Towards Explaining Satellite Based Poverty Predictions with Convolutional Neural Networks

Summary, in English

Deep convolutional neural networks (CNNs) have been shown to predict poverty and development indicators from satellite images with surprising accuracy. This paper presents a first attempt at analyzing the CNNs responses in detail and explaining the basis for the predictions. The CNN model, while trained on relatively low resolution day- and night-time satellite images, is able to outperform human subjects who look at high-resolution images in ranking the Wealth Index categories. Multiple explainability experiments performed on the model indicate the importance of the sizes of the objects, pixel colors in the image, and provide a visualization of the importance of different structures in input images. A visualization is also provided of type images that maximize the network prediction of Wealth Index, which provides clues on what the CNN prediction is based on.

• Author: Hamid Sarmadi; Thorsteinn Rognvaldsson; Nils Roger Carlsson; Mattias Ohlsson; Ibrahim Wahab; Ola Hall
• Editor: Yannis Manolopoulos; Zhi-Hua Zhou
• Department/s: Centre for Environmental and Climate Science (CEC); Department of Human Geography
• Publishing year: 2023
• Language: English
• Publication/Series: 2023 IEEE 10th International Conference on Data Science and Advanced Analytics, DSAA 2023 - Proceedings
• Links: http://dx.doi.org/10.1109/DSAA60987.2023.10302541
• Document type: Conference paper
• Publisher: IEEE - Institute of Electrical and Electronics Engineers Inc.
• Topic: Social Sciences Interdisciplinary
• Keywords: Deep Convolutional Neural Networks; Explainable AI; Poverty prediction; Satellite Images
• Conference name: 10th IEEE International Conference on Data Science and Advanced Analytics, DSAA 2023
• Conference date: 2023-10-09 - 2023-10-12
• Conference place: Thessaloniki, Greece
• Status: Published
• ISBN/ISSN/Other: ISBN: 9798350345032