One of the challenging field in Artificial Intelligence is Face recognition,  People can recognize the emotion others by looking at other faces, and not more than a second it will know whether it's angry, happy, or sad. But not with a computer, it need a complex algorithm to know whether the image contain happy faces, sad or else.

So when someone come with other ideas about Face Recognition surely it make one step closer to make a really smart computer (software).  That's happen  not so long ago, Researchers  At Department of Artificial Intelligence (DIA) of Universidad Politécnica de Madrid’s School of Computing (FIUPM) have, in conjunction with Madrid’s Universidad Rey Juan Carlos, has developed an algorithm that make a software capable of processing 30 images per second to recognize a person's facial expressions in real time, and categorize it into six prototype expression, anger, disgust , fear, happiness, sadness and surprise.

This software although it can process a sequence of images, and recognizing the expression,  even still a prototype, it can be run on desktop computer or even a laptop. But of course it won't run on our pentium 166 Mhz.

The prototype analyzes the face of person sitting in front a camera connected to a computer, then the system analyses the person's face  through several boxes, each “attached” to or focusing on part of the user’s face. These boxes monitor the user’s facial movements until they manage to determine what the facial expression is by comparison with the expressions captured from different people (333 sequences) from the Cohn-Kanade database.

So far it has 89% success chances, not bad for a prototype.

One thing that comes to mind is this software has a really good chance to adopt in video chatting to analyze whether our chatting friend is happy or not, or perhaps got to do with e-commerce solutions, and also put that in front of the cashier store, to check the people and analyze whether the customer happy or feel bad after they do some shopping.

The results of this research were published in the January issue of Pattern Analysis and Applications in an article authored by Luis Baumela and Enrique Muñoz, of the DIA, and José Miguel Buenaposada, of the Universidad Rey Juan Carlos de Madrid’s College of Computer Systems Engineering. http://www.springerlink.com/content/q075h33723m475k1/?p=d626d623f4294fde878ed45706cd3971π=2