Intelligent video surveillance mechanisms for abnormal activity recognition in real-time: a systematic literature review

Abstract

Video surveillance plays a crucial role in securing indoor and outdoor locations in today's unreliable world, particularly in real-time applications for behaviour detection, comprehension, and labelling activities as normal or suspicious. For example, in the development of automated video surveillance systems, smart video reconnaissance systems based on picture recognition and activity recognition that detect violent behaviours is basic to forestalling wrongdoings and giving public security. According to the literature, Artificial Intelligent, Machine Learning, and deep portrayal-based approaches have been effectively utilized in image recognition and human activity observation tasks. In this literature review, a 3D convolution neural network based on deep learning is used as the proposed methodology. Thus, this article completed a Systematic Literature Review (SLR) in light of intelligent video surveillance to real-time identify abnormal activities from 2016 to 2021. In this current study, 50 research papers were considered and based on the screen filtering, the most suitable 16 papers were filtered based on intelligent video surveillance and real-time abnormal activities. Furthermore, this study identifies potential areas for improvement in intelligent video surveillance systems that can enhance public safety and security, underscoring the importance of ongoing research in this field.