Instructions
-----------------------
Dependency
-----------------------
BM3D: http://www.cs.tut.fi/~foi/GCF-BM3D/
LASIP Image Restoration Toolbox: http://www.cs.tut.fi/~lasip/2D/
Gabriel Peyre Toolbox: div.m, getoptions.m grad.m https://www.ceremade.dauphine.fr/~peyre/codes/
Matlab Toolbox: dst.m, dct.m, kaiser.m
Misc: BM3D also depends on check_order.m if you run into any error: https://searchcode.com/codesearch/view/13666439/
BM3D Configuration
-----------------------
CBM3D comes with lots of parameters. Here is the parameter we have used for better result:
transform_2D_HT_name = 'dct';
transform_2D_Wiener_name = 'dst';
%%%% Hard-thresholding (HT) parameters:
N1 = 4;
Nstep = 1;
N2 = 8;
Ns = 49;
tau_match = 3000*3;
lambda_thr2D = 0;
lambda_thr3D = 2.7;
beta = 2.0;
%%%% Wiener filtering parameters:
N1_wiener = 4;
Nstep_wiener = 1;
N2_wiener = 8*2;
Ns_wiener = 39;
tau_match_wiener = 400*3;
beta_wiener = 2.0;
%%%% Block-matching parameters:
stepFS = 1; %% step that forces to switch to full-search BM, "1" implies always full-search
smallLN = 'not used in np'; %% if stepFS > 1, then this specifies the size of the small local search neighb.
stepFSW = 1;
smallLNW = 'not used in np';
thrToIncStep = 1; %% used in the HT filtering to increase the sliding step in uniform regions
We also modified CBM3D so that it takes two color spaces, one for hard thresholding and one for Wiener filter. For hard thresholding, we use yCbCr and opp for Wiener.
You need to modify CBM3D.m in order to reproduce the same result. For convenience, we have put a patch file to CBM3D.m in 3rdparty/BM3D that performs the above changes.
Execution
-----------------------
Please execute:
apps/demosaic/demo_demosaic.m
没有合适的资源?快使用搜索试试~ 我知道了~
资源推荐
资源详情
资源评论
收起资源包目录
基于matlab的ADMM噪声Bayer图像的联合去噪.zip (208个子文件)
weights.caffemodel 2.16MB
weights.caffemodel 2.14MB
weights.caffemodel 1.57MB
convert_to_lmdb 3KB
create_net 6KB
demosaick 11KB
.gitignore 7B
demosaicnet.iml 455B
psnr_comparison_Kodak_sigma15_algos_3.jpg 172KB
psnr_comparison_Kodak_sigma0_algos_3.jpg 171KB
psnr_comparison_Kodak_sigma5_algos_3.jpg 165KB
psnr_comparison_Kodak_sigma25_algos_3.jpg 159KB
psnr_comparison_McM_sigma0_algos_3.jpg 153KB
psnr_comparison_McM_sigma5_algos_3.jpg 148KB
psnr_comparison_McM_sigma25_algos_3.jpg 142KB
psnr_comparison_McM_sigma15_algos_3.jpg 140KB
CBM3D.m 28KB
VBM3D.m 26KB
BM3D.m 22KB
CVBM3D.m 21KB
BM3D_CFA.m 17KB
BM3DSHARP.m 17KB
BM3DDEB_init.m 17KB
BM3DDEB.m 17KB
pd.m 12KB
natsort.m 12KB
Demo_IDDBM3D.m 9KB
analyze.m 6KB
function_LPAKernelMatrixTheta.m 6KB
function_CreateLPAKernels.m 5KB
natsortfiles.m 5KB
natsortfiles_doc.m 5KB
compareAlgorithms.m 4KB
function_stdEst2D.m 4KB
demo_demosaic.m 4KB
joint_demosaicing_denoising_admm.m 4KB
call_flexisp.m 3KB
combineResultImages.m 3KB
function_Window2D.m 2KB
demosaickingMatrix.m 2KB
demo_BM3DSAPCA.m 2KB
crossChannelMatrix.m 1KB
observationMat.m 1KB
call_joint_demosaicing_denoising_admm.m 1KB
compute_operator_norm.m 1020B
generate_bayer_mask.m 938B
generate_bayer_mask.m 938B
generate_bayer.m 821B
generate_bayer.m 821B
addMosaicMatrix.m 797B
comppsnr.m 794B
comppsnr.m 782B
gammaCorrection.m 711B
buildCRGB.m 618B
readGammaTable.m 579B
darkChannelMatrix.m 577B
vec2rgb.m 531B
rgb2bayer3d.m 428B
testAddMosaicMatrix.m 376B
vec2gray.m 354B
bayer2bayer3d.m 343B
rgb2vec.m 338B
comppsnr_rgb.m 330B
analysis.mat 1KB
bm3d_thr_colored_noise.mexa64 144KB
bm3d_thr_sharpen_var.mexa64 132KB
bm3d_wiener_video.mexa64 109KB
bm3d_thr_video.mexa64 100KB
bm3d_wiener_colored_noise.mexa64 100KB
bm3d_CFA_thr.mexa64 71KB
bm3d_thr.mexa64 54KB
bm3d_CFA_wiener.mexa64 42KB
bm3d_wiener_color.mexa64 40KB
bm3d_thr_color.mexa64 39KB
bm3d_wiener.mexa64 38KB
bm3d_thr_colored_noise.mexglx 118KB
bm3d_thr_sharpen_var.mexglx 106KB
bm3d_wiener_video.mexglx 90KB
bm3d_thr_video.mexglx 86KB
bm3d_wiener_colored_noise.mexglx 82KB
bm3d_CFA_thr.mexglx 47KB
bm3d_thr.mexglx 40KB
bm3d_wiener_color.mexglx 32KB
bm3d_CFA_wiener.mexglx 31KB
bm3d_thr_color.mexglx 31KB
bm3d_wiener.mexglx 28KB
bm3d_thr_colored_noise.mexmaci 145KB
bm3d_thr_sharpen_var.mexmaci 133KB
bm3d_wiener_video.mexmaci 109KB
bm3d_thr_video.mexmaci 101KB
bm3d_wiener_colored_noise.mexmaci 97KB
bm3d_thr.mexmaci 57KB
bm3d_wiener_color.mexmaci 45KB
bm3d_thr_color.mexmaci 41KB
bm3d_wiener.mexmaci 41KB
bm3d_thr_colored_noise.mexmaci64 133KB
bm3d_thr_sharpen_var.mexmaci64 121KB
bm3d_wiener_video.mexmaci64 93KB
bm3d_thr_video.mexmaci64 89KB
bm3d_wiener_colored_noise.mexmaci64 89KB
共 208 条
- 1
- 2
- 3
资源评论
- Make程序设计2023-08-10不错的资源,是我想要的,感谢博主分享! #完美解决问题 #运行顺畅 #内容详尽
唐先生的博客
- 粉丝: 2211
- 资源: 528
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- 126641091882064LSPatch.apk
- 5G网络基础培训课件.zip
- 2024-spring-HIT-CS-大作业
- yolo目标检测项目实验
- downloadFile-1.hc
- C++课程设计:基于Qt的航班信息管理系统
- ADS7822UVerilog驱动,前面传的有点问题
- 基于python的高性能爬虫程序,使用了多线程+缓存+xpath实现的,这里以彼-岸图库为例,实现,仅用于学习交流
- 中分辨率成像光谱仪(MODIS)烧毁面积产品信息MODIS-C6-BA-User-Guide-1.2.pdf
- Screenshot_20240427_172613_com.huawei.browser.jpg
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功