Fusing camera and WiFi sensors for opportunistic localization
Faculty of Sciences. Mathematics and Computer Science
Faculty of Applied Engineering Sciences
UBICOMM 2011, the fifth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies
There has been a lot of research done towards both camera and Wi-Fi tracking respectively, both these techniques have their benefits and drawbacks. By combining these technologies, it is possible to eliminate their respective weaknesses, to increase the possibilities of the system as a whole. This is accomplished by fusing the data from Wi-Fi and camera before inserting it in a particle filter. This will result in a more accurate and robust localization system. The measurement model for Wi-Fi data uses a difference feature vector for comparing data to the fingerprint. The images taken from the camera are analysed, and filtered to detect human shapes. In this paper it is proven that an increased accuracy can be achieved by fusing the sensor data of both Wi-Fi and camera.