0

I have this project in which i have a huge database of images divided into certain set.

Now I have this one image which is not exactly the same, but is nearly the same as one or many images present in database. I need to predict that this image may belong to which set.

I must admit that I am very new to image processing. However I have a little knowledge about SURF. SIFT, image fingerprinting, but I am not able to select which way to go.

I was thinking if I could save certain features of image in a database then again find some key points of example image then compare it in database and set some threshold to compare them. Please help me how should I start.

Shubham Paramhans
  • 85
  • 9
  • possible duplicate of [Checking images for similarity with OpenCV](http://stackoverflow.com/questions/11541154/checking-images-for-similarity-with-opencv) – Sam Aug 08 '13 at 09:54
  • There is a ton of similar questions (and answers) on this topic, here and in the wild... – Sam Aug 08 '13 at 09:55

1 Answers1

0

I think, you need to assign some hash code to every image from database (e.g http://www.phash.org/ ) and then search rough using this hash. After rough search, you can apply more expensive methods like descriptors comparsion.

Andrey Smorodov
  • 10,649
  • 2
  • 35
  • 42