Research On High-speed Sewing Machine Needles Appearance Quality Detection System Based On Image Processing Technology

It has brought a lot of things for the convenience of our modern life by the science andtechnology. Many repetitive mechanical works have replaced by machines. It saves the cost ofproduction, what’s more, it also improves people’s working efficiency. In apparel, footwear, andother knitting industry, the sewing machine needle as the extensive use of key sewing parts, itsquality has a much impact on the sewing effects and product quality.Currently, the existing detection technology mainly relies on manual inspection in thesewing machine needle industry,but the manual inspection have many disadvantages,such aslow efficiency, high labor intensity, easy fatigue. The biggest drawback of it is unable to conductan all-side inspection, and it leads to unqualified products into the market.According to the actual needs of enterprises for the production of a sewing machine needle,we designed a reasonable, effective, high-speed sewing machines needle detection system. Wecombined with the research group in the accumulation of image processing technology, andkeep track of the latest research results of domestic and international field of image processing.The system solved the problem of the quality of the sewing machine needle, and greatlyincreased the rate of qualified products.Chapter1, we introduce the background of the paper, the purpose of the paper and thecurrent development of image processing technology, comes up with the designed plan of thehigh-speed detection system based on the image processing technology, and introduced in detail.Chapter2, the chapter elaborates on the selection of the camera, the lens, the light, interfacecards, and other hardware in the machine vision systems. And it also elaborates the constitutes ofthe system, the transmission process of the system signal, the main function as well as theprocess of image processing.Chapter3, the chapter illustrates various defects of the sewing machine needle, whilecompares various methods of image segmentation and threshold algorithm, and selects qualitytesting algorithm. We tell the story of the visual detection system design and implementation process by the HALCON and the example of the nonporous needle.Chapter4, we carry out numerous experiments and analysis by the detection system of themachine needle, test the function of each part, verify the results of the detection system. Theexperimental results show that the system can achieve the detection of the quality of the needle,deal with the unqualified products, achieve the desired objectives.