A new spontaneous pupillary oscillation-based verification system

• Published in: Expert systems with applications Vol. 40; no. 13; pp. 5352 - 5362
• Main Authors: Villalobos-Castaldi, Fabiola M., Suaste-Gómez, Ernesto
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01.10.2013; Elsevier
• Subjects: Applied sciences; Artificial intelligence; Biometric authentication; Biometrics; Computation; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Exact sciences and technology; Expert systems; Memory and file management (including protection and security); Memory organisation. Data processing; Methodology; Oscillations; Pattern recognition. Digital image processing. Computational geometry; Pupillary hippus; Pupils; Software; Spatio-temporal biometric template; Spontaneous; Spontaneous pupillary oscillations; Video-based verification system; Spontaneous pupillary oscillations; Biometric authentication; Video-based verification system; Pupillary hippus; Spatio-temporal biometric template; Biometrics; Capability index; Liveness; Video signal; Error rate; Addressing; Contraction; Pupil; Luminance; Dilatation; User interface; Image sequence; Authentication; Database; Computer security
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2013.03.042

•It is introduced a new pupillary-based verification system.•It is presented the initial identity authentication results based on the spatio-temporal features of the spontaneous pupillary oscillations.•It is introduced a new pupillary-based verification system.•The authentication results based on the spatio-temporal features.•It is described the methodology to compute a spatio-temporal biometric template.•This biometric has the capability to provide enough discriminative information.•The obtained EER was in the order of 0.2338%. A novel pupillary-based verification system is introduced, along with the early identity authentication results and analysis, based on the spatio-temporal features computed from the spontaneous pupillary oscillations. The authors demonstrate that this biometric trait has the capability to provide enough discriminative information to authenticate the identity of a subject. A new methodology to compute the spatio-temporal biometric template recordings of the pupil area changes, in a video-oculography sequence under constant luminance level, is also introduced in this paper. According to the authors’ knowledge, there is no evidence that other attempts were made, addressing this methodology to distinguish individuals based on the spatio-temporal representations, computed from the normal dilation-contraction behavior of the pupil. In this work, liveness will be detected by using the information obtained from the spontaneous pupillary oscillation mechanism. Preliminary experiments were conducted by using a particular own collected database, resulting in a (Equal Error Rate) in the order of 0.2338%.