assume that I have a set of image and segmented image of a object in image (training set). Now, I want to measure features of object in image based on the training data. But I don't know how many features in image. And which is best feature for image segmentation? Gradient feature or entropy feature or ....Could you explain help me ?

  • 1
    $\begingroup$ When you say I don't know how many features are in an image, this makes no sense. Images don't have features inherently. You, as an engineer, select features you want to extract from images. Can you elaborate on this point? $\endgroup$ – Phonon May 16 '14 at 20:56
  • $\begingroup$ @Phonon: It means that image has many features ex mean, std, edge,...But which is more popular for image classification? $\endgroup$ – John May 17 '14 at 13:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.