Search academic texts

Registration Number

0221U101248, 0119U100755 , R & D reports

Title

Methods and means of structural and statistical identification of hierarchical objects by the characteristic points of their contours.

Head

Zahorodnia Diana I., Кандидат технічних наук

Registration Date

18-01-2021

Organization

Ternopil National Economic University

Description

The object of research is the process of processing visual information and decision-making in conditions of uncertainty in automated video surveillance systems. The subject of the research is methods and means of object identification that increase the efficiency of automated video surveillance systems. The purpose of scientific work is to develop methods and tools aimed at improving the efficiency of automated video surveillance systems by reducing the amount of data processed. Research methods - methods of system analysis, methods of contour selection, wavelet analysis, functional analysis, cluster analysis, methods of mathematical statistics, methods of modeling and programming. The results of the study and their novelty: as a result of scientific work for the first time developed a method of selecting characteristic contour points with adjustable detail of objects, based on wavelet analysis of the curvature function, which reduced the number of contour points in accordance with the processing; received further development: 1) the method of contour segmentation based on the Canney method, a distinctive feature of which is the use of wavelet functions to emphasize the differences in image intensity, which allowed to obtain a sequence of contour preparations of a hierarchical object with adjustable detail; 2) combined adaptive method of structural-statistical identification of objects based on the application of a hierarchical sequence of identification vectors, calculated on the basis of characteristic points of the contour of the object with adjustable detail, which increased the efficiency of DIA with high reliability; 3) the method of data classification based on the methods of cluster analysis, which allowed to reduce the processing time of the frame with a slight decrease in reliability, which increased the efficiency; 4) information technology of decision-making on the form of objects of hierarchical structure on the basis of methods of segmentation of DIA.

Product Description

popup.authors

Datsko Mykola M
Dorosh Vitalii I
Zahorodnia Diana I.
Zolotukhin Denys V.
Kit Ivan R.
Kaniovskyi Andrii A.
Lipyanina-Honcharenko Hrystyna V.
Lendiuk Dmytro T.
Pastukh Taras I.
Sydor Andriy I.