Face Recognition Systems - Reliability and Credibility

Abstract

The biometric systems are part of our everyday lives. Many of us are in touch with them every day in order to carry out our work. However, we are unaware about their important qualities: reliability and integrity. Those qualities may in many cases affect our satisfaction with these tools and their eventual deployment to designated areas. The face recognition technology has emerged as an attractive solution to address many contemporary needs for identification and verification of identity claims. Face recognition systems are based on the anthropological minutiae on the face. One of the easiest methods used during the 1960s is the graphical method based on descriptive geometry principles made by the Soviets. This chapter may be divided into two main parts – measuring faces by the A4Vision biometric system and measuring faces by the analytical-statistical method. The chapter introduces reliability of the A4Vision system tested in laboratory conditions and shows the bases of this system – the analytical-statistical method applied to measurement of changes in the face of ten different women, and ten emotional changes of one woman, which were measured to obtain better understanding of programming requirements for better reliability of those systems.