Open Access3D fingerprint modelling and synthesis
Author(s) -
Long Shuqin,
Li Shuiwang,
Zhao Qijun,
Song Wanzhong
Publication year - 2015
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
eISSN - 1350-911X
pISSN - 0013-5194
DOI - 10.1049/el.2015.0382
Subject(s) - fingerprint (computing) , artificial intelligence , computer science , principal component analysis , fingerprint recognition , pattern recognition (psychology) , process (computing) , computer vision , set (abstract data type) , sampling (signal processing) , fingerprint verification competition , filter (signal processing) , programming language , operating system
Two‐dimensional (2D) synthetic fingerprint images have been successfully used for evaluating large‐scale automated fingerprint identification systems (AFISs). However, they are limited in assessing the whole process involved in an AFIS, particularly fingerprint image acquisition and fingerprint deformation. Hence, it is desired to develop synthetic 3D fingerprints and 3D fingerprint phantoms. In this reported work, a first attempt to establish a statistical shape model of 3D fingerprints is made by first re‐sampling and aligning a set of training 3D fingerprint data and then applying the principal component analysis method. On the basis of the proposed model, synthetic 3D fingerprint shapes can be generated randomly. By further mapping fingerprint textures onto the shapes, synthetic 3D fingerprints can be obtained. Example results are provided demonstrating the effectiveness of the proposed model and method. The model and code will be publicly available for academic purposes.