Vision based vehicle speed detection system

Abstract

In the past few decades, the number of vehicles out on the roads has been increasing at a rate that does not seem to be slowing down any time soon. With so many vehicles in use, naturally the chances of accidents or mishaps occurring are also greatly increased. To reduce the chances of such events, it is necessary that traffic rules be strictly enforced and anyone that breaks these important traffic rules is caught and punished for it as soon as possible, An important factor in many accidents is excessive speeding which is why it is needed that over speeders be caught efficiently and reliably. The main methods of catching : over speeders rely on using specific techniques and technologies to measure the speed of moving vehicles. The existing traffic surveillance systems typically rely on methods that are both expensive to implement and come with several limitations which do not allow them to be useful in many different scenarios. The results of such scenarios are inaccurate calculations of speed which defeats the purpose of such expensive equipment and techniques in the first place. With this project, the aim is to create a system that is inexpensive, has a high accuracy and addresses the major problems with existing solutions. The system uses a combination of various image processing techniques to calculate the speed of moving vehicles and is both expandable and scalable without much expense.