Research Programme

Video Analytics for Anomaly Detection

(2021-2022)

Computer Vision Edge Computing

Internet of Things (IoT) Infrastructure / Systems (IS) Data Science (DS) Human Computer Interaction (HCI)

Project Overview

Combines pre-trained vision models with edge processing to flag unusual activity for smart city deployments in farms, castles, car parks, and bus stops.

Most commercial camera systems can only detect a limited set of predefined objects, and streaming video to the cloud requires substantial bandwidth. This project addresses these limitations by combining pre-trained vision models to identify complex anomalies through edge processing rather than full real-time video analysis.

The system performs lightweight edge analysis to catch the first signs of anomalies—such as unusual animal movement, after-hours vehicle entry, or risky behaviour—before running more advanced inference. The approach reduces bandwidth costs and increases responsiveness for both rural and urban deployments.

Use Cases

The research explores four real-world smart city applications:

Farm security in Monmouthshire: Theft prevention and worker safety monitoring in agricultural settings
Raglan Castle: Vandalism detection and child safety at heritage sites
Blaenau Gwent car parks: Usage monitoring and public transport incentivisation
Bus stop monitoring: Detection of anti-social behaviour at public transport locations

Objectives

Review anomaly detection literature from camera feeds
Measure processing trade-offs between edge and cloud architectures
Explore environmental factors affecting detection performance, particularly low-light conditions
Develop adaptive systems for anomaly selection based on deployment context
Create techniques leveraging existing deep learning approaches with standard cameras