A Mobile System for Vision Based Road Sign Inventory

Publication from Digital

Paletta L., Seifert C., Benesova W., Andreu J-P., Lipetski Y., Hoedl E., Jeitler A.

5th International Symposium on Mobile Mapping Technology (MMT 07),Padua Italy May , 2007


For road sign inventory and maintenance, we propose to use a mobile
 system based on a handheld device, GPS sensor, a camera, and a standard
 mobile GIS software. Camera images are then analysed via object recognition
 algorithms which results in an automated detection, i.e., localisation
 and classification of the signs. We present here the localisation
 of points and regions of interest, the fitting of geometrical constraints
 to the extracted set of interest points, and the matching of content
 information from the visual information within the sign plate. From
 the preliminary operational state of the vision based road sign detection
 system we conclude that the selected methodology is efficient enough
 to achieve the requested high quality in object detection and classification.
 This work is funded by the European Commissions IST project DETECT
 under grant number IST-2001-32157, and the Austrian Joint Research
 Project Cognitive Vision, sub-projects S9103-N04 and S9104-N04.

Keywords: mobile geo-information systems, road sign inventory, visual object recognition