小波提升方法在移动机器人导航图像去噪中的应用

英文篇名：The Application of LWT in the Image Denoise of Mobile Robot Visual Navigation in the Complex Environment
中文刊名：长沙大学学报
英文刊名：Journal of Changsha University
作者：阚晶
英文作者：KAN Jing(School of Information Science and Engineering ; Central South University ; Hunan 410083 ; China)
中文关键词：复杂环境 ; 移动机器人 ; 视觉导航 ; 小波提升
英文关键词：complex environment ; mobile robot ; visual navigation ; LWT
出版日期：2006-09-30
机构：中南大学信息科学与工程学院湖南长沙410083
年：2006
期：05
出版单位：长沙大学学报

摘要

小波提升算法能够将复杂的滤波过程分解成多个简单步骤,且分解的每一步都可逆.针对光照不均匀、烟雾和灰尘等复杂环境下,移动机器人视觉系统所提取的图像有大幅度噪声,实验利用小波提升算法的快速去噪,并通过仿真实验验证了该方法的有效性.实验结果证明该方法既能够计算快捷精确,又能够节省内存,能够提高移动机器人视觉导航系统的整体反映速度.
The lifting wavelet algorithm can be divided into several simple processes and every step has its inverse transformation.The images contain lots of noises based on complex environment such as illumination non-uniformity,smog and dust.With the method of LWT,some the results of experiments have indicated the processing speed of the mobile robot system can be raised accurately and the rapidity and real-timeliness of mobile robot system can be improved by the image denoise method of mobile robot navigation based on LWT under the Complex Environment.

引文

[1]SWELDENS W.Lossless image compression using integer tointeger wavelet transforms[J].International conference on im-age processing,1997,1(1):596-600.
    [2]SWELDENS W.The lifting scheme:A new philosophy inbiorthogonal wavelet constructions[A].Proceedings of SPIEon Wavelet Applications in Signal and Image Processing III[C].San Diego:SPIE,1995,(2):68-79.
    [3]SWELDENS W.The lifting scheme:A construction of secondgeneration wavelets[J].SIAM Journal Mathematical Analysis,1997,29(2):511-546.
    [4]Kovacevic J.et a1.Wavelet Families of Increasing Order inArbitrary Dimensions[J].IEEE Trans.on Image Processing,2000,9(3):480-496.
    [5]蔡自兴.智能控制———基础及应用[M].北京:国防工业出版社,1998.
    [6]段晨东.第二代小波降噪及其在诊断系统中的应用[J].小型微型计算机系统,2004,(7):1342-1343.
    [7]陈香朋,曹思远.第二代小波变换及其在地震信号去噪中的应用[J].石油物探,2004,(6):547-549.
    [8]郭晶,孙伟娟.小波分析理论与MATLAB 7实现[M].北京:北京电子工业出版社,2005.