Design And Implementation Of Vehicle-mounted MOST Camera And Face Recognition Interface

With the continuous development of automotive electronics embedded products, avariety of automotive electronic equipment continue to enrich the user’s driving experience.The development of in-car communication systems started with the CAN-Bus, which hasbeen used in production vehicles since the1990s. At the same time, the demand for in-vehicleentertainment devices becomes more and more urgent. MOST (Media Oriented SystemsTransport) is a communication system, which was developed to meet these demands. Sincethe MOST network technology was proposed in1998, it has gone through three generationsof specification development process, including MOST25, MOST50and MOST150. It has acomplete development framework and interface during the development process, to facilitatethe expansion of a node when you add a new function block. MOST bus has thecharacteristics of high speed, anti-interface, light quality, etc. Therefore it has graduallybecome a competitive vehicle network.Face recognition technology plays an irreplaceable role in our daily life as a matureresearch and it can provide a safe environment for daily life as a security verification method.So the article puts forward the following ideas: adding the camera and face recognition nodesto the MOST ring network, and the MOST asynchronous data is transferred between the twonodes.Firstly, this article discusses the MOST network topology, MOST node structure and thecorresponding relationship of the OSI seven layer networks; since this article focuses onasynchronous data transmission, therefore discusses the MOST50frame agreement,asynchronous data domain structure in detail; then introduces the MOST High protocol,Layer1and Layer2.Secondly, this article puts forward the general design and realization scheme of thesystem: it expounds the MOST camera and face recognition nodes of the overall systemstructure design. The camera node includes a USB camera, a network interface controllerOS81092and a microprocessor S3C6410; the face recognition node includes a LCD touchscreen, a network interface controller OS81092and a microprocessor S3C6410.Thirdly, this paper studies the selection and configuration of the master control module, intelligent network interface controller module, transceiver module, the pin connectionbetween OS81092and S3C6410and camera module. This paper sets up the system divisionof labor for the two nodes from the perspective of system software design; and expounds theconfiguration of the asynchronous data files and the immigration from single chipmicrocomputer to the Linux operating system; then introduces the steps of face recognition,training, recognize faces implemented by the face recognition node and the camera node;Finally it lists the control commands transmission mechanism, the establishment of theasynchronous data transmission connection and the realization of the data transmissionmechanism.Finally, MOST network analyzer optolyzer provided by Microchip and the USART serialinterface debugging tools are used to confirm the practicability and reliability of this systemfrom the perspective of multiple. From what has been discussed above, this system has addedthe camera and face recognition node to the MOST ring network successfully which has madecontribution to the development of MOST network technology.