作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2007, Vol. 33 ›› Issue (23): 22-24. doi: 10.3969/j.issn.1000-3428.2007.23.008

• 博士论文 • 上一篇    下一篇

基于Perona-Malik模型的各向异性主动轮廓外力场

宁纪锋1,2,吴成柯1,刘侍刚1   

  1. (1. 西安电子科技大学综合业务网国家重点实验室,西安 710071;2. 西北农林科技大学信息工程学院,杨凌 712100)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-12-05 发布日期:2007-12-05

Anisotropic External Force Field for Active Contour Based on Perona-Malik Model

NING Ji-feng1,2, WU Cheng-ke1, LIU Shi-gang1   

  1. (1. National Key Laboratory of Integrated Service Networks, Xidian University, Xi’an 710071; 2. College of Information Engineering, Northwest Agriculture & Forest University, Yangling 712100)
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-12-05 Published:2007-12-05

摘要: 利用经典的Perona-Malik各向异性去噪模型具有保护图像边界信息的特点,将经过Perona-Malik模型处理后图像的负梯度作为外力场,研究其对主动轮廓法分割结果的影响,提出了一种主动轮廓外力场模型PMF。理论分析和实验结果表明,PMF模型不仅能够保持图像的边界信息,克服了传统外力场不能进入图像凹部的缺陷,而且对初始曲线的约束较少。由于PMF是基于去噪模型而得,因此具有较好的鲁棒性。

关键词: 主动轮廓模型, PMF模型, 各向异性扩散, 图像分割

Abstract: The Perona-Malik model is a classical method of anisotropic removing noises. It has the advantage of remaining the edge map of image. The negative gradient of restored image by Perona-Malik model is defined as external forces, and the segmentation results that affect active contour model are studied. Accordingly, an external force field for active contour model——PMF is presented. Theoretical analysis and experimental results show that PMF can retain the edge information of image and enter the edge’s concaves entirely. PMF has large capture range with few restrictions to initial curves. Moreover, because PMF is derived from Perona-Malik, it is robust to the noise.
【Key words】active contour model; PMF model; anisotropic diffusion; image segmentation

Key words: active contour model, PMF model, anisotropic diffusion, image segmentation

中图分类号: