Research And Implementation Of Mobile Augmented Reality System Based On Android Platform

Augmented reality means that makes the virtual information witch computer generated into the real world.In recent years,the number of mobile-based augmented reality systems has been increasing.Mobile devices are easy to carry and have location services and sensor functions.According to the differences between the tracer registration way,the mobile augmented reality system can be divided into augmented reality system which based on the computer vision and augmented reality system combined with sensor and LBS.This paper has studied and designed to achieve thease two systems.The main research results are as follows:(1)The key of augmented reality system which based on computer vision is to solve the problem of virtual fusion.The illumination consistency can better eliminate the abrupt sense of the virtual model.At present,the study of illumination consistency is mostly concentrated in PC terminal.The method of constructing lighting model is more complex and the algorithm is more computationally intensive,so it is hard to use directly on the mobile terminal.Using mobile light sensor can better build realistic scene lighting model,but the current implementations of augmented reality systems developed using mobile sensors are complex and do not facilitate porting and functional expansion.What is more,the system only consider the situation that the marker recognition success.This paper uses the way of combining Android Studio,Unity3D and EasyAr SDK to implement the system.This way is more intuitive and convenient.Firstly,on the one hand,using EasyAr SDK and the Unity3D to set up the AR environment;On the other hand,using Android Studio to registers and monitors lighting sensor to construct real world light model.And then through the Android Studio and Unity3D communication mechanism to transmit the maximum intensity data.At the same time,the gyroscope is used to obtain the mobile phone's posture in the three-dimensional world.Finally,the light source direction and light source intensity of the virtual world are changed dynamically in the Unity3D,and then render the shadows.The results show that this system has a better effect on the consistency of light,and also has higher real-time and better fluency.(2)Mobile augmented reality system which combining sensor and LBS can better replace the traditional 2D navigation system.However,most of the current system's orientation data is directly acquired by the direction sensor,and its data is not stable.Most system users can not use their own POI.In this paper,a mobile augmented reality navigation system based on campus is implemented by the combination of sensor and LBS.The system firstly uses the magnetic field and acceleration sensor in the Android device to obtain the direction data,and introduces low-pass filtering for the instability of the sensor data for processing optimization.Secondly,combining with the Amap SDK for obtaining location information and calculate the AR rotation matrix.Then design the database to store POI information.The final results show that the user-defined POI tag information can be displayed more accurately on the AR view.The user can easily interact with the tag,and is more intuitive and directional than the traditional map navigation.