ACCURATE LOCALIZATION IN DENSE URBAN AREA USING GOOGLE STREET VIEW IMAGES

ACCURATE LOCALIZATION IN DENSE URBAN AREA USING
GOOGLE STREET VIEW IMAGES
 Mahdi Salarian Andrea Manavella Rashid Ansari  
 Email: [e-mail] Email [e-mail] Email: [e-mail]
 School of Electrical and Computer Engineering
University of Illinois at Chicago
Chicago, IL  
 
ABSTRACT
Accurate information about the location and orientation of a  
camera in mobile devices is central to the utilization of
location-based services (LBS). Most of such mobile devices  
rely on GPS data but this data is subject to inaccuracy due to
imperfections in the quality of the signal provided by  
satellites. This shortcoming has spurred the research into  
improving the accuracy of localization. Since mobile devices
have camera, a major thrust of this research has been seeks to  
acquire the local scene and apply image retrieval techniques
by querying a GPS-tagged image database to find the best  
match for the acquired scene.. The techniques are however  
computationally demanding and unsuitable for real-time  
applications such as assistive technology for navigation by  
the blind and visually impaired which motivated out work.  
To overcome the high complexity of those techniques, we  
investigated the use of inertial sensors as an aid in image-
retrieval-based approach. Armed with information of media  
other than images, such as data from the GPS module along  
with orientation sensors such as accelerometer and gyro, we  
sought to limit the size of the image set to c search for the  
best match. Specifically, data from the orientation sensors  
along with Dilution of precision (DOP) from GPS are used to  
find the angle of view and estimation of position. We present  
analysis of the reduction in the image set size for the search  
as well as simulations to demonstrate the effectiveness in a  
fast implementation with 98% Estimated Position Error.
Index Terms— mobile device, LBS, Geotagged-images,