Algorithms for selecting the outlines of objects images in intellectual video surveillance systems

DOI: 10.31673/2412-9070.2020.053539

Abstract

Intensive development of means of receiving and transmitting digital images creates the problem of processing huge amounts of video information flows. There is a wide range of tasks in which images are considered as a source of information on the basis of which it is necessary to make a decision. Important tasks to be solved by intelligent video surveillance systems are: identification of objects and determination of their trajectories; measuring the speed of objects; detection of alarming events in the tasks of object-territorial protection in real time.
One of the main operations in intelligent video surveillance systems in image processing for further analysis is the selection of contours of images of objects, because the contour contains all the necessary information to recognize objects by their shape. This approach allows you to not consider the internal points of the image and, thus, significantly reduce the amount of information processed. This makes it possible to analyze images in real time.
Contour analysis is a set of methods for selecting, describing and processing image contours that allows you to describe, store, compare and search for objects presented in the form of their external contours, as well as effectively solve the main problems of pattern recognition — transfer, rotate and zoom image of the object. In this case, the contour means a space-length gap, difference or abrupt change in brightness values.
The purpose of the publication is to consider the algorithms for selecting the contours of images of objects in the problems of detecting alarming events by intelligent video surveillance systems.
The problem of selection of contours of images of objects in problems of detection of disturbing events by intelligent systems of video surveillance is considered. In order to improve the basic characteristics of intelligent video surveillance systems, algorithms for contouring images of objects are proposed to ensure the detection of four types of alarming events: the appearance and presence of the object in the surveillance zone, moving the object in the forbidden direction, leaving the object and overturning the object.

Keywords: intelligent video surveillance systems; video analytics; alarm detection; digital image processing; object detection; object tracking; object image contours; contour analysis.

References
1. Ainsworth T. Buyer Beware // Security Oz. 2002. Vol. 19. P. 18–26.
2. Сальников И. И. Критерии отнесения устройств и систем обработки информации к интеллектуальным // XXI век: итоги прошлого и проблемы настоящего плюс. Пенза: Изд-во Пенз. гос. технол. акад., 2012. С. 11–15.
3. Крючкова Л. П., Кременський М. С. Методи виявлення тривожних подій в інтелектуальних системах відеоспостереження // Сучасний захист інформації. 2019. №3. С. 64–69.
4. Heikkila M., Pietikainen M. A texture-based method for modeling the background and detecting moving objects // IEEE Transactions on Pattern Analysis and Machine Intelligence. 2006. vol. 28, no. 4. Р. 657–662.
5. Сакович И. О., Белов Ю. С. Обзор основных методов контурного анализа для выделения контуров движущихся объектов // Инженерный журнал: наука и инновации: електрон. версія журн. 2014. Вып. 12. URL: http://engjournal.ru/catalog/it/hidden/1280.html
6. Сирота А. А., Соломатин А. И. Статистические алгоритмы обнаружения границ объектов на изображениях // Вестник ВГУ. 2008. № 1. С. 58–64 (Сер. Системный анализ и информационные технологии).
7. Хачумов М. В. Сжатие, передача и распознавание контуров ригидных объектов, описанных цепными кодами // Современные наукоемкие технологии: електрон. версія журн. 2020. № 8. С. 79–85. URL: http://www.top-technologies.ru/ru/article/view?id=38177 (дата обращения: 24.09.2020).
8. Shih Frank. Image processing and pattern recognition: fundamentals and techniques // IEEE Press, 2010. 537 p.
9. Гонсалес Р. С., Вудс Р. Э. Цифровая обработка изображений: 3-е изд., испр. и доп. Москва: Техносфера, 2012. 1104 с.