Density Measurements from a Top-View Position using a Time-of-Flight Camera

Inhalt:
This paper proposes two methods that utilize depth images to calculate the densities among crowds of people. A Time-of-Flight (ToF) camera provides these depth images of the scene from a Top-View position and therefore supplies distance information from the camera to each pixel in the scene. The proposed density measurements are performed by two individual methods. The first one defines density by the estimation of the number of persons in relation to the available space, as seen by the camera. Therefore, a people detection algorithm is used. The second method contrarily bases on the calculation of the occupancy of a certain model plane. Here, depth images and a model plane are projected into a 3D voxel scene. The occupied space is defined by voxels that comply the model plane’s constraints and are located above the model plane’s height. Finally, density is measured by the relation of occluded space in contrast to the space provided by the model plane. Experiments have proven the applicability of the proposed methods in different crowding situations.