Convolution in Image Processing

Question

Does anyone know how to write arg min code in matlab with explanation?

I am referring to a paper for my research. I am not sure whether I can paste the paper link here. At first, I try to do feature extraction of finger using Gabor Filter. According to the paper, after feature extraction level, I need to do the following using formula to get my desired image:

OriCode(x, y) = arg j min {I(x, y) ∗ G(x, y, θj)}

∗ stands for convolution, I(x,y) stands for input image, G(x,y,θj) stands for gabor filtered image

There is actually a small j between arg and min. OriCode ranges from 0 to J-1 whereby j ranges from 0 to J. Here,J is set to 6.

I try to use conv2 to convolute my original input image and gabor filtered image. End up I am getting a blank white image. And what is the arg min used for? Please help me. Thanks.

Answer 1

I'm going to use a grayscale image for simplicity. Suppose you have a 3D matrix and each slice is the result of filtering the image with the Gabor filter for that angle.

I = randn(256, 256); % Just some random image
G = gabor_filter; % Assume G is 256x256xNum_Angles, e.g. 256x256x6 if you had 6 angles
Y = zeros(256, 256, size(G, 3)); % Preallocate a temporary matrix

for idx = 1:size(G, 3)
    Y(:, :, idx) = conv2(I, G(:, :, idx), 'same');
end

Ymin = min(Y, [], 3); % This takes the minimum along the 3rd dimension