纺织学报 ›› 2021, Vol. 42 ›› Issue (12): 196-204.doi: 10.13475/j.fzxb.20210204309

• 综合述评 • 上一篇    

棉花中异性纤维检测图像分割和边缘检测方法研究进展

任维佳1,2, 杜玉红1,2(), 左恒力1,2, 袁汝旺1,2   

  1. 1.天津工业大学 机械工程学院, 天津 300387
    2.天津工业大学 天津市现代机电装备技术重点实验室, 天津 300387
  • 收稿日期:2021-02-15 修回日期:2021-09-06 出版日期:2021-12-15 发布日期:2021-12-29
  • 通讯作者: 杜玉红
  • 作者简介:任维佳(1993—),男,博士生。主要研究方向为图像处理及模式识别、异性纤维检测。
  • 基金资助:
    国家自然科学基金青年基金项目(51205288);天津市自然科学基金项目(18JCYBJC20200)

Research progress in image segmentation and edge detection methods for alien fibers detection in cotton

REN Weijia1,2, DU Yuhong1,2(), ZUO Hengli1,2, YUAN Ruwang1,2   

  1. 1. School of Mechanical Engineering, Tiangong University, Tianjin 300387, China
    2. Tianjin Key Laboratory of Advanced Mechatronics Equipment Technology, Tianjin 300387, China
  • Received:2021-02-15 Revised:2021-09-06 Published:2021-12-15 Online:2021-12-29
  • Contact: DU Yuhong

RichHTML

30

PDF (PC)

157

摘要：

为进一步提高棉花中异性纤维的检测效率,对异性纤维图像处理方法进行探究。通过分析不同异性纤维图像边缘检测方法的定位精度、背景模糊以及受噪声影响情况发现,不同图像分割方法中异性纤维边缘连续性和分割效果存在差异性。统计了常见异性纤维图像边缘检测法和图像分割方法,分析了各类处理方法的优势和局限性,归纳了适用于各类异性纤维的检测方法,总结了现有研究中存在的问题和不足。研究认为:目前对不同种类异性纤维检测适用的图像处理方法不同,还无法同时检测出全部种类的异性纤维;应根据实践中具体异性纤维的种类、含量、物理特性等选择适合的检测算法并进行算法融合,开发普适性好的算法以降低成本和减少计算量。

关键词: 异性纤维, 图像预处理, 边缘检测, 图像分割, 在线检测

Abstract:

In order to further improve the detection efficiency for picking up alien fibers among cotton, image processing methods for detecting alien fibers were reviewed. This paper analyzed the inaccurate location, background blur and the influence of noise in edge detection methods, and studied the edge continuity and segmentation effect of different alien fibers in the image segmentation methods. The common edge detection methods and image segmentation methods for alien fibers among cotton were discussed, advantages and limitations of various processing methods were analyzed, and the detection methods applicable to various alien fibers were summarized, pointing out the existing problems and deficiencies in current practice. It is concluded that different image processing methods are currently applied to detect different types of alien fibers, and it is not possible to detect all types of alien fibers at the same time. The paper highlighted that the suitable detection algorithms should be selected and combined according to the specific types, contents, physical characteristics of alien fibers to develop an universal algorithm in order to reduce the cost and calculation burden.

Key words: alien fiber, image pre-processing, edge detection, image segmentation, online inspection

中图分类号: 

  • TS111.9

图1

异性纤维检测方法示意图"

图2

Kirsch算子卷积模板方向"

表1

各边缘检测算子的优缺点"

算子 优缺点比较
Roberts 对目标与背景差异性大且噪声较少的图像处理效果较好,但Roberts算子边缘定位效果不理想,提取边缘较粗
Prewitt 对渐变灰度值和噪声较多的异性纤维图像处理效果较好
Sobel 对渐变灰度值和噪声较多的图像处理效果较好,平滑效果较Prewitt算子好,Sobel算子边缘定位效果不理想
Kirsch 对渐变灰度值和噪声较多的异性纤维图像处理效果较好
Canny 该方法一直是经典边缘检测算法,对噪声抑制效果较好,定位精度较高,能够检测到真正的弱边缘,对细小异性纤维的检测效果较好
Laplacian 对图像中阶跃性边缘点定位准确,对噪声敏感,易丢失部分边缘的方向信息,边缘检测不连续
LOG LOG算子预先对图像进行高斯平滑处理,在一定程度上克服了噪声的影响,多用来判断边缘像素位于明区还是暗区

