src
所属分类:图形图像处理
开发工具:Visual C++
文件大小:33KB
下载次数:264
上传日期:2010-06-04 09:54:00
上 传 者:
najialuo
说明: 非局部均值算法(NLM)去除含噪声图像中的噪声,达到图像恢复的效果,经过试验,效果较为明显
(Non-local means algorithm (NLM) to remove the image noise with noise, to the effect of image restoration, after testing, the effect is more obvious)
文件列表:
src (0, 2010-04-10)
src\denoise.cpp (5696, 2010-04-10)
src\io_tiff.h (540, 2010-03-08)
src\mt19937ar.h (158, 2010-03-01)
src\diffnoise.c (4796, 2010-03-08)
src\addnoise.c (3154, 2010-04-06)
src\estnoise.cpp (2208, 2010-03-08)
src\mt19937ar.c (4789, 2010-03-08)
src\makefile (1527, 2010-03-08)
src\io_tiff.c (12540, 2010-03-08)
src\nlmeans_lib.cpp (24148, 2010-03-08)
src\GPLv3.txt (35068, 2010-03-02)
src\nlmeans_lib.h (1175, 2010-03-01)
% non-local means denoising
# ABOUT
* Author : Antoni Buades
* Author : Nicolas Limare
* Copyright : (C) 2009, 2010 IPOL Image Processing On Line http://www.ipol.im/
* Licence : GPL v3+, see GPLv3.txt
# OVERVIEW
This source code provides an implementation of the non-local means
denoising algorithm, as described in IPOL
http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/
4 programs are provided:
* `denoise`, denoise an image with the nl-means algorithm
* `estnoise`, estimate the noise standard deviation
* `addnoise`, add a gaussian noise to an image
* `diffnoise`, visualise the noise removed
`addnoise` uses the Mersenne Twister pseudo-random number
generator, by Makoto Matsumoto and Takuji Nishimura. The code is
included, with its owm distribution licence.
-> http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html
All these programs read and write TIFF images, but can be easily
adapted to any other file format.
Only 8bit RGB or grayscale TIFF images are handled. Other TIFF files
are implicitly converted to 8bit.
# REQUIREMENTS
The code is written in ANSI C and C++, and should compile on any
system with an ANSI C/C++ compiler.
The libtiff header and libraries are required on the system for
compilation and execution. See http://www.remotesensing.org/libtiff/
# COMPILATION
Simply use the provided makefile, with the command `make`.
Alternatively, you can manually compile
cc -c io_tiff.c; \
c++ io_tiff.o nlmeans_lib.cpp estnoise.cpp -ltiff -o estnoise
cc -c io_tiff.c; \
c++ io_tiff.o nlmeans_lib.cpp denoise.cpp -ltiff -o denoise
cc io_tiff.c mt19937ar.c addnoise.c -ltiff -o addnoise
cc io_tiff.c diffnoise.c -ltiff -o diffnoise
# USAGE
`addnoise` takes 3 parameters: `addnoise in.tiff out.tiff sigma`
* `in.tiff` : input clean image
* `out.tiff` : output noisy image
* `sigma` : standard deviation of teh noise (for the R, G and B channels)
`estnoise` takes 1 parameter: `estnoise image.tiff`
* `image.tiff` : the noisy image to measure
`denoise` takes 3 or 5 parameters:
`denoise in.tiff out.tiff sigma`
`denoise in.tiff out.tiff rsigma gsigma bsigma`
* `in.tiff` : input noisy image
* `out.tiff` : output denoised image
* `[rgb]sigma` : the noise standard deviation on the R, G and B channels
`diffnoise` takes 4 or 6 parameters:
`diffnoise noisy.tiff clean.tiff out.tiff sigma`
`diffnoise noisy.tiff clean.tiff out.tiff rsigma gsigma bsigma`
* `noisy.tiff` : input noisy image
* `clean.tiff` : input denoised image
* `out.tiff` : visualisation of the removed noise, (noisy - clean) / sigma
* `[rgb]sigma` : the noise standard deviation on the R, G and B channels
# ABOUT THIS FILE
Copyright 2009, 2010 IPOL Image Processing On Line http://www.ipol.im/
Author: Nicolas Limare
Copying and distribution of this file, with or without modification,
are permitted in any medium without royalty provided the copyright
notice and this notice are preserved. This file is offered as-is,
without any warranty.
近期下载者:
相关文件:
收藏者: