DocumentCode
568863
Title
TrueSight A Pedestrian Navigation System Based in Automatic Landmark Detection and Extraction on Android Smartphone
Author
Ferreira, Alessandro Luiz Stamatto ; Santos, Selan Rodrigues dos ; de Miranda, Leonardo Cunha
Author_Institution
Dept. de Inf. e Mat. Aplic., Univ. Fed. do Rio Grande do Norte (UFRN), Natal, Brazil
fYear
2012
fDate
28-31 May 2012
Firstpage
91
Lastpage
99
Abstract
From time to time someone gets lost and askhimself "How do I get there?" With the advent of the GPS this question can be answered. However due to difficulties such as lack of precision, possibility of inaccurate maps,network dependency, and cost lead the pursuit of analternative solution. In order to locate himself the person can use a different method: using a smartphone camera his position is recognized visually, based in environment references, and then an arrow pointing the right direction appears in a map in the display. This method was implemented in the application framework of Android,using OpenCV and its implementation of the SURFalgorithm. The final application is named TrueSight and westudy its viability and limitations. The authors conclude thata vision-based navigation system is viable, but database improvements and exhibition could make it better.
Keywords
cameras; computer vision; feature extraction; object detection; operating systems (computers); pedestrians; smart phones; visual databases; Android smartphone; GPS; OpenCV; SURF algorithm; TrueSight; automatic landmark detection; automatic landmark extraction; database improvements; open source computer vision library; pedestrian navigation system; smartphone camera; vision-based navigation system; Androids; Global Positioning System; Google; Humanoid robots; Image segmentation; Robustness; Visualization; android; landmark recognition; localization; mobile device; navigation;
fLanguage
English
Publisher
ieee
Conference_Titel
Virtual and Augmented Reality (SVR), 2012 14th Symposium on
Conference_Location
Rio Janiero
Print_ISBN
978-1-4673-1929-4
Type
conf
DOI
10.1109/SVR.2012.14
Filename
6297564
Link To Document