FTVd
所属分类:图形图像处理
开发工具:matlab
文件大小:158KB
下载次数:25
上传日期:2013-11-26 22:13:21
上 传 者:
tianyumi
说明: 快速TV算法用于图像去模糊、图像去噪等,其中TV模式包括两种
(TV went fast algorithm for image blur, image denoising, which includes two TV mode)
文件列表:
FTVd_v4.1 (0, 2013-10-25)
FTVd_v4.1\READ_FTVd_v3.0.txt (12358, 2013-09-11)
FTVd_v4.1\Untitled.asv (535, 2013-10-10)
FTVd_v4.1\Untitled.m (950, 2013-10-24)
FTVd_v4.1\Untitled2.asv (591, 2013-10-25)
FTVd_v4.1\Untitled2.m (591, 2013-10-25)
FTVd_v4.1\demoAll.m (201, 2013-09-11)
FTVd_v4.1\demoMTVL1.m (2882, 2013-09-11)
FTVd_v4.1\demoMTVL2.m (2565, 2013-09-11)
FTVd_v4.1\demoTVL1.asv (1770, 2013-09-11)
FTVd_v4.1\demoTVL1.m (1770, 2013-09-11)
FTVd_v4.1\demoTVL2.asv (1735, 2013-09-18)
FTVd_v4.1\demoTVL2.m (1677, 2013-10-10)
FTVd_v4.1\images (0, 2013-09-11)
FTVd_v4.1\images\Thumbs.db (7168, 2013-09-11)
FTVd_v4.1\images\lena256.png (122250, 2013-09-11)
FTVd_v4.1\solvers (0, 2013-09-13)
FTVd_v4.1\solvers\FTVd_v4.asv (1349, 2013-09-13)
FTVd_v4.1\solvers\FTVd_v4.m (1349, 2013-09-13)
FTVd_v4.1\solvers\coresolvers (0, 2013-10-25)
FTVd_v4.1\solvers\coresolvers\ADM2MTVL1.m (10952, 2013-09-11)
FTVd_v4.1\solvers\coresolvers\ADM2MTVL2.m (8793, 2013-09-11)
FTVd_v4.1\solvers\coresolvers\ADM2TVL1.m (5856, 2013-09-11)
FTVd_v4.1\solvers\coresolvers\ADM2TVL2.m (6004, 2013-10-24)
FTVd_v4.1\solvers\coresolvers\ADM_TVL2_test.m (5835, 2013-09-11)
FTVd_v4.1\solvers\utilities (0, 2013-09-11)
FTVd_v4.1\solvers\utilities\imfilter33.m (2295, 2013-09-13)
FTVd_v4.1\solvers\utilities\snr.m (484, 2013-09-11)
FTVd_v4.1\test_0515.asv (759, 2013-09-11)
FTVd_v4.1\test_0515.m (3931, 2013-09-11)
************************************************************************
FTVd_v4: Fast Total Variation Deconvolution (version 4)
************************************************************************
Copyright (C) 2009 Junfeng Yang, Yin Zhang, Wotao Yin and Yilun Wang
1). Get Started
===================
Run any demo code listed below:
demoTVL2 : Demo TV/L2 solve
demoMTVL2 : Demo multichannel TV/L2 solve
demoTVL1 : Demo TV/L1 solve
demoMTVL1 : Demo multichannel TV/L1 solve
or run demoAll to demostrate all.
2). Introduction
====================
FTVd_v4 keeps all the features of FTVd_v3.0 and is a major upgrade.
In this version, we implemented the Alternating Direction Method (ADM),
which dates back to Glowinski and Marrocco (1975) and Gabay and Mercier
(1976).
Compared with ver 3.0, this version avoids the ill-conditioning caussed by
large penalty paramter by considering the augmented Lagrangian and ADM.
As a result, FTVd_v4 has a much simply parameter selection rule and
behaves very efficiently and roubustly.
For more introductory material about FTVd, see READ_FTVd_v3.0, which is a
copy of README in FTVd_v3.0.
3). Usage
====================
FTVd_v4 is called in the following way:
out = FTVd_v4(Bn,H,mu,str),
where
Bn -- a blurry and noiy observation,
H -- a convolution kernel,
mu -- Lagrangian multiplier (positive number)
str -- a string used to specify a model.
* str = 'L2' or 'l2' to specify a L2 model (for white noise),
* str = 'L1' or 'l1' to specify a L1 model (for impulsive noise).
Users can also specify algorithm paramters and call FTVd_v4 like
out = FTVd_v4(Bn,H,mu,str,opts),
where opts is a structure with some fields. For example, when L1 model
is to be solved, users may specify fields listed below:
% opts -- a structure containing algorithm parameters {default}
* opst.beta1 : a positive constant {5}
* opst.beta2 : a positive constant {20}
* opst.gamma : a constant in (0,1.618] {1.618}
* opst.maxitr : maximum iteration number {500}
* opst.relchg : a small positive parameter which controls
stopping rule of the code. When the
relative change of X is less than
opts.relchg, then the code stops. {1.e-3}
* opts.print : print inter results or not {0}
For L2 models, .beta1 and .beta2 should be replaced by one field .beta. For
more explanation about FTVd, see each core solver in folder "coresolver"
and also check READ_FTVd_v3.0.
4). References
====================
For algorithmic details, such as continution on penalty parameters and
optimality conditions, see references:
[1] Y. Wang, J. Yang, W. Yin and Y. Zhang, "A New Alternating Minimiza-
tion Algorithm for Total Variation Image Reconstruction",
SIAM Journal on Imaging Sciences, 1(3), 248-272, 2008.
[2] J. Yang, W. Yin, Y. Zhang and Y. Wang, "A Fast Algorithm for Edge-
Preserving Variational Multichannel Image Restoration", SIAM Journal
on Imaging Sciences, 2(2), 569-592, 2009.
[3] J. Yang, Y. Zhang and W. Yin, "An efficient TVL1 algorithm for
deblurring multichannel images corrupted by impulsive noise",
SIAM Journal on Scientific Computing, 31(4), 2842-2865, 2009.
[4] M. Tao, J. Yang and B. He, "Alternating direction algorithms for total
variation deconvolution in image reconstruction ", TR09-18, Department
of Mathmatics, Nanjing University, August, 2009, Available at Optimization
online: http://www.optimization-online.org/DB_HTML/2009/11/2463.html
5). Contact Information
=======================
FTVd is available at: http://www.caam.rice.edu/~optimization/L1/ftvd/
Please feel free to e-mail the following authors with any comments
or suggestions:
Junfeng Yang, Depart. Math., Nanjing Univ.,
6). Copyright Notice
====================
FTVd is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 3 of the License, or (at your option)
any later version.
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
General Public License for more details at
.
近期下载者:
相关文件:
收藏者: