基于机器视觉的纱线外观质量检测-纺织工程专业论文.docxVIP

II II Abstract New measurement of yarn appearance quality based on machine vision New measurement of yarn appearance quality based on machine vision Abstract With the development of textile technology, fabric develops in the direction of refinement, and the requirement of yarn increases with it. This requires a high precision testing method, yarn appearance quality detection system based on machine vision, can not only have high detection accuracy, but also can directly detect yarn appearance size and obtain the scientific detection results. In this paper, hardware and software of the yarn appearance quality detection system based on machine vision system are introduced in detail. The hardware Configuration contain a camera (S2-1y-05H40), a telecentric lens (MGTL60C), a LED light sourse (OPT-LS83-W), a image capture card (Xcelera-CL LX1 Base), Camera Link and a PC with core i7 , 3.40 GHZ frequency and 4GB Physical memory; HALCON software is used to realizmorphological opening operatione image acquisition, image processing, and the corresponding calculation in this new measurement system. Threshold segmentation and morphological processing are used to obtain the main body of yarn and complete hairiness in this paper. Threshold is used to realize the separation of yarn and background, morphological opening operation can eliminate the protrusion and the loop fiber surrounding the main body of raw silk in the binary image, morphological closing operation can fill the small holes and connect the hairiness almost broke away. By calculating the change in diameter of yarn, the coefficient of variation (CV value) of diameter is adopted to characterize the yarn evenness, and the number of different levels of thick place, thin place and nep is adopted to characterize the yarn common defects. Two kinds of yarns (same yarn count but different quality grade and same quality grade but different yarn count) are measured to obtain results, which are compared with the results of USTE