how to find connected component in an image

i'm doing a project to recognize kannada text,the first step says find connected components from a binarynimage.i tried doing it using bwconncomp but i'm not able to display the image.soo can u please help me with dis.

Answer 1

[Walter Roberson]

bwconncomp should work, or you could use bwlabel, or you could use regionprops

bwconncomp has an example where it reconstructs the binary image with one of the components missing.

For the moment I suggest you use bwlabel() and imshow() the result, as then the different components should come out in different colors.

Answer 2

[Image Analyst]

See my Image Segmentation Tutorial http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862

In short

% Do connected components labeling:
labeledImage = bwlabel(binaryImage);
% Let's assign each blob a different color to visually show the user the distinct blobs.
coloredLabels = label2rgb (labeledImage, 'hsv', 'k', 'shuffle'); % pseudo random color labels
% coloredLabels is an RGB image.  We could have applied a colormap instead (but only with R2014b and later)
imshow(coloredLabels);

[deeksha h r]

Error using bwlabel Expected input number 1, BW, to be two-dimensional.

Error in bwlabel (line 65) validateattributes(BW, {'logical' 'numeric'}, {'real', '2d', 'nonsparse'}, ...

Error in analysterer (line 3) labeledImage = bwlabel(binaryImage); this is the error i'm getting for using the above code,nd this error appears for many of the syntaxes i used.

[Image Analyst]

What does this say

whos binaryImage

It must be logical. How did you generate your binary image?

[Walter Roberson]

your binaryImage needs to be 2 dimensional. Watch out that your original image might be RGB when you are expecting it to be grayscale . In particular, if it is a JPEG image then even if it looks gray, it is much much more likely that it is RGB in which all of the colors happen to be shades of gray.

[Image Analyst]

try this possible fix

if ndims(binaryImage) > 2
    % It's color
    binaryImage = binaryImage(:,:,2) > 128;
else
    % It's gray scale, but maybe not 0 and 1.
    % Make sure it's logical, not 0 and 255 or something.
    binaryImage = binaryImage > 0;
end