Vehicle Flow Measuring Based on Temporal Difference Image

The data of vehicle flow is very important to control and lead to traffic in intelligent transportation systems (ITS).We present a method for detecting vehicles in image sequences without prior knowledge about the position of the road. A single camera placed in a moving vehicle provides image data. The image seen from camera located beside or upon the road can be used for vehicle detection ,facility of car or car group measurement, parking car detection, etc. Moreover the image seen from camera located in vehicle can be used for preceding car detection, measurement of the distance to preceding car, obstacle detection, lane detection, etc .To detect passing vehicles, we use temporal differencing and projected motion. We tested our algorithms by means of many different experiments. These experiments illustrate that the correct recognition rate of this system reaches up to 90%.And the combination of phase control of road control, which can intelligent lead to traffic by real-time and satisfies the requirements of practical application.