图3

图像分割效果对比"

图4

文献[39]检测效果"

[1] 杜玉红, 董超群, 任维佳, 等. 改进Faster RCNN模型在棉花异性纤维识别中的应用[J]. 激光与光电子学进展, 2020, 57(12):132-141.
DU Yuhong, DONG Chaoqun, REN Weijia, et al. Application of improved Faster RCNN model for foreign fiber recognition in cotton[J]. Advances in Lasers & Optoelectronics, 2020, 57(12):132-141.
[2] 马琳, 于永玲. SECUROMATSCFO异性纤维检测清除系统工作原理分析[J]. 棉纺织技术, 2004, 32(2):89-90.
MA Lin, YU Yongling. Analysis of working principle of SECUROMATSCFO foreign fiber detection and removal system[J]. Cotton Textile Technology, 2004, 32(2):89-90.
[3] 吕晓英, 蒋春玲. 浅谈棉花中异性纤维的危害及杜绝措施[J]. 中国纤检, 2013(Z1):66-67.
LÜ Xiaoying, JIANG Chunling. Talking about the harm of foreign fiber in cotton and its elimination mea-sures[J]. China Fiber Inspection, 2013(Z1):66-67.
[4] 董超群, 杜玉红, 任维佳, 等. 应用光学成像技术检测棉花中异性纤维的研究进展[J]. 纺织学报, 2020, 41(6):183-189.
DONG Chaoqun, DU Yuhong, REN Weijia, et al. Research progress in optical imaging technology for detecting foreign fibers in cotton[J]. Journal of Textile Research, 2020, 41(6):183-189.
[5] 师红宇, 管声启. 基于视觉注意计算模型的棉花异性纤维检测[J]. 东华大学学报(自然科学版), 2016, 42(3):400-405.
SHI Hongyu, GUAN Shengqi. Detection of cotton foreign fiber based on visual attention computation model[J]. Journal of Donghua University (Natural Science), 2016, 42(3):400-405.
[6] 何晓昀, 韦平, 张林, 等. 基于深度学习的籽棉中异性纤维检测方法[J]. 纺织学报, 2018, 39(6):131-135.
HE Xiaoyun, WEI Ping, ZHANG Lin, et al. Detection method of foreign fibers in seed cotton based on deep-learning[J]. Journal of Textile Research, 2018, 39(6):131-135.
[7] 冈萨雷斯. 数字图像处理[M]. 3版. 阮秋琦, 译. 北京: 电子工业出版社, 2011: 26-29.
RAFAEL C GONZALEZ. Digital image processing[M]. 3rd ed. RUAN Qiuqi, Translating. Beijing: Publishing House of Electronics Industry, 2011: 26-29.
[8] 王成军, 严晨. 机器视觉技术在分拣系统中的应用研究综述[J]. 制造技术与机床, 2020(5):32-37.
WANG Chengjun, YAN Chen. A review of the application of machine vision technology in sorting system[J]. Manufacturing Technology and Machine Tools, 2020(5):32-37.
[9] CHERRI A K, KARIM M A. Optical symbolic substitution: edge detection using prewitt, sobel, and roberts operators[J]. Applied Optics, 1989, 28(21):4644-4648.
doi: 10.1364/AO.28.004644
[10] 王小东, 赵仁宏. 基于Kirsch算子图像分割的FPGA设计与实现[J]. 影像科学与光化学, 2019, 37(4):332-335.
WANG Xiaodong, ZHAO Renhong. Design and implementation of FPGA based on Kirsch operator image segmentation[J]. Imaging Science and Photochemistry, 2019, 37(4):332-335.
[11] 王飞, 靳向煜. 基于边缘检测的原棉杂质图像识别方法适用性分析[J]. 现代纺织技术, 2019, 27(5):39-43.
