Computer Gets Help from the Human Brain
The interface between a human brain and a Computer are, in general, designed to help people mainly who are disabled, whereas the latest project is making use of the inverse interface in which a computer is taking help from the human brain for completing its tasks which they cannot manage on their own.
The interface was used by researchers for sorting through satellite images to make the surface-to-air missiles faster than any other machine or human analyst that may manage to complete the task alone.
An associate professor at Columbia University, Paul Sajda stated, “With Google, there is a need to type in words to describe what exactly is running in our mind and suppose I’m looking for something ‘funny looking’.”
She describes that computers are struggling in classifying the images according to the type of concept of the abstract, where this work can be finished by the human brain almost instantly. Electrical signals within our brain react within no time that a person could not even realize that he has recognized an odd or unusual image.
An electroencephalogram (EEG) cap is used in an instrument called C3Vision (Cortically Coupled Computer Vision) to scout out the activities of the brain as the person wearing it is shown about 10 images for a second. Machine-learning algorithms are trained for detecting the neurological signals that signify interest in an image is used for the analyzing the activities of this brain. By monitoring these signals, the system rapidly ranks the images according to the level of interest they appear to the viewer. Then the search is refined by bringing back other images that are similar to those which are ranked high in the priority list. Sajda said, “A search tool helps you to find images that are very similar to the ones which have to steal your attention,”.
The conscious brain could not be able to register a “Hit” to the speed at which it works. Whereas the neurological visual pathways react more faster than the brain. the brain generates some electrical signals which the 64 EEG electrodes within the cap could detect and decode that is on the edge of the subconscious.
Sajda along with his colleagues at Columbia have started up an offshoot company called Neuro-matters to commercialize the technology with $4.6 million in funding from Defense Advanced Research Projects Agency. Not only for military applications, even neuro-marketing and advanced gaming interfaces could be possibly included in this applications which make use of the computer – brain interfaces. “It could be used to get demographic feedback on how much an advert could grab people’s attention,”. For more information, click here.