Transfer Learning from Deep Features for Remote Sensing and Poverty Mapping

TL;DR

This paper introduces a transfer learning method using CNNs trained on nighttime light data to extract socioeconomic indicators from satellite imagery, significantly aiding poverty mapping in data-scarce regions.

Contribution

It presents a novel transfer learning approach that leverages nighttime light proxies to train CNNs for poverty prediction from satellite images, overcoming data scarcity issues.

Findings

Learned features effectively predict poverty levels.

Model approaches the accuracy of field survey data.

Transfer learning reduces need for extensive labeled data.

Abstract

The lack of reliable data in developing countries is a major obstacle to sustainable development, food security, and disaster relief. Poverty data, for example, is typically scarce, sparse in coverage, and labor-intensive to obtain. Remote sensing data such as high-resolution satellite imagery, on the other hand, is becoming increasingly available and inexpensive. Unfortunately, such data is highly unstructured and currently no techniques exist to automatically extract useful insights to inform policy decisions and help direct humanitarian efforts. We propose a novel machine learning approach to extract large-scale socioeconomic indicators from high-resolution satellite imagery. The main challenge is that training data is very scarce, making it difficult to apply modern techniques such as Convolutional Neural Networks (CNN). We therefore propose a transfer learning approach where nighttime light intensities are used as a data-rich proxy. We train a fully convolutional CNN model to predict nighttime lights from daytime imagery, simultaneously learning features that are useful for poverty prediction. The model learns filters identifying different terrains and man-made structures, including roads, buildings, and farmlands, without any supervision beyond nighttime lights. We demonstrate that these learned features are highly informative for poverty mapping, even approaching the predictive performance of survey data collected in the field.