Alerting the drivers about road signs with poor visual saliency

Author

Simon, Ludovic ; Tarel, Jean-Philippe ; Brémond, Roland

Author_Institution

Lab. for Road Oper., Perception, Simulation & Simulators, Univ. Paris Est, Paris, France

fYear

2009

fDate

3-5 June 2009

Firstpage

48

Lastpage

53

Abstract

This paper proposes an improvement of advanced driver assistance system based on saliency estimation of road signs. After a road sign detection stage, its saliency is estimated using a SVM learning. A model of visual saliency linking the size of an object and a size-independent saliency is proposed. An eye tracking experiment in context close to driving proves that this computational evaluation of the saliency fits well with human perception, and demonstrates the applicability of the proposed estimator for improved ADAS.

Keywords

computer vision; driver information systems; learning (artificial intelligence); object detection; road safety; support vector machines; SVM learning; driver alert system; driver assistance system; human perception; object size; road safety; road sign image; visual saliency; Displays; Humans; Image processing; Joining processes; Laboratories; Machine learning; Magnetic heads; Object detection; Road safety; Support vector machines; Advanced Driver Assistance Systems; Conspicuity; Eye-Tracking; Head Up Display; Human Vision; Image Processing; Machine Learning; Object Detection; Road Safety; Road Signs; SVM; Visual Saliency;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Vehicles Symposium, 2009 IEEE

Conference_Location

Xi´an

ISSN

1931-0587

Print_ISBN

978-1-4244-3503-6

Electronic_ISBN

1931-0587

Type

conf

DOI

10.1109/IVS.2009.5164251

Filename

5164251