Answer the question
In order to leave comments, you need to log in
How to compare 2 pictures for similarity?
My task is to compare 2 pictures and say if they are similar. First, I find the histograms of each image and compare them with a measure of similarity. Then I compare the obtained distance with a certain threshold, if it is less than the threshold, then I calculate the AKAZE descriptor, filter by the hamming distance and the RANSAC filter. then I check that after filtering the connections there are more than 6, if more, then I say that the images are similar. how do you like my approach? is he correct?
Answer the question
In order to leave comments, you need to log in
Maybe a neural network? And according to the histogram, completely different pictures can match.
https://github.com/JohannesBuchner/imagehash - source in python, but described in some detail, can be transferred to another PL if desired.
It checks exactly the visual similarity of the pictures - quite accurately ..
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question