欢迎访问作物学报，今天是
作物学报

作物学报 ›› 2011, Vol. 37 ›› Issue (07): 1274-1279.doi: 10.3724/SP.J.1006.2011.01274

• 耕作栽培·生理生化 • 上一篇    下一篇

基于YCbCr颜色空间和Fisher判别分析的棉花图像分割研究

刘金帅,赖惠成*,贾振红   

  1. 新疆大学信息科学与工程学院, 新疆乌鲁木齐 830046
  • 收稿日期:2010-11-02 修回日期:2011-03-28 出版日期:2011-07-12 网络出版日期:2011-04-12
  • 通讯作者: 赖惠成, E-mail: lai@xju.edu.cn, Tel:13999152197
  • 基金资助:

    本研究由科技部国际合作项目(2009DFA128707)项目资助。

Image Segmentation of Cotton Based on YCbCcr Color Space and Fisher Discrimination Analysis

LIU Jin-Shuai,LAI Hui-Cheng,JIA Zhen-Hong   

  1. College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
  • Received:2010-11-02 Revised:2011-03-28 Published:2011-07-12 Published online:2011-04-12
  • Contact: 赖惠成, E-mail: lai@xju.edu.cn, Tel:13999152197

PDF

1704

被引次数

22

摘要： 棉花的分割是采棉机器人研究的关键技术。本文分别在HSV、HIS和YCbCr颜色空间下，首先根据棉花的颜色信息与背景颜色信息的差距，对样本图像中的各个对象(棉絮、棉枝、土壤等)分类; 其次根据分类结果分别提取各类在各颜色空间下的样本像素值; 再根据类间离散度最大和类内离散度最小的准则计算出Fisher判别向量和各类的质心; 最后按照像素值离各质心最近的准则进行图像分割。结果表明, 在YCbCr颜色空间下产生的分割噪声最小，选取此颜色空间，采用贴标签的方法自适应去噪。实验仿真表明，本方法可有效避免阳光直射和阴影的干扰，对各种情况都能准确分割，分割准确率达90.44%。

关键词: 棉花分割, Fisher线性判别分析, YCbCr颜色空间, 贴标签去噪

Abstract: For cotton harvesting robot, the cotton image segmentation is one of the key technologies. In this paper, under HSV, HIS, and YCbCr color spaces respectively, according to the difference between cotton color and background color, the various objects(cotton batting, cotton branches, soil etc.)in the sample images were classified, and then the pixel value of every category in different samples was extracted based on the classification result. In the following, the rule that the dispersion is biggest between different classes and smallest within the same class was used to calculate the Fisher discrimination vector and the center of mass in every class. Finally, image segmentation was carried out based on the criterion of pixel value close to the center of mass. The result showed that the least segmentation noise was obtained in the YCbCr color space, in which the method of labeling for self-adapting denoising was need. The simulation showed that the cotton could be separated exactly from the background by the above algorithm whether the cotton was exposed to the sunlight or the shadow. A total of that 136 cotton images were segmented with an accuracy of 90.44% in YCbCr color space.

Key words: Cotton segmentation, Fisher linear discrimination analysis, YCbCr color space, Labeling denoising

[1]Wang L(王玲), Ji C-Y(姬长英). Technical analysis and expectation for cotton harvesting based on agricultural robot. Cotton Sci (棉花学报), 2006, 18(2): 124–128 (in Chinese with English abstract)
[2]Zhang Y-J(章毓晋). Image Enginerring(II) Image Analysis
[图像工程(中册)图像分析], 2nd edn. Beijing: Tsinghua University Press, 2005. pp 73–105(in Chinese)
[3]Wang Y(王勇), Shen M-X(沈明霞), Ji C-Y(姬长英). Model and analysis of ripe cotton color for different parts in picking period. Trans CSAE (农业工程程学报), 2007, 23(4): 183–185 (in Chinese with English abstract)
[4]Wang Y(王勇), Shen M-X(沈明霞), Ji C-Y(姬长英). Using color data and shape properties for cotton fruit recognition. Transact Chin Soc Agric Machinery (农业机械学报), 2007, 38(11): 77–79, 87 (in Chinese with English abstract)
[5]Wang L(王玲), Wang P(王萍), Chen B-L(陈兵林), Liu S-J(刘善军), Ji C-Y(姬长英). Image segmentation technique of field cotton based on color threshold. Acta Agron Sin (作物学报), 2010, 36(3): 502–507 (in Chinese with English abstract)
[6]Wei J-D(韦皆顶), Fei S-M(费树岷), Wang M-L(汪木兰), Yuan J-N(袁建宁). Research on the segmentation strategy of the cotton images on the natural condition based upon the HSV color-space model. Cotton Sci (棉花学报), 2008, 20(1): 34–38 (in Chinese with English abstract)
[7]Chen G(陈果). The image’s segmentation by threshold based on Fisher function.Chin J Sci Instrum (仪器仪表学报), 2003, 24(6): 564–567 (in Chinese)
[8]Fisher R A. The use of multiple measurements in taxonomic problems. Annals Eugenics, 1936, 7: 179–188
[9]Duchene J, Leclercq S. An optimal transformation for discriminant and principal component analysis. IEEE Trans Pattern Analysis Machine Intelligence, 1988, 10(6): 978–983
No related articles found!
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
京ICP备15008789号-2