Error while run the code

function veins = miura_max_curvature(img, fvr, sigma)

winsize = ceil(4*sigma);
[X,Y] = meshgrid(-winsize:winsize, -winsize:winsize);

the above function is called in the main script

function [region, edges] = lee_region(img, mask_h, mask_w)
% Localise the finger region

% Parameters:
%  img    - Input vascular image
%  mask_h - Height of the mask
%  mask_w - Width of the mask

% Returns:
%  region - Binary mask indicating the finger region
%  edges  - Matrix containing two rows, first row corresponds to the 
%           y-positions of the upper finger edge and the second row
%           corresponds to the y-positions of the lower finger edge.

% Reference:
% Finger vein recognition using minutia-based alignment and local binary
%   pattern-based feature extraction
% E.C. Lee, H.C. Lee and K.R. Park
% International Journal of Imaging Systems and Technology
%   Volume 19, Issue 3, September 2009, Pages 175-178   
% doi: 10.1002/ima.20193

% Author:  Bram Ton <[e-mail]>
% Date:    20th March 2012
% License: Simplified BSD License

[img_h, img_w] = size(img);

% Determine lower half starting point
if mod(img_h,2) == 0
    half_img_h = img_h/2 + 1;
else
    half_img_h = ceil(img_h/2);
end

% Construct mask for filtering
mask = zeros(mask_h,mask_w);
mask(1:mask_h/2,:) = -1;
mask(mask_h/2 + 1:end,:) = 1;

% Filter image using mask
img_filt = imfilter(img, mask,'replicate'); 
%figure; imshow(img_filt)

% Upper part of filtred image
img_filt_up = img_filt(1:floor(img_h/2),:);
[~, y_up] = max(img_filt_up); 

% Lower part of filtred image
img_filt_lo = img_filt(half_img_h:end,:);
[~,y_lo] = min(img_filt_lo);

% Fill region between upper and lower edges
region = zeros(size(img));
for i=1:img_w
    region(y_up(i):y_lo(i)+size(img_filt_lo,1), i) = 1;
end

% Save y-position of finger edges
edges = zeros(2,img_w);
edges(1,:) = y_up;
edges(2,:) = round(y_lo + size(img_filt_lo,1));

the above function is called in the main function

img = im2double(imread('veinvein.jpg')); % Read the image
img = imresize(img,0.5);               % Downscale image

fvr = lee_region(img,4,20);    % Get finger region

%% Extract veins using maximum curvature 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; 

%% 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;

above code is the main script

Error in miura_max_curvature (line 27)

winsize = ceil(4*sigma);

how to solve the following error

Unable to perform assignment because the size of the left side is 44-by-483 and the size of the right side is 44-by-161-by-3.

Answer 1

[Image Analyst]

Bran Ton wrote some non-robust code when he did this:

[img_h, img_w] = size(img);

Never do that with an image. If img happens to be a color image, img_w will give you the number of columns times the number of color channels (3) instead of the number of columns like you were expecting. He should have done this:

[img_h, img_w, numberOfColorChannels] = size(img);

When you do it like I said, what does numberOfColorChannels end up being? 1 or 3?

For further reference see: Steve Eddins's blog about the size function when used with images.