Interdigital palm region for biometric identification

• First experimental study on the ridge pattern distribution in the ‘interdigital palm region’ for biometrics.
• Novel classification methodology of palms according to five classes.
• Study of complementarity of the interdigital and traditional palm regions.
• Evaluation of personal recognition using interdigital region reaching EER = 0.01% (on database on 416 subjects).
• Interdigital palm region image database from 416 subjects in this paper is made publicly available.

The interdigital palm region represents about 30% of the palm area and is inherently acquired during palmprint imaging, nevertheless it has not yet attracted any noticeable attention in biometrics research. This paper investigates the ridge pattern characteristics of the interdigital palm region for its usage in biometric identification. An anatomical study of the interdigital area is initially carried out, leading to the establishment of five categories according to the distribution of the singularities and three regions of interest for biometrics. With the identified regions, our study analyzes the matching performance of the interdigital palm biometrics and its combination with the conventional palmprint matching approaches and presents comparative experimental results using four competing feature extraction methods. This study has been carried out with two publicly available databases. The first one consists of 2,080 images of 416 subjects acquired with a touchless low-cost imaging device focused on acquiring the interdigital palm area. The second database is the publicly available BiosecurID hand database which consists of 3,200 images from 400 users. The experimental results presented in this paper suggest that features from the interdigital palm region can be used to achieve competitive performances as well as offer significant improvements for conventional palmprint recognition.