The Detection System Of Yarn Based On Digital Image Processing Technology

With the development of modern technology and people's living level, people call for increasingly high quality of the dry goods on daily life. As a textile giant, China plays an irreplaceable role in textile industry. However, Chinese soft goods are mostly ecumenical and inferior because of the lack of innovating consciousness. They possess great disparity compare to up-market fabrics from foreign countries. During the processing of spinning and weaving, the tensility of yarn is an important parameter. It directly relates to the quality of products, production efficiency and smooth subsequent production process. Therefore, people pay more attention to test and control of the yarn's tensility.The yarn between the two pulleys has a certain degree of sag because of its own gravity. This issue by detecting the level of sag to detect yarn tension. As a result of the quality of general yarn is very light, the level of sag is very small. From the middle of the yarn to exert a downward direction force. This force performing a "magnifying glass" role. It makes the sag of yarn significantly. Solved the problem that detecting the level of sag to detect yarn tension is heavily dependent on the weight of its own. With the help of the force, the yarn between the two pulleys will become approximately linear. The anger between the two lines, the line of force and the yarn tension have corresponding relation.Soft aspects, the paper introduces the algorithm based on Hough Transform line detection to detect anger between the straight line of yarn and the horizontal line. By detection the angle between the straight line of yarn and the horizontal line to calculate the tension of yarn. We shift the image domain to parameter space useing Hough. Hough transform has a strong anti-interference ability. Therefore, this detection algorithm have high accuracy on detecting yarn tension. About the issue of Detection disconnect of yarn, this paper offers two solutionsIt is divided into two situations. And in each case provides a possible algorithm. In terms of hardware, a digital image processing system is constructed base on DSP. After CCD collecting the image of the yarn, we using the video decoder chip to translate the analog image to digital signal. DSP deals with the digital signal which is sent through the port VPO. We can restore the analog image from the digital signal which has been processed by the DSP using the incoder.The experimental results achieved the expected goal, which proves the design scheme is practicable.