English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Mapping the spatial distribution of poverty in developing countries remainsan important and costly challenge. These "poverty maps" are key inputs forpoverty targeting, public goods provision, political accountability, and impactevaluation, that are all the more important given the geographic dispersion ofthe remaining bottom billion severely poor individuals. In this paper we trainConvolutional Neural Networks (CNNs) to estimate poverty directly from high andmedium resolution satellite images. We use both Planet and Digital Globeimagery with spatial resolutions of 3-5 sq. m. and 50 sq. cm. respectively,covering all 2 million sq. km. of Mexico. Benchmark poverty estimates come fromthe 2014 MCS-ENIGH combined with the 2015 Intercensus and are used to estimatepoverty rates for 2,456 Mexican municipalities. CNNs are trained using the 896municipalities in the 2014 MCS-ENIGH. We experiment with several architectures(GoogleNet, VGG) and use GoogleNet as a final architecture where weights arefine-tuned from ImageNet. We find that 1) the best models, which incorporatesatellite-estimated land use as a predictor, explain approximately 57% of thevariation in poverty in a validation sample of 10 percent of MCS-ENIGHmunicipalities; 2) Across all MCS-ENIGH municipalities explanatory powerreduces to 44% in a CNN prediction and landcover model; 3) Predicted povertyfrom the CNN predictions alone explains 47% of the variation in poverty in thevalidation sample, and 37% over all MCS-ENIGH municipalities; 4) In urban areaswe see slight improvements from using Digital Globe versus Planet imagery,which explain 61% and 54% of poverty variation respectively. We conclude thatCNNs can be trained end-to-end on satellite imagery to estimate poverty,although there is much work to be done to understand how the training processinfluences out of sample validation.

Poverty Mapping Using Convolutional Neural Networks Trained on High and Medium Resolution Satellite Images, With an Application in Mexico