期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2017
卷号：6
期号：7
页码：982-985
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Synthetic Aperture Radar (SAR) is a coherent radar system that is widely used to produce high-resolution images of the earth. Due to its comprehensible nature, SAR images are adversely stimulated by speckle noise which is a multiplicative noise and is difficult to eliminate by classic de-noising filters. The commonly used non-local means filter is not optimal for these images because image noise prevents precise determination of the correct coefficients for averaging, leading to over-smoothing and other artifacts. So, we implement an adaptive version of non-local means approach to de-noise such images to improve their appearance. This adaptive method addresses this problem by first smoothing the noisy image, then applying k- means approach to smoothed image which divides the image into various clusters and then applying non-local means filter to different clusters. We show that this adaptive non-local means approach provided more efficient results in de-noising the SAR images as compared to old non-local means filter.
关键词：De-noising; non-local means (NLM); speckle; Synthetic Aperture Radar (SAR)
