Dynamic Behavior Analysis in Compressed Fingerprint Videos

Abstract

Traditional fingerprint acquisition is limited to single-image capture and processing. With the advent of faster capture hardware, faster processors, and advances in video compression standards, newer systems can capture and exploit video signals for tasks that are difficult using a single image. In this paper, we propose the use of fingerprint video sequences to investigate detection of two aspects of dynamic behavior of fingerprints. Specifically, we are interested in the detection of distortion of fingerprint impressions due to excessive force and the detection of positioning of fingers during image capture. These two issues often lead to difficulties in establishing a precise match between images acquired from a finger. The proposed techniques use fingerprint video sequences to investigate dynamic characteristics of fingerprints across frames. Our approach directly works on MPEG-{1, 2} encoded fingerprint video bitstreams. The inter-field flow estimate is used to investigate temporal characteristics of the behavior of the fingerprints. The joint temporal and motion analysis leads to reliable detection and characterization of relative finger position and pressure in streamed sequences. A significant advantage of our approach for distortion analysis is that it does not involve decompressing the video stream. The proposed methods have been tested on the NIST-24 live-scan fingerprint video database and the results are promising. We also describe a new concept called the "resultant biometrics" - a new type of biometrics which has both a physiological, physical (e.g., force, torque, linear motion, rotation) component and/or temporal characteristic, added by a subject to an existing biometric. This resultant biometric is both desirable and efficient in terms of easy modification of compromised biometrics and is harder to produce with spoof body parts.