[1]张新宇,陈桓,汪天富,等.基于各向异性扩散的超声图像混合滤波方法[J].深圳大学学报理工版,2008,25(4):397-402.
 ZHANG Xin-yu,CHEN Huan,WANG Tian-fu,et al.A hybrid filter based on anisotropic diffusion for medical ultrasound images[J].Journal of Shenzhen University Science and Engineering,2008,25(4):397-402.
点击复制

基于各向异性扩散的超声图像混合滤波方法()
分享到:

《深圳大学学报理工版》[ISSN:1000-2618/CN:44-1401/N]

卷:
第25卷
期数:
2008年4期
页码:
397-402
栏目:
电子光学与信息工程
出版日期:
2008-10-31

文章信息/Info

Title:
A hybrid filter based on anisotropic diffusion for medical ultrasound images
文章编号:
1000-2618(2008)04-0397-06
作者:
张新宇陈桓汪天富王睿陈思平
深圳大学信息工程学院,深圳市生物医学工程重点实验室,深圳 518060
Author(s):
ZHANG Xin-yuCHEN HuanWANG Tian-fuWANG Ruiand CHEN Si-ping
College of Information Engineering,Shenzhne University,Shenzhen 518060,P.R.China
关键词:
医学超声图像图像降噪各向异性扩散各向同性扩散中值滤波
Keywords:
medical ultrasound image image denoising anisotropic diffusion isotropic diffusionmedian filter
分类号:
TP 391.41
文献标志码:
A
摘要:
提出基于各向异性扩散的混合滤波新方法,以图像梯度为依据,利用各向异性扩散、各向同性扩散和自适应脉冲噪声滤波对医学超声图像进行分段滤波.为提高模型的稳定性与适用性,通过实验优化模型中的高斯平滑系数σ、步长△t、梯度阈值YminYmax4个主要参数,并确定各参数默认值及其修正范围.实验结果表明,该方法在去噪效果、滤波速度、边缘保护、稳定性和适用性方面优于现有的单一滤波方法,具有较好的应用前景.
Abstract:
A hybrid filtering method based on anisotropic diffusion for ultrasound speckle suppression was proposed.This method was designed to take advantage of the different denoising properties of three techniques:isotropic diffusion filter,improved AD filter and adaptive impulse noise filter.The hybrid filtering method was a combination of three filters working according to the different ranges of local gradient.The optimal value of four key parameters,including Gaussian smoothing coefficient σ,step size △t,gradient threshold Ymin and Ymax, in the hybrid method were determined in the experiments on hundreds of images.The results have shown that,in comparison with median filtering and speckle reducing anisotropic diffusion (SRAD) filtering,the hybrid method had better performance in noise reduction,edge preservation,stability and suitability,and it was approximately 50 times faster than the SRAD,indicating that it is an efficient and effective method of medical ultrasonic image filtering.

参考文献/References:

[1]Loupas T,Mcdicke W N,Allan P L.用于医学超声图像散斑噪声抑制的自适应中值滤波器[J].IEEE电路系统汇刊,1989,36(1):129-135 (英文版).
[2]Hao X H,Gao S K,Gao X R.用于超声散斑噪声抑制的多尺度非线性阈值新方法[J].IEEE医学影像汇刊,1999,18(9):787-794 (英文版).
[3]Yu Y J,Acton S T.抑制散斑噪声的各向异性扩散方法[J].IEEE图像处理汇刊,2002,11(11):1260-1270 (英文版).
[4]Barcelos Z A C,Boaventura M,Silva E.用于去噪和边缘检测的具有良好平衡性的流体方程[J].IEEE图像处理汇刊,2003,12(7):751-763 (英文版).
[5]Lee J S.基于局部统计量的数字图像增强和噪声滤波[J].IEEE模式分析和机器智能汇刊,1982,2(2):286-294 (英文版).
[6]Malik J,Perona P.基于各向异性扩散的尺度空间和边缘检测[J].IEEE模式分析和机器智能汇刊,1990,12(7):629-639 (英文版).
[7]Witkin A P.尺度空间滤波[C]//第8届人工智能国际联合会议论文集.卡尔斯鲁厄(德国):RFG,1983:1019-1022 (英文版).
[8]胡旺,冯伟森,李志蜀.针对脉冲噪声的双窗口自适应中值滤波方法[J].四川大学学报:工程科学版,2006,38(4):131-135.
[9]吕忆松,陈亚珠,郭玉红.基于各向异性散布的医学图像非线性滤波法[J].北京生物医学工程,2003,6(22):81-84.


[1]Loupas T,Mcdicke WN,Allan PL.An adaptive weighted median filter for speckle suppression in medical ultrasonic images[J].IEEE Transcations on Circuits System,1989,36(1):129-135.
[2]Hao X H,Gao S K,Gao X R.A novel multiscale nonlinear thresholding method for ultrasonic speckle suppressing[J].IEEE Transcations on Medical Imaging,1999,18(9):787-794.
[3]Yu Y J,Acton S T.Speckle reducing anisotropic diffusion[J].IEEE Transcations on Image Processing,2002,11(11):1260-1270.
[4]Barcelos Z A C,Boaventura M,Silva E.A well-balanced flow equation for noise removal and edge detection[J].IEEE Transcations on Image Processing,2003,12(7):751-763.
[5]Lee J S.Digital image enhancement and noise filtering by using local statistics[J].IEEE Transcations on PAMI,1982,2(2):286-294.
[6]Malik J,Perona P.Scale-space and edge detection using anisotropic diffusion[J].IEEE Transcations on PAMI,1990,12(7):629-639.
[7]Witkin A P.Scale-space filtering[C]//Proceedings of 8th International Joint Conferenc Artifical Intelligence.Karlsrvhe(Germany):RFG,1983:1019-1022.
[8]HU Wang,FENG Wei-sen,LI Zhi-shu.An adaptive median filter with double windows for impulsive noises[J].Journal of Sichuan University:Engineering Science Edition,2006,38(4):131-135 (in Chinese).
[9]LU Yi-song,CHEN Ya-zhu,Guo Yu-hong.A medical image nonlinear filtering method based on anisotropic diffusion[J].Beijing Biomedical Engineering,2003,6(22):81-84 (in Chinese).

备注/Memo

备注/Memo:
收稿日期:2008-02-25;修回日期:2008-08-13
基金项目:国家自然科学基金资助项目(60772147)
作者简介:张新宇(1976-),女(汉族),四川省成都市人,深圳大学讲师、博士.E-mail:xyzhang9@szu.edu.cn
通讯作者:汪天富 (1966-), 男(汉族),深圳大学教授、博士.E-mail:tfwang@szu.edu.cn
更新日期/Last Update: 2008-11-26