Road traffic signs detection and classification for blind man navigation system

This paper presents the method to find the road traffic signs detection and classification in vision-based robot guidance system that apply for blind man navigation system which can have two main roles that first for road traffic signs detection and next for signs classification. Traffic signs recognition is a less studied field even though it provided road user with very valuable information about the road profile in order to make walking safer and easier. The algorithm described in this paper take advantage of sign features that their color and shapes are very different from natural environments. The system is divided into three parts, first for detected and improved the raw sign image and second part for shape analysis with a continuous thinning algorithms and image coding algorithm and finally for the image recognition and decision by fuzzy logic and back propagation neural network (BNN) technique to display the right task. Some results from natural scenes are shown that system performance can work well and is valid to detect other kinds of signs that would train the mobile robot to perform some task at that place.