Research And Design Of Twist Meter Based On Solios Board Image Processing

The quality of the fabric has always been the most concern in the fabric production process.In order to improve its overall quality,it will be from the yarn carding,parallel,textile and other production processes to detect,and yarn twist detection is one of the important guarantee of fabric quality.Yarn twist stability will affect a number of aspects of the performance of the yarn breaking strength,elongation at break,the yarn linear density will have a huge impact,but also affect the fabric feel,hygroscopicity and Breathability and other properties.However,From the current development point of view,most of the domestic textile enterprises in the detection or through artificial or semi-automatic mechanization to be detected.In this case caused by the detection rate is very slow,the face of a large number of work will be the case of errors,affecting its quality,and now the continuous development of social science and technology in the process,information technology has also been widely used,with machine vision instead of artificial visual become a hot topic at home and abroad.Therefore,this paper analyzes the current situation of yarn twist detection at home and abroad,and understands the working principle of yarn twist meter,and develops the detection machine based on machine vision twist.This paper mainly studies the twist of the yarn.The purpose of the study is to use the method of image processing to detect the twist,and to design the on-line twist meter to detect the yarn twist accurately and quickly.The system is divided into two parts: twist detection hardware platform construction and software detection system.The hardware part mainly includes the lens,the CCD camera,the selection of the acquisition card,as well as the design of the illumination system and the design of the camera synchronization signal;The software detection part mainly transmits the image data information collected by the camera to the computer,the computer carries on the algorithm processing to the stored data,and calculates and displays the twist value of the yarn.The software program uses the Matrox MIL10.0 library function,and the compiler chooses Visual Studio 2010.The yarn software system includes yarn twist detection algorithm research,,we mainly discuss the digital morphological method and the Fourier analysis method.Finally,the two methods are tested and the Fourier analysis method is determined.The method can filter the Fourier spectrum by Fourier transform and construct the appropriate low-pass filter,which can effectively remove the smaller defects and keep the image information better.The experiment proves that the method is not Need to twist,do not destroy the original yarn,directly through the image acquisition and processing to complete the twist detection,simple work,with high detection accuracy and low error rate.