Deep learning‐based approach to latent overlapped fingerprints mask segmentation. Issue 11 (1st November 2018)

• Record Type: Journal Article
• Authors: Stojanović, Branka; Marques, Oge; Nešković, Aleksandar
• Is Part Of: IET image processing. Volume 12:Issue 11(2018)
• Page Start: 1934
• Page End: 1942
• Publication Date: 2018-11-01
• Subjects: fingerprint identification -- image classification -- image segmentation -- learning (artificial intelligence) -- feedforward neural nets; deep learning‐based approach -- overlapped region class -- single region class -- background class -- image block classification -- convolutional neural networks -- subsequent verification -- fully automated fingerprint separation solutions -- criminal investigations -- civil applications -- latent overlapped fingerprints mask segmentation; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2017.1227
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : Overlapped fingerprints can be potentially present in several civil applications and criminal investigations. Segmentation of overlapped fingerprints is a required step in the process of fingerprint separation and subsequent verification. Overlapped fingerprint segmentation is performed manually (and the resulting manually drawn masks are a required additional input) in all of the overlapped latent fingerprints separation approaches in the literature, which make them only semi‐automatic. This study proposes a novel overlapped fingerprint mask segmentation approach, thereby filling that gap in the development of fully automated fingerprint separation solutions. The proposed method uses convolutional neural networks to classify image blocks into three classes – background, single region, and overlapped region. The proposed approach shows satisfactory performance on three different datasets and opens the door for full automation of fingerprint separation algorithms, which is a very promising research area.