A video surveillance system for fingerprint acquisition

In fingerprint-based watch-list systems used by law enforcement, immigration, and military agencies, the targeted users (such as criminals, illegal immigrants, terrorists) have strong incentive to cheat the fingerprint system. A simple but effective method to cheat the system is to present the wrong fingers or even not a finger (e.g., palm). The current solution to prevent such behavior is to let the official closely monitor the fingerprint acquisition process. However, automatic solution is desired because the number and scale of fingerprint-based watch-list systems are rapidly increasing while human labor is expensive. We propose a video surveillance system to ensure the validity of input fingers. The proposed system captures the region around the fingerprint scanner using a webcam, determines whether the posture of the hand is valid, and estimates finger position (left index, right middle). Experimental results demonstrate the validity of the proposed algorithm.