Page 30 - Research and Innovation Pulse
P. 30
Graph-Based Temporal Attention Network for Anomaly
Chapter 7: The Economic Impact of Weather and Climate
Recognition in Internet of Things Video Surveillance
Author(s): Author(s):
Tol, R.S.J. (Department of Economics, University of Sussex; Institute for Environmental Studies, Vrije Universiteit; Ullah, W. (Mohamed Bin Zayed University of Artificial Intelligence, Department of Machine Learning), Khan, L.U. (Abu
Department of Spatial Economics, Vrije Universiteit; Tinbergen Institute; CESifo; Payne Institute for Earth Resources, Dhabi University, Department of Computer Science and Information Technology), Guizani, M. (Mohamed Bin Zayed
Colorado School of Mines; College of Business, Abu Dhabi University) University of Artificial Intelligence, Department of Machine Learning), Wang, C.-D. (Sun Yat-sen University), Wu, D. (Sun
Yat-sen University)
Index Terms:
Index Terms:
Climate change; D24; economic growth; error-correction model; O44; O47; Q54; stochastic frontier analysis; weather
shocks Anomaly detection; Benchmarking; Graphic methods; Network security; Real time systems; Security systems; Anomaly
recognition; Graph convolution network; Graph-based; Intelligent security; Memory network; Public safety; Surveillance
Abstract: systems; Surveillance video; Temporal attention memory; Video surveillance; Convolution
I propose a new conceptual framework to disentangle the impacts of weather and climate on economic activity Abstract:
and growth: A stochastic frontier model with climate in the production frontier and weather shocks as a source of
inefficiency. I test it on a sample of 160 countries over the period 1950–2014. Temperature and rainfall determine Accurate anomaly recognition in video surveillance is critical to ensuring public safety, infrastructure protection,
production possibilities in both rich and poor countries, positively in cold countries and negatively in hot ones. and intelligent security operations. However, existing methods often fail to generalize across diverse and dynamic
Weather anomalies reduce inefficiency in rich countries but increase inefficiency in poor and hot countries and more environments due to their limited capacity to model complex spatio-temporal patterns. In this article, we propose
so in countries with low weather variability. The climate effect is larger than the weather effect. a novel graph-based temporal attention-memory network (G-TAMNet) that significantly advances video anomaly
detection by capturing intricate temporal dependencies as well as spatial relationships. By integrating temporal self-
Read the paper attention into a graph convolutional network (GCN) framework, our model enhances the learning of salient and
contextually relevant features that traditional approaches tend to overlook. Furthermore, to enable deployment in
resource-constrained settings, such as Internet of Things (IoT)-based surveillance systems, we incorporate model
quantization strategies that reduce computational overhead without compromising detection accuracy. Extensive
evaluations of three widely used benchmarks - UCF-Crime, LAD-2000, and RWF-2000 - demonstrate the effectiveness
of the proposed G-TAMNet model, achieving accuracy, precision, recall, and F1 -scores of 54.3%/61.1%/59.01%/61.1%,
79.3%/69.1%/70.0%/69.1%, and 94.0%/94.0%/94.2%/94.1%, respectively. These results reflect consistent performance
gains of 3.3%, 9.9%, and 0.74% in accuracy over existing state-of-the-art methods on the corresponding datasets.
Such improvements underscore the robustness, scalability, and practical viability of G-TAMNet for real-time anomaly
detection in resource-constrained IoT surveillance systems. In addition, its reliable generalization across diverse
environments highlights the potential of the model to advance intelligent security analytics and transform real-world
applications in the domain of information forensics.
Read the paper
Abu Dhabi University | Research and Innovation Pulse Newsletter May Edition | 31
