Image processing code explanation.

    lambda  = 4;%8
    theta   = 0;
    psi     = [0 pi/2];
    gamma   = 0.5;
    bw      = 1;
    N       = 12;

    bp_filter_input_image = sharpened_original_image;
    bp_filtered_image = zeros(size(bp_filter_input_image, 1),
                        size(bp_filter_input_image, 2), N);
    img_out_disp = zeros(size(bp_filter_input_image, 1), 
                   size(bp_filter_input_image, 2), N);

    %   display 12 images in one window
    figure;
    for n=1:N
        mean_filter = BP_fn(bw,gamma,psi(1),lambda,theta) + 
                      1i * BP_fn(bw,gamma,psi(2),lambda,theta);
        % filter output to the n-th channel
        bp_filtered_image(:, :, n) = imfilter(bp_filter_input_image, 
                                     mean_filter, 'symmetric');
        % next orientation
        theta = theta + pi/N;
        % default superposition method, L2-norm
        image_vector = [];
        image_vector = sum(abs(bp_filtered_image(:,:,n)).^2, 3).^0.5;
        % normalize
        img_out_disp(:,:,n) = image_vector./max(image_vector(:));
        %result show
        str=sprintf('BP theta=pi/%d',n);
        subplot(3,4,n),imshow(img_out_disp(:,:,n));xlabel(str); 
    end

I have the following questions:

(1) What is the target/end-result of this entire routine?

(2) What is going on in the following line of code? What is 1i?

mean_filter = BP_fn(bw,gamma,psi(1),lambda,theta) + 
              1i * BP_fn(bw,gamma,psi(2),lambda,theta);

(3) What is going on in the following line of code?

image_vector = sum(abs(bp_filtered_image(:,:,n)).^2, 3).^0.5;

[Ba Ba Black Sheep!]

function H = BP_fn(bw,gamma,psi,lambda,theta)
    sigma = lambda/pi*sqrt(log(2)/2)*(2^bw+1)/(2^bw-1);
    sigma_x = sigma;
    sigma_y = sigma/gamma;

      sz = fix(8*max(sigma_y,sigma_x));
      if mod(sz,2)==0, sz = sz+1;end

      [x y] = meshgrid(-fix(sz/2):fix(sz/2),fix(sz/2):-1:fix(-sz/2));

      % Rotation 
      x_theta = x*cos(theta)+y*sin(theta);
      y_theta = -x*sin(theta)+y*cos(theta);

      H = exp(-0.5*(x_theta.^2/sigma_x^2+y_theta.^2/sigma_y^2))
              .*cos(2*pi/lambda*x_theta+psi);

[Mallikarjun umadi]

can anyone suggest code for skin disease identification using matlab code?only for herpes, psoriasis and dermatitis

Answer 1

[Image Analyst]

According to the comments it displays 12 filtered images. Can't you ask the author? Why are you running code if you have no idea what it's supposed to do?

1i is the imaginary variable sqrt(-1).

The line of code:

image_vector = sum(abs(bp_filtered_image(:,:,n)).^2, 3).^0.5;

is a bit strange. bp_filtered_image(:,:,n) is a 2-D complex valued image. Then they take the square magnitude of it. Not sure why they're summing along the third dimension. Then they take the square root to get the magnitude of it, like you'd get from just using abs() or norm() I think.

I don't know what BP_fn() is since the code for that was not included.

[Ba Ba Black Sheep!]

> Can't you ask the author?

I mailed the author and he doesn't respond.

> Why are you running code if you have no idea what it's supposed to do?

I am doing a thesis. Trying to understand this and some other parts of the source code so that I can use it in my own project.

> I don't know what BP_fn() is since the code for that was not included.

BP_fn() included.