Generating Interpretable Poverty Maps using Object Detection in Satellite Images

• URL: http://arxiv.org/abs/2002.01612v2
• Date: Tue, 18 Feb 2020 02:02:57 GMT
• Title: Generating Interpretable Poverty Maps using Object Detection in Satellite Images
• Authors: Kumar Ayush, Burak Uzkent, Marshall Burke, David Lobell, Stefano Ermon
• Abstract summary: We demonstrate an interpretable computational framework to accurately predict poverty at a local level by applying object detectors to satellite images. Using the weighted counts of objects as features, we achieve 0.539 Pearson's r2 in predicting village-level poverty in Uganda, a 31% improvement over existing (and less interpretable) benchmarks.
• Score: 80.35540308137043
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate local-level poverty measurement is an essential task for governments and humanitarian organizations to track the progress towards improving livelihoods and distribute scarce resources. Recent computer vision advances in using satellite imagery to predict poverty have shown increasing accuracy, but they do not generate features that are interpretable to policymakers, inhibiting adoption by practitioners. Here we demonstrate an interpretable computational framework to accurately predict poverty at a local level by applying object detectors to high resolution (30cm) satellite images. Using the weighted counts of objects as features, we achieve 0.539 Pearson's r^2 in predicting village-level poverty in Uganda, a 31% improvement over existing (and less interpretable) benchmarks. Feature importance and ablation analysis reveal intuitive relationships between object counts and poverty predictions. Our results suggest that interpretability does not have to come at the cost of performance, at least in this important domain.

Related papers

• Fairness and representation in satellite-based poverty maps: Evidence of urban-rural disparities and their impacts on downstream policy [5.456665139074406]
  This paper investigates disparities in representation, systematic biases in prediction errors, and fairness concerns in satellite-based poverty mapping across urban and rural lines. Our findings highlight the importance of careful error and bias analysis before using satellite-based poverty maps in real-world policy decisions.
  arXiv  Detail & Related papers  (2023-05-02T21:07:35Z)
• Learning and Reasoning Multifaceted and Longitudinal Data for Poverty Estimates and Livelihood Capabilities of Lagged Regions in Rural India [22.98110639419913]
  The project aims to examine the poverty situation of rural India for the period of 1990-2022 based on the quality of life and livelihood indicators. The project also intends to examine causation and longitudinal analysis to examine the reasons for poverty.
  arXiv  Detail & Related papers  (2023-04-27T05:33:08Z)
• Interpreting wealth distribution via poverty map inference using multimodal data [0.0]
  We propose a pipeline of machine learning models to infer the mean and standard deviation of wealth across multiple populated places. These models leverage seven independent and freely available feature sources based on satellite images, and metadata collected via online crowd-sourcing and social media. Our results recover the local mean and variation of wealth, and correctly capture the positive yet non-monotonous correlation between them.
  arXiv  Detail & Related papers  (2023-02-17T11:35:44Z)
• Graph-based Village Level Poverty Identification [52.12744462605759]
  The development of the Web infrastructure and its modeling tools provides fresh approaches to identifying poor villages. By modeling the village connections as a graph through the geographic distance, we show the correlation between village poverty status and its graph topological position. We propose the first graph-based method to identify poor villages.
  arXiv  Detail & Related papers  (2023-02-14T06:58:40Z)
• Building Coverage Estimation with Low-resolution Remote Sensing Imagery [65.95520230761544]
  We propose a method for estimating building coverage using only publicly available low-resolution satellite imagery. Our model achieves a coefficient of determination as high as 0.968 on predicting building coverage in regions of different levels of development around the world.
  arXiv  Detail & Related papers  (2023-01-04T05:19:33Z)
• Predicting Poverty Level from Satellite Imagery using Deep Neural Networks [0.0]
  I develop deep learning computer vision methods that can predict a region's poverty level from an overhead satellite image. I explore the impact that data quantity and data augmentation have on the representational power and overall accuracy of the networks.
  arXiv  Detail & Related papers  (2021-11-30T18:57:24Z)
• Interpretable Poverty Mapping using Social Media Data, Satellite Images, and Geospatial Information [0.0]
  We present a interpretable and cost-efficient approach to poverty estimation using machine learning and readily accessible data sources. We achieve an $R2$ of 0.66 for wealth estimation in the Philippines, compared to 0.63 using satellite imagery.
  arXiv  Detail & Related papers  (2020-11-27T05:24:53Z)
• Accurate and Robust Feature Importance Estimation under Distribution Shifts [49.58991359544005]
  PRoFILE is a novel feature importance estimation method. We show significant improvements over state-of-the-art approaches, both in terms of fidelity and robustness.
  arXiv  Detail & Related papers  (2020-09-30T05:29:01Z)
• Predicting Livelihood Indicators from Community-Generated Street-Level Imagery [70.5081240396352]
  We propose an inexpensive, scalable, and interpretable approach to predict key livelihood indicators from public crowd-sourced street-level imagery. By comparing our results against ground data collected in nationally-representative household surveys, we demonstrate the performance of our approach in accurately predicting indicators of poverty, population, and health.
  arXiv  Detail & Related papers  (2020-06-15T18:12:12Z)
• Efficient Poverty Mapping using Deep Reinforcement Learning [75.6332944247741]
  High-resolution satellite imagery and machine learning have proven useful in many sustainability-related tasks. The accuracy afforded by high-resolution imagery comes at a cost, as such imagery is extremely expensive to purchase at scale. We propose a reinforcement learning approach in which free low-resolution imagery is used to dynamically identify where to acquire costly high-resolution images.
  arXiv  Detail & Related papers  (2020-06-07T18:30:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.