Efficient detection of speed limit signs within obscure environment

摘要

An automatic speed limit signs detection and recognition system would be important in a driver assistance system and intelligent transport system. It can be reduced the car accidents if the drivers pay attention to the signs. However, those signs might be located in the obscure environment that makes the trouble for driver to detect and recognize. This work presents the technique to detect and recognize the speed limit sign based on obscure environment using the intra-frame intersection technique. Also the HSV, color segmentation and the dilation with erosion technique are employed. The results show that this technique can provide 100% accuracy to detect and recognize the number of speed limit signs in the condition of 35% obscure environment.