In our everyday lives, our face is an important part of who we are and how people identify us. It would be very difficult for us to interact with and identify others if everyone looked the same. The face is arguably a person's most unique physical characteristic. Other than identical twins, each person's face has some unique characteristics that can be distinguished, but it is a lot more complex for the computer to quantify these features for recognition.

Facial recognition generally consists of two parts; face detection and face identification. Face detection is the computer's ability to be able to detect a portion of the video image as an individual's face from a complex background. This extraction alone is a challenge for the software algorithm. Once the location and area of the face within an image is identified, the algorithm will need to extract enough critical features within the face and compare these features from a database to accurately identify the individual. The commonly used technique for this analysis process is called Local Feature Analysis. There are other techniques for facial identification such as the eigen face technique, but the LFA and its variants are more popularly used by most practical implementations today.

Advantages

• Easy to use and requires no special training or equipment.
• Relatively inexpensive hardware compared to other biometrics.
• Completely non-intrusive for user to use.

Disadvantages

• It cannot be used to distinquish identical twins.
• It can have high False Acceptance Ratio for people with similar features such as family members and relatives.
• A person's facial features can change over time.
• Environmental lighting condition might influence the recognition process.

For more information on this technology, please look at our face recognition reference books section for our recommendations.

At BioVeriCom, Inc., we are developing facial recognition technology to augment other biometric technology for additional security. This product is still under development.