arXiv subjects

Edward Oughton

Publications and source records attributed to Edward Oughton.

5 recordsLinked to original sources

arXiv2026

Infra-Bench CLS: A Global, Open-Source Benchmark for Critical Infrastructure Classification with Earth Observation Foundation Models

Critical infrastructure location data is often incomplete and unevenly distributed globally, especially in developing regions. Earth observation foundation models are proposed as a new step in enabling us to more efficiently understand the natural and built environment, raising questions as to their effectiveness in performing challenging downstream tasks. Yet, foundation models remain largely untested for detecting and classifying the facility-scale critical infrastructure that underpins a range of important societal and economic functions. Subsequently, Infra-Bench CLS is introduced as a benchmark to test foundation models on 18,756 Sentinel-1 SAR and Sentinel-2 multispectral facility-scale critical infrastructure asset images covering seven continents and 13 infrastructure classes, with results reported for the 10 retained classes. Using linear probing and fine-tuning for two training dataset levels (1.0x and 0.3x), seven foundation models are evaluated (SatlasPretrain S2, SatlasPretrain S1, CROMA, Prithvi-EO-2.0, AlphaEarth Foundations, OlmoEarth v1.1-Base, and DINOv3 ViT-L/16). When comparing macro F1 scores to a ResNet-18 supervised baseline of 39.2 percent, the best foundation model achieved 57.9 percent, a 48 percent improvement. Top performing classes were airports (F1 85.3 percent), train stations (F1 82.1 percent), and data centers (F1 77.6 percent). By contrast, many of the power sector classes perform poorly (F1 27.5-46.2 percent). These findings suggest foundation models can enable superior critical infrastructure classification, but future work should evaluate performance on higher-resolution imagery, particularly for poorly performing sectors, such as power.

cs.CV↗

arXiv2025

Assessing the Sensitivities of Input-Output Methods for Natural Hazard-Induced Power Outage Macroeconomic Impacts

It is estimated that over one-fourth of US households experienced a power outage in 2023, costing on average US $\$150$ Bn annually, with $87\%$ of outages caused by natural hazards. Indeed, numerous studies have examined the macroeconomic impact of power network interruptions, employing a wide variety of modeling methods and data parameterization techniques, which warrants further investigation. In this paper, we quantify the macroeconomic effects of three significant natural hazard-induced US power outages: Hurricane Ian (2022), the 2021 Texas Blackouts, and Tropical Storm Isaias (2020). Our analysis evaluates the sensitivity of three commonly used data parameterization techniques (household interruptions, kWh lost, and satellite luminosity), along with three static models (Leontief and Ghosh, critical input, and inoperability Input-Output). We find the mean domestic loss estimates to be US $\$3.13$ Bn, US $\$4.18$ Bn, and US $\$2.93$ Bn, respectively. Additionally, data parameterization techniques can alter estimated losses by up to $23.1\%$ and $50.5\%$. Consistent with the wide range of outputs, we find that the GDP losses are highly sensitive to model architecture, data parameterization, and analyst assumptions. Results sensitivity is not uniform across models and arises from important a priori analyst decisions, demonstrated by data parameterization techniques yielding $11\%$ and $45\%$ differences within a model. We find that the numerical value output is more sensitive than intersectoral linkages and other macroeconomic insights. To our knowledge, we contribute to literature the first systematic comparison of multiple IO models and parameterizations across several natural hazard-induced long-duration power outages, providing guidance and insights for analysts.

econ.GN↗

arXiv2023

The contribution of US broadband infrastructure subsidy and investment programs to GDP using input-output modeling

More than one-fifth of the US population does not subscribe to a fixed broadband service despite broadband being a recognized merit good. For example, less than 4% of citizens earning more than US \$70k annually do not have broadband, compared to 26% of those earning below US \$20k annually. To address this, the federal government has undertaken one of the largest broadband investment programs ever via The Bipartisan Infrastructure Law, with the aim of addressing this disparity and expanding broadband connectivity to all citizens. We examine broadband availability, adoption, and need for each US state, and then construct an Input-Output model to explore the potential structural macroeconomic impacts of broadband spending on Gross Domestic Product (GDP) and supply chain linkages. In terms of macroeconomic impact, the total potential indirect contribution to US GDP by the program could be as high as US \$84.8 billion, \$32.7 billion, and \$9.78 billion for the Broadband Equity, Access, and Deployment program, the Affordable Connectivity Program, and additional programs, respectively. Thus, overall, the broadband allocations could expand US GDP by up to US \$127.3 billion (0.10% of annual US GDP over the next five years). Moreover, the broadband packages within the Bipartisan Infrastructure Law could create up to 230,000 jobs (0.14% labor market increase). We contribute one of the first economic impact assessments of the US Bipartisan Infrastructure Law to the literature.

econ.GN↗

arXiv2022

Assessing thermal imagery integration into object detection methods on ground-based and air-based collection platforms

Object detection models commonly deployed on uncrewed aerial systems (UAS) focus on identifying objects in the visible spectrum using Red-Green-Blue (RGB) imagery. However, there is growing interest in fusing RGB with thermal long wave infrared (LWIR) images to increase the performance of object detection machine learning (ML) models. Currently LWIR ML models have received less research attention, especially for both ground- and air-based platforms, leading to a lack of baseline performance metrics evaluating LWIR, RGB and LWIR-RGB fused object detection models. Therefore, this research contributes such quantitative metrics to the literature. The results found that the ground-based blended RGB-LWIR model exhibited superior performance compared to the RGB or LWIR approaches, achieving a mAP of 98.4%. Additionally, the blended RGB-LWIR model was also the only object detection model to work in both day and night conditions, providing superior operational capabilities. This research additionally contributes a novel labelled training dataset of 12,600 images for RGB, LWIR, and RGB-LWIR fused imagery, collected from ground-based and air-based platforms, enabling further multispectral machine-driven object detection research.

cs.CV↗

arXiv2021

Policy options for digital infrastructure strategies: A simulation model for broadband universal service in Africa

Internet access is essential for economic development and helping to deliver the Sustainable Development Goals, especially as even basic broadband can revolutionize available economic opportunities. Yet, more than one billion people still live without internet access. Governments must make strategic choices to connect these citizens, but currently have few independent, transparent and scientifically reproducible assessments to rely on. This paper develops open-source software to test broadband universal service strategies which meet the 10 Mbps target being considered by the UN Broadband Commission. The private and government costs of different infrastructure decisions are quantified in six East and West African countries (C\^ote D`Ivoire, Mali, Senegal, Kenya, Tanzania and Uganda). The results provide strong evidence that `leapfrogging` straight to 4G in unconnected areas is the least-cost option for providing broadband universal service, with savings between 13-51% over 3G. The results also demonstrate how the extraction of spectrum and tax revenues in unviable markets provide no net benefit, as for every $1 taken in revenue, a $1 infrastructure subsidy is required from government to achieve broadband universal service. Importantly, the use of a Shared Rural Network in unviable locations provides impressive cost savings (up to 78%), while retaining the benefits of dynamic infrastructure competition in viable urban and suburban areas. This paper provides evidence to design national and international policies aimed at broadband universal service.

cs.CY↗