Video Monitoring System with Human Pose Analysis Guard-N 4.0

Yevhen Petrov

Citation: Yevhen Petrov, "Video Monitoring System with Human Pose Analysis Guard-N 4.0", Universal Library of Engineering Technology, Special Issue.

Copyright: This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The article analyzes the features inherent in the architecture of the software suite Guard-N 4.0, oriented toward automating monitoring in video surveillance systems. The relevance of the study is determined by the need to increase the efficiency of security services by minimizing the human factor and automatically detecting potentially dangerous situations. The scientific novelty lies in a hybrid methodology: neural network extraction of a human skeletal model is integrated with subsequent deterministic pose analysis based on geometric predicates, which ensures high performance on mass-market hardware configurations. The work sequentially describes the main components of the solution — from video stream capture and preprocessing to pose classification and the operator notification mechanism. As the methodological basis, the results of other studies were used and analyzed. Special attention is devoted to the algorithm for recognizing key body states (falling, squatting, shooter pose, raised hands). The aim of the study is to demonstrate the effectiveness of the proposed approach for motion capture and body positioning in three-dimensional space; to this end, methods of computer vision, machine learning, and multithreaded programming are employed. In conclusion, testing results are presented, and the practical viability of the proposed architecture in real operating conditions is confirmed. The material will be of interest to engineers, security system developers, and video analytics specialists.


Keywords: Anomaly Detection, Behavior Analysis, Computer Vision, Pose Recognition, OpenCV, Python, Skeletal Tracking.

Download doi https://doi.org/10.70315/uloap.ulete.2023.005