A Framework for Automatic Detection of Traffic Violations

Automatic detection of violations is a wellstudied area of research. Detection of traffic regulations violations and recognition of the violating vehicles using automated techniques can efficiently replace the requirement of traffic police with virtual traffic police. Detection of violations includes detecting motorcyclists without helmets and triple riding. Also, the automated number plate recognition system plays a crucial role in identifying defaulters and simplifying traffic jams. This paper proposes automated techniques for the detection and classification of vehicles violating traffic rules and also for tracking stolen vehicles or uncertified vehicles.