计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (23): 167-170.

• 图形、图像、模式识别 • 上一篇    下一篇

结合先验形状和Mumford-Shah模型的活动轮廓分割

郑洁雪1,杨 玲1,王中科2   

  1. 1.成都信息工程学院 电子工程学院,成都 610225
    2.成都信息工程学院 网络工程学院,成都 610225
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-08-11 发布日期:2011-08-11

Combining prior shape and Mumford-Shah model for active contour segmentation

ZHENG Jiexue1,YANG Ling1,WANG Zhongke2   

  1. 1.College of Electronic Engineering,Chengdu University of Information Technology,Chengdu 610225,China
    2.College of Networks Engineering,Chengdu University of Information Technology,Chengdu 610225,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-08-11 Published:2011-08-11

摘要: 基于先验形状和Mumford-Shah模型的活动轮廓分割是一种抗噪声干扰、稳定的图像分割方法。该模型采用水平集方法,并结合活动轮廓模型、先验形状和Mumford-Shah模型来控制曲线演化。特定目标的先验知识可以有效地指导目标准确分割,经过主成分分析(PCA)法可以得到感兴趣对象形状的主要信息。通过对不同图片分割实验表明,针对特定的形状,该方法对杂乱背景、部分遮挡、缺失和强噪声的图片依然能得到满意的结果。

关键词: 图像分割, 先验形状, 主动轮廓模型, Mumford-Shah模型, 水平集

Abstract: The active contour segmentation based on prior shape and M-S model has been widely used in image segmentation because of their computational efficiency and stability.Based on level set method,the novel model using the active contour model,prior shape and Mumford-Shah model can control evolvement of curve.Prior knowledge on certain object is a useful guidance in image segmentation.It exploits the prior knowledge of desired object’s shape through a Principle Component Analysis(PCA) method.By numerical experiments presented in the paper,the proposed approach is robust to cluttered background,partial occlusions,missing and noise with the specific shape.

Key words: image segmentation, shape prior, active contour model, Mumford-Shah model, level set