Research And Implementation Of Key Technologies Of Intelligent Vehicle Occupant Monitoring System（OMS）

In recent years,with the continuous development of computer vision and smart cars,smart cockpits have become another hot development direction in the automotive market after smart driving.Intelligent cockpit The intelligent cockpit is also called the intelligent vehicle occupant monitoring system（OMS）.It not only includes some technologies of ADAS assisted driving,but also proposes a number of new technologies for passengers,such as emotion recognition,forgotten objects detection,gesture recognition and gesture recognition.and other smart technologies.In addition to ensuring the safety of the driver and satisfying the safety of the driver,driving a car also guarantees the safety of the passengers in the car and the ride experience.Therefore,according to the definition of the intelligent vehicle occupant monitoring system,from the perspective of passenger safety and car experience,it is proposed to meet the needs of passengers based on the research of human gesture recognition technology and target detection technology,and to expand the technical system of the intelligent vehicle occupant monitoring system.The paper The main work is as follows:1.Aiming at the problem that the height of children generally does not meet the height of the co-pilot airbag,which will cause unnecessary secondary injuries to children in a car accident,a co-pilot child identification algorithm based on OpenPose human gesture recognition technology is proposed to realize the detection of co-pilot children.warning function.The skeleton key point information of all passengers in the car is obtained through the OpenPose gesture recognition algorithm,and the passenger skeleton information of the copilot is separated and processed by the traditional image processing method;seat height to determine the category of passengers.2.Aiming at the human-computer interaction function requirements of passenger multimedia in the car,an in-vehicle projection gesture interaction algorithm based on Mediapipe hand key point detection is proposed,which realizes the passenger’s humancomputer projection interaction in the car to complete some multimedia functions.Through the combination of the projector and the camera,the projector projects the screen on the rear roof,the camera obtains the image of the rear roof,and the passengers click or drag the virtual buttons on the projection screen on the rear roof.The in-vehicle projection gesture interaction algorithm of Mediapipe hand key point detection analyzes the passenger’s gesture,and generates a corresponding response signal and transmits it to the vehicle system to trigger the corresponding function.3.Based on the research results of the key technology of the intelligent vehicle occupant monitoring system（OMS）in this subject,combined with other functions and requirements of the intelligent vehicle occupant monitoring system,the intelligent vehicle occupant monitoring system is designed and implemented.Using TCP communication mode and multiprocess programming mode,the research results of the project are displayed visually,making certain contributions to the research and development of intelligent cockpit.To sum up,in view of the safety of passengers in the car cabin and the practical problems of human-computer interaction,this paper proposes two algorithms to realize passenger seat height analysis and gesture projection interaction combined with the characteristics of key points of the human body,and integrate them into the intelligent car occupant monitoring system.Actual driving scenarios are rigorously evaluated and analyzed.Finally,the effectiveness of the algorithm is verified according to a large number of real vehicle scene data.