opencv - How to remove similar images based on hog features? -


i have 5000 images , each image can generate vector 1000 dimensions(hog feature), of images similar want remove similar ones. there way achieve this?

===============================================================

as @thedarkside ofthemoon suggested, let me explain little bit more trying do. using svm + hog features image classification. have prepared training data of training images similar want remove similar ones reduce computation cost. don't know if removal of similar images has side effect on final classification rate criteria of 'similarity' must found. that's trying do.

in way(not using hog features) can compute color histogram each image , compare against others.

like,

  1. get first image , compute histogram,

  2. now each other images calculate histogram , compare first one.

if find close match on histogram can discard it. , using cv_comp_correl match in range of 0-1.


Comments

Popular posts from this blog

javascript - RequestAnimationFrame not working when exiting fullscreen switching space on Safari -

Python ctypes access violation with const pointer arguments -