计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (21): 205-209.

• 图形、图像、模式识别 • 上一篇    下一篇

结合分水岭机制的有监督图像背景分割算法

温 雯1,郝志峰1,2,邵壮丰3   

  1. 1.广东工业大学 计算机学院,广州 510006
    2.华南理工大学 计算机科学与工程学院,广州 510641
    3.中国电信广东网络操作维护中心,广州 510110
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-07-21 发布日期:2011-07-21

Supervised background segmentation algorithm combined with watershed mechanism

WEN Wen1,HAO Zhifeng1,2,SHAO Zhuangfeng3   

  1. 1.School of Computer,Guangdong University of Technology,Guangzhou 510006,China
    2.School of Computer Science and Engineering,South China University of Technology,Guangzhou 510641,China
    3.Network Operation and Maintenance Center of Guangdong Telecom,Guangzhou 510110,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-07-21 Published:2011-07-21

摘要: 传统的分水岭分割算法属于无监督的图像分割算法,分割获得的子区域往往不具备现实的语义信息。在分水岭分割的基础上,利用子区域像素值的高斯统计性质,提出了一种有监督的图像背景学习方法。该算法能够通过对少量人工标注的图像样本的学习,获得刻画背景子区域规律的统计模型。在此基础上对新图片中隶属于背景的子区域进行判断和合并,从而达到区分目标与背景的目的。实验验证了算法的有效性。

关键词: 背景学习, 有监督, 分割, 分水岭算法

Abstract: The traditional watershed segmentation algorithm is a kind of unsupervised segmentation algorithms,which produces sub-regions without semantic representation.A supervised image segmentation algorithm is proposed,which is based on Gaussian statistical property of sub-regions obtained by watershed segmentation.The proposed algorithm can learn the statistical model of background with a few labeled images,and then correctly separates the objects from background by merging the sub-regions which are judged members of the background.Experiments verify the validity of the proposed method.

Key words: background learning, supervised, segmentation, watershed algorithm