Facial Recognition : In Softwares and Living Creatures

Facial Recognition is an important and interesting feature of many animals and simulated Facial Recognition done by the softwares to detect One from others. Human has the ability of Facial Recognition of own species (like you can detect me looking different than your neighbor) as well as cross species (your dog does not look the same to you like your friend’s dog of same species, age and sex). Your dog will get scared and bark if you put a mud mask and colors on them and appear behind him / her. Yeap, dog has the exceptional ability of smell sensation, still they requires the help of Facial Recognition. Very minute or sophisticated differences on the facial features can only be detected by few experienced peoples. Mother of an identical twin can recognize two different persons.

Basics of Facial Recognition

Facial Recognition has various aspects, psychological (you can not accept that your wife’s suddenly, Miraculously looking different overnight and has been like your neighbor’s wife – you will scream if you see a different looking face beside you is sleeping), social interaction (a street dog you regular pat and give food will look at you face and waive its tail), religious and cultural (many tribes, monks uses colors, masks to hide their real look) and the idea later ported to softwares – Facial Recognition Softwares.

So, Facial Recognition is not so simple really. The ability to detect and distinguish between faces (Facial Recognition in human usually termed as visual perception) by the human brain acquired during the first months of life. The higher center for Facial Recognition is on the occipital lobe of the brain, with aging it deteriorates. A partial or complete absence of this ability Facial Recognition in human is called prosopagnosia. Facial Recognition for human needs 100% accuracy for being a normal human.

On industrial context, facial recognition is one of the biometric methods. It is used for safety, criminology and forensic usages for the purpose of identification of individuals. For a proper Facial Recognition, a normal human brain needs two ears and two eyes to be visible within one picture. This is the basic reason why the legal photos for proper Facial Recognition for identification is well known for being looking bad, artistically.

Facial Recognition Softwares

Facial Recognition softwares are used for automatically identifying or verifying a person from a digital image or a video frame. Linear Discriminate Analysis, Elastic Bunch Graph Matching and many algorithms defines how a Facial Recognition Software will work.

Most image organizing softwares has some basic Facial Recognition algorithm with over 20% error rate (false positive and false negative), which is actually far from being a standard. Professional Facial Recognition Softwares, claimed to have around 0.1% error rate. Facial Recognition Softwares can not work independently with 100% accuracy unlike fingerprint softwares as a simple dummy can fool most Facial Recognition Softwares.