pre-processing steps in palm vein image ????

17 次查看（过去 30 天）
mona zizo
mona zizo 2012-11-7
回答： Sowjanya M 2017-3-25
how can I show the vascular in binary palm vein image I make this code and the vascular not apper in binary >> I=imread('001_l_460_01.jpg'); >> figure,imshow(I); >> level=graythresh(I); >> bw=im2bw(I,level); >> figure,imshow(bw); >> roipoly(bw); >> x=[408 514 708 598] x = 408 514 708 598 >> y=[313 96 213 421] y = 313 96 213 421 >> BW=poly2mask(x,y,576,768); >> h=fspecial('unsharp'); >> I2=roifilt2(h,I,BW); >> figure,imshow(I2); what is the wrong???
  2 个评论
Image Analyst
Image Analyst 2012-11-7
And where did you upload 001_l_460_01.jpg to?
mona zizo
mona zizo 2012-11-8
I have CASIA-Multi-Spectral-PalmprintV1 database

请先登录，再进行评论。

请先登录，再回答此问题。

回答（3 个）

Image Analyst
Image Analyst 2012-11-7
What makes you think that some automatically picked threshold is going to be any good? Chances are it is not. And what makes you think doing an unsharp mask filter on a binary image is a good thing to do? I don't see the rationale for doing that.
  2 个评论
mona zizo
mona zizo 2012-11-8
thax what is your opnion to do it???
Image Analyst
Image Analyst 2012-11-8
That's not my field. Though I do know there has been research at UCLA on identifying vein patterns to identify criminals. There's probably more research that you know of or you can search Vision Bib

请先登录，再进行评论。

Shahmeer Rosli
Shahmeer Rosli 2015-11-30
Hello I would to ask. are my code for Palm Vein are correct? and is this all for the future extraction? and 1 more question what is the best classifier that suit with my future extraction. thank you
clear; close all; clc; clear variables; %% % Howto use the miura_* scripts.
img = im2double(imread('RightFive.bmp')); % Read the image
% img=img(:,:,1);
img=imcrop(img,[199.5 121.5 248 253]);
figure(1); imshow(img); title('Original Captured Image') %%
% img = imresize(img,0.8); % Downscale image
% Get the valid region, this is a binary mask which indicates the region of % the finger. For quick testing it is possible to use something like: % fvr = ones(size(img)); fvr = lee_region(img,40,200); % Get finger region
figure(2); imshow(fvr) title('Detected Finger Region') %% Extract veins using maximum curvature method % Filter Kernels Method sigma = 3; % Parameter v_max_curvature = miura_max_curvature(img,fvr,sigma);
% Binarise the vein image md = median(v_max_curvature(v_max_curvature>0)); v_max_curvature_bin = v_max_curvature > md;
figure(3); imshow(v_max_curvature_bin) title('Binarised Veins Extracted by Maximum Curvature Method')
%% thinning K=bwmorph(v_max_curvature_bin,'thin',Inf); figure(4) , imshow(K); % set(gcf,'position',[1 1 600 600]); title('Thinning');
%% CLEAN bw =bwmorph(K,'clean'); figure(5), imshow(bw), title ('Clean');
%% v_max_curvature_bin3= bwmorph(bw,'shrink'); figure, imshow(v_max_curvature_bin3)
v_max_curvature_bin1= bwmorph(bw,'bridge');
figure, imshow(v_max_curvature_bin1), title ('bridge');
v_max_curvature_bin2 = bwmorph(bw,'thin',Inf);
figure, imshow(v_max_curvature_bin2), title ('thin');
Visualise
Overlay the extracted veins on the original image
overlay_max_curvature = zeros([size(img) 3]);
overlay_max_curvature(:,:,1) = img;
overlay_max_curvature(:,:,2) = img + 0.4*v_max_curvature_bin;
overlay_max_curvature(:,:,3) = img;
figure;
imshow(overlay_max_curvature)
title('Maximum curvature method')
Sowjanya M
Sowjanya M 2017-3-25
I want Matlab coding for combined features of finger texture and vein images.

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by