Project Abstract

Abstract:With the proliferation of closed-circuit television (CCTV) systems in various domains such as public safety, transportation, and retail, the need for effective anomaly detection methods has become increasingly crucial. This paper presents a comprehensive review of data analysis techniques employed in CCTV anomaly detection systems. We explore various approaches, including machine learning algorithms, deep learning models, and statistical methods, highlighting their strengths and limitations. Additionally, we discuss the challenges associated with real-world CCTV data, such as variations in lighting conditions, occlusions, and scene complexity. Furthermore, we examine the importance of dataset curation, feature selection, and model evaluation in ensuring the robustness and generalization of anomaly detection systems. By synthesizing existing research and identifying areas for future exploration, this paper aims to provide insights and guidance for researchers and practitioners in the field of CCTV anomaly detection.

Keywords: ARTIFICIAL INTELLIGENCE, MACHINE LEARNING ALGORITHMS, ANOMALY DETECTION

 Conference Details

Session: Presentation Stream 5 at Presentation Slot 7

Location: GH011 at Tuesday 7th 13:30 – 17:00

Markers: Nader Al Khatib (GTA), Tom Owen

Course: MSc Advanced Computer Science, Masters PG

Future Plans: I’m looking for work