[1]王益艳,陈阳,罗立民.基于结构保持的增广型各向异性PDE滤波[J].东南大学学报(自然科学版),2014,44(4):717-722.[doi:10.3969/j.issn.1001-0505.2014.04.007]
 Wang Yiyan,Chen Yang,Luo Limin.Extended anisotropic PDE filtering based on structure remain[J].Journal of Southeast University (Natural Science Edition),2014,44(4):717-722.[doi:10.3969/j.issn.1001-0505.2014.04.007]
点击复制

基于结构保持的增广型各向异性PDE滤波()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
44
期数:
2014年第4期
页码:
717-722
栏目:
计算机科学与工程
出版日期:
2014-07-16

文章信息/Info

Title:
Extended anisotropic PDE filtering based on structure remain
作者:
王益艳13陈阳12罗立民12
1东南大学计算机科学与工程学院, 南京 210096; 2东南大学计算机网络与信息集成教育部重点实验室, 南京 211189; 3四川文理学院物理与机电工程学院, 达州 635000
Author(s):
Wang Yiyan13 Chen Yang12 Luo Limin12
1School of Computer Science and Engineering, Southeast University, Nanjing 210096, China
2Key Laboratory of Computer Network and Information Integration of Ministry of Education, Southeast University, Nanjing 211189, China
3School of Physics and Mechanical and Electronic Engineering, Sichuan University of Arts and Science, Dazhou 635000, China
关键词:
偏微分方程 各向异性 扩散权函数 差曲率 结构张量
Keywords:
partial differential equation anisotropy diffusion-weighted function difference curvature structure tensor
分类号:
TP391.9
DOI:
10.3969/j.issn.1001-0505.2014.04.007
摘要:
基于偏微分方程模型的图像恢复框架,从各向异性扩散的角度,提出了一种基于结构保持的增广型PDE图像滤波算法.首先定义了一种新的扩散权函数,采用差曲率代替传统的梯度算子作为边缘判别器,以克服后者对噪声敏感的不足.并根据图像结构张量信息分别设计了梯度和切线2个不同方向的扩散系数,在图像平坦区域,梯度和切线2个方向具有同等扩散,而在图像边缘处只沿着切线方向扩散.同时,引入数据保真项,有利于在滤波时保持图像细节.实验结果表明,无论对灰度图像还是医学图像,该算法可比现有其他几种扩散模型取得更好的主观视觉效果和客观性能评价指标.
Abstract:
According to the image restoration framework of the partial differential equation(PDE)model, an extended PDE image filtering algorithm based on structure remains is proposed from the viewpoint of anisotropic diffusion. First, a new diffusion-weighted function is designed by taking the differential curvature instead of the conventional gradient operator as edge indicator, overcoming the disadvantage of being sensitive to noise. And then, different gradient and tangential diffusion coefficients are designed respectively according to the structure tensor information of image. In the smooth region of image, diffusion is executed with equal diffusion along both the gradient and tangential directions, while on the edge of the image, diffusion is executed only along the tangential direction. Meanwhile, the data-fidelity term is introduced, which is beneficial to maintaining image details during filtering. The experimental results demonstrate that for both gray image and medical image, the new algorithm is superior to the conventional diffusion models in terms of objective performance evaluation and subjective visual effect.

参考文献/References:

