Identity Verification Using Combined Biometric Signature Comparison Algorithms

This paper describes a method for verifying a person’s identity based on their biometric signature. As part of our work, we collected the most important publications of the last years in this area and then attempted to implement and run them on real data. We established a method for combining the results of different algorithms into a single result that minimizes False Positives while allowing for a slightly larger False Negatives. We also added a proprietary method based on a modified DTW algorithm. We have tested everything on a database of over 1.3 million signatures available for our research. In this paper we present the test results and quality parameters of our final algorithm.