Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (36): 195-197.DOI: 10.3778/j.issn.1002-8331.2008.36.056

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

Image denoising method with multi-window based on Support Vector Clustering

QI Qi,JIANG Jia-fu,HE Wei   

  1. School of Computer and Communication Engineering,Changsha University of Science and Technology,Changsha 410076,China
  • Received:2007-12-25 Revised:2008-03-03 Online:2008-12-21 Published:2008-12-21
  • Contact: QI Qi

基于支持向量聚类的多窗口图像去噪方法

齐 琦,蒋加伏,何 伟   

  1. 长沙理工大学 计算机与通信工程学院,长沙 410076
  • 通讯作者: 齐 琦

Abstract: The paper presents an image denoising method based on Support Vector Clustering(SVC) aiming at salt and pepper noise.The pixel point is marked as signal,the points of possible positive noise and the points of possible negative noise using local statistical characteristics.In the latter two categories,authors use iterating SVC on the gray value of pixels and deal it with noise filtering.Thus,the position of the noise will be located and the noise will be handled.Experiments show that the method presented in this paper has a very good effect on image denoising achieving 70%,especially has the better effect on the detail preserving.

Key words: Support Vector Clustering(SVC), image denoising, salt and pepper noise

摘要: 提出一种针对椒盐噪声的SVC多窗口图像去噪方法。利用局部统计特性将像素点标记为信号点、可能的正噪声点及可能的负噪声点。在后两类中根据灰度值不同迭代使用支持向量聚类确定出噪声点的位置,并对其进行多窗口滤波。实验证明该方法在噪声率达到70%以上时具有很好的去噪效果,尤其在保持图像细节方面效果显著。

关键词: 支持向量聚类, 图像去噪, 椒盐噪声