Modeling land cover change in Houston with Deep Learning

Authors: Benoit Parmentier*, National Socio-Environmental Synthesis Center, