Introduction
The Facial Recognition Software is a program that identifies people using face images
taken of them with a camera or video and comparing the measurements of the face to
measurements of other faces in a database. The facial recognition software was created in
the 1960s. At first, the software calculated distances and ratios of points in a face including
the eyes, nose, ears and mouth. Then other features were taken in consideration such as the
hair color and lip thickness. In 1993 the Face Recognition Technology Evaluation
(FERET) was created and it resulted in the development of face recognition mathematical
formulas and algorithms to identify individuals. In 2002 the Face Recognition Vendor
Tests (FRVT) were performed in order to determine if these software’s could be
commercially available and to provide training of how to use these systems. Lastly, in
2006 the latest face recognition software became available including the high-resolution
face images and 3D scans.
In this report, I will explain how the facial recognition software works, who uses it and
how it is used, what problems are intended to solve, and its limitations.
How does Facial Recognition Software works
The Facial Recognition Software is a computer program used to identify people. The
software can recognize a person within a group of people or alone, captures the image of
the persons face through a digital image, and identify who is the person by running the
image or video through an image database. Some of the different types of techniques and
software used to identify a person are as follows:
Traditional Facial Recognition Software:
The traditional facial recognition software works by capturing an image of a person’s face
and measuring the nodal points of the person’s face. The nodal points are based on the
facial features and landmarks of the face such as the length and width of the nose, the
depth of the eye sockets and the distance between the eyes. The numbers of nodal points in
a human face that the software identifies are 80. Then, number of nodal points is
transferred to the database as a numerical code called faceprint. In the database, the
faceprint from the individual is compared to other numerical codes (faceprint) stored in it
in order to identify the person.
Three-Dimensional (3D) Facial Recognition Software:
The three-dimensional (3D) facial recognition software captures a three-dimensional (3D)
image of the person’s face. Then, measures the curves of the face on a sub-millimeter scale
and creates a template which is translated into a code that represents the features on a
