Is it possible to work on image dataset of different sizes for CAD system using supervised learning modules, i.e., I have a dataset of brain tumour images with a different size of tumours, they are cropped based on the area of thses tumours. texture features should be extracted from these images and feeded to SVM classifier. is it applicable? or they should be in fixed sized?

  • 1
    If your features are scale-invariant, you don't need to resize the images – ignatius Dec 3 at 15:28
  • is the uniform local binary patterns considered scale-invariant? – gin Dec 4 at 16:15
  • I don't think so... If I can remember, they are rotation invariant – ignatius Dec 4 at 16:18
  • You can perform several test by computing uLBPs on different scales and guess if they are scale invariant.... – ignatius Dec 4 at 16:18
  • great, what feature descriptors you suggest for scale invariant? – gin Dec 4 at 16:30

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.