The Design And Implementation Of Driver Status Detection System Based On Kinect

With the development of economy and society,the transportation industry has developed rapidly,and the number of cars is rapidly increasing in our country.As a result,the number of traffic accidents has increased rapidly and has been on the rise.As 85% of traffic accidents are caused by human elements,it can be seen that real-time monitoring of the driver’s status and timely warning are of great significance.At present,the application of driver status detection is still relatively small,and there are common problems such as poor real-time performance,low accuracy,high environmental requirements,single detection methods,and tedious issues.In order to solve the above problems,this paper researches and analyzes the research status at home and abroad,designs and implements a driver state detection system based on Kinect,which can early warning the abnormal state of the driver and has a very good practical value for on-board vehicle.The driver state detection system solves three problems of night detection,detection methods and detection performance.In this paper,the Kinect camera is used as an image acquisition device,which is not affected by the light conditions and can be detected simultaneously during the day and night.In terms of detection methods,this article comprehensively detects the driver’s blinking frequency,closed eye time,and yawning frequency,effectively improving the accuracy of the detection.In terms of detection performance,this paper studies the AAM face feature point algorithm and proposes an improved method to solve the problem of inaccurate local positioning and low real-time performance.The driver state detection system,which is divided into three parts in this paper: Kinect image acquisition terminal,state detection terminal and background supervision platform.The main tasks include the analysis of system requirements,the formulation of the overall system plan,and the use of hierarchical and structured ideas based on the overall plan for detailed design and implementation of the system’s functional modules.The functional modules include an image acquisition module of a state detection terminal,a face recognition module,a status analysis module,an alarm module,a user management module of a background monitoring system,a real-time display module,and a picture management module.Finally,in order to verify the feasibility of the system,the system was tested for functionality and performance.The analytic result shows that under different scenarios,the rate of accuracy for driver state detection reaches 80%,the driver’s status can be judged and take corresponding measures within 1s.