Road sign recognition is a part of driver support systems. Its main aim is the increase of traffic safety by calling the driver's attention to the presence of key traffic signs. Additionally, a vision-based system able to detect and classify traffic signs from road images in real-time would also be useful as a support tool for guidance and navigation of intelligent vehicles. This paper proposes a new method for the detection and recognition of traffic signs using self-organising maps (SOM). This method first detects potential road signs by analysing the distribution of red pixels within the image, and then identifies these road signs from the distribution of dark pixels in their pictograms. Additionally, a novel hybrid system combining programmable hardware and artificial neural networks for embedded machine vision is introduced, and a prototype of this system is used in the implementation of the application. The experiments indicate a good performance of the new approach using SOM in both speed and classification accuracy. (C) 2008 Elsevier B.V. All rights reserved.
|Number of pages||10|
|Journal||Image and Vision Computing|
|Publication status||Published - 4 May 2009|
- road sign detection
- road sign recognition
- self-organising map