计算机与现代化 ›› 2011, Vol. 1 ›› Issue (3): 63-65,7.doi: 10.3969/j.issn.1006-2475.2011.03.017

• 图像处理 • 上一篇    下一篇

一种基于边缘检测的改进的中值滤波去噪方法

张毛女,柳 薇   

  1. 华南师范大学计算机学院,广东 广州 510631
  • 收稿日期:2010-11-24 修回日期:1900-01-01 出版日期:2011-03-18 发布日期:2011-03-18

An Improved Median Filter Denoising Based on Edge Detection

ZHANG Mao-nü, LIU Wei   

  1. School of Computer Science, South China Normal University, Guangzhou 510631, China
  • Received:2010-11-24 Revised:1900-01-01 Online:2011-03-18 Published:2011-03-18

摘要: 中值滤波是一种去除椒盐噪声的有效算法,在现实生活中应用相当广泛。传统的中值滤波在去噪的同时很容易丢失图像的边缘信息。本文提出一种基于边缘检测的改进的中值滤波去噪算法。它先将含噪图像的边缘信息检测出来,然后将边缘信息保存下来,再对原含噪图像用中值滤波进行图像去噪,然后对于保存下来的边缘信息用小模板进行去噪,再用该小模板去噪后的边缘信息区替换中值滤波后的边缘信息。最后通过实验验证,此去噪算法可以在去噪的同时更好地保护图像的边缘信息。

关键词: 中值滤波, 椒盐噪声, 去噪

Abstract: To remove impulse noise, median filter is a valid method which is applied in a wide range of real life. Conventional median filter in denoising images loses the edge information easily while images are denoised. This paper presents an improved edge detection-based median filter algorithm. It first detects edge information of noisy image, and then edge information is saved, and then the original noisy image is denoised with median filter, for the edge information which is preserved before denosing; with a small template to remove the noise, then uses it to replace the edge information which is denoised by median fliter. Finally, through experimental verification, the denoising algorithm can be better in preserving edge information while images are denoised.

Key words: median filter, salt &, pepper noise, denoising

中图分类号: