• Debraj Roy

Mapping shelter deprivation in India using satellite images

Updated: Nov 12, 2018

An approach that combines machine learning with high-resolution satellite imagery to provide new data on socioeconomic indicators of poverty and wealth..

an approach that combines machine learning with high-resolution satellite imagery to provide new data on socioeconomic indicators of poverty and wealth.

Slums, characterized by sub-standard housing conditions, are a common in fast growing Asian cities. The elimination of poverty worldwide is the first of 17 UN Sustainable Development Goals for the year 2030. To track progress towards this goal, we require frequent and reliable data on the distribution of poverty than traditional data collection methods can provide. However, reliable and up-to-date information on their locations and development dynamics is scarce. New methods are emerging on how to measure poverty, consumption expenditure, and asset wealth from high-resolution satellite imagery that may be able to provide systematic and objective-based measures on where these households are more likely to be found when other data is not available. Despite numerous studies, the task of delineating slum areas remains a challenge and no general agreement exists about the most suitable method for detecting or assessing detection performance.

“The United Street Sellers Republic — the USSR — is the second-largest economy in the world after the United States.”- Robert Neuwirth

In a recent paper, a team of researchers from the Netherlands escience Center, Computational Science Lab at the University of Amsterdam and the International Institute for Geo-Information Science and Earth Observation (ITC) have applied a standard computer vision method (Bag of Visual Words framework and Speeded-Up Robust Features) for image-based classification of slum and non-slum areas in Kalyan and Bangalore, India, using very high resolution RGB images. To delineate slum areas, image segmentation is performed as pixel-level classification for three classes: Slums, Built-up and Non-Built-up. For each of the three classes, image tiles were randomly selected using ground truth observations. A multi-class support vector machine classifier has been trained on 80% of the tiles and the remaining 20% were used for testing. The final image segmentation has been obtained by classification of every 10th pixel followed by a majority filtering assigning classes to all remaining pixels. The results demonstrate the ability of the method to map slums with very different visual characteristics in two very different Indian cities.

The Road Ahead

The study indicates that if we can predict shelter deprivation based on publicly available satellite imagery, this means that when survey data does not exist or is incomplete, we can use earth observation data to making objective decisions about where to focus poverty alleviation programs.

"The SDGs are a bold commitment to finish what we started, and end poverty in all forms and dimensions by 2030. This involves targeting the most vulnerable, increasing access to basic resources and services, and supporting communities affected by conflict and climate-related disasters." UN-Sustainable Development Goal

In the future the team is planning to use temporal satellite imagery to help us understand the dynamic nature of poverty. This would help to detect seasonal variations in vulnerability. For example, we could extract information from satellite imagery that is correlated with changing agricultural yields, inflation, or seasonal floods and droughts. The satellite images coupled with computational models such as this could also predict when households or communities move out of extreme poverty and how slums grow and emerge.

145 views0 comments