WANG Fei, JIN Xiangyu. Applicability analysis of image recognition method for raw cotton impurity based on edge detection[J]. Advanced Textile Technology, 2019, 27(5):39-43.
[12] LU S, XIN B, DENG N, et al. Investigation of cross-sectional image analysis method to determine the blending ratio of polyester/cotton yarn[J]. Journal of Microscopy, 2020, 279(1):16-25.
doi: 10.1111/jmi.v279.1
[13] 王冬, 尹伯彪, 刘翔, 等. 棉花中白色异性纤维的线扫描激光成像检测方法[J]. 农业工程学报, 2015, 31(9):310-314.
WANG Dong, YIN Bobiao, LIU Xiang, et al. Linear scanning laser imaging method for detection of white foreign fibers in cotton[J]. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(9):310-314.
[14] CHEN G, LIU S, LIU J, et al. An accurate detection system of foreign fibers in cotton[J]. Sensor Letters, 2014, 12(3/4):838-842.
doi: 10.1166/sl.2014.3160
[15] CHEN X, CHEN L, WANG S, et al. Research on destriping based on rough set[J]. Journal of Computer and Communications, 2015, 3(11):8-12.
[16] WANG X, YANG W, LI Z. A fast image segmentation algorithm for detection of pseudo-foreign fibers in lint cotton[J]. Computers & Electrical Engineering, 2015, 46:500-510.
[17] 夏彬, 王泽武, 高慧. 棉纤维视觉检测中图像分割方法适用性分析[J]. 中国棉花加工, 2019(3):7-10.
XIA Bin, WANG Zewu, GAO Hui. Applicability analysis of image segmentation method in cotton fiber vision detection[J]. Cotton Processing in China, 2019(3):7-10.
[18] HUANG Y, CHEN G X, CHEN Q F. A novel approach of edge detection based on gray correlation degree and kirsch operator[J]. Applied Mechanics & Materials, 2015, 731:169-172.
[19] CHEN Z, XU W, LENG W, et al. A new high-speed foreign fiber detection system with machine vision[J]. Mathematical Problems in Engineering, 2010(SPECa):242-256.
[20] 向军, 张杰, 潘如如, 等. 采用纹理平滑的印花织物图案轮廓提取[J]. 纺织学报, 2017, 38(11):162-167.
XIANG Jun, ZHANG Jie, PAN Ruru, et al. The contour extraction of printed fabric patterns with smooth texture[J]. Journal of Textile Research, 2017, 38(11):162-167.
[21] 王欣. 基于机器视觉的棉花伪异性纤维识别研究[D]. 北京:中国农业大学, 2015: 27-31.
WANG Xin. Research on cotton pseudo foreign fiber recognition based on machine vision[D]. Beijing: China Agricultural University, 2015: 27-31.
[22] 陈铭明. 基于FPGA与DSP的棉花异纤检测系统研究[D]. 南京:南京理工大学, 2015: 40-45.
CHEN Mingming. Research on cotton foreign fiber detection system based on FPGA and DSP[D]. Nanjing: Nanjing University of Science and Technology, 2015: 40-45.
[23] ZHENG K Y, YAO Y. Automatic defect detection based on segmentation of pulsed thermographic images[J]. Chemometrics and Intelligent Laboratory Systems, 2017, 162:35-43.
doi: 10.1016/j.chemolab.2017.01.008
