Lie detection using image processing

Eye Blink Pattern can be used for the identification of various human behaviors. This paper proposes a method for lie detection and detecting the truth of an alibi through image processing techniques. The lie detection mechanism is based upon the experimentally proven hypothesis derived from eye blink literature that liars experience more cognitive demand than truth tellers. Thus, their lies are associated with a significant decrease in eye blinks, directly followed by an increase in eye blinks when the cognitive demand ceases, after the lie has been told. Blink detection is done with the help of MATLAB using the HAAR Cascade algorithm. The decision regarding the open or closed state of the eyes is made by implementing skin detection algorithm on the histogram back projections of the detected eye images.