吉林大学学报(信息科学版) ›› 2019, Vol. 37 ›› Issue (2): 148-154.

• • 上一篇    下一篇

Lab 空间的改进k-means 算法彩色图像分割

霍凤财1a,1b,孙雪婷1a,任伟建1a,1b,杨迪2,于涛3   

  1. 1. 东北石油大学a. 电气信息工程学院; b. 黑龙江省网络与智能控制重点实验室,黑龙江大庆163318;
    2. 大庆油田有限责任公司第二采油厂规划设计研究所,黑龙江大庆163461;
    3. 大庆油田有限责任公司第四采油厂规划设计研究所,黑龙江大庆163511
  • 收稿日期:2018-11-08 出版日期:2019-03-25 发布日期:2019-06-10
  • 作者简介:霍凤财( 1976— ) ,男,黑龙江安达人,东北石油大学副教授,博士,硕士生导师,主要从事智能算法和图像处理等研究,(Tel) 86-13936895698( E-mail) huofc@126. com。
  • 基金资助:
    国家自然科学基金资助项目( 61374127; 51404073) ; 国家自然科学基金优秀青年科学基金资助项目( 61422301) ; 中国博士
    后科学基金资助项目( 2014M550180) ; 黑龙江省教育厅科学技术研究基金资助项目( 12541090)

Improved k-Means Algorithm Based on Lab Space for Color Image Segmentation#br#

HUO Fengcai1a,1 b,SUN Xueting1a,REN Weijian1a,1 b,YANG Di2,YU Tao3   

  1. 1a. School of Electrical Engineering and Information; 1b. Heilongjiang Provincial Key Laboratory of Networking and Intelligent Control,Northeast Petroleum University,Daqing 163318,China;
    2. Institute of Planning and Design for Second Oil Production Plants,Daqing Oil Field Company,Daqing 163461,China;
    3. Institute of Planning and Design for Fourth Oil Production Plants,Daqing Oil Field Company,Daqing 163511,China
  • Received:2018-11-08 Online:2019-03-25 Published:2019-06-10

摘要: 为减弱经典k-means 算法中RGB( Red Green Blue) 空间各个颜色分量高度线性相关以及欧氏距离的尺度相关性对图像分割结果产生的影响,并克服RGB 空间色彩分布不匀的缺陷,提出了一种基于Lab 颜色空间的改进k-means 聚类彩色图像分割方法。首先,将颜色空间从RGB 转换为Lab 空间,每个像素点都可以由L、a、b 3 分量组合进行表示。其次,用马氏距离替换欧氏距离进行改进,应用改进后的k-means 算法对图像像素点进行聚类,从而实现分割目的。通过实验证明该改进算法比经典k-means 算法具有更好的分割效果和准确度。

关键词: 聚类, 图像分割, 颜色空间, 马氏距离

Abstract: In order to reduce the influence of the high linear correlation of each color component in the RGB( Red Green Blue) space and the scale correlation of the Euclidean distance to the image segmentation results in classical k-means algorithm,the Lab color space can overcome the defect of uneven color distribution in RGB space,an improved k-means clustering method for color image segmentation based on Lab color space is proposed. Firstly,the color space is converted from RGB to Lab,and each pixel can be represented by the combination of L,a and b. Secondly,mahalanobis distance is used to replace Euclidean distance,and the improved k-means algorithm is used to cluster the pixels of the image to achieve the purpose of segmentation.Finally,experiment results show that the improved algorithm has better segmentation effect and accuracy than the classical k-means algorithm.

Key words: cluster, image segmentation, color space, mahalanobis distance

中图分类号: 

  • TP18