Mapping of Intellect Vision to Picture Pixels Using EEG Based Model

In recent times, crime detection and investigation has seen major innovation largely due to the need to stay a step ahead of the perpetrators of the crime. An area that hasn’t seen the light of the day is criminal identification. Criminal identification relies largely on the verbal description of witnesses and the interpretation of sketching artists, both of which can be erroneous. To address this need, we propose a thought-to-picture interpreter. It requires a person to visualize a picture in his mind while the same picture is mapped onto the screen. It consists of analysing datasets obtained from Electroencephalography (EEG) signals and mapping it to Red, Green, Blue (RGB) pixels of an image with the help of an algorithm. To collect the EEG data, an emotive headset will be used. A combination of basic colour grids is shown to the subject. The subject later visualizes the same grid in his mind and the EEG signals are recorded. This is correlated to the actual pixels in the image. This process is to be repeated to train the system and achieve the desired results. The system can be extended further, to produce any image that is envisioned, without necessitating the use of an existing database for matching. Further, portability of the device can be enhanced using wireless microcontrollers. Keywords-Crime detection, Electroencephalography, RGB, Machine learning, Artificial Intelligence, Emotive