PGRID: Power Grid Reconstruction in Informal Developments Using High-Resolution Aerial Imagery

TL;DR

The paper addresses the lack of high-resolution power grid maps in refugee camps and informal settlements, hindering energy access planning. It proposes PGRID, a pipeline that uses high-resolution overhead imagery to detect electrical poles and segment distribution lines, followed by post-processing to produce a unified grid layout. The method combines pole detection via FCN8 with a four-component loss and line segmentation via asymmetric DeepLabV3 with patch-wise labeling, trained with limited labels and hard-negative mining, and evaluated on benchmark datasets and Turkana camps. Results show F1 scores of about 0.71 for pole detection and 0.82 for line segmentation in Turkana, and demonstrate that PGRID can augment existing power-grid datasets, enabling precise, open-access grid maps for humanitarian use.

Abstract

As of 2023, a record 117 million people have been displaced worldwide, more than double the number from a decade ago [22]. Of these, 32 million are refugees under the UNHCR mandate, with 8.7 million residing in refugee camps. A critical issue faced by these populations is the lack of access to electricity, with 80% of the 8.7 million refugees and displaced persons in camps globally relying on traditional biomass for cooking and lacking reliable power for essential tasks such as cooking and charging phones. Often, the burden of collecting firewood falls on women and children, who frequently travel up to 20 kilometers into dangerous areas, increasing their vulnerability.[7] Electricity access could significantly alleviate these challenges, but a major obstacle is the lack of accurate power grid infrastructure maps, particularly in resource-constrained environments like refugee camps, needed for energy access planning. Existing power grid maps are often outdated, incomplete, or dependent on costly, complex technologies, limiting their practicality. To address this issue, PGRID is a novel application-based approach, which utilizes high-resolution aerial imagery to detect electrical poles and segment electrical lines, creating precise power grid maps. PGRID was tested in the Turkana region of Kenya, specifically the Kakuma and Kalobeyei Camps, covering 84 km2 and housing over 200,000 residents. Our findings show that PGRID delivers high-fidelity power grid maps especially in unplanned settlements, with F1-scores of 0.71 and 0.82 for pole detection and line segmentation, respectively. This study highlights a practical application for leveraging open data and limited labels to improve power grid mapping in unplanned settlements, where the growing number of displaced persons urgently need sustainable energy infrastructure solutions.

Figures (9)

• Figure 1: Profile views (left) better support mapping individual lines and tower structures, while overhead imagery (right) better supports mapping the layout of a power grid.
• Figure 2: Illustration of strict (left) and all (right) matching variants for evaluating performance of pole predictions. Ground truth poles are shown at the center in blue, with predicted poles in green surrounding the ground truth pole. The dashed circle represents the threshold ($th$). In the strict match variant, the ground truth pole is matched to its closest prediction, so long as the prediction is within the threshold. In the all variant, the ground truth pole is matched to all predicted poles within the threshold. The all match is a better approach to evaluate performance for cases where mask predictions for poles is a non-continuous blob.
• Figure 3: Spatially distinct geographic train and test splits for model training. Models were trained on the Kakuma Camp (right) and tested on the Kalobeyei Camp (left).
• Figure 4: Test set sample images with predicted poles from the trained pole detection model shown as red blobs.
• Figure 5: Left: Sample images showing electrical lines (top), ground truth lines buffered by 2 m (middle) and model predictions (bottom). Right: Sample ground truth power line (blue) alongside on predicted power line (green). Electrical poles are shown as circles.