Latent dactyloscopy pairing: presentation attained through feedback from EPITOME. (24th March 2021)

• Record Type: Journal Article
• Authors: Pugalenthi, R.; Vinnarasi, F. Sangeetha Francelin; Kumar, P. Mohan; Renjith, R. Arokia
• Is Part Of: International journal of bio-inspired computation. Volume 17:Number 2(2021)
• Page Start: 85
• Page End: 94
• Publication Date: 2021-03-24
• Subjects: candidate record -- feature filtration -- latent fingerprint pairing -- epitome feedback; Biologically-inspired computing -- Periodicals; Computational biology -- Periodicals; 572.0285
• Languages: English
• ISSNs: 1758-0366

Abstract

Latent fingerprints are recognised as evidence against offenders in law enforcement agencies. The automatic fingerprint identification system (AFIS) method can be applied to rolled and plain fingerprints, which gives results of high accuracy. The latent accuracy will be less due to background noise and poor ridge quality. Also, pairing of latent imaging is a challenging issue. Keeping these in view, this work suggests the pairing of top-down data or feedback from an epitome to filter and to develop a latent pairing accuracy. The exhibited results show that the combination of the proposed feedback pattern with a latent pair of state-of-the-art develops a finding accuracy for WVU datasets and latent NIST SD27 against a background dataset of 100X epitome by 1 to 3.5 percentage. This proves that the combined feedback pattern with the latest results can be carried out in real-time.