Font Size: 

The merging development in geospatial information technology has raised the development in Mobile Mapping System (MMS). In 2007, Google Inc. established and launched its Google Street View (GSV) Services which provide volumetric street view panoramas and associated geo-referenced spatial data to users in internet communities. Users can access these panoramas and positional data by appropriate approaches to expand and extend the added values in various applications for three-dimensional geospatial data acquisition.

Since GSV produces volumetric street view panoramas continuously and each panorama involves abundant features, it is our purpose to explore and mine useful information from the big imagery data produced from GSV streamlines. This research aims on developing application programs that integrate Google Maps Application Programming Interface (API), JavaScript, and C/C++ routines for accessing the orientation parameters of GSV panoramas in order to extract features of interest (FOI) and to determine georeferenced positions of these FOI for promoting added values in mobile services, such as location-based services (LBS), 3-D street-view model reconstruction, virtual 3-D city touring. Techniques in automatically identifying, extracting, and positioning of FOI, such as traffic signs, traffic lights, and street signs from GSV panoramas are primarily concerned for low-cost, volumetric and quick establishment of infrastructural database in public transportation network. An experimental study area in Taichung City was selected for testing the developed software system and establishing the database of selected GSV panoramas and extracted FOI for future spatial analysis and geospatial statistical analysis using Geographic Information Systems (GIS).

google street view; traffic infrastructure; feature extraction; big imagery data;