Research Of Yarn Quality Detection Based On Computer Image Processing

The index of yarn filament irregularity is one of main indices to measure yarn quality. At present, there are mainly three methods to detect yarn irregularity: length measurement and weight measurement method, instruments measurement method and visual measurement method. These methods each have advantages and disadvantages. With the development of computer image processing technique, taking full advantage of digital image processing technique to overcome disadvantages of other detect methods and improving the level of yarn quality detection has become an important research field and has important significance to enhance yarn quality.This paper mainly studies appearance quality detection in yarn detection. The paper proposes a method that makes use of the computer image processing and analysis technique in quality detection of yarn blackboard and yarn filament irregularity analysis.In this method, yarn images are obtained by scanning blackboard which is twisted by the yarn. By designing a multi-structural element auto-adaptive determination weight's morphology filter to preprocess the image, an image with minimal distortion is got and target signal is completely detached from background signal, i.e. wiping off the yawp and standing out the target. Then maximum-square-error method is used to determine the threshold value automatically, and binarize the image according to the advised threshold. Afterward, the feature of the image is extracted and analyzed, i.e. extracting every target signal's feature values from post-preprocessing image and then calculating the discriminant index, such as yarn average diameter, CV value,thin and thick node, severe thin and thick node, ring,nep numbers and so on by extracted feature values. Here, the method based on abut yarn fragments area to detect and count the nep numbers is adopted. At last, the experiment results are analyzed and discussed. The experiments indicate that using computer image processing method to detect yarn quality has the advantage of veracity, quantified indices and has some applied value to realize yarn quality automatic detection.