一种复杂图像背景中残留液晶识别方法

黄妍妍; 高宝红; 张海涛; 康军广

doi:10.3788/YJYXS20142906.1024

您当前的位置：

首页 >

文章列表页 >

一种复杂图像背景中残留液晶识别方法

成像技术与图像处理 | 更新时间：2020-08-12

- 一种复杂图像背景中残留液晶识别方法
- Identification method of rudimental liquid crystal in complex background image
- 液晶与显示 2014年29卷第6期页码：1024-1031
- 作者机构：
  
  1. 河北工业大学微电子研究所天津,300130
  2. 河北工业大学网络管理中心天津,300130
- 作者简介：
- 基金信息：
  
  02国家重大专项项目（No.2009zx02308）();河北省高等学校科学技术研究项目（No.Z2011293）();河北省自然科学基金项目（No.E20142021
- DOI：10.3788/YJYXS20142906.1024
  中图分类号： TP141.9;TP751.1
- 收稿日期：2014-02-27，
  
  修回日期：2014-04-11，
  
  网络出版日期：2014-05-06，
  
  纸质出版日期：2014-12-05
- 稿件说明：
移动端阅览
黄妍妍, 高宝红, 张海涛, 康军广. 一种复杂图像背景中残留液晶识别方法[J]. 液晶与显示, 2014,29(6): 1024-1031

HUANG Yan-yan, GAO Bao-hong, ZHANG Hai-tao, KANG Jun-guang. Identification method of rudimental liquid crystal in complex background image[J]. , 2014,29(6): 1024-1031
黄妍妍, 高宝红, 张海涛, 康军广. 一种复杂图像背景中残留液晶识别方法[J]. 液晶与显示, 2014,29(6): 1024-1031 DOI： 10.3788/YJYXS20142906.1024.

HUANG Yan-yan, GAO Bao-hong, ZHANG Hai-tao, KANG Jun-guang. Identification method of rudimental liquid crystal in complex background image[J]. , 2014,29(6): 1024-1031 DOI： 10.3788/YJYXS20142906.1024.

摘要

LCD液晶屏是对表面洁净度有严格要求的部件，基于数字图像处理的残留液晶识别方法能够定量、客观地评价表面的洁净度。针对复杂背景图像中残留液晶难以准确识别的问题，本文提出并实现了有效的图像处理流程和算法，包括分割、边缘连接、图像填充、去除多余边界以及噪声等操作。提出了一种边缘连接算子，在对图像进行分割后，将不连续的图像边界连接为闭合，为图像填充奠定基础。提出并实现了一个基于二叉分类方法的模式分类器，用于对污染物几何形状进行自动识别。实验表明，针对单一背景残留液晶的识别与分析方法能够精确获得残留液晶边缘，并且控制了噪声的干扰；针对复杂背景残留液晶识别与分析方法能够对背景存在明显差异的特定类型图像进行准确的处理。

Abstract

LCD screen requires strict surface cleanliness. The inspection method based on digital image processing focusing on the residual liquid crystal can give quantitative and objective assessment on the components surface cleanliness. According to the difficulty to identify the rudimental liquid accurately in a complex background image

this paper proposes an effective image processing flow and algorithm

including the operation of segmentation

edge connection

image filling

redundant boundary and noise removing

etc. An edge connecting operator is designed to connect the discontinuous boundary in the image after the segmentation. A classifier is designed and implemented based on binary classification method for automatic identification of geometry types of contaminants. Experiments show that the residual liquid crystal in the simple background image can accurately inspected

and the interference of the noise is excluded. The identification and analysis method for the complex background image can accurately process particular types of images that their backgrounds have significant differences.

关键词

Keywords

references

李广福,刘玉岭,李薇薇. LCD残留液晶的清洗[J]. 清洗世界, 2006,22(9):24-28. Li G F, Liu Y L, Li W W. Removal of rudimental LC on LCD[J].Cleaning World,2006,22(9):24-28.(in Chinese)[2] Huang Y Y, Gao N, Yu M, et al.Detection of surface cleanliness based on image recognition[C].ⅡTA '08. Proceedings of the 2008 Second International Symposium on Intelligent Information Technology Application,Washington, DC, USA: IEEE Computer Society,2008: 360-365.[3] 于洪春,邓意成,郑喜凤. 面积约束下的最优阈值法分割LED像素点阵[J]. 液晶与显示, 2012,27(5):653-657. Yu H C,Deng Y C, Zheng X F. Segmentation of LED pixel matrix by using optimal threshold method with area constraint[J].Chinese Journal of Liquid Crystals and Displays,2012,27(5):653-657.(in Chinese)[4] 丘文涛,赵建,刘杰. 结合区域分割的SIFT图像匹配方法[J]. 液晶与显示, 2012,27(6):827-831. Qiu W T, Zhao J, Liu J. Image matching algorithm combining sift with region segmentation[J].Chinese Journal of Liquid Crystals and Displays,2012,27(6):827-831.(in Chinese)[5] 靳永亮, 王延杰, 刘艳滢, 等. 红外弱小目标的分割预检测[J]. 光学精密工程, 2012, 20(1): 171-178. Jin Y L, Wang Y J, Liu Y Y, et al.Pre-detection method for small infrared target[J].Optics and Precision Engineering,2012, 2012, 20(1): 171-178.(in Chinese)[6] 吴笑天, 鲁剑锋, 贺柏根,等. 雾天降质图像的快速复原[J]. 中国光学, 2013, 6(6): 892-899. Wu X T, Lu J F, He B G, et al.Fast restoration of haze-degraded image[J].Chinese Optics,2013, 6(6) :892-899.(in Chinese)[7] Wang X F, Zhang X Y, Gao R N. Research of image edge detection based on mathematical morphology[J].International Journal of Signal Processing, Image Processing and Pattern Recognition,2013,6(5):227-236.[8] 黄妍妍,姜乐涛,于明,等. LCD表面液晶污染的非接触检测新算法[J]. 液晶与显示, 2009,24(1): 135-138. Huang Y Y, Jiang Y T, Yu M, et al.New method of non-contact inspection for liquid crystal residues on lcds surface[J].Chinese Journal of Liquid Crystals and Displays,2009, 24(1): 135-138.(in Chinese)[9] Rafael C G.Digital Image Processing[M].3rd ed. Pearson, 2013.[10] Zhao H,Chen Z X. A simple hole filling algorithm for binary cell images[J].Applied Mechanics and Materials,2013, 433-435:1715-1719.[11] Spasojevic S S,Susic M Z, Durovic Z M. Recognition and classification of geometric shapes using neural networks[C].2012 11th Symposium on Neural Network Applications in Electrical Engineering(NEUREL),2012:71-76.[12] Gueheneuc Y G,Albin-Amiot H. A pragmatic study of binary class relationship[C].18th IEEE International Conference on Automated Software Engineering,Montreal, Canada: ASE 2003, 2003:277-280.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

TFT-LCD表面Mura缺陷的AOI检测研究进展

基于OpenCV的双波长光纤导引光在高精度装调上的应用

基于FPGA的分段近似双边滤波算法设计与实现

基于改进YOLOv7的小目标焊点缺陷检测算法

健身行为的人体姿态估计及动作识别