[1] Mohammad R H, Ahmad M O, Wang Chunyan. An edge-adaptive Laplacian kernel for nonlinear diffusion filters[J]. IEEE Transactions on Image Processing, 2012, 21(4):1561-1572.
[2] Kang Y N. Variational PDE-based image segmentation and inpainting with applications in computer graphics[D]. Los Angeles, CA, USA: University of California, 2008.
[3] 冯象初, 王卫卫. 图像处理的变分和偏微分方程方法[M]. 北京:科学出版社, 2009.
[4] Guichard F, Moisan L, Morel J M. A review of PDE models in image processing and image analysis[J]. Journal de Physique Ⅳ, 2002, 92(12):137-154.
[5] Perona P, Malik J. Scale-space and edge detection using anisotropic diffusion[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990, 12(8):629-639.
[6] Catte F, Coll T, Lions P, et al. Image selective smoothing and edge detection by nonlinear diffusion[J]. SIAM Journal on Numerical Analysis, 1992, 29(1):182-193.
[7] Alvarez L, Lion P, Morel J. Image selective smoothing and edge detection by nonlinear diffusion[J]. SIAM Journal on Numerical Analysis, 1992, 29(3):845-866.
[8] Alvarez L, Mazorra L. Signal and image restoration using shock filters and anisotropic diffusion[J]. SIAM Journal on Numerical Analysis, 1994, 31(2):590-605.
[9] Gilboa G, Sochen N, Zeevi Y Y. Image enhancement and denoising by complex diffusion processes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(8): 1020-1036.
[10] Gilboa G, Sochen N, Zeevi Y Y. Regularized shock filters and complex diffusion[C]//Lecture Notes in Computer Science. Berlin, Germany, 2002: 399-413.
[11] You Y L, Xu W, Tannenbaum A, et al. Behavioral analysis of anisotropic diffusion in image processing[J]. IEEE Transactions on Image Processing, 1996, 5(11):1539-1553.
[12] Chen Qiang, Montesinos P, Sun Q S, et al. Adaptive total variation denoising based on difference curvature[J]. Image and Vision Computing, 2010, 28(3):298-306.
[13] 何宁, 吕科, 王袆雪. 基于结构保持的MR图像运动伪影快速抑制方法[J]. 电子学报, 2013, 41(7): 1319-1323.
  He Ning, Lü Ke, Wang Yixue. The method of MR images motion artifacts fast suppression based on structure remains[J]. Acta Electronica Sinica, 2013, 41(7): 1319-1323.(in Chinese)
[14] Liu Feng, Liu Jingbo. Anisotropic diffusion for image denoising based on diffusion tensors[J]. Journal of Visual Communication and Image Representation, 2012, 23(3):516-521.
[15] Xiong Bo, Yin Zhouping. A universal denoising framework with a new impulse detector and nonlocal means[J]. IEEE Transactions on Image Processing, 2012, 21(4):1663-1675.

相似文献/References:

[1]施建勇,雷国辉,宋雄伟,等.初始应力各向异性的边界面方程[J].东南大学学报(自然科学版),2004,34(3):379.[doi:10.3969/j.issn.1001-0505.2004.03.022]
 Shi Jianyong,Lei Guohui,Song Xiongwei,et al.Bounding surface equation of initial stress anisotropy[J].Journal of Southeast University (Natural Science Edition),2004,34(4):379.[doi:10.3969/j.issn.1001-0505.2004.03.022]
[2]曾玲玲,洪振舜,刘松玉.考虑固结路径影响的天然沉积土不排水剪切试验研究[J].东南大学学报(自然科学版),2012,42(4):744.[doi:10.3969/j.issn.1001-0505.2012.04.032]
 Zeng Lingling,Hong Zhenshun,Liu Songyu.Experimental study on undrained shear behavior of natural sedimentary clays under various consolidation stress paths[J].Journal of Southeast University (Natural Science Edition),2012,42(4):744.[doi:10.3969/j.issn.1001-0505.2012.04.032]
[3]徐承龙.一类重特征偏微分方程的可解性[J].东南大学学报(自然科学版),1987,17(2):126.[doi:10.3969/j.issn.1001-0505.1987.02.014]
 Xu Chenglong(Departmen of Mathematics and Mechanics).The Solvability for a Class of Partial Diffferential Eqnatious with Double Characteristies[J].Journal of Southeast University (Natural Science Edition),1987,17(4):126.[doi:10.3969/j.issn.1001-0505.1987.02.014]

备注/Memo

备注/Memo:
收稿日期: 2013-12-27.
作者简介: 王益艳(1982—),男,博士生,讲师; 罗立民(联系人),男,博士,教授,博士生导师,luo.list@seu.edu.cn.
基金项目: 国家重点基础研究计划(973计划)资助项目(2010CB732503)、国家自然科学基金资助项目(81370040, 31100713)、江苏省自然科学基金资助项目(BK2011593)、四川省教育厅一般科研资助项目(13ZB0098).
引用本文: 王益艳,陈阳,罗立民.基于结构保持的增广型各向异性PDE滤波[J].东南大学学报:自然科学版,2014,44(4):717-722. [doi:10.3969/j.issn.1001-0505.2014.04.007]
更新日期/Last Update: 2014-07-20