Gao X.1,Shevtsova N.2, Hong K.1, Batty S.1, Podladchikova L.2, Golovan A.2, Shaposhnikov D.2, Gusakova V.2
1School of Computing Science, Middlesex University
London, United Kingdom
2A.B. Kogan Research Institute for Neurocybernetics,
Rostov State University, Rostov-on-Don, Russia
A new approach has been developed for accurate and fast recognition of traffic signs based on human vision models. It applies colour appearance model CIECAM97 to segment traffic signs from the rest of scenes. This model takes viewing conditions into account and can recognise colours as accurate as an average observer invariant of lighting sources, background, and surrounding colours. A Behavioural Model of Vision (BMV) is then used to identify the signs after segmented images are converted into gray-level representation. The BMV is transform invariant and can recognise most signs with distorted shapes. Two standard traffic sign databases are established. One is British traffic signs and the other is Russian traffic signs. Preliminary results show that around 90% signs taken from the British road with various viewing conditions have been correctly identified.