Convolution Neural Network For Face Recognition And Application On Fatigue Driving Detecting

Convolution neural network is a deep machine learning method originated from artificial neural network.It has the characteristic of local connection and weight sharing,and can realize the automatic extraction of features.It improves the problem of feature extraction in traditional recognition method.Therefore,convolution neural network is widely used in natural language processing,speech recognition,recommender system,computer vision and other fields.Fatigue driving detection technology based on the external characteristics of the driver has made some progress in many aspects,but the driver's facial feature extraction method needs to be further improved,while the driver's eye location time is longer,affecting the system recognition rate.This paper applies the convolution neural network to the face recognition and improves pupil location algorithm to overcome the problem of large calculation of the traditional one.And according to the characteristics of the driver's eyes in different states with different ratio of width and height,a simple and feasible method of eye state judgment is achieved,and the fatigue state of the driver is judged by PERCLOS algorithm.Studying from the experiment on the ORL face data set,the recognition rate of convolution neural network model constructed in this paper is 85%,and the average time is 20 ms.The experiment finds that the accuracy and average time of the improved Hough transform method for the driver's eyes are 92% and 29 ms respectively and the recognition rate of eye state judgment method is 83.9%.By contrast,the fatigue driving detection method used in this paper can achieve better results than the traditional method.A fatigue driving detection prototype system based on face recognition is designed,which realizes the functions of driver's facial feature detection,eye location,eye state judgment and fatigue judgment.The experiment results show that the average recognition rate of fatigue driving is 87.5%,and the response time of fatigue judgment is 17 ms,which has good application value.