Application And Research Of Automatic Detection System Of Tire Cord Quality In Vision Technology

In recent years, with the development of the automotive industry in China, people have paid more and more attention to the quality and safety on cars, especially the tire. Tire carcass is made of rubber cord with several layers, if the rubber surface covers defects like the ply drum kits, bare surface, bubbles and some others, they will directly affect the tire quality badly, even a serious threat to personal safety and loss in economic. Therefore, testing for the quality of the tire cord is a very important part to the tire production.However, there is nearly no research on the testing tire cord quality in domestic until now. And it's expensive and complicated to operate for the foreign relevant products. Based on these factors, domestic manufacturers still use manual inspection on the tire cord. The traditional method of manual inspection has some problems, such as:employees work is intensive, test results are easily affected by subjective factors, there is high rate of missed and false detection, etc. On this basis, and by analysising the basic methods of automatic detection to fabric defect and combining with the actual cord structure, we presents the quality inspection system based on machine vision technology. Use machine vision replacing the artificial vision, we improve the inspection speed, reduce the labor costs, and improve the reliability of the inspection.This paper mainly studies the part of the system's software, including the image preprocessing, thresholding, feature extraction and defects recognition. In this paper,we propose to use median filter to remove the texture background; then use threshold to divide the defect of the cord, the advantages of this approach are that we wipe out the interfere from the cord texture on defect, meanwhile, we improve the processing speed for the post-treatment. There is significantly difference between the normal texture image and image with defect, by comparing the feature value we can recognize the defect from the image. We extracted feature of the cord image by using gray level co-occurrence matrix. According to the related principle, we select some effective and independent parameters from those various GLCM features as the basis for the defect identification. We use the Euclidean distance on the defect recognition of the Cord image. Finally, we give the outputs by combining of MATALAB with VB.We proposed the detection system of tire cord quality in vision technology in this paper which has high processing speed, and success to discriminate the defect.