Facial Recognition System

Information security is an important field in keeping track of everything related to and not limited to the virtual world which is the internet. And of the important aspect of the virtual reality world is that you have to know what and who you are dealing with in order to not being a victim of the ‘Ghost’ in the other end of the virtual realities world. Other than that, it is also crucial for us to immediately identify a person so we have time to judge whether to trust or not to trust. That sort of situation gave a birth to a facial recognition system.
Facial recognition system has been used by organizations like Federal Bureau of Investigation (FBI), Central Intelligence Agency (CIA), Facebook and other big companies such as Apple, ASUS and

For example, it can calculate the Euclidean distance between the eyes, the size or the shape of the eyes, the nose, cheekbones, and the jaw [1]. All this extracted features will then be used to search other image which has similar features or landmarks [2]. This method used by a Facial Recognition system is called a traditional approach. The algorithm using this kind of method then classified into two major approaches which is geometric and photometric. A geometric-based algorithm will classify and identify a face based on the landmarks while a photometric-based algorithm will transform an image into values and then compares the values by subtracting it with

LDA aims to reduce dimensionality while at the same time trying to preserve as much of the class discriminatory information as possible. In general, LDA will classify objects into one of two or more groups based on a set of features that describe the objects.

2.3.1.3 Elastic Bunch Graph Matching (EBGM)

EBGM is a much more advanced face recognition algorithm. The EBGM algorithm needs example of what the jets looks like to locate the landmarks in an image. Therefore we select the jets manually from the image and the algorithm will automatically generate a graph called face bunch graph, where each node of the bunch graph corresponds to a facial landmark and contains a bunch of model jets that has been extracted from the model image [8].

2.3.2 3-Dimensional Face Recognition

The performance of the 3 Dimensional Face Recognition was first tested in 2006 during the Face Recognition Vendor Test event where 3D face scans was done on six 3-D algorithms. According to Jonathon Philips, the 3-Dimensional face recognition can