纺织学报 ›› 2021, Vol. 42 ›› Issue (06): 91-96.doi: 10.13475/j.fzxb.20201105806

• 纺织工程 • 上一篇    下一篇

基于改进帧间差分法的经纱撞筘拥纱在线检测

夏旭文, 孟朔, 潘如如(), 高卫东   

  1. 生态纺织教育部重点实验室(江南大学), 江苏 无锡 214122
  • 收稿日期:2020-11-30 修回日期:2021-03-14 出版日期:2021-06-15 发布日期:2021-06-28
  • 通讯作者: 潘如如
  • 作者简介:夏旭文(1995—),男,硕士生。主要研究方向为纺织图像处理。

On-line detection of warp collision and reed embedding based on improved inter-frame difference method

XIA Xuwen, MENG Shuo, PAN Ruru(), GAO Weidong   

  1. Key Laboratory of Eco-Textiles (Jiangnan University), Ministry of Education, Wuxi, Jiangsu 214122, China
  • Received:2020-11-30 Revised:2021-03-14 Published:2021-06-15 Online:2021-06-28
  • Contact: PAN Ruru

RichHTML

6

PDF (PC)

50

摘要：

为实时检测浆纱过程中经纱撞筘现象,开发了基于机器视觉方法的浆纱机经纱撞筘拥纱实时自动检测系统。通过自主搭建的图像采集系统捕捉经纱通过筘齿完整图像。首先裁剪初始图像获取检测区域,然后通过高斯滤波进行图像平滑以降低图像细节层次;然后扩大帧间差,加入Canny边缘算法和膨胀的数学形态学运算来实现对帧间差分法的改进,最后通过改进帧间差分法实现浆纱机经纱撞筘拥纱的识别。实验结果表明:所开发的检测系统可准确地识别浆纱机经纱撞筘拥纱和对拥纱进行跟踪,并且达到实时检测的作用,可以为实际生产提供指导。

关键词: 撞筘检测, 图像处理, 帧差法, 高斯滤波, 形态学运算

Abstract:

In order to monitor the warp bumping phenomenon in the warp sizing process, a real-time automatic detection system for warp reed collision of the sizing machine based on machine vision was developed. The complete image of warp yarn passing through the reed was captured through the developed image acquisition system. The initial image was cropped to obtain the detection area, and before the image through Gaussian filtering was smoothed to reduce the level of image details. Then, the inter-frame difference was enlarged, and the Canny edge algorithm and expanded mathematical morphology operation were added to complete the improvement of the inter-frame difference method. Recognition and tracking of the warp yarn collision and reed holding of the sizing machine were achieved by the improved inter-frame difference method. The experimental results show that the developed detection system can accurately identify the warp in the sizing machine and track the reed-holding yarn, fulfilling the real-time detection function, which can guide the actual production.

Key words: reed detection, image processing, frame difference method, gaussian filter, morphological operations

中图分类号: 

  • TS111.8

图1

图像采集系统"

图2

采集的部分经纱图像"

图3

裁剪后图像"

图4

图像平滑效果对比"

图5

传统帧间差分法示意图"

图6

改进帧间差分法示意图"

表1

不同帧间差和车速下撞筘检测准确率"

车速/(m·min-1) 帧间差t/帧 准确率/%
20 14.40
30 16.52
40 21.34
10 50 24.51
60 28.89
80 37.34
100 40.97
20 30.01
30 40.87
40 53.18
30 50 59.08
60 64.51
80 75.43
100 79.17
20 45.04
30 60.72
40 68.37
50 50 77.07
60 82.17
80 88.68
100 89.75
20 46.63
30 69.08
40 70.54
70 50 84.02
60 87.07
80 90.89
100 91.34

图7

不同帧差下差分图像和二值化图像"

图8

不同阈值δ下平均误检率及响应时间 "

图9

不同车速下系统平均响应时间"

