Driver mouth monitoring method based on machine vision

When we use the machine vision to monitor the driving behavior of a driver,identifying the mouth state is one of the key technologies. In fact, when a driver drives in a normal, talking or dozing state, his/her mouth opening degree is quite different. Based on this fact,the driver mouth contour and locality were extracted by a Fischer classifier to form a mouth region geometric feature group as eigenvalues.The eigenvalues made up an eigenvector as the input of a three-level BP neural network which might give the output in one of three above-mentioned spirit states.The experiment results show that this new method can monitor the driver mouth region accurately and quickly.