How to analyze one image 100 times using for loop?

Hello,

I have an image. I want to divide the image into 100 frames and analyze each frame separately. Please find my code below:

img = imread('1F1.jpg');
k = 1;
    for j = 1: 100
            dblSubtractedImage = img(j) - img(k);
            s = dblSubtractedImage*5;
            L = mat2gray(s);
            imshow(L);
            im = imclearborder(255-imclearborder(255-s));
            hc = sum(im,2);
            mask = hc >= 14500;
            fr = find(mask, 1, 'first');
            lr = find(mask, 1, 'last');
            cr = im(fr:lr, :);
            D = std2(cr);
            meanIntensityValue = mean2(cr);
            % imshow(cr)
            K = mat2gray(cr);
            min_image = min(K(:));
            max_image = max(K(:));
            % figure
            imshow(K)
            [rows, columns, numberOfColorBands] = size(K);
            if numberOfColorBands > 1
                grayImage = K(:, :, 2);
            end
            binaryImage = grayImage~= 40;
            area2 = sum( binaryImage,'all');
    end
    


AverageArea = (sum(area2))/100;
AverageStandardDeviation = (sum(D))/100;
AverageIntensityValue = (sum(meanIntensityValue))/100;

The first frame is used as a reference. The first frame is subtracted from all the other frames subsequently. The resulting images are then masked using a threshold which is shown above. Few other parameters such as the mean, standard deviation and the area of each of the resulting image are found. At the end of the for loop, the average value of the parameters above are found. However, I am getting the error message below. Any help would be appreciated. Thank you.

>> f

Index exceeds the number of array elements (0).

Error in mat2gray (line 38)

if limits(2)==limits(1) % Constant Image

Error in f (line 17)

K = mat2gray(cr);

[darova]

I think something is wrong here

img = imread('1F1.jpg');
k = 1;
for j = 1: 100
        dblSubtractedImage = img(j) - img(k);

[KSSV]

YOu have only one image......how you can take it as 100 frames?

[Warid Islam]

Hi KSSV,

I have the other 100 frames too. Do I need to read all the images in the file and then perform the code?

[Warid Islam]

Hi KSSV,

All I need is to automate the process. I have attached 4 images of the same sample at different frames. The first image (1F1.jpg) would be subtracted from all the other images. There are around 200 images similar to the ones below at different frames. I was wondering if it is possible to use a for loop and perform the operation.

Answer 1

[Gaurav Garg]

Hi,

I understand that you wish to analyze each frame separately. However, there’s some problem in the code you have written.

In line 4, you subtract the 2 images and store the output in sblSubtractedImage. Now, when you display sblSubtractedImage, you get 0 as output. Hence, the variables you use thereafter are also 0 and variable cr stores a 0 ×1 empty uint8 column vector. And therefore, you get error while using mat2gray.

What is happening in your case is, you have saved the file normally (using Windows option to save file) and you are reading subtracting first 2 pixels which are both white. However, if you use imwrite (MATLAB function) to save the image you will get the desired portion and this should solve your error.

[Saher77]

I am doing my first experiment in deep learning. I tried some code where I am unable an error using mat2gray. Here's my code

%Extract training features using CNN
%get the network weights for the 2nd convolutional layer
w1=net.Layers(2).Weights;
%scale and resize the weights for visualization
w1=mat2gray(w1);
w1=imresize(w1,5);

Error is :

Index exceeds the number of array elements (0).

Error in mat2gray (line 38)

if limits(2)==limits(1) % Constant Image

Error in trial1 (line 63)

w1=mat2gray(w1);

Can anyone help me get out o this?

[Walter Roberson]

You would get that error under the circumstance that net.Layers(2).Weights was empty -- in such a case w1 would be empty, and the attempt to find min() and max() of the array would come out empty.

You can fix the immediate problem by using

w1 = rescale(w1);

as rescale() does not give an error message when the input is empty.

... After that you would get an error in the imresize() because imresize() does not work on empty arrays. But you would have fixed the mat2gray() problem.