纺织学报 ›› 2021, Vol. 42 ›› Issue (03): 82-88.doi: 10.13475/j.fzxb.20200700407

• 纺织工程 • 上一篇    下一篇

基于改进图像阈值分割算法的纱线疵点检测

李东洁1,2(), 郭帅1, 杨柳1   

  1. 1.哈尔滨理工大学 自动化学院, 黑龙江 哈尔滨 150080
    2.哈尔滨理工大学 切削加工及制造智能化技术教育部重点实验室, 黑龙江 哈尔滨 150080
  • 收稿日期:2020-07-01 修回日期:2020-10-21 出版日期:2021-03-15 发布日期:2021-03-17
  • 作者简介:李东洁(1981—),女,教授,博士。主要研究方向为微纳机器人技术及基于图像的缺陷检测。E-mail: dongjieli2013@163.com
  • 基金资助:
    国家自然科学基金面上项目(51975170);黑龙江省普通高校基本科研业务费专项资金项目(LGYC2018JQ017)

Yarn defect detection based on improved image threshold segmentation algorithm

LI Dongjie1,2(), GUO Shuai1, YANG Liu1   

  1. 1. College of Automation, Harbin University of Science and Technology, Harbin, Heilongjiang 150080, China
    2. Key Laboratory of Intelligent Technology for Cutting and Manufacturing, Ministry of Education,Harbin University of Science and Technology, Harbin, Heilongjiang 150080, China
  • Received:2020-07-01 Revised:2020-10-21 Online:2021-03-15 Published:2021-03-17

摘要:

针对纺织行业纱线疵点检测方法可靠性差、灵敏度低、检测速度低的问题,提出一种基于数字图像处理的纱线疵点判定方法。首先,搭建纱线图像采集系统,完成纱线图像采集;其次,针对纱线边缘信息难处理以及传统双边滤波对椒盐噪声处理效果差的问题,对双边滤波进行改进,改进后的双边滤波可有效保存纱线边缘信息;再者,针对传统阈值分割计算量大、最佳阈值难以寻找的问题,对传统阈值分割算法进行改进,改进的阈值分割算法在保证处理效果的同时提高了整体算法的处理速度;最后,采用亚像素对纱线边缘进行计算,提高了纱线疵点检测的精确度。实验结果验证了算法的有效性及可靠性,该算法在提高精确度的同时将检测速度提高了20%以上,对提高纱线质量检测的准确性具有重要意义。

关键词: 疵点检测, 纱线疵点, 阈值分割, 亚像素点判断, 图像处理

Abstract:

Aiming at the problems of poor reliability, low sensitivity and low speed of yarn defect detection in the textile industry, a new yarn defect detection method based on digital image processing was proposed. A yarn image acquisition system is built to obtain yarn image. In view of the difficulty of yarn edge information processing and the poor effect of traditional bilateral filtering on pepper-and-salt noise processing, the bilateral filtering was worked on for improvement, and the improved bilateral filtering was shown to be effective for preserving the yarn edge data. Furthermore, aiming at the problem of large amount of calculation and difficulty in finding the optimal threshold, the optimal threshold calculation method of traditional threshold segmentation algorithm is improved. The improved threshold segmentation algorithm not only ensures the processing effect, but also improves the processing speed of the whole algorithm. Sub-pixel is used to calculate the yarn edge and improve the accuracy of yarn defect detection. The experimental results verified the effectiveness and reliability of the algorithm, and increased the detection speed by more than 20% while improving the accuracy, which is of great significance for improving the accuracy of yarn quality detection.

