What is this technology about?
It is a security oriented technology that identifies individuals from data obtained from their faces. Contemporaries include fingerprint scanner and iris recognition systems.
Among the different biometric techniques, facial recognition may not be the most efficient and reliable, but one remarkable advantage is the absence of cooperation required from the subject.
This makes identification of criminals in a crowd easier.
Recognition algorithms can be divided into two main types; geometric, which looks at distinguishing features, or photometric, which is a statistical approach that distills an image into values and compares the values with templates to eliminate variances.
An emerging trend which claims to have improved accuracy is the 3D facial recognition that uses sensors to capture the shape of the face. It is not affected by lighting and can identify faces from a range of viewing angles including a profile view.
However the recognition system is not perfect, especially in recognizing faces tilted more than 200 to the sides. Other conditions that affect efficiency include poor resolution images, poor lighting, sunglasses and long hair.
According to a source, a big smile can render the system less effective.
Some mobile phone manufacturing companies have adopted that as part of their security system, however there usually is an alternative if it fails as it is usually the most underdeveloped or efficient.
Even FBI uses it alongside other biometric technologies like iris recognition and fingerprint scanners.
Facial recognition, though fanciful, has not contributed much with regards to security when it comes to identifying live persons, but lots are underway in terms of recognizing persons in pictures. Picasa [Google] and facebook are a few of the many software giants that use it.
Even mobile apps have been designed to recognize faces from pictures and arrange them accordingly.
This implies that facial recognition will remain relevant especially in areas not exclusively tied to security. However much work needs to be done before it becomes very effective for standalone use with regards to security and identification of live and moving persons.
