Real-time recognition of U.S. speed signs
Authors:
- 1 June 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 19310587,p. 518-523
Abstract
In this paper a camera-based system for detection, tracking, and classification of U.S. speed signs is presented. The implemented application uses multiple connected stages and iteratively reduces the number of pixels to process for recognition. Possible sign locations are detected using a fast, shape-based interest operator. Remaining objects other than speed signs are discarded using a classifier similar to the Viola-Jones detector. Classification results from tracked candidates are utilized to improve recognition accuracy. On a standard PC the system reached a detection speed of 27 fps with an accuracy of 98.8%. Including classification, speed sign recognition rates of 96.3% were achieved with a frame rate of approximately 11 fps and one false alarm every 42 s.Keywords
This publication has 20 references indexed in Scilit:
- Robust on-vehicle real-time visual detection of American and European speed limit signs, with a modular Traffic Signs Recognition system2007
- Fast and Robust Traffic Sign Detection2006
- Robust road sign detection and recognition from image sequences2005
- Traffic Sign Recognition Based On Color Image Evaluationion2005
- Efficient recognition of speed limit signs2005
- Fast shape-based road sign detection for a driver assistance system2005
- Traffic sign shape classification evaluation I: SVM using distance to borders2005
- A graphics hardware implementation of the generalized Hough transform for fast object recognition, scale, and 3D pose detection2004
- Generic sign board detection in images2003
- Traffic sign recognition in noisy outdoor scenes2002