纺织工程学报2024，Vol.2Issue(4)：61-68,8.
基于图像处理的纱线条干均匀度检测
Yarn strand evenness detection based on image processing
摘要
Abstract
To address the issues of uneven lighting and tilt affecting the results in yarn strand evenness detection,a method for detecting yarn strand evenness based on homomorphic filtering and Hough transform is explored.Firstly,the homomorphic filtering method is used to separate the image into illumination and reflectance components to suppress the illumination component and highlight the reflectance component,thereby eliminating uneven lighting and emphasizing the strand part.Secondly,the tilt of yarn in the image is detected using the Hough transform to rotate and correct the tilt angle to a horizontal position.Finally,the main part of the yarn in noise-free signal images is extracted through image segmentation to calculate the average diameter of the collected yarn.Experimental results show that the deviation of the CV value of the yarn strand obtained through image processing compared to the results from existing strand meters is less than 4%,validating the accuracy of the image processing method and providing reference for the future development of a complete visual detection system.关键词
光照不均/纱线倾斜/条干均匀度/直径提取/CV值Key words
uneven lighting/yarn tilt/strand evenness/diameter extraction/CV value分类
轻工纺织引用本文复制引用
王畅巍,柯薇,刘威,邓文韬..基于图像处理的纱线条干均匀度检测[J].纺织工程学报,2024,2(4):61-68,8.基金项目
湖北省技术创新专项2019AAA005. （）
