| Summary: | This thesis intends to start a new path on geolocation, using an approach that combines GPS information with an image analysis component. The main goal is to demonstrate that geolocation can be more accurate with the help of an image analysis system, which adds value to the coordinates read from the GPS, reducing the error through the identification of entities on some captured images. With this module, it would be possible for a device to know its relative position, even in points where the GPS information is not available. This would help to fix problems like the well known Urban Canyons, where the GPS information is lost for moments leaving geolocation softwares with no clue about their position, or on highways parallel exit roads where GPS Navigation Systems frequently get confused, placing the vehicle on the wrong road.
|
|---|