Automated Fingerprint Identification System: with and without the Possibility of Correction of a Digitalised Image

Abstract

According to the fact that systems for automatic processing of biometric data are constantly advancing in terms of speed and reliability, as well as in terms of adding new processing capabilities, the question of choosing the appropriate system becomes more important. In this paper the idea is to present the technical and technological solutions of the Automated Fingerprint Identification System with different operating principles, with and without the possibility of correction or coding of a digitized image. Comparisons of different systems were performed in test and production environments. The test database with 10 000 records and about half a million records of dactyloscoped persons in the production of database for testing the performance of search was used. The results have shown that there exists a statistically significant difference (p < 0.001) between examined systems in face fingerprint search according to latent fingerprint databases (which means indirect verification). In the production environment, it was found that there exists statistically significant difference (p < 0.001) in the direct and indirect verification showing advantages and disadvantages of the compared systems.