Non-intrusive video-detection for traffic flow observation and surveillance is the primary alternative to conventional inductive loop detectors. Video Image Detection Systems (VIDS) can derive traffic parameters by means of image processing and pattern recognition methods. Existing VIDS emulate the inductive loops. We propose a trajectory based recognition algorithm to expand the common approach and to obtain new types of information (e.g. queue length or erratic movements). Different views of the same area by more than one camera sensor are necessary, because of the typical limitations of single camera systems, resulting from occlusions by other cars, trees and traffic signs. A distributed cooperative multi-camera system enables a significant enlargement of the observation area. The trajectories are derived from multi-target tracking. The fusion of object data from different cameras will be done by a tracking approach. This approach opens up opportunities to identify and specify traffic objects, their location, speed and other characteristic object information. The system creates new derived and consolidated information of traffic participants. Thus, also descriptions of individual traffic participants are possible.

Non-intrusive video-detection for traffic flow observation and surveillance is the primary alternative to conventional inductive loop detectors. Video Image Detection Systems (VIDS) can derive traffic parameters by means of image processing and pattern recognition methods. Existing VIDS emulate the inductive loops. We propose a trajectory based recognition algorithm to expand the common approach and to obtain new types of information (e.g. queue length or erratic movements). Different views of the same area by more than one camera sensor is necessary, because of the typical limitations of single camera systems, resulting from the occlusion effect of other cars, trees and traffic signs. A distributed cooperative multi-camera system enables a significant enlargement of the observation area. The trajectories are derived from multi-target tracking. The fusion of object data from different cameras is done using a tracking method. This approach opens up opportunities to identify and specify traffic objects, their location, speed and other characteristic object information. The system provides new derived and consolidated information about traffic participants. Thus, this approach is also beneficial for a description of individual traffic participants. Keywords: Multi-camera system, fixed-viewpoint camera, cooperative distributed vision, multi-camera orientation, multi-target tracking

In this paper, we describe an automatic approach for a terrestrial line scanner calibration. The system is calibrated with prior knowledge of the exterior orientation of the camera in an unknown coordinate system. This data is acquired by tracking the moving sensor platform with an infrared camera tracking system (ARTtrack2) along a calibration scene. The calibration of a line scanner includes the determination of the transformation parameters to the known control point coordinate system (translation and rotation) and three parameters of the one-dimensional distortion model (simplified BROWN) for the sensor array. The focal length of the camera is assumed to be known. After a first rectification of the data with the tracking observations, photogrammetric targets are automatically detected and decoded. Finally, the distortion parameters of the camera and the transformation between world- and tracking coordinates are iteratively estimated.

Trajectories extracted from an object tracker are a valuable source of information to derive describing parameters for traffic situation analysis. In order to ensure the quality of these parameters, the tracker's output has to be evaluated. This is a challenging task, since the choice of the metrics and the choice of the matching --- of ground truth tracks and the output of the tracker --- influences the results considerably. We present the evaluation of a Kalman-filter based tracker using ground truth data of traffic scenes. The Kalman-Filter is object space based and uses observations obtained from multiple camera views. In the evaluation, a two level approach is employed. The first level assesses the ability to correctly locate the objects. The second one evaluates the consistent identification of objects throughout the scene. Matching of tracks is done using a spatial overlap measure. Subsequently, several metrics are applied, whose properties are outlined. The results for simulated and real traffic data are presented. Finally, the implications of aggregation and normalization of these metrics in order to estimate the algorithm's performance are discussed.

The use of non-intrusive video-detection for traffic flow observation and surveillance is the primary alternative to conventional inductive loop detector. A Video Image Detection System (VIDS) can derive traffic parameters by means of image processing and pattern recognition methods. Classic systems like the inductive (double) loops or microwave radar detectors are able to measure presence and length of a vehicle but also speed and time gap to the preceding vehicle. Existing VIDS emulates the inductive loop by virtual loops and derives the traffic parameters in a similar way. Additional benefits of the VIDS, e.g. large area observation of traffic flow or detection of special behaviors of single vehicles can not be exploiting in this way. To expand the common approach also for new types of information (queue length or erratic movements), we use a trajectory based recognition algorithm, which is related to the detection and tracking of individual vehicles in an image sequence. It will be shown as an essential application for this algorithm how the usual characteristic traffic parameters can be derived from the measured trajectories again. A direct comparison between the video based and the classic inductive loop measurement of the traffic parameters is also possible.