基于Zernike 矩相似度联合的图像滤波.docVIP

Journal of Image and Signal Processing图像与信号处理, 2013, 2, 1-7 /10.12677/jisp.2013.21001 Published Online January 2013 (/journal/jisp.html) Image Denoising by Zernike-Moment-Similarity Collaborative Filtering* Xiuchun Xiao , Jianhuang Lai 1,2 1 1 School of Information Science and Technology, Sun Yat-sen University, Guangzhou 2 College of Information, Guangdong Ocean University, Zhanjiang Email: stsljh@, springxxc@163.com Received: Oct. 22 , 2012; revised: Nov. 6 , 2012; accepted: Nov. 15 , 2012 nd th th Abstract: Because similarity function defined in non-local means filter is subject to image noise, it cannot robustly represent the real similarity between pixels. Zernike moments are good statistics of the pixels in image patch, and have rotation-invariant feature, so they can be utilized to describe image feature while resistance to noise. In this paper, Zernike-moment-similarity is defined according to the difference of Zernike moments instead of pixel intensity, and then the intensity of the processed pixel is estimated by weighting the intensities of the local window according to the collaborative Zernike-moment-similarity. Simulation experiment results and analysis demonstrate that the presented algorithm can achieve better performance and higher PSNR than the current algorithms which directly adopting inten- sity difference as its similarity function. Keywords: Image Denoising; Multilateral Filtering; Non-Local Means Filtering; Similarity Measure; Zernike Moments 基于 Zernike矩相似度联合的图像滤波* 肖秀春 1,2，赖剑煌 1 1 中山大学信息科学与技术学院，广州 2 广东海洋大学信息学院，湛江 Email: stsljh@, springxxc@163.com 收稿日期：2012年 10月 22日；修回日期：2012年 11月 6日；录用日期：2012年 11月 15日 摘 要：非局域均值滤波(non-local means filtering, NLMF)采用图像块间灰度差测度像素间相似性，由于灰度差 易受噪声影响，这种相似性测度缺乏鲁棒性。图像块的 Zernike矩是块内像素灰度的统计量，且具有旋