计算机应用 ›› 2012, Vol. 32 ›› Issue (07): 1885-1889.DOI: 10.3724/SP.J.1087.2012.01885

• 图形图像技术 • 上一篇    下一篇

基于双阈值及迭代法的图像脉冲噪声检测算法

杨润玲,周军妮,魏蕊   

  1. 西安建筑科技大学 信息与控制工程学院,西安710055
  • 收稿日期:2011-12-29 修回日期:2012-02-15 发布日期:2012-07-05 出版日期:2012-07-01
  • 通讯作者: 杨润玲
  • 作者简介:杨润玲(1976-),女,陕西西安人,讲师,硕士,主要研究方向:图像处理、模式识别;周军妮(1977-),女,陕西西安人,讲师,硕士,主要研究方向:图像处理、模式识别;魏蕊(1981-),男,陕西西安人,助教,博士研究生,主要研究方向:智能信息处理。

Effective detection algorithm based on two-threshold and iteration for images with impulse noise

YANG Run-ling,ZHOU Jun-ni,WEI Rui   

  1. School of Information and Control Engineering, Xi'an University of Architecture and Technology, Xi'an Shaanxi 710055, China
  • Received:2011-12-29 Revised:2012-02-15 Online:2012-07-05 Published:2012-07-01
  • Contact: YANG Run-ling

摘要: 为了减少图像中的脉冲噪声对后续图像处理的影响,针对脉冲噪声的特点,提出了双阈值和迭代法的噪声检测算法。双阈值选取方法理论可靠,两次迭代保证了噪声点检测具有较高的正确率,最后的选择性中值滤波算法也使得图像的细节不被模糊。实验结果表明,所提算法具有较强的自适应性、较低的噪声漏检率以及较好的滤波效果。

关键词: 双阈值, 脉冲噪声, 标识矩阵, 噪声检测, 滤波

Abstract: An effective noise detection algorithm based on two-threshold and iteration for noise images with impulse noise was proposed to decrease the influence of impulse noise. The selective method of two-threshold was reliable in theory, and the two-step iteration method guaranteed the accuracy of noise detection. Moreover, the median filtering algorithm with selected pixels made sure that the details of image were not blurred at the same time. The experimental results show that the proposed algorithm is more robust and adaptive, and it has lower leak-detection and better filtering effect.

Key words: two-threshold, impulse noise, marked matrix, noise detection, filtering

中图分类号: