Face recognition is a very challenging problem. Images that appear like humans can be very difficult to match under different environment changes (lighting, background, clothing, etc). Furthermore, recognizing faces that might have altered (shaved off beard, sunglasses, chin and nose operation) requires expanding the basic image with more possible alternations. Therefore from one image of a terrorist, five thousand can be easily generated. With so many images, easily approaching the number of millions, a feasible framework capable of processing the images almost in real-time is required.
It is also worth noting that to date, no work using a distributed system in face recognition technique has been found and this work investigates the aspects of such system. The work shows how such system is modeled, and its workings.