0

I'm just starting a project using OpenCv to extract SIFT features from multiple images. Then I will make a database for training machine with SVM.

Untill now, I studied this code which is very basic. But i would like to know, how should i proceed to convert the matrix of features into a normalized, readable archive for SVM?

What is the best approach to extract features from hundreds of images? (later objective)

Thank you!

Mailerdaimon
  • 6,003
  • 3
  • 35
  • 46
user2967127
  • 41
  • 1
  • 8
  • Have you tried to use search? Such question have been asked not once, e.g. http://stackoverflow.com/questions/20041355/how-to-use-sift-surf-as-features-for-a-machine-learning-algorithm/20044974#20044974 – old-ufo Nov 21 '13 at 12:20

1 Answers1

0

If you can elaborate more on the specific problem you are trying to solve, it would be easier to answer.

However, I can suggest using the Bag of Words approach of which you can read about here: http://gilscvblog.wordpress.com/2013/08/23/bag-of-words-models-for-visual-categorization/

GilLevi
  • 2,117
  • 5
  • 22
  • 38