A Method for Vehicle Count in the Presence of Multiple-Vehicle Occlusions in Traffic Images

Abstract

This paper proposes a novel method for accurately counting the number of vehicles that are involved in multiplevehicle occlusions, based on the resolvability of each occluded vehicle, as seen in a monocular traffic image sequence. Assuming that the occluded vehicles are segmented from the road background by a previously proposed vehicle segmentation method and that a deformable model is geometrically fitted onto the occluded vehicles, the proposed method first deduces the number of vertices per individual vehicle from the camera configuration. Second, a contour description model is utilized to describe the direction of the contour segments with respect to its vanishing points, from which individual contour description and vehicle count are determined. Third, it assigns a resolvability index to each occluded vehicle based on a resolvability model, from which each occluded vehicle model is resolved and the vehicle dimension is measured. The proposed method has been tested on 267 sets of real-world monocular traffic images containing 3074 vehicles with multiple-vehicle occlusions and is found to be 100% accurate in calculating vehicle count, in comparison with human inspection. By comparing the estimated dimensions of the resolved generalized deformable model of the vehicle with the actual dimensions published by the manufacturers, the root-mean-square error for width, length, and height estimations are found to be 48, 279, and 76 mm, respectively.

Real-time hardware–software embedded vision system for ITS smart camera implemented in Zynq SoC

“…division of a large uniform object). In the literature more advanced solutions such as 3D deformable models [50] were also described. However, the computational complexity limits their usage in embedded devices.…”

Backward-Simulation Particle Smoother with a hybrid state for 3D vehicle trajectory, class and dimension simultaneous estimation

“…The algorithm in [24] estimates the dimension by means of a Generalized Deformable Model (box shaped) parametrized with respect to the vanishing point of the images; the authors focus on the problem of vehicle inter-occlusion, and their system neglects the tracking and dimension estimation problems. The algorithm proposed by [10] provides a unified framework to estimate both the 3D vehicle trajectory and the vehicle dimensions by means of a deformable 3D box model, but its authors do not face the class estimation problem.…”

Occlusion and Abandoned Object Detection for Surveillance Applications

“…Methods to solve the occlusion problem in multiple interacting objects tracking have been previously presented in [8,12,15,16]. Sensor based occlusion tracking methods [8,18] cannot handle the difference between moving and nonmoving obstacles.…”

“…In general, solutions to both pedestrian and vehicle occlusion problems extend from spatial to temporal domains can be broadly classified into many types. Some types like active contour -based [2], stereo vision-based, region based, model based [16], feature based [12] are also seen in the literature. Object tracking in [17] overcome occlusion by fusing multiple camera inputs, but it cannot handle complete occlusion.…”