Investigation of Dependence of Brightness-Structural Features of Images Generated by the Vision System on Navigation Parameters of Unmanned Aerial Vehicles and their Use for Selection the Object of Reference

The aim of the article is to substantiate the use of brightness-structural features of images to select a reference object. This goal is achieved by studying the dependence of brightness-structural features of images on the sighting geometry. The solution to the first problem is based on the study of the brightness properties of images for their subsequent segmentation. The study was performed in the MATLAB software environment. The influence of sighting geometry on the distribution of image brightness was determined for sighting angles from -30˚ to -60 ˚ and altitudes from 500 to 1500 meters. The solution to the second problem is based on the study of the structural properties of images segmented by brightness. The solution to the third problem is to assess the efficiency of selecting a reference object on a segmented image using structural-brightness features. The most significant result is the substantiation of the use of brightness-structural features of images and the definition of the conditions for their use to select a reference object. The significance of the results lies in the analysis of brightness-structural features of images with different object content from the sighting geometry. The novelty of the work lies in the development of the procedure for forming current images and the decision function using the brightness-structural features of images. The difference from known works is the use of a new set of informative features in the formation of a segmented image, which is necessary for making a decision on the selection of a reference object.