Hello, first of all thank you for this good work, however I have a couple of queries when making the 1:1 comparison, it gives me a similarity below 30% in most cases. But when I see the matches of the minutiae, they are very similar, what could I be doing wrong?
Loading...
Template 1: size 2976 bytes, #fingerprints 1
Template 2: size 2640 bytes, #fingerprints 1
Template 3: size 2896 bytes, #fingerprints 1
Matching...
Similarity of 1 and 2: 28% [in 531us]
Similarity of 1 and 3: 20% [in 521us]
What would be a good percentage of coincidence, or what would have to be improved for a higher level of similarity?
based on this paper
Minutiae-based Matching Scheme for Fingerprint Identification using the Grassfire Spot Matching
they use 10 minutiae to match fingerprints, we can take this as correct valué for this implementation ?
Fingerprints have usually 20 to 30 minutiae and two fingerprints are considered to be matched if they have more than 10 matched minutiae.
Thanks so much @neilharan
Hello, first of all thank you for this good work, however I have a couple of queries when making the 1:1 comparison, it gives me a similarity below 30% in most cases. But when I see the matches of the minutiae, they are very similar, what could I be doing wrong?
What would be a good percentage of coincidence, or what would have to be improved for a higher level of similarity?
based on this paper
Minutiae-based Matching Scheme for Fingerprint Identification using the Grassfire Spot Matching
they use 10 minutiae to match fingerprints, we can take this as correct valué for this implementation ?
Thanks so much @neilharan