Adaptive Image Despeckling with Center Affine Filter
摘要：Due to the coherent nature of scattering phenomenon, coherent imaging systems such as radar, sonar, and ultrasound systems are inherently corrupted by a granular multiplicative noise, called speckle noise. Such speckles visually degrade the image quality, leading to the poor performance of further operations on coherent tomography images. Though a lot of efforts have been made, it is still an open problem of preserving details while reducing speckles. On one hand, speckle formation is data-dependent random process. On the other hand, either the earlier analysis results of speckles by Raghavan and Beauchemin et al, or the recent analysis of Gragnaniello et al tells that the ratio of arithmetic to geometric mean is a suitable measure to separate speckles from useful image features. Based on these two facts, we study a data-related adaptive Center Affine Filter (CAF) for removing speckles from images. The designed CAF is capable of adaptively suppressing the speckle noises while maintaining the useful details in an image.
专家简介：郑丽颖，哈尔滨工程大学计算机科学与技术学院教授、博士生导师，黑龙江省计算机学会智能人机交互专业委员会委员（副主任）、科技部重点专项评审专家、国家科技奖励评审专家、国家自然科学基金通信评委、本领域知名期刊审稿人。作为项目负责人主持“国家自然科学基金”、“中国博士后科学基金”等基金项目、以及多项横向研究项目。出版译著1部（“装备科技译著出版基金”资助）、获得多项授权的发明专利、获得黑龙江省科技进步三等奖等科研奖项，研究成果曾发表在IEEE Transactions on Image Processing、Signal Processing 等国内外期刊上。