OPTIMIZED INTELLIGENT MODULES FOR VIDEO SURVEILLANCE AND MONITORING

Abstract

Platform of optimized intelligent modules, designed for intelligent video surveillance and control systems, is an outstanding tool for implementing effective security measures and improving monitoring processes. This paper examines the development and optimization of several important modules, including a facial recognition system, line intersection detection algorithms, methods for detecting the presence of a person in specified zones and an algorithm for searching for people. The most productive algorithms, such as Faster R-CNN(Faster Region-based Convolutional Neural Network), YOLO(You Only Look Once), SSD(Single-Shot Detector), were researched, thereby work was carried out to improve and optimize the YOLO algorithm. In addition to the use of this algorithm, optimizations were carried out, including image processing methods (including scaling) and a frame skipping mechanism using parallel computing, which significantly reduced the computational load. The resulting platform provides users with the ability to effectively monitor and analyze video streams, automatically identify potential threats and events, which makes it the optimal solution for ensuring security in a variety of applications, including public places, enterprises and critical infrastructure facilities. The results of this paper provide new prospects for improving video surveillance and control systems, contributing to an increase in the level of security and efficiency of actions.