POSSIBILITIES OF VIDEO INFORMATION PROCESSING FOR INTELLIGENT VIDEO SURVEILLANCE SYSTEMS IN FOG ENVIRONMENTS USING THE CONCEPT OF THE INTERNET OF THINGS
Is devoted to the development of methods and tools for processing video information in foggy environments using neural networks and the concept of the Internet of Things (IoT). The main aspects that determine the choice of processing tools for the video surveillance operator are considered, taking into account the peculiarities of the tasks assigned to him. Solutions are proposed to increase the speed of video information processing in conditions of restrictions on their reliability and cost. An overview of possible solutions is presented to the developer of an intelligent video surveillance system. Non-trivial methods are proposed for solving the assigned tasks. We offer two-level processing of high quality video information for intelligent video surveillance systems. Methods of organizing interaction with graphic stations and high-performance computing servers are considered. Performance metrics are proposed. The efficiency indicators of the proposed methods for processing video information in foggy environments were measured and the superiority in comparison with other methods of processing video information was proved. Specific examples are given.
Authors: A. N. Subbotin, Zhukova N. А.
Direction: Informatics, Computer Technologies And Control
Keywords: Video processing in foggy environments, neural networks, foggy calculations, video processing, the concept of the Internet of Things
View full article