On monitoring development using high resolution satellite images

We develop a machine learning based tool for accurate prediction of development and socio-economic indicators from high resolution day-time satellite imagery. The indicators that we use are derived from the Census 2011 [The Ministry of Home Affairs, Government of India, 2011] and the NFHS-4 [The Ministry of Health and Family Welfare, Government of India, 2016] survey data. We use a deep convolutional neural network to build a model for regression of asset indicators from satellite images. We show that the direct regression of asset indicators gives superior R2 scores compared to that of transfer learning through night light data, which is a popular proxy for economic development used world wide. We also use the asset prediction model for accurate transfer learning of other socio-economic and health indicators which are not intuitively related to observable features in satellite images, or are not always well correlated with each other. The tool can be extended to monitor the progress of development of a region over time, and to flag potential anomalies because of dissimilar outcomes due to different policy interventions in a geographic region by detecting sharp spatial discontinuities in the regression output.

[1]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[2]  Thomas M. Stoker,et al.  There Is No Free House: Ethnic Patronage in a Kenyan Slum , 2019, American Economic Journal: Applied Economics.

[3]  P. Sutton,et al.  Shedding Light on the Global Distribution of Economic Activity , 2010 .

[4]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[5]  P. Sutton,et al.  Using Nighttime Satellite Imagery as a Proxy Measure of Human Well-Being , 2013 .

[6]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[7]  Sendhil Mullainathan,et al.  Machine Learning: An Applied Econometric Approach , 2017, Journal of Economic Perspectives.

[8]  T. Fujii,et al.  Is Predicted Data a Viable Alternative to Real Data? , 2016, The World Bank Economic Review.

[9]  A. Tatem,et al.  Using remotely sensed night-time light as a proxy for poverty in Africa , 2008, Population health metrics.

[10]  Wilhelm Burger,et al.  Digital Image Processing - An Algorithmic Introduction using Java , 2008, Texts in Computer Science.

[11]  S. Shalev-Shwartz,et al.  Stochastic Gradient Descent , 2014 .

[12]  J. Kleinberg,et al.  Prediction Policy Problems. , 2015, The American economic review.

[13]  P. Sutton,et al.  The Night Light Development Index (NLDI): a spatially explicit measure of human development from satellite data , 2012 .

[14]  Edgar Erdfelder,et al.  G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences , 2007, Behavior research methods.

[15]  M. Pal,et al.  Role of individual and household level factors on stunting: A comparative study in three Indian states , 2007, Annals of human biology.

[16]  Budhendra L. Bhaduri,et al.  A global poverty map derived from satellite data , 2009, Comput. Geosci..

[17]  Susan Athey,et al.  The State of Applied Econometrics - Causality and Policy Evaluation , 2016, 1607.00699.

[18]  Alois Knoll,et al.  Gradient boosting machines, a tutorial , 2013, Front. Neurorobot..

[19]  A. Storeygard,et al.  The View from Above: Applications of Satellite Data in Economics , 2016 .

[20]  Andrea Vedaldi,et al.  Visualizing Deep Convolutional Neural Networks Using Natural Pre-images , 2015, International Journal of Computer Vision.

[21]  D. Donaldson,et al.  Evolving Comparative Advantage and the Impact of Climate Change in Agricultural Markets: Evidence from 1.7 Million Fields around the World , 2014, Journal of Political Economy.

[22]  Robin Burgess,et al.  Environmental policy and the economic downturn , 2010 .

[23]  Ryan E. Wiegand.  Introductory Econometrics: Using Monte Carlo Simulation With Microsoft Excel , 2006 .

[24]  Marta C. González,et al.  Using Convolutional Networks and Satellite Imagery to Identify Patterns in Urban Environments at a Large Scale , 2017, KDD.

[25]  S. Jayachandran.  Air Quality and Early-Life Mortality: Evidence from Indonesia's Wildfires , 2008 .

[26]  Didier Sornette,et al.  Dynamics and spatial distribution of global nighttime lights , 2013, EPJ Data Science.

[27]  R. Barro,et al.  A New Data Set of Educational Attainment in the World, 1950-2010 , 2010 .

[28]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[29]  Robert W. C. Brown.  Comparing Census Data in 90 Countries , 1971 .

[30]  P. Novosad,et al.  Politics and Local Economic Growth: Evidence from India , 2017 .

[31]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.

[32]  Sang Michael Xie,et al.  Combining satellite imagery and machine learning to predict poverty , 2016, Science.

[33]  Quanshi Zhang,et al.  Interpreting CNN knowledge via an Explanatory Graph , 2017, AAAI.

[34]  K. Roychowdhury,et al.  Night Lights and Economic Activity in India: A study using DMSP-OLS night time images , 2011 .

[35]  W. Nordhaus,et al.  Using luminosity data as a proxy for economic statistics , 2011, Proceedings of the National Academy of Sciences.

[36]  Alessandro Tarozzi,et al.  Stunting among Children Facts and Implications , 2013 .

[37]  M. Lazar.  Shedding Light on the Global Distribution of Economic Activity , 2010 .

[38]  Christopher D. Elvidge,et al.  Estimating the Information and Technology Development Index (IDI) Using Nighttime Satellite Imagery , 2010 .

[39]  Christopher D. Elvidge,et al.  Estimation of Mexico's Informal Economy and Remittances Using Nighttime Imagery , 2009, Remote. Sens..

[40]  Bolei Zhou,et al.  Understanding Intra-Class Knowledge Inside CNN , 2015, ArXiv.

[41]  Andrew Zisserman,et al.  Return of the Devil in the Details: Delving Deep into Convolutional Nets , 2014, BMVC.

[42]  Arabinda Ghosh,et al.  Open Defecation and Childhood Stunting in India: An Ecological Analysis of New Data from 112 Districts , 2013, PloS one.

[43]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[44]  Hal R. Varian,et al.  Big Data: New Tricks for Econometrics , 2014 .