HOW TO IDENTIFY MEAN AND STANDARD DEVIATION FOR THIS CODE?

Read image by its file name

  I = imread('.');
           I=imresize(I,[256 256]);
            I=im2double(I);
            imshow(I)
            
%         Convert RGB to HSV    
              img1=rgb2hsv(I);
            figure(), subplot(2, 2, 1), imshow(I), title('Original')
            subplot(2, 2, 2), imshow(img1(:,:,1)), title('Hue'), colorbar
            subplot(2, 2, 3), imshow(img1(:,:,2)), title('Saturation'), colorbar
            subplot(2, 2, 4), imshow(img1(:,:,3)), title('Value of Brightness'), colorbar
            
            H=img1(:,:,1).*255;
            S=img1(:,:,2).*255;
            V=img1(:,:,3).*255;
      
%   H(:,:,1) = H(:,:,1) * 2.5;


%         min. and max value of hsv   
           
             Hmin = min(H(:));
             Smin = min(S(:));
             Vmin = min(V(:));
             Hmax = max(H(:));
             Smax = max(S(:));
             Vmax = max(V(:));


            %Set the hue value to zero if it
            %less than 50 or great than 150
            H((H < 50) | (H > 150)) = 0;
            %Set the hue value of wheat straw
            % pixel to zero
              H(H > 49 & H < 60 & S > 5 & S < 50 & V > 150) = 0;
            %Thresholding
            T = 49; %T can be any value in [1, 49]
            t = T./255;
            BW = im2bw(H, t);
            %Delete the objects less than 100 pixels
            BW = bwareaopen(BW, 100);
            %Show the identification result
            figure, imshow(BW);

         % Calculate the area, in pixels, of binary image.
numberOfPixels1 = sum(BW(:));
% Another way to calculate it that takes fractional pixels into account.
numberOfPixels2 = bwarea(BW);

  % CALCULATE PARAMETERS        
  I =rgb2gray(I);
  I=double(I);
  BW =(BW);
%  % % Find the mean squared error
  mse = sum((I(:)-BW(:)).^2) / numel(I);
%  
%  % % now find the psnr, as peak=255
  psnr = 10*log10(255*255/mse);  

[Adam Danz]

TLDR

Mean and std of what variable?

[tashu Dabariya]

I and BW

[tashu Dabariya]

% CALCULATE PARAMETERS        
  I =rgb2gray(I);
  I=double(I);
  BW =(BW);
%  % % Find the mean squared error
  mse = sum((I(:)-BW(:)).^2) / numel(I);
%  
%  % % now find the psnr, as peak=255
  psnr = 10*log10(255*255/mse)
is it right way to calculatet psnr and mse?
for mean and std variable might be I and BW

Answer 1

[Adam Danz]

It looks like you want to calculate the mean squared error between the I and BW arrays which can be done using immse() from the image processing toolbox.

https://www.mathworks.com/help/images/ref/immse.html

err = immse(I,double(BW)); 

[tashu Dabariya]

error showed it must be have the same class

I is image is original

and BW is binary

for same class i convert original image(rgb) to gray

I =rgb2gray(I);
  I=double(I);
  BW=(BW);

is it right?

I am working on greenness identification i want to know how to find mean and std for identification

[Adam Danz]

I saw in your code, you converted I to double: I=double(I); In my answer, I also converted BW to double which would avoid that error.

[Adam Danz]

tashu, this is a different question than what you started with and I suggest writing a new question with a more focused title so it attracts experts in this area.

Also, providing code is a big help but please format it by using the format buttons while your editing the question. I've formatted your code so far but it saves us some time if you do it yourself. Thanks.