15 November 2007 SAR image speckle noise reduction based on second generation bandelets and a kernel-based possibilistic C-means clustering
Author Affiliations +
Proceedings Volume 6787, MIPPR 2007: Multispectral Image Processing; 67872L (2007) https://doi.org/10.1117/12.750722
Event: International Symposium on Multispectral Image Processing and Pattern Recognition, 2007, Wuhan, China
Abstract
Synthetic Aperture Radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. Speckle noise of SAR affects image quality and image interpretation seriously. To alleviate deleterious effects of speckle, various ways have been devised to suppress it. An ideal algorithm should smooth the speckle without blurring edges and fine details. But most classical algorithms cannot satisfy these two demands very well. Due to the property of SAR images speckles is multiplicative noise, it difficult to estimate the variance of the high-frequency subband coefficients. Most classical approaches such as wavelet thresholding or shrinkage scheme of Donoho and Johnstone are not suitable for SAR images speckle noise removal. In this paper, a novel approach to SAR image speckle reduction is presented, which is based on second generation bandelets and a kernel-based possibilistic C-means clustering algorithm (BKPCM).
© (2007) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Shuang Liu, Shuang Liu, Licheng Jiao, Licheng Jiao, Guofeng Zhang, Guofeng Zhang, } "SAR image speckle noise reduction based on second generation bandelets and a kernel-based possibilistic C-means clustering", Proc. SPIE 6787, MIPPR 2007: Multispectral Image Processing, 67872L (15 November 2007); doi: 10.1117/12.750722; https://doi.org/10.1117/12.750722
PROCEEDINGS
6 PAGES


SHARE
Back to Top