PD-SEG: Population Disaggregation Using Deep Segmentation Networks For Improved Built Settlement Mask

TL;DR

This paper introduces PD-SEG, a deep segmentation network-based method for high-resolution population disaggregation in developing countries, improving accuracy over existing datasets by incorporating satellite imagery and POI data.

Contribution

The paper presents a novel deep segmentation approach that combines satellite imagery and POI data for precise population estimation at 30m resolution in developing nations.

Findings

Achieved more accurate population masks compared to existing datasets.

Effectively excluded non-residential areas using POI data.

Demonstrated improved spatial resolution in population estimates.

Abstract

Any policy-level decision-making procedure and academic research involving the optimum use of resources for development and planning initiatives depends on accurate population density statistics. The current cutting-edge datasets offered by WorldPop and Meta do not succeed in achieving this aim for developing nations like Pakistan; the inputs to their algorithms provide flawed estimates that fail to capture the spatial and land-use dynamics. In order to precisely estimate population counts at a resolution of 30 meters by 30 meters, we use an accurate built settlement mask obtained using deep segmentation networks and satellite imagery. The Points of Interest (POI) data is also used to exclude non-residential areas.