Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (21): 185-187.DOI: 10.3778/j.issn.1002-8331.2010.21.053

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

Color image segmentation based on JSEG and watershed methods

SUN Wei-fang1,2,DUAN Zhi-yong1,ZHU Zhong-jie2   

  1. 1.Physical Engineering College,Zhengzhou University,Zhengzhou 450001,China
    2.Ningbo Key Lab of Digital Signal Processing(DSP),Zhejiang Wanli University,Ningbo,Zhejiang 315100,China
  • Received:2009-08-14 Revised:2009-11-18 Online:2010-07-21 Published:2010-07-21
  • Contact: SUN Wei-fang



  1. 1.郑州大学 物理工程学院,郑州 450001
    2.浙江万里学院 宁波市数字信号处理(DSP)重点实验室,浙江 宁波 315100
  • 通讯作者: 孙卫芳

Abstract: The JSEG is a classic algorithm for color image segmentation.However,it need iterate computing the local J-values and performing region growing under predefined multiple scales,which makes the segmentation process complex and computationally expensive.Hence,to overcome this drawback,a novel method based on joint JSEG and watershed technologies is proposed.The new method implements spatial segmentation for only once with the watershed algorithm after the J-image is acquired,which can effectively reduce the computational load while keeping good perceptual results.Thus,the segmentation efficiency is improved.

Key words: image segmentation, Joint Systems Engineering Group(JSEG) algorithm, watershed algorithm, region merging

摘要: 针对JSEG算法在分割过程中需要在多个尺度下反复进行局部J值计算和区域生长,分割过程繁琐、算法复杂度高的不足,提出一种结合分水岭与JSEG的图像分割新算法。新算法在计算得到图像J图后,通过引入分水岭算法直接对J图进行空域分割,在基本保证良好分割效果的前提下,有效降低了原JSEG算法的复杂度、提高了分割效率。

关键词: 图像分割, 静态图像压缩标准(JSEG)算法, 分水岭算法, 区域合并

CLC Number: