Abstract
In forensic applications the evidential value of palmprints is obvious according to surveys of law enforcement agencies which indicate that 30 percent of the latents recovered from crime scenes are from palms. Consequently, developing forensic automatic palmprint identification technology is an urgent and challenging task which deals with latent (i.e., partial) and full palmprints captured or recovered at 500 ppi at least (the current standard in forensic applications) for minutiae-based offline recognition. Moreover, a rigorous quantification of the evidential value of biometrics, such as fingerprints and palmprints, is essential in modern forensic science. Recently, radial triangulation has been proposed as a step towards this objective in fingerprints, using minutiae manually extracted by experts. In this work we help in automatizing such comparison strategy, and generalize it to palmprints. Firstly, palmprint segmentation and enhancement are implemented for full prints feature extraction by a commercial biometric SDK in an automatic way, while features of latent prints are manually extracted by forensic experts. Then a latent-to-full palmprint comparison algorithm based on radial triangulation is proposed, in which radial triangulation is utilized for minutiae modeling. Finally, 22 latent palmprints from real forensic cases and 8680 full palmprints from criminal investigation field are used for performance evaluation. Experimental results proof the usability and efficiency of the proposed system, i.e, rank-1 identification rate of 62% is achieved despite the inherent difficulty of latent-to-full