Enhancing the KidSat Model: Integrating Geographical Encoding and Data Quality Assessment for Childhood Poverty Prediction

2026-07-09

AI Summary

This study addresses key challenges in predicting child poverty from satellite imagery—namely sparse survey data, poor image quality due to cloud cover, and the absence of explicit spatial structure in predictive models. Building upon the KidSat framework, the authors introduce three critical enhancements: re-aggregating Demographic and Health Surveys (DHS) data to construct an optimized target matrix, implementing a two-stage image quality filtering mechanism, and integrating DINOv2 visual embeddings with spherical harmonic-based geographic encoding to strengthen spatial representation. Experimental results across 33 African countries demonstrate that the proposed approach reduces the mean absolute error (MAE) in cluster-level predictions of severe deprivation from 0.2167 to 0.1658, achieving an 18.83% relative error reduction and confirming the efficacy of gradient-boosted tree models (XGBoost/LightGBM) as prediction heads.

Abstract

Accurate poverty mapping using satellite imagery is often hindered by (i) noisy and sparse survey-derived supervision, (ii) image quality issues such as cloud cover and image corruption, and (iii) lack of explicit spatial structure in image-only models. Building on the KidSat framework, we develop an enhanced pipeline that improves predictive accuracy via refined data preprocessing, systematic image quality assessment, and mathematically defined geographic encoding. First, we refine the fine-tuning target matrix by resolving high-cardinality sparsity and reducing one-hot dimensionality from 103 to 51 via DHS re-aggregation. Second, we introduce a simple two-stage quality-screening procedure to filter heavily clouded or corrupted observations. Third, we fuse DINOv2 visual embeddings with Spherical Harmonics (SH) location features. Across extensive experiments, these changes reduce MAE from 0.2167 to 0.1759, corresponding to an 18.83% relative reduction on the cluster-level severe-deprivation proportion scale. When extended from 16 to 33 African countries, the best-performing configuration achieves an overall MAE of 0.1658. We find that SH features consistently improve performance over the image-only backbone, whereas higher-capacity coordinate Multi Layer Perception augmentation (SH+SIREN) can underperform without carefully designed objectives. Finally, gradient-boosted tree heads (XGBoost/LightGBM) most effectively exploit nonlinear interactions in the fused visual-geographic representation. These findings provide a scalable and principled recipe for improving satellite-based socioeconomic predictions using only publicly accessible data.

Problem

• poverty prediction
• satellite imagery
• data quality
• spatial structure
• noisy supervision

Innovation

• geographic encoding
• data quality assessment
• Spherical Harmonics
• satellite-based poverty prediction
• DINOv2 embeddings