Application And Research Of Image Processing Technology In Fabric Defect Detection

Now the computer technology has developed from research to application stage, more factors of production realize the automation under the support of high-tech, especially the development of the image processing technology provides key technical support for industrial production of detecting link. In the textile industry there will be found wrong knitting, leakage fabric, stains and other defects in the direct production of cloth, it needs to detect staff spend a lot of time and effort to examine the cloth, and testing experience, emotional, physical condition, this kind of low efficiency of manual test fabric and textile market in long-term in a state of disharmony, in urgent need of the industrialization of the inspection to replace. So automatic fabric inspection systems will become the main trend of the development of the textile industry. At the same time, the fabric defect detection methods become the hotspot.The key steps of defect detection are feature extraction, which are commonly used in the method of gray level statistics, texture structure, etc. Then the algorithm of defect detection system has been moving towards the more powerful development of the neural network technology. These systems can detect a variety of fabric elements, but for monochromatic fabrics, Image processing algorithm is still technical support. Through the research of the image processing technology, in-depth understanding of the methods of image processing, the threshold segmentation, noise filtering, image enhancement and other commonly used algorithms are given introduced and analysis in this paper, on the basis of this, this article expands the main content.Firstly, a defect detection method put foward for specific cloth. The method to image processing for theoretical basis, the integration of a variety of algorithms, and to the image enhancement, threshold segmentation and noise processing, defect orientation and development aspects of the innovation, the final completion of specific fabric image defect extraction. The fourth chapter of the paper will explain in detail the overall process of the method;Then Based on the OpenCV library function and image processing technology, a fabric defect detection system is designed and realized, for the verification of the proposed method. The software can cowboy and similar texture of cotton fabric image processing detected fabric image in the presence of defects and defect location, size and other information is presented to the users in the test report form.The fabric defect detection method has achieved certain results, but there are many places need to be improved. The design and development of the defect detection software also need to be further optimized to meet the actual industrial production, in the end of the paper, gives some algorithms and systems to improve the idea.