Why do i get an error when rgb image (tiff) is converted to grayscale ? The value of such image in workspcespace is like <556x663x4 uint8>. Rgb handles only images of type <556x663x3 uint8>.

I am doing a project in which the grayscale value for a particular set of image is calculated. In an image(tiff format), there six sub images which are obtained by cropping the original image in microsoft office. After cropping during the processing of these six images, I face the issue of that some of the images not being read by matlab because of value of image is 556x663x4 uint8. The image being read are of the type 556x663x3 uint8. So, I want to know whether the issue is with cropping of tiff or with something else. THanks in advance.

[KSSV]

Try readRGBAImage

Check the link as well: https://in.mathworks.com/matlabcentral/answers/66807-what-does-4d-of-an-image-indicates

[Sumit Chaudhary]

I have read about RGBA. The main point is that the original image(taken form laserjet Hp scanner) has six sub images and while processing some are processed while others are not because of the 4th dimension. Why after cropping are the six sub images are not of the same type ? Matlab forum do not support the tiff images that is why not uploaded the images,otherwise it easier to deal with the situation.

[KSSV]

You zip the image and upload.

[Sumit Chaudhary]

Thanks for the suggestion. In the zip file, I have included the original image with six sub images and code for gray-scale value.

Answer 1

[KSSV]

    You can crop the image in MATLAB itself and do what you want. Check the below code.

      I = imread('original.tif') ;  % your original image 
      av = zeros(1,6) ;
      for i = 1:6   % loop for each block 
          imshow(I) ;
          [Image, rect] = imcrop(I);  % select the image you want by dragging a rectangle 
          imwrite(Image,strcat(num2str(i),'.tif')) ;  % save the cropped image 
          [pixelCount, grayLevels] = imhist(Image);
          bar(grayLevels, pixelCount);
          grid on;
          av(i) = mean2(Image); 
      end

[Sumit Chaudhary]

Thanks for the help.

[KSSV]

Thanks is accepting the answer. Enjoy MATLAB. :)