A, The MULTIPARAMETER BIOMETRIC PERSONALITY IDENTIFICATION

Abstract

The article is devoted to the development of a system for biometric identification of a person by face, fingerprints and voice. As informative signs of biometric identification of a person by face, two-dimensional and three-dimensional characteristics of a person's face were used, taking into account the area and volume. A sophisticated identification algorithm has been developed to take into account such phenomena as portrait shift, different photo scales and the tilt of the identified person.

For biometric identification of a person by fingerprints, an FPM10A scanner and an Arduino microcontroller were used. Identification signs are based on the analysis of the structure of papillary patterns on the finger: type and type of papillary pattern; direction and steepness of streams of papillary lines; the structure of the central pattern of the pattern; delta structure; the number of papillary lines between the center and the delta and many other signs.

Each print can contain up to 70 or more minutes. For biometric identification of a person by voice, the MFCC and PLP algorithms are used for digital processing and analysis of audio recordings. Various algorithms are used for acoustic analysis of speech: hidden Markov models, a model of a mixture of Gaussian distributions. The result of determining the tone of speech and the content of speech for the purposes of identification by voice is obtained. The “Multiparameter automated system of biometric identification of a person” has been developed on the Visual FoxPro DBMS.