[24] 李晓慧, 陈智勇, 韩珑枝. LoG图像分割方法在棉纤维检验中的应用研究[J]. 中国纤检, 2018(1):93-95.
LI Xiaohui, CHEN Zhiyong, HAN Longzhi. Application of LoG image segmentation method in cotton fiber tes-ting[J]. China Fiber Inspection, 2018(1):93-95.
[25] 马永杰, 陈梦利. 基于改进拉普拉斯-高斯算子的阴影消除方法[J]. 激光与光电子学进展, 2020, 57(12):105-113.
MA Yongjie, CHEN Mengli. Shadow removal method based on improved laplace-gaussian operator[J]. Laser & Optoelectronics Progress, 2020, 57(12):105-113.
[26] 王秋萍, 张志祥, 朱旭芳. 图像分割方法综述[J]. 信息记录材料, 2019, 20(7):12-14.
WANG Qiuping, ZHANG Zhixiang, ZHU Xufang. Overview of image segmentation methods[J]. Information Recording Materials, 2019, 20(7):12-14.
[27] ZHANG Y, SMITH P W. Robust and efficient detection of non-lint material in cotton fiber samples[C]// YUPENG, SMITH P W. IEEE 6th International Conference on Computer Society. Orlando: IEEE, 2002: 51-56.
[28] PEKER K A. Contaminant and foreign fiber detection in cotton using gaussian mixture model[C]// OZSAN G. IEEE 8th International Conference on Application of Information & Communication Technologies (AICT). Astana: IEEE, 2014: 1-4.
[29] GUO Yingying, WANG Xinjie, ZHAI Yusheng, et al. A novel method for identification of cotton contaminants based on machine vision[J]. International Journal for Light and Electron Optics, 2014, 125(6):1707-1710.
doi: 10.1016/j.ijleo.2013.08.043
[30] WANG Xin, YANG Wenzhu, LI Zhenbo. A fast image segmentation algorithm for detection of pseudo-foreign fibers in lint cotton[J]. Computers & Electrical Engineering, 2015, 46:500-510.
[31] FANG Jingjing, JIANG Yu, YUE Jun, et al. A hybrid approach for efficient detection of plastic mulching films in cotton[J]. Mathematical and Computer Modelling, 2013, 58(3/4):834-841.
doi: 10.1016/j.mcm.2012.12.018
[32] 刘健庄, 栗文青. 灰度图象的二维Otsu自动阈值分割法[J]. 自动化学报, 1993(1):101-105.
LIU Jianzhuang, LI Wenqing. Two-dimensional Otsu automatic thresholding method for gray image segmentation[J]. Acta Automatica Sinica, 1993(1):101-105.
[33] 王昊鹏, 冯显英, 李丽. 基于改进二维最大类间方差法的白色异性纤维检测算法[J]. 农业工程学报, 2012, 28(8):214-219.
WANG Haopeng, FENG Xianying, LI Li. A white heterotropic fiber detection algorithm based on the improved two dimensional maximum inter-class variance method[J]. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(8):214-219.
[34] SAHDRA G S, KAILEY K S. Detection of contaminants in cotton by using ydbdr color space[J]. Computer Technology & Applications, 2012, 3(3):1118-1124.
[35] GULATI H, SINGH P. SWT approach for the detection of cotton contaminants[J]. International Journal of Computer Science Trends and Technology (IJCST), 2015, 3(2):224-229.
