Transfer Learning from Deep Features for Remote Sensing and Poverty Mapping

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.

In Proc. 30th AAAI Conference on Artificial Intelligence

References in corpus (1)

• Fully Convolutional Networks for Semantic Segmentation

Cited by in corpus (16)

• Deep Visual Domain Adaptation: A Survey
• ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
• Towards Better Exploiting Convolutional Neural Networks for Remote Sensing Scene Classification
• Semantic Labeling in Very High Resolution Images via a Self-Cascaded Convolutional Neural Network
• Mapping Informal Settlements in Developing Countries using Machine Learning and Low Resolution Multi-spectral Data
• Deep Learning Based Multi-Label Text Classification of UNGA Resolutions
• A Generalizable and Accessible Approach to Machine Learning with Global Satellite Imagery
• Transfer Learning for Unseen Robot Detection and Joint Estimation on a Multi-Objective Convolutional Neural Network
• AI for Anticipatory Action: Moving Beyond Climate Forecasting
• Gradual Domain Adaptation via Self-Training of Auxiliary Models
• A Distributionally Robust Optimization Method for Adversarial Multiple Kernel Learning
• On evaluating CNN representations for low resource medical image classification
• Measuring Economic Activities of China with Mobile Big Data
• Clustering augmented Self-Supervised Learning: Anapplication to Land Cover Mapping
• Deep-learning coupled with novel classification method to classify the urban environment of the developing world
• Adversarial Transfer Learning for Cross-domain Visual Recognition