Human Identification Based on Color Stimuli

Human identification by using Electroencephalogram is becoming promising field and reliable to improve security systems. It is difficult to acquire EEG at a certain mental condition always such as concentration or relaxation. This paper represents a simple model to identify individuals and finding most effective primary color by using features of EEG by means of color stimuli. A comparison between primary and secondary colors for identification has also been made. Standard additive primary colors blue, green, red and one secondary color yellow were selected for experiment. Four neural networks were built by extracting various features of EEG in the domain of time and frequency. All artificial neural networks showed satisfactory performance with minimum mean square error for identification. Among the four selected colors blue color based ANN showed minimum mean square error of 6.238×10^-08.