Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (19): 172-173.DOI: 10.3778/j.issn.1002-8331.2009.19.053

• 图形、图像、模式识别 • Previous Articles     Next Articles

Color image segmentation based on improved watershed transform

JIANG Rong-rong1,2,XIAO Sha-li1,GAO Jia-li3   

  1. 1.Key Lab of Optoelectronic Technology & Systems of Education Ministry of China,Chongqing University,Chongqing 400030,China
    2.Laboratory & Facility Management Division,Chongqing Technology and Business University,Chongqing 400067,China
    3.The Engineering Training Center,Chongqing Institute of Technology,Chongqing 400054,China
  • Received:2008-05-15 Revised:2008-08-04 Online:2009-07-01 Published:2009-07-01
  • Contact: JIANG Rong-rong

改进分水岭算法的彩色图像分割技术

姜蓉蓉1,2,肖沙里1,高家利3   

  1. 1.重庆大学 光电技术与系统教育部重点实验室,重庆 400030
    2.重庆工商大学 实验与实验设备管理处,重庆 400067
    3.重庆工学院 工程训练中心,重庆 400054
  • 通讯作者: 姜蓉蓉

Abstract: With regard to the over-segmentation of the convention watershed transform,the paper proposes an algorithm for eliminating irrelevant minima in the resulting gradient images.The paper uses gradient-threshold-controlled segmentation function for watershed transformation.The experimental results show that the proposed algorithm can effectively improve segmentation accuracy for different image natures.

Key words: watershed transform, gradient threshold, over-segmentation, color image segmentation

摘要: 针对传统的分水岭变换可能产生严重的过渡分割,提出的分水岭变换算法在梯度图中去除了无相关点,用梯度临界值控制分割函数式。实验结果显示该算法对具有不同特征的图像均可有效地改进分割的精确性。

关键词: 分水岭变换, 梯度临界值, 过渡分割, 彩色图像分割