Corresponding detection processes were designed based on the characteristics of various typical defects,such as scratches,bubbles,dark spots,and color differences.This paper improves the contrast adjustment algorithm and proposes an adaptive grayscale transformation method with a Sigmoid function as the mapping curve to enhance point and line defects such as bubbles and scratches,and a local adaptive histogram equalization algorithm is used to highlight the color difference defect areas.Finally,with the combination of morphology processing,edge detection,and other algorithms,the detection of corresponding defects was achieved.The experimental results show that this method has a detection rate of 90％for various types of point,line,and color difference defects on the light plate.
万方数据
维普
