The unprecedented increase in the number of vehicles has led to increased number of road accidents. Detecting and catching traffic violators has become a necessity for authorities all over the world in order to reduce chances of accidents and ensuring road safety. Red light violation and speeding is a major cause of road collisions. The offenders tend to violate traffic rules if they know that they can get away with it without getting caught.
The research shows that a 15 year old technique of detecting red light violation used single camera for capturing image of vehicles and applying image processing techniques. The basic technique used in those systems was frame differencing where a comparison was made between the current picture frame and the previous one. Once detected, the video frames were used to capture its license plate. However, there were certain problems seen with such systems. False detection was observed due to changes in daylight. In certain cases, the system falsely detected RLVD when a vehicle was stationary at red light or if a non-vehicle object like a pedestrian or bird crossed the border line.