首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于并行组合加权均值级联形态滤波算法的散斑噪音压缩
引用本文:蒋立辉,王骐,王春晖,尚铁良.基于并行组合加权均值级联形态滤波算法的散斑噪音压缩[J].光子学报,1999,28(10):923-927.
作者姓名:蒋立辉  王骐  王春晖  尚铁良
作者单位:哈尔滨工业大学光电子技术研究所,哈尔滨,150006;可调谐激光技术国家重点实验室
摘    要:本文提出一种新的用于散斑噪音压缩的并行组合加权均值级联形态滤波算法。首先采用3×1和1×3两个线型结构元素分别进行形态开一闭、闭一开运算,然后利用散斑噪音的负指数统计规律加权求平均,最后再采用5×1和1×5两个线型结构元素重复上述运算。模拟结果表明此算法既有效地压缩了散斑噪音又保持了图象的几何结构,并且通过比较证明其优于F.Safa的多方向形态滤波算法。

关 键 词:散班噪音  形态滤波  并行处理  加权平均
收稿时间:1999-05-31
修稿时间:1999-05-31

SPECKLE NOISE SUPPRESSING BASED ON A PARALLEL COMBINATION WEIGHTED AVERAGE MULTISTAGE MORPHOLOGICAL FILTERING ALGORITHM
Jiang Lihui,Wang Qi,Wang Chunhui,Shang Tieliang.SPECKLE NOISE SUPPRESSING BASED ON A PARALLEL COMBINATION WEIGHTED AVERAGE MULTISTAGE MORPHOLOGICAL FILTERING ALGORITHM[J].Acta Photonica Sinica,1999,28(10):923-927.
Authors:Jiang Lihui  Wang Qi  Wang Chunhui  Shang Tieliang
Institution:1. National Key Laboratory of Tunable Laser Technology, Harbin Institute of Technology, Harbin 150006;2. Institute of optronicc, Harbin Institute of Technology, Harbin 150006
Abstract:A parallel combination weighted average multistage morphological filtering algorithm isproposed in this paper.It first makes the morphological open-close and close-opefl transformation byusing 3 ? 1 and 1 ? 3 Iinear structure elements respectively, then takes a weighted average over theabove transformational result according to the statistical law of speckle,repeats the previous processesby using 5 ? 1 and 1 ? 5 linear structure elements in the end.Simulation result shows the algorithm cannot only effectiveIy suppress speckle noise in images but also preserve the geometrical features ofimages.Through comparison,it is proved that the algorithm is superior to F.Safa algorithm,
Keywords:Speckle noise  Morphological filter  Weighted averages  Parallel processing
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《光子学报》浏览原始摘要信息
点击此处可从《光子学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号