Vision based systems using cameras are widely used in various industry verticals. The basic functionality of a vision system is to detect an object and then recognize it. In current automotive industry these systems are being used in applications like collision avoidance, intelligent cruise control, blind spot detection, parking assistance, identify pedestrians, etc.During driving conditions, 2 wheeler and 4 wheeler drivers are required to concentrate more on the road because of which they may tend to miss the traffic signs; more often during night times. The traffic signs can be one way, U-turn, right/left turn prohibited, bump ahead and etc. So there is a need for a system which detects, recognizes and intimates these signs to the driver.Extensive research is being carried out in the area of recognizing a traffic sign, but more focused research is done for a complex problem of detecting the sign in the real world environment. The proposed image processing algorithm for detection of the traffic signs in the recorded images from the camera helps the vision system to intimate through visible or audible alerts. Such an integration of the vision system with the vehicle keeps the driver informed about the traffic signs ahead, providing better driver assistance and safety.