Poverty Mapping Using Convolutional Neural Networks Trained on High and Medium Resolution Satellite Images, With an Application in Mexico.
Boris BabenkoJonathan HershDavid NewhouseAnusha RamakrishnanTom SwartzPublished in: CoRR (2017)
Keyphrases
- satellite images
- convolutional neural networks
- convolutional network
- high resolution
- multispectral
- satellite imagery
- change detection
- remote sensing
- digital elevation models
- high resolution satellite images
- united states
- satellite data
- google earth
- data sets
- urban areas
- land cover
- remote sensing images
- hyperspectral
- landsat tm
- quantitative analysis
- training set
- feature extraction
- three dimensional
- image processing