Design And Implementation Of Vehicle Target Recognition System Based On Android System

The advances in technology and the improvement of China's comprehensive national strength have promoted the development of the automotive industry,and the automobile holding volume and the automobile production has increased a lot.The growth of traditional automotive industry has reached an unprecedented peak.With the development of artificial intelligence,intelligent car field has gradually been the concern of researchers.Autopilot combines the information industry with the automobile industry,and it is predicted that in the next 10-20 years,auto driving will have a profound impact on the traffic field.Because the automatic driving technology has not been widespread now,so,the main direction of intelligent transportation is to assist driving.Computer image processing technology is one of the key technologies in the field of automatic driving.The realization of the technology depends on the video stream captured by the camera.Vehicle target recognition is the branch of computer image processing and one of the key technologies in the field of automatic driving.This paper studies the algorithm of vehicle target recognition,selects the frame difference method as the recognition algorithm of this system,and implements the system development with the help of the Android platform.Android as one of today's most popular mobile phone operating system,Google provides an open source solution for Android,mobile phone manufacturers can use Android as a software development platform,in recent years,applications based on Android has developed rapidly,Intelligent transportation fields based on Android system has a very broad prospect.Using Android system to assist driving also becomes a development trend.This topic is based on Android for vehicle target recognition.Using mobile camera to collect the surrounding environment of vehicles and identify the target vehicles in the range of perception,the function of this system can be used as a sub module to the Android automatic driving system.In order to reduce the frequency of traffic accidents,and to follow the current development direction,in this paper,the author analyzes the commonly used vehicle recognition algorithm,designs a vehicle target recognition software based on Android operating system,with the open source computer vision library,OpenCV.In the environment of Android Srudio integrated development,a vehicle target recognition software based on Android operating system is designed and implemented,and the vehicle target recognition system based on Android is realized.The system can help the driver to better complete the driving operation through the debugging program.In the smart phone,using the camera with the smart phone to capture the information of the lane surrounding the camera vision,the information of the lane is transmitted to the vehicle identification module in the form of frame sequence.The vehicle recognition module uses the frame difference method to make the two consecutive frames in the video stream and get the moving target.Real time output to the phone screen.The main contents of this paper are as follows:1.We completed the construction of the Android studio platform,and used the open source OpenCV software to assist the frame difference method to process the video stream.2.Read the frame sequence from the Android mobile phone and input it to the recognition module,and use the frame difference method to complete the target vehicle recognition.3.The vehicle target recognition system based on Android is implemented in this paper,and it can be applied to Android smartphone,and the test result is good.