Speed-bump Detection using Otsu's Algorithm and Morphological Operation

Driver Assistance System (DAS) is one of the most mandatory subsystems of Intelligent Transport System (ITS). Speed Bump Detection (SBD) is a subset of DAS that supports the driver to recognise the presence of speed bump and alert the driver. Under the critical condition, it has to operate on the vehicle directly for safety. In the research work, two methods of Speed Bump Detection have proposed (i) Otsu’s threshold method and (ii) Morphological Operation method. These methods are superior to the existing methods because it does not involve any network, so it is free from GPS error, network overload, delay and false alarm. It also detects the speed bump with low projected where the existing method using sensors fails. The proposed Otsu’s method is straight forward, efficient and gives 74.6% accuracy for all types of road condition. The performance of Morphological or structural operation method achieves better result 85.8 % detection ratio. The two proposed methods give more than 90 % detection ratio for properly painted road and optical illusion type speed bump.