Research On Technologies Of Online Surface Quality Inspection For Strip Steel Based On Machine Vision

In recent years, along with the rapid development of digital image gathering and processing technology, online surface inspection systems for strip steel have been widely used. Surface quality inspection technology based on machine vision is nondestructive, intelligent, precision and fast, which makes it a trend in defect detecting technology. Based on machine vision technology, the study focuses on the core strip surface defect detection algorithms and key software technology, main works and achievements of the thesis are as follows:1. With a view to the characteristics of surface defects for strip steel and the demand of system performance, an overall technology scheme, especially on Linear CCD and FPGA+DSP architecture designed on the system layer is presented. Taking the characteristics of surface defects into account, optimal configuration of receiving mode containing bright-field and dark-field is analyzed to make defects effectively detected. The algorithm flow of image-processing applied to the surface defect of strip steel is fully discussed, feasible image processing algorithm and feature extraction algorithm is presented at each processing step in the paper.2. Present a new image denoising method based on the WBCT which is an improvement on wavelet. Experiment results clearly demonstrated that algorithm can get higher PSNR and the better visual.3. A wavelet based maximum modulus algorithm is proposed for edge protection, which can coordinate well the edge detection precision and denoising effect to extract the surface defect edges on multi-scale. Experimental results showed that the new algorithm can exclude well the fault detect edge from detection with the particulars of defect edge kept up, thus providing better edge detectivity for the subsequent online processing in surface defect detection.4. Research on time domain feature and frequency domain feature extraction method of strip steel surface defect. In order to solve the problem about shape feature extraction which is a high difficulty technology, a new shape extraction method based on wavelets packet was presented and the experiment results clearly demonstrated the capability of the proposed scheme can effectively deal with extracted shape feature.