Real-time video of traffic scenes contain a wealth of information not available from conventional point detectors. In addition to the instantaneous, wide-area coverage provided by image data, image sequences capture the dynamic aspects of the traffic. Initially, researchers concentrated on minimizing hardware complexity, and thus cost, at the expense of sophisticated algorithms that could more fully exploit the information inherent in image data. If image data could be processed in real-time to produce a track file for each object of interest, then the traffic flow through the scene would be fully characterized for traffic management purposes. This paper presents the status of work in process at the Environmental Research Institute of Michigan (ERIM) to develop real-time image processing algorithms for detecting and tracking vehicles in actual traffic settings. The image processing techniques for detecting and tracking will be illustrated, the corresponding computational resources will be described, and preliminary results on typical video sequences will be presented.

Dynamic traffic information from remote video monitors

A monitoring system for local traffic control based on vehicle tracking

An integrated system for ground plane obstacle detection based on Alpha Data Parallel system with four processors is presented in this paper. We partially use on-line calibration to make the system self adaptable and use the projective transformation method to find the occupied space and/or free space in front of a mobile robot moving on the ground plane. We investigate stereo correspondence problem in the framework of color stereo vision to decrease the ambiguities and inconsistencies of available information in images. The results show our system is efficient and robust.

Real time obstacle detection using Alpha parallel systems

We describe a real-time optical device for vehicle detection and recognition on roads. With vehicle traffic on public roads steadily increasing, there is a growing need for efficient monitoring and control of traffic flow. There is need for traffic counting (vehicle detection), estimation of vehicle speed and for vehicle classification (i.e. passenger car, truck). An optical sensor is very well suited to this problem giving detailed information from vehicles seen from a position beside or above the road. We implemented a device for this task operating in real-time with image scanning rates of about 100 Hz. Such rates are necessary for vehicle speeds up to 100 km/h. We solved the problem of high data rates by reducing the sensor's field of view down to two parallel receptor columns with a well known spacing and by sampling the grayvalues from the columns with a rate well fitted to the minimum of the required information. The principal idea of the applied image processing is to use change detection by means of difference pictures, embedded in a dynamic scene driven control. For the reliable detection and segmentation of moving vehicles a simple vehicle model and some rough estimates of the traffic flow are used. The segmented image data of the vehicles together with the derived information are collected for further inspection and processing. The system is realized on the Visual Interpretation System for Technical Applications (VISTA) developed by the Fraunhofer Institute IITB /1/. Most data processing is performed by software. The system achieves well the picture scanning rate of 100 Hz. For experimental purposes all image processing data are continuously displayed on a video monitor.

Matthias Burkhard

Papers

Real-Time Road Surveillance By An Optical Sensor