资源简介
当前最成功的三种图像去噪方法Matlab运行程序比较,包括著名的BLS-GSM,Non local means以及BM3D(Block Matching 3D)方法。运行目录下的Denoising_Demo.m即可得到结果。
代码片段和文件信息
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Code arranged by Jiang Lei(jianglei730@gmail.com) %
% Tested in MATLAB 7.6 %
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%% Test Denosing Methods %%%%%%%%%%%%%%%%%
%%%%%%% Read Original Image %%%%%%%%
clc;
close all;
% choose one image file to test
file_name = ‘lena1.png‘;
% file_name = ‘barco.png‘;
% file_name = ‘boat.png‘;
% file_name = ‘fingerprint.png‘;
% file_name = ‘flinstones.png‘;
% file_name = ‘house.png‘;
% file_name = ‘peppers256.png‘;
% specify noise magnitude
sig = 10;
path(path ‘.\Sample Images‘);
origin_img = imread(file_name);
% origin_img = imcrop(origin_img [50 180 50 50]);
figure imagesc(origin_img); colormap gray; axis image;
title(‘Original Image‘);
disp(‘Display Original Image‘)
disp(‘%%% Press Any Key to Continue %%%‘);
pause;
origin_img = double(origin_img);
[hw] = size(origin_img);
%%%%%%% Generate Random Noise ******
randn(‘state‘ 0);
noise = randn(size(origin_img));
noise = noise/sqrt(mean2(noise.^2));
noise_img = origin_img + sig*noise;
figure imagesc(noise_img); colormap gray; axis image; drawnow;
title(‘Noise Image‘);
disp(‘Display Noise Image‘)
%%%%%%%%%%%%%%%%%%%%%%%%%%%% BLS-GSM methods %%%%%%%%%%%%%%%%%%%%%%%%%%%%
disp(‘ ‘);
disp(‘%%% Press Any Key to Continue %%%‘);
pause;
cd BLS-GSM;
path(path ‘.\Simoncelli_PyrTools‘);
path(path ‘.\Added_PyrTools‘);
path(path ‘.\denoising_subprograms‘);
disp(‘Testing BLS-GSM Denoising Methods...‘);
% Noise Parameters (assumed additive Gaussian and independent of the
% original image)
sig = 10; % standard deviation
PS = ones(size(origin_img)); % power spectral density (in this case flat i.e. white noise)
seed = 0; % random seed
% Pyramidal representation parameters
Nsc = ceil(log2(min(h w)) - 4); % Number of scales (adapted to the image size)
Nor = 3; % Number of orientations (for X-Y separable wavelets it can only be 3)
repres1 = ‘uw‘; % Type of pyramid (shift-invariant version of an orthogonal wavelet in this case)
repres2 = ‘daub1‘; % Type of wavelet (daubechies wavelet order 2N for ‘daubN‘; in this case ‘Haar‘)
% Model parameters (optimized: do not change them unless you are an advanced user with a deep understanding of the theory)
blSize = [3 3]; % n x n coefficient neighborhood of spatial neighbors within the same subband
% (n must be odd):
parent = 0; % including or not (1/0) in the neighborhood a coefficient from the same spatial location
% and orientation as the central coefficient of the n x n neighborhood but
% next coarser scale. Many times helps but not always.
