计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (14): 184-186.

• 图形、图像、模式识别 • 上一篇    下一篇

基于方差和边缘插值的邻近点图像修复算法

周春霞,吴锡生   

  1. 江南大学 信息工程学院,江苏 无锡 214122
  • 收稿日期:2007-08-30 修回日期:2007-11-23 出版日期:2008-05-11 发布日期:2008-05-11
  • 通讯作者: 周春霞

Adjacent pixels image inpainting based on deviation and edge interpolation

ZHOU Chun-xia,WU Xi-sheng   

  1. College of Information Engineering,Jiangnan University,Wuxi,Jiangsu 214122,China
  • Received:2007-08-30 Revised:2007-11-23 Online:2008-05-11 Published:2008-05-11
  • Contact: ZHOU Chun-xia

摘要: 提出了一种复杂度较低的结合去噪的图像修复新方法,将图像修复和去噪相结合,达到了更好的修复图像的目的。图像去噪算法通过检验每一个像素点的被腐蚀程度得到了一种基于邻近像素点并保留图像对象边缘和细节的算法;图像修复算法通过定义修复区域像素点的优先度得到了一种邻近像素点的图像修复算法。经实验验证,该方法具有良好的性能。

关键词: 图像修复, 优先度, 图像去噪, 邻近像素点, 边缘

Abstract: It is ready to propose a new method of image inpainting and noise removing without complex theory.It combines the inpainting with removing noise to get the better effect.The algorithm of noise removing test whether the whole image points is influenced by noise to express the method of reserving objects’ edge and detail in image based on vicinity points;and the algorithm of image inpainting introduces a parameter Point Priority(PP) in inpainting region so as to express the inpainting algorithm based on vicinity points.The algorithm has better performance validated by experiments.

Key words: image inpainting, priority, noise removing, vicinity points, edge