Kernel estimates as general concept for the measuring of pedestrian density

The standard definition of pedestrian density produces scattered values, hence, many approaches have been developed to improve the features estimated density. This paper provides a review generally applied methods and presents general framework based on various kernels that bring desired properties estimates (e.g., continuity) incorporate ordinarily used methods. kernel concept considers each as source distribution, parametrized by type Gauss, cone) size. quantitative parametric study performed experimental data illustrates parametrization brings features, for instance, conic with base radius in (0.7, 1.2) m smooth values retain trend features. correspondence between non-kernel (namely Voronoi diagram customized inverse distance nearest pedestrian) is achievable wide range parameter. Thereby generality supported.