[1] 朱苏康, 高卫东. 机织学[M]. 北京: 中国纺织出版社, 2015:75.
ZHU Sukang, GAO Weidong. Weaving science[M]. Beijing: China Textile & Apparel Press, 2015:75.
[2] 陈长虎, 程西楼, 黄彦萍. 细号高密集聚纺织物浆纱工艺探讨[J]. 棉纺织技术, 2015, 43(1):61-64.
CHEN Changhu, CHENG Xilou, HUANG Yanping. Discussion on the sizing process of fine-number and high-density poly-textile[J]. Cotton Textile Technology, 2015, 43(1):61-64.
[3] 常永和, 薛元, 韩晨晨, 等. 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(9):146-152.
CHANG Yonghe, XUE Yuan, HAN Chenchen, et al. On-line detection of force signal of ring spinning guide plate and analysis of spinning tension[J]. Journal of Textile Research, 2018, 39(9):146-152.
[4] CHAN J P, PALMER G S. Machine vision-applications in industry[C]// IEEE Colloquium on Application of Machine Vision. London: IET, 1995:1-6.
[5] 王振亚, 曾黄麟. 一种基于帧间差分和光流技术结合的运动车辆检测和跟踪新算法[J]. 计算机应用与软件, 2012, 29(5):117-120.
WANG Zhenya, ZENG Huanglin. A new algorithm for moving vehicle detection and tracking based on the combination of inter-frame difference and optical flow technology[J]. Computer Applications and Software, 2012, 29(5):117-120.
[6] 苏佳, 高丽慧. 采用帧差法和相关滤波改进的TLD跟踪算法[J]. 计算机工程与设计, 2020, 41(6):1694-1700.
SU Jia, GAO Lihui. Improved TLD tracking algorithm using frame difference method and related filtering[J]. Computer Engineering and Design, 2020, 41(6):1694-1700.
[7] DAI Kexue. Prospects and current studies on back-ground subtraction techniques for moving objects detection from surveillance video[J]. Journal of Image and Graphics, 2006, 11(7):917-919.
[8] SANDFORD A, SARKER T, BERNIER T. Effects of geometric distortions, gaussian blur, and contrast negation on recognition of familiar faces[J]. Visual Cognition, 2017:1-16.
[9] 高海壮, 段先华. 基于帧差法和混合高斯的海上运动目标检测[J]. 计算机与数字工程, 2019, 47(5):1140-1144.
GAO Haizhuang, DUAN Xianhua. Detection of sea moving targets based on frame difference method and Gaussian mixture[J]. Computer and Digital Engineering, 2019, 47(5):1140-1144.
[10] 孙晓杰, 梁义. 基于双阈值的图像边缘实时检测系统[J]. 仪表技术与传感器, 2019(8):83-86.
SUN Xiaojie, LIANG Yi. Real-time image edge detection system based on dual thresholds[J]. Instrumentation Technology and Sensors, 2019(8):83-86.
[11] 李航. 统计学习方法[M]. 北京: 清华大学出版社, 2012:18-19.
LI Hang. Statistical learning methods[M]. Beijing: Tsinghua University Press, 2012:18-19.
[1] 江燕婷, 严庆帅, 辛斌杰, 高琮, 施楣梧. 纺织品单向导水性能测试方法分析[J]. 纺织学报, 2021, 42(05): 51-58.
[2] 李东洁, 郭帅, 杨柳. 基于改进图像阈值分割算法的纱线疵点检测[J]. 纺织学报, 2021, 42(03): 82-88.
[3] 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.
[4] 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.
[5] 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.
[6] 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.
[7] 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.
[8] 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.
[9] 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.
[10] 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.
[11] 陆奕辰 王蕾 唐千惠 潘如如 高卫东. 应用图像处理的纱线黑板毛羽量检测与评价[J]. 纺织学报, 2018, 39(08): 144-149.
[12] 王雯雯 高畅 刘基宏. 应用卷积神经网络的细纱断纱锭位识别[J]. 纺织学报, 2018, 39(06): 136-141.
[13] 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.
[14] 王雯雯 刘基宏. 应用优化霍夫变换的细纱断头检测[J]. 纺织学报, 2018, 39(04): 36-41.
[15] 王传桐 胡峰 徐启永 吴雨川 余联庆. 改进频率调谐显著算法在疵点辨识中的应用[J]. 纺织学报, 2018, 39(03): 154-160.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
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