Premium
The Effect of Image Alterations on Identification Using Palmar Flexion Creases
Author(s) -
Cook Tom,
Sutton Raul,
Buckley Kevan
Publication year - 2013
Publication title -
journal of forensic sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.715
H-Index - 96
eISSN - 1556-4029
pISSN - 0022-1198
DOI - 10.1111/1556-4029.12249
Subject(s) - minutiae , identification (biology) , artificial intelligence , computer science , fingerprint (computing) , biometrics , matching (statistics) , pattern recognition (psychology) , rotation (mathematics) , computer vision , filter (signal processing) , noise (video) , translation (biology) , image (mathematics) , fingerprint recognition , mathematics , statistics , biochemistry , botany , messenger rna , gene , biology , chemistry
Palmprints are identified using matching of minutia points, which can be time consuming for fingerprint experts and in database searches. This article analyzes the operational characteristics of a palmar flexion crease ( PFC ) identification software tool, using a dataset of 10 replicates of 100 palms, where the user can label and match palmar line features. Results show that 100 palmprint images modified 10 times each using rotation, translation, and additive noise, mimicking some of the characteristics found in crime scene palmar marks, can be identified with a 99.2% genuine acceptance rate and 0% false acceptance rate when labeled within 3.5 mm of the PFC . Partial palmprint images can also be identified using the same method to filter the dataset prior to traditional matching, while maintaining an effective genuine acceptance rate. The work shows that identification using PFC s can improve palmprint identification through integration with existing systems, and through dedicated palmprint identification applications.