boundary = 1; % Boundary mirror extension to avoid boundary artifacts
covariance = 1; % Full covariance matrix (1) or only diagonal elem
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
文件 21697 2009-06-17 14:55 Denoise Methods\BM3D\BM3D.m
文件 17116 2009-06-17 14:54 Denoise Methods\BM3D\BM3DDEB.m
文件 17668 2009-06-17 14:58 Denoise Methods\BM3D\BM3DSHARP.m
文件 180038 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_thr.mexa64
文件 198707 2009-05-27 10:31 Denoise Methods\BM3D\bm3d_thr.mexglx
文件 200704 2007-01-17 14:17 Denoise Methods\BM3D\bm3d_thr.mexw32
文件 195584 2008-10-17 14:52 Denoise Methods\BM3D\bm3d_thr.mexw64
文件 110712 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_thr_color.mexa64
文件 158284 2008-11-18 11:41 Denoise Methods\BM3D\bm3d_thr_color.mexglx
文件 151552 2008-11-18 17:34 Denoise Methods\BM3D\bm3d_thr_color.mexw32
文件 124928 2008-10-17 15:32 Denoise Methods\BM3D\bm3d_thr_color.mexw64
文件 175956 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_thr_colored_noise.mexa64
文件 260360 2008-11-18 11:44 Denoise Methods\BM3D\bm3d_thr_colored_noise.mexglx
文件 217088 2008-11-18 19:14 Denoise Methods\BM3D\bm3d_thr_colored_noise.mexw32
文件 196608 2008-10-16 19:06 Denoise Methods\BM3D\bm3d_thr_colored_noise.mexw64
文件 144444 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_thr_sharpen_var.mexa64
文件 235818 2008-11-18 11:49 Denoise Methods\BM3D\bm3d_thr_sharpen_var.mexglx
文件 188416 2008-11-18 17:11 Denoise Methods\BM3D\bm3d_thr_sharpen_var.mexw32
文件 158208 2008-10-17 12:26 Denoise Methods\BM3D\bm3d_thr_sharpen_var.mexw64
文件 204696 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_thr_video.mexa64
文件 166577 2008-11-20 12:11 Denoise Methods\BM3D\bm3d_thr_video.mexglx
文件 196608 2007-03-12 00:03 Denoise Methods\BM3D\bm3d_thr_video.mexw32
文件 193536 2008-11-19 23:25 Denoise Methods\BM3D\bm3d_thr_video.mexw64
文件 122697 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_wiener.mexa64
文件 129108 2009-05-27 10:31 Denoise Methods\BM3D\bm3d_wiener.mexglx
文件 131072 2007-03-06 15:51 Denoise Methods\BM3D\bm3d_wiener.mexw32
文件 144896 2008-10-17 14:56 Denoise Methods\BM3D\bm3d_wiener.mexw64
文件 122779 2009-06-17 13:08 Denoise Methods\BM3D\bm3d_wiener_color.mexa64
文件 170317 2008-11-18 11:41 Denoise Methods\BM3D\bm3d_wiener_color.mexglx
文件 159744 2008-11-18 17:50 Denoise Methods\BM3D\bm3d_wiener_color.mexw32
............此处省略516个文件信息
相关资源
- [精通MATLAB GUI设计第3版][陈垚光毛涛涛
- 张正友标定程序—MATLAB
- MATLAB 深度学习简介(ppt)
- Census及Hamming距离三维立体重建Matlab程
- NMFSC处理程序,非负矩阵分解matlab程序
- [MATLAB精品丛书][高清文字版]MATLAB语言
- Matlab凸优化工具箱
-
Matlab分享系列 - 3 - Matlab_Simuli
nk代码 - 数字信号处理及其MATLAB实现
- Matlab中unwrap函数C代码实现
- Matlab神经网络工具箱使用教程
- 基于LABVIEW和MATLAB混合编程障碍物识别
- 减速器优化
- 摄像头数字识别matlab程序.zip
- matlab预测控制工具箱使用手册
- PCA人脸识别论文附MATLAB程序
- matlab聚类分析工具箱很好用
- MATLAB多元线性回归
- 基于MATLAB的图像融合算法
- 非常好用的matlab图像拼接程序
- MATLAB(DCT DWT)图像数字水印的方案毕
- MATLAB s函数最详细指南编写教程24998
- L-K金字塔光流法matlab代码实现
- 几乎包含IEEE所有节点的牛拉潮流计算
- 目标检测的DPM模型在windows下的实现
- 发动机万有特性曲线matlab源程序
- matlab实现VQ说话人识别系统.zip
- pbm/pgm/ppm图片的读写Matlab
- 基于MATLAB数字图像处理的实现和仿真
-
matlab simuli
nk simevent 建模入门
评论
共有 条评论