[36] KAUR Damandeep, KHULLAR Sunil. Principal component analysis for detecting foreign fiber contaminant in cotton with comparative analysis[J]. International Journal of Computer Science and Communication Engineering, 2014, 3(1):51-55.
[37] KAN D. An online image segmentation method for foreign fiber detection in lint[C]// LI D, YANG W. 6th Ifip International Conference on Computer and Computing Technologies in Agriculture. Berlin: Springer, 2010: 701-709.
[38] 王昊鹏, 冯显英, 王娜, 等. 基于改进的粒子群多阈值算法的白色异性纤维检测[J]. 农业工程学报, 2013, 29(23):153-158.
WANG Haopeng, FENG Xianying, WANG Na, et al. Detection of white foreign fibers based on improved particle swarm multiple threshold algorithm[J]. Transactions of the Chinese Society of Agricultural Engineering, 2013, 29(23):153-158.
[39] 师红宇, 管声启, 吴宁. 棉花中异性纤维的图像多分辨率差分检测方法[J]. 纺织学报, 2014, 35(5):13-18.
SHI Hongyu, GUAN Shengqi, WU Ning. Image multi-resolution differential detection method for heterogenous fiber in cotton[J]. Journal of Textile Research, 2014, 35(5):13-18.
[40] AGRAWAL P, OJHA G, BHATTACHARYA M. A generic algorithm for segmenting a specified region of interest based on chanvese's algorithm and active contours[M]// New Delhi: Springer India, 2016: 239-247.
[41] 李国辉, 苏真伟, 夏心怡. 基于不规则成像机器视觉的棉花白色异纤检测算法[J]. 农业机械学报, 2010, 41(5):164-167.
LI Guohui, SU Zhenwei, XIA Xinyi. A method of measuring cotton white foreign fiber based on machine vision of irregular imaging[J]. Transactions of the Chinese Society for Agricultural Machinery, 2010, 41(5):164-167.
[42] 张馨, 李道亮, 杨文柱, 等. 高分辨率棉花异性纤维彩色图像的快速分割方法[J]. 农业机械学报, 2011, 42(1):159-164,192.
ZHANG Xin, LI Daoliang, YANG Wenzhu, et al. A fast segmentation method for high resolution cotton foreign fiber color image[J]. Transactions of the Chinese Society for Agricultural Machinery, 2011, 42(1):159-164,192.
[43] WANG Guanjun, GAO Gaili, LI Daoliang, et al. Rational position of apparatus for removing cotton foreign fiber in dynamic equilibrium[J]. Applied Mechanics and Materials, 2012, 252:44-50.
doi: 10.4028/www.scientific.net/AMM.252
[44] 李丽. 皮棉中异性纤维检测算法的研究[D]. 济南:山东大学, 2013: 30-57.
LI Li. Research on foreign fiber detection algorithm in lint[D]. Jinan: Shandong University, 2013: 30-57.
[45] 刘加岭. 基于图像视觉的棉花异性纤维识别关键技术研究[D]. 济南:山东大学, 2020: 37-54.
LIU Jialing. Research on the key technology of cotton foreign fiber recognition based on image vision[D]. Jinan: Shandong University, 2020: 37-54.
[46] 汪峰, 许忠保, 王正龙, 等. 基于纹理特征的棉亚麻纤维识别技术[J]. 棉纺织技术, 2016, 44(4):1-5.
WANG Feng, XU Zhongbao, WANG Zhenglong, et al. Identification technology of cotton flax fiber based on textural feature[J]. Cotton Textile Technology, 2016, 44(4):1-5.
[1] 吕文涛, 林琪琪, 钟佳莹, 王成群, 徐伟强. 面向织物疵点检测的图像处理技术研究进展[J]. 纺织学报, 2021, 42(11): 197-206.
[2] 胡群, 张宁, 潘如如. 应用均值漂移的印花面料交互式换色方法[J]. 纺织学报, 2021, 42(11): 97-102.
[3] 庹武, 王哓玉, 高雅昆, 于媛媛, 郝潇潇, 刘永亮, 郭鑫. 基于改进边缘检测算法的服装款式识别[J]. 纺织学报, 2021, 42(10): 157-162.
[4] 徐晋, 杨鹏程, 肖渊, 胥光申. 织物表面导电线路喷射打印中微滴关键参数的视觉测量[J]. 纺织学报, 2021, 42(07): 137-143.
[5] 张文昌, 单忠德, 卢影. 基于机器视觉的纱笼纱杆快速定位方法[J]. 纺织学报, 2020, 41(12): 137-143.
[6] 裘柯槟, 陈维国, 周华, 应双双. 成像技术在纺织品颜色测量中的应用进展[J]. 纺织学报, 2020, 41(09): 155-161.
[7] 邵金鑫, 张宝昌, 曹继鹏. 基于图像处理与深度学习方法的棉纤维梳理过程纤维检测识别技术[J]. 纺织学报, 2020, 41(07): 40-46.
[8] 张建新, 李琦. 基于机器视觉的筒子纱密度在线检测系统[J]. 纺织学报, 2020, 41(06): 141-146.
[9] 董超群, 杜玉红, 任维佳, 赵地. 应用光学成像技术检测棉花中异性纤维的研究进展[J]. 纺织学报, 2020, 41(06): 183-189.
[10] 张缓缓, 赵妍, 景军锋, 李鹏飞. 基于亚像素边缘检测的纱线条干均匀度测量[J]. 纺织学报, 2020, 41(05): 45-49.
[11] 路浩, 陈原. 基于机器视觉的碳纤维预浸料表面缺陷检测方法[J]. 纺织学报, 2020, 41(04): 51-57.
[12] 高畅 刘基宏. 基于奇部Gabor滤波器的纱管余纱量检测[J]. 纺织学报, 2018, 39(10): 138-142.
[13] 常永和 薛元 韩晨晨 杨瑞华 高卫东 张毅. 环锭纺导纱板受力信号在线检测及纺纱张力分析[J]. 纺织学报, 2018, 39(09): 146-152.
[14] 何晓昀 韦平 张林 邓斌攸 潘云峰 苏真伟. 基于深度学习的籽棉中异性纤维检测方法[J]. 纺织学报, 2018, 39(06): 131-135.
[15] 徐启永 胡峰 王传桐 吴雨川. 改进频率调谐显著算法在疵点图像分割中的应用[J]. 纺织学报, 2018, 39(05): 125-131.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
[1] 曹建达;顾小军;殷联甫. 用BP神经网络预测棉织物的手感[J]. 纺织学报, 2003, 24(06): 35 -36 .
[2] 【作者单位】：中国纺织工程学会秘书处【分类号】：+【DOI】：cnki:ISSN:0-.0.00-0-0【正文快照】：　　香港桑麻基金会设立的“桑麻纺织科技奖” 0 0 年提名推荐工作;在纺织方面院士;专家和有关单位的大力支持下;收到了 个单位 (人 )推荐的 位候选人的. 2003年桑麻纺织科技奖获奖名单[J]. 纺织学报, 2003, 24(06): 107 .
[3] 【分类号】：Z【DOI】：cnki:ISSN:0-.0.00-0-0【正文快照】：　　一;纺　纱模糊控制纺纱张力的研究周光茜等 ( - )………………原棉含杂与除杂效果评价方法的研究于永玲 ( - )……网络长丝纱免浆免捻功能的结构表征方法李栋高等 ( - )……………. 2003年纺织学报第二十四卷总目次[J]. 纺织学报, 2003, 24(06): 109 -620 .
[4] 黄立新. Optim纤维及产品的开发与应用[J]. 纺织学报, 2004, 25(02): 101 -102 .
[5] 张治国;尹红;陈志荣. 纤维前处理用精练助剂研究进展[J]. 纺织学报, 2004, 25(02): 105 -107 .
[6] 秦元春. 纺织工业发展方向初探[J]. 纺织学报, 2004, 25(02): 108 -110 .
[7] 高伟江;魏文斌. 纺织业发展的战略取向——从比较优势到竞争优势[J]. 纺织学报, 2004, 25(02): 111 -113 .
[8] 姚玉元;陈文兴;张利;潘勇. 催化氧化型消臭蚕丝纤维的研究[J]. 纺织学报, 2004, 25(03): 7 -8 .
[9] 林红;陈宇岳;任煜;仲志锋;王红卫. 经等离子体处理的蚕丝纤维结构与性能[J]. 纺织学报, 2004, 25(03): 9 -10 .
[10] 潘旭伟;顾新建;韩永生;程耀东. 面向协同的服装供应链快速反应机制研究[J]. 纺织学报, 2006, 27(1): 54 -57 .
京ICP备14006648号　 京公网安备11010502044800号
版权所有 © 2021 《纺织学报》编辑部
地址： 北京市朝阳区延静里中街3号主楼6层《纺织学报》杂志社　邮政编码：100025　
电话：010-65017711,65917740　传真：010-65016539　Email:fangzhixuebao@vip.126.com
本系统由北京玛格泰克科技发展有限公司设计开发