Employing Night-Time Light Images for Wealth Assessment in India: A Machine Learning Perspective

摘要

With the urbanization upsurge and rapid development, India is the country with dense population of urban dwellers. However, disparity among various states in terms of infrastructures, per-capita wealth and socio-economic dynamics is still the serious issue that hinders the development process. In this light, wealth assessment for various states becomes crucial for effective policy implementation. Although, collecting data about economic status of Indian families is highly cost extensive, motivating remote sensing as a cheaper yet robust way of measuring economic livelihood data. In this work, we combine publicly available night time light imagery which are good proxy measure for economic activities, along with recent survey data to develop machine learning based models that predict per-capita consumption in India. We have presented state-wise economic status for different states and showed the effectiveness of the proposed scheme by comparing with the ground survey data.