Viewpoint independent vehicle speed estimation from uncalibrated traffic surveillance cameras

We present here a prototype of an algorithm for vehicle speed estimation. Different from previous approaches, our algorithm requires no road markers and fewer manual calibrations. Based on specific projection rules, we find a relation between the in-camera coordinate and the real world coordinate. A non-linear regression is employed to estimate the model parameters. This model enables us to estimate the real world position of the vehicles directly from a video sequence taken by a surveillance camera. The algorithm shows its ability to produce accurate estimations in our experiments.

I. Introduction

Nowadays, most of the traffic speed sensors deployed are magnetic based appliances or radar systems, infrared sensors [7]–[9]. However, it might be impractical or too expensive to install them at all locations. However, it is commonly noticed that a large number of CCTV cameras are already deployed for traffic surveillance and management. This common phenomenon motivates us to investigate the possibility of using computer vision techniques to build more easy-to-set (and use) traffic surveillance systems. And the idea of this paper is to make full use of these videos from these un-calibrated cameras and make them become additional speed sensors which can increase the capability of traffic management system at low costs.