My code doest work well in loop,but works fine while doing manually.please find the error

The code runs well with first image but while adding the second image in the folder the DWT filter part doesnt do well with it and gives out the error as shown below.Basically my code shoule be reading a image1 from a folder and process it and wait until the second image is uploaded in the folder and while it detects image2 in the folder it should process and so on.Please may I know where the error is.

The code is

clc;
close all;
clear ;

waitMax = 20;
k = 1;
while k < 6    
    jpgFilename = sprintf('image%d.jpg', k);
    fullFileName = fullfile('C:\Users\New User1\Desktop\Project Matlab\', jpgFilename);
    if exist(fullFileName, 'file') %if the image exist in the folder
        %       rgbImage = imread(fullFileName); 
        %       figure, imshow(rgbImage)
        
        myimg = imread(fullFileName);
        J = im2gray(myimg);
        mygrayimg = imresize(J,[256 256],'nearest'); 
        New2 = im2double(mygrayimg);
        subplot(2,3,1);
        imshow(New2); 
        title('Original Image');
        
        %Adding Gaussian noise with mean 0 and variance 0.01
        
        gau = imnoise(mygrayimg, 'gaussian', 0, 0.01);
        subplot(2,3,2);
        imshow(gau);
        New = im2double(gau);
        title('Gaussian Noise Image');
        
        
        % Perform Coiflet wavelet denoising
        % Set the denoising parameters
        waveletType = 'coif5';  % Coiflet wavelet type
        level =4;              % Wavelet decomposition level
        threshold = 's';    % Thresholding method
        thrSettings = 1.231;       % Thresholding settings
        % Perform wavelet decomposition
        [C, S] = wavedec2(New, level, waveletType);
        % Compute the noise estimate using universal thresholding
        sigma = median(abs(C)) / 0.6745;
        % Apply the thresholding
        D = wthresh(C, threshold, sigma * thrSettings);
        % Perform inverse wavelet transform to obtain the denoised image
        denoisedImage = waverec2(D, S, waveletType);
        
        % Display the denoised image
        subplot(2,3,6);
        imshow(denoisedImage);
        title('DWT Coif Denoised Image');
        
        mse_coif = immse(denoisedImage , New2)
        psnr_coif = psnr(denoisedImage , New2) 
        
    else
        for count=1:waitMax
            %wait 20 seconds to the image 
            pause(1)
            if exist(fullFileName, 'file')
                k=k-1;
                break; %Exit this loop (return to the main loop)
            end
        end
    end
    k=k+1;
end

[Walter Roberson]

Something must be assigning to either "abs" or "median"

[HARSHA BHONSLE]

how can i change it? it works well manually but not automatically!

[Walter Roberson]

After you have run and reached the error, please show

which abs median

I suspect that there are additional lines in your code that you did not show us

Answer 1

[Aniket]

Hi @HARSHA BHONSLE,

I tried to reproduce the issue by uploading 2 images into the folder, but the given code ran properly without any errors.

As mentioned by Walter, the error message you are encountering suggests that "abs" or "median" must be assigning itself to some variable and is being treated as an array instead of built-in function.

You can check what function or file is MATLAB referring to using the below command.

which abs median 

Additionally, I recommend including the following checks to ensure there are no issues with the input image and algorithm:

1. Show image in figure window using “imshow” function just after reading it.
2. Before proceeding with the wavelet decomposition, you can check if the image contains any ‘NaN’ or ‘Inf’ values.
3. Ensure that the wavelet decomposition is performed correctly. You can check if the wavelet decomposition vector returned is empty as shown below:

[C, S] = wavedec2(New, level, waveletType); 
if isempty(C) 
    error('Wavelet decomposition failed.'); 
end 

You may add try/catch blocks if any of the above checks fail.

Hope this helps!