Paper Title
Traffic Accidents Reduction by Facial Expression Identification
Abstract
One of issue of road traffic safety is reduction of the number of traffic accidents. Currently preventive safety
technologies of vehicle are highlighted to be one of solution of reducing the number of traffic accidents. Out of its
technology, diver’s state adaptive safety system may be one of promising candidates. Therefore, identifying driver’s
psychosomatic states is indispensable to establish those kind of safety functions. A state of anger often has happened in
traffic jam or aggressive driving which may result in severe traffic accidents. This research adopted Kohonen neural network
as classification algorithm to identify anger state of driver by using six facial expression. Six types of facial expression are
ordinary,anger, drowsiness, sorrow, delight and surprise which is thought to express almost human emotions. Therefrom this
research established to identify anger state of driverby using six facial expressions. Then we proposed driver’s anger state
safety function which is one element of driver’s psychosomatic state adaptive safety system in cooperation with artificial
intelligence function. Finally, this research calculated reduction effect of the number of traffic accidents by using function of
detecting driver’s anger. Finally, this research verified validity of calculation by referring reduction rate of ESC.
Keywords- Traffic accident reduction, Anger state of driver, Driver’s state monitoring, Kohonen Neural Network, ASV