Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (26): 155-157.DOI: 10.3778/j.issn.1002-8331.2010.26.048

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

Improved anisotropic diffusion image smoothing method

YANG Xu-lang,HOU Yu-qing,CHEN Shen   

  1. School of Information Science and Technology,Northwest University,Xi’an 710127,China
  • Received:2009-11-03 Revised:2010-02-22 Online:2010-09-11 Published:2010-09-11
  • Contact: YANG Xu-lang

一种改进的各向异性扩散图像平滑方法

杨旭朗,侯榆青,陈 燊   

  1. 西北大学 信息科学与技术学院,西安 710127
  • 通讯作者: 杨旭朗

Abstract: An improved anisotropic smoothing method based on the characteristics of human visual system is proposed.The four-order partial differential information of image is merged into this method to avoid the“blocky effects”,the direction of diffusion is increased and the numerical model is improved.Experimental results show that this method can obtain good results in noise elimination and edge reservation.

Key words: anisotropic diffusion, human visual system, image smoothing

摘要: 在人类视觉系统特性基础上提出了一个改进的各向异性平滑方法。加入图像四阶偏微分信息避免“阶梯效应”,扩展演化方向,改进数值模型。实验结果表明,该方法在噪声消除和边缘保留方面能获得较好的效果。

关键词: 各向异性扩散, 人类视觉系统, 图像平滑

CLC Number: