How to use K-Mean Algorithm to segment and cluster Binary image

3 views (last 30 days)
hi please i need your help to segment and cluster binary image using K-means algorithm i need to understand how to deal with the features of the binary image please help me i am in deep trouble i cant finish my research thank you in advanced

Accepted Answer

Image Analyst
Image Analyst on 10 Apr 2017
What would your feature(s) be for the image? The locations? The gray levels (doesn't make sense though). Why do you think you need kmeans()? Why don't you let us see your binary image and tell us how many classes you want, and which blobs you want in each class. We can't say much otherwise.
  5 Comments
Image Analyst
Image Analyst on 6 May 2017
I don't conduct free consultations on the side, only here. I'm attaching a kmeans demo and a document thresholding demo. Maybe they will help.

Sign in to comment.

More Answers (1)

moahaimen talib
moahaimen talib on 6 May 2017
Dear sir i am using this image i need i made a binary i need to extract each character then recognize using deep neural first i read your answer of line segmentation this is the url
https://www.mathworks.com/matlabcentral/answers/181941-i-need-a-matlab-code-for-segmentation-of-text-lines-in-a-scanned-document-image-using-projection-pro
then i need to make these algorithm steps
1-Lines Segmentation 2-Words Segmentation 3-Characters Segmentation 4-Characters Segmentation
i wish you can help me with that sir thank you

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!