Preparation of a Database of Faces to Obtain Reliable Results of Human Identification System at the Entrance

Nowadays, biometric identification methods are widely used in the fields of security and access control systems, based on the unique physical and behavioral characteristics of a person. One of these methods is facial recognition system that allows you to automate access to protected objects by using the computer vision system. Creating a database of faces of people who have an access to the object is an important part of the facial recognition system. Therefore, the preparation of a database to improve the reliability of the results of human faces identification is an urgent task. In this paper, the principal component method is used to implement the face recognition process and the viola-Jones algorithm is used to detect faces in images. The Euclidean metric was used to calculate the distances between the test images and the database images. As a result, it is shown that preprocessing the database can improve recognition results.