Research And Development Of An Machine-Vision-Based On-line Inspection System For Board Edge Size

During the productive process of furniture board,manual detection method is ofter taken to measure the size of the board after edge sealing,which makes it difficult to guarantee the efficiency of detection.According to the actual situation of projects,this paper designs an online detection system for edge-sealed furniture board,which is based on machine vision.Through the system,measurements and detection of furniture boards of different types and sizes can be real time and online.According to the technical index of plate size detection,the lighting scheme of the light source was determined,and the selection of key equipment such as industrial camera and lens was completed according to the actual situation,and the overall hardware scheme of the system was determined.A subpixel edge detection algorithm based on quadratic curve fitting is proposed.Median filter plate image is used for denoising to reduce the influence of noise on subsequent edge extraction.In order to improve the detection accuracy,Laplacian operator is used to find the pixel-level edge of the image,and on this basis,the method of quadratic curve fitting is used to find the sub-pixel edge of the plate image.This paper studies the camera advance calibration method of the system,and analyzes the characteristics of the camera imaging of the system,and puts forward the dimension superposition method based on the camera position invariance to calculate the size of the plate,which can ensure the accuracy,and reduce the complexity of the algorithm and the detection time of the system.The system software is designed and developed,and the dimensional measurement experiment is carried out based on the system.The experimental results show that the dimensional detection system has higher accuracy and detection speed,and can meet the requirements of dimensional detection accuracy and real-time performance in the industrial production process.