Facial recognition, or face recognition, is a type of biometric that identifies individuals in a digital image. Biometrics refers to the identification by their characteristics or traits. This process is performed by analysing and comparing an individual’s patterns. Facial recognition systems are commonly used for security purposes, but are increasingly implemented in other areas. For instance, Microsoft Kinect motion gaming system uses facial recognition to differentiate between the players.
There are three different techniques used in facial recognition:
- Traditional Recognition
- Three-Dimensional Recognition
- Skin Text Analysis
Number One – Traditional Recognition
The traditional facial recognition system uses algorithms identifying facial features. It extracts landmarks and features from an image of the subject’s face. For example, the algorithm analyses the position, size and shape of your eyes, nose, cheekbones and jaw. Recognition algorithms can be divided into two main approaches: geometric and photometric. Geometric, looks at the distinguishing features of the face. Photometric, however, distils the image into values and then compares those with a template to eliminate inconsistencies.
Number Two – Three-Dimensional Recognition
Using three-dimensional recognition is an emerging trend, claimed to improve recognition accuracy. This technique uses 3D sensors to capture information about the face’s shape. The information is then used to identify distinctive features of that face, such as the outline of eye sockets, chin, nose and mouth.
One major advantage of this type of facial recognition is that, unlike most techniques, it is not affected by changes in lighting. Furthermore, this system can identify a person without having a profile shot of the person’s face. Using built-in sensors, 3D Recognition works by projecting a structured light onto a person’s face and creating a profile for the person.
Number Three – Skin Texture Analysis
Skin texture analysis is another emerging technique that uses the visual details of the skin to identify the individual. This technique turns the unique lines, patterns and spots on a person’s face into a mathematical equation. Tests have shown that, with the addition of this type of analysis, performance in recognising faces has improved by twenty-five percent.
Facial recognition has become the future of security. As an example, London police installed facial recognition technology into all cameras in the Borough area. In the seven years that this technology has been in use, the crime rate on the streets has decreased by thirty-four percent. Remember, using this type of technology is no longer limited to casinos and street corners, install it at work and home, and ensure your safety.
I am Greg Jones, a mortgage bond specialist for a local financial institution. In my spare time, I love to research and blog about control systems integration and home security features.
