Road Sign Detection and Recognition in Adverse Case using Pattern Matching

Application of new technology in building human comforts and automation is growing fast, particularly in automobile industry. Automatic detection and recognition of traffic signs for assisting driver to ensure a safe travel have been givenpractical importance for intelligent traffic system.The proposed method detects the location of the traffic sign in thecaptured image, based on its geometrical characteristics and using color information. Such signs are then recognized using a pattern matching with the normalized cross correlation method and tracked through a sequence of images. The algorithm is tested using image set of different traffic signs and non traffic signs taken under various adverse conditions such as, various backgrounds, lighting conditions, orientation and distances. Experimental result shows the better performance in the detection and recognition of road signs with recognition rate of 90%.Computational time is also quite low which makes it applicable for the real time system.