RESEARCH OF THE VISIBILITY OF POLYGONAL MODELS BASED ON POINT GEOMETRIC MODELS OF A SET OF CAMERA POSITIONS

Abstract

The problem of visibility of objects often arises in such tasks as monitoring of buildings using unmanned aerial vehicles, placement of video surveillance cameras and light sources indoors and outdoors, analysis of terrain in geographic information systems, application of paint to the surface. This problem can be considered in terms of finding a plurality of sensor positions that provide maximum capture of terrain points or objects by this sensor. As a rule, the problem of object placement does not have an exact solution, so in practice it uses optimization methods, which most often find suboptimal solutions. This leads to the need to conduct studies of the visibility of geometric models taking into account the position of the camera to determine the set of points with the maximum overview ability. The paper investigates an approach to determining the set of points that provide the maximum view of an object, which is given in the form of a polygonal model, based on dividing the set of candidate survey points into a regular grid and solving the problem of covering such a set. The paper studies the influence of the main parameters of the method, namely, the number of iterations of the algorithm to achieve a given visibility coefficient, the required degree of space discretization (spatial resolution of the grid) and the minimum number of camera positions that provide the maximum view of the object surface. The presented approach consists of the following stages: emission of rays along a single hemisphere from each polygon of the object model; discretization of the obtained set of survey points; an iterative algorithm for determining a set of points that provide an overview of the object surface with a given coefficient of the model's visibility.Keywords: placement, video surveillance, analysis of terrain, visibility of geometric models, set coverage problem, point set, regular grid, model visibility ratio.