Key words: defect detecting, yarn defect, threshold segmentation, sub-pixel point judgment, image processing

中图分类号: 

  • TH74

表1

CCD相机内部参数"

参数 数值 参数 数值
ax/mm 520.547 9 k1 -0.803 4
ay/mm 805.510 6 k2 2.613 0
u0/mm 370 k3 0.000 0
v0/mm 325 p1 -0.005 3
μ 0.00 p2 -0.001 5

图1

采集到的纱线图像"

图2

纱线灰度图像处理图"

图3

范围核逼近曲线"

图4

纱线滤波后图像"

图5

纱线二值图像处理图"

图6

纱线数学形态学处理图"

图7

亚像素边缘检测"

图8

纱线检测结果"

图9

纱线直径对比图"

表2

样本测试结果"

疵点
类型
传统方法 亚像素方法
平均直
径/mm
直径偏差
率/%
平均处理
时间/s
平均直
径/mm
直径偏差
率/%
平均处理
时间/s
正常 0.149 6 2.46 0.125 0.147 9 1.30 0.107
粗节 0.223 1 52.81 0.132 0.221 8 51.91 0.105
细节 0.102 5 29.79 0.122 0.104 6 28.35 0.102

图10

纱线控制界面"

[1] ZHANG B, TANG C. Study on detection method of low contrast fabric defects image based on regional variation saliency[C] //2019 IEEE 4th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC).Chengdu: Institute of Electrical and Electronics Engineers, 2019: 1103-1106.
[2] CHENG L, ZHAO L Y, CHEN L, et al. Digital image processing of cotton yarn seriplane[C] //2010 International Conference on Computer and Information Application.Tianjin: Institute of Electrical and Electronics Engineers, 2010: 274-277.
[3] CARVALHO V, CARDOSO P, BELSLEY M, et al. Development of a yarn evenness measurement and hairiness analysis system[C] // IECON 2006 - 32nd Annual Conference on IEEE Industrial Electronics. Paris: Institute of Electrical and Electronics Engineers, 2006: 3621-3626.
[4] 徐国香. UST-4条干仪的测试及应用[J]. 毛纺科技, 2007,35(9):54-57.
XU Guoxiang. Test and application of UST-4 dry instrument[J]. Wool Textile Journal, 2007,35(9):54-57.
[5] 盛国俊. 光电式纱线质量检测系统[D]. 北京: 清华大学, 2009: 2-4.
SHENG Guojun. Photo electric yarn quality detection system[D]. Beijing:Tsinghua University, 2009: 2-4.
[6] OZKAYA Y, ACAR M, JACKSON M. Digital image processing and illumination techniques for yarn characterization[J]. Journal of Electronic Imaging, 2005,14(2):1-13.
doi: 10.1117/1.1900134 pmid: 17235365
[7] GUHA A, AMARNATH C, PATERIA S, et al. Measurement of yarn hairiness by digital image processing[J]. Journal of The Textile Institute, 2010,101(3):214-222.
[8] 章国红, 辛斌杰, 张瑞. 基于图像分析技术的纱线黑板条干检测[J]. 棉纺织技术, 2016,44(12):19-24.
ZHANG Guohong, XIN Binjie, ZHANG Rui. Yarn strip detection based on image analysis technology[J]. Cotton Textile Technology, 2016,44(12):19-24.
[9] FABIJANSKA A, JACKOWSKA-STRUMILLO L. Image processing and analysis algorithms for yarn hairiness determination[J]. Machine Vision and Applications, 2012,23(3):527-540.
[10] VLADIMIR G, IONOV Evgen, NAING Lin Aung, Automatic detection and classification of weaving fabric defects based on digital image processing[C] //2019 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering. Saint Petersburg: Institute of Electrical and Electronics Engineers, 2019: 2218-2221.
[11] ZHANG Z Y. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000,22(11):1330-1334.
[12] LENG X, JI K, XING X. Hybrid bilateral filtering algorithm based on edge detection[J]. IET Image Processing, 2016,10(11):809-816.
[13] LI S Y, XU B G, FU H, et al. A two-scale attention model for intelligent evaluation of yarn surface qualities with computer vision[J]. Journal of The Textile Institute, 2018,109(6):798-812.
[14] LI S Y, FENG J, XU B J, et al. Integrated digital system for yarn surface quality evaluation using computer vision and artificial intelligence[J]. IPCV, 2012,1:472-476.
[15] LI Z J, PAN R R, GAO W D. Formation of digital yarn black board using sequence images[J]. Textile Research Journal, 2016,86(6):593-603.
[16] 迟开龙, 潘如如, 刘基宏, 等, 基于数字图像处理的纱线条干均匀度检测初探[J]. 纺织学报, 2012,33(12):19-24.
CHI Kailong, PAN Ruru, LIU Jihong, et al. Primary discussion on detection of yarn evenness based on digital image processing[J]. Journal of Textile Research, 2012,33(12):19-24.
[17] 李变变, 李忠健, 潘如如, 等. 纱线条干均匀性序列图像测量方法[J]. 纺织学报, 2016,37(11):26-31.
LI Bianbian, LI Zhongjian, PAN Ruru, et al. Measurement of yarn evenness using sequence images[J]. Journal of Textile Research, 2016,37(11):26-31.
[1] 唐千惠, 王蕾, 高卫东. 基于图像处理的织物保形性检测[J]. 纺织学报, 2021, 42(03): 89-94.
[2] 孟朔, 夏旭文, 潘如如, 周建, 王蕾, 高卫东. 基于卷积神经网络的机织物密度均匀性检测[J]. 纺织学报, 2021, 42(02): 101-106.
[3] 朱磊, 任梦凡, 潘杨, 李博涛. 基于相似性定位和超像素分割的织物疵点检测[J]. 纺织学报, 2020, 41(10): 58-66.
[4] 周文明, 周建, 潘如如. 应用上下文视觉显著性的色织物疵点检测[J]. 纺织学报, 2020, 41(08): 39-44.
[5] 杨恩君, 廖义辉, 刘安东, 俞立. 基于低秩分解的织物疵点检测[J]. 纺织学报, 2020, 41(05): 72-78.
[6] 王文胜, 李天剑, 冉宇辰, 卢影, 黄民. 筒子纱纱笼纱杆的定位检测方法[J]. 纺织学报, 2020, 41(03): 160-167.
[7] 张铮烨, 辛斌杰, 邓娜, 陈阳, 邢文宇. 基于边界跟踪测量麻纤维横截面参数的算法研究与应用[J]. 纺织学报, 2020, 41(02): 39-43.
[8] 张缓缓, 马金秀, 景军锋, 李鹏飞. 基于改进的加权中值滤波与K-means聚类的织物缺陷检测[J]. 纺织学报, 2019, 40(12): 50-56.
[9] 吴义伦, 李忠健, 潘如如, 高卫东, 张宁. 应用色纺纱图像的纬编针织物外观模拟[J]. 纺织学报, 2019, 40(06): 111-116.
[10] 黄嘉俊, 柯薇, 王静, 邓中民. 基于计算机视觉的牛仔服装色差检测评级系统[J]. 纺织学报, 2019, 40(05): 163-169.
[11] 王文帝, 辛斌杰, 邓娜, 李佳平, 刘宁娟. 单一视角下自适应阈值法的纱线毛羽识别及其应用[J]. 纺织学报, 2019, 40(05): 150-156.
[12] 蔡逸超, 周晓, 宋明峰, 牟新刚. 应用多尺度多方向模板卷积的筒子纱缺陷检测[J]. 纺织学报, 2019, 40(04): 152-157.
[13] 徐洋, 朱治潮, 盛晓伟, 余智祺, 孙以泽. 基于机器视觉的鞋面特征点自动识别改进方法[J]. 纺织学报, 2019, 40(03): 168-174.
[14] 杜帅, 李岳阳, 王孟涛, 罗海驰, 蒋高明. 基于改进局部自适应对比法的织物疵点检测[J]. 纺织学报, 2019, 40(02): 38-44.
[15] 巫莹柱 单颖法 黄伯熹 林广茂 梁家豪 张晓利. 聚对苯二甲酸丙二醇酯与聚对苯二甲酸丁二醇酯混纺纤维的智能识别[J]. 纺织学报, 2018, 39(09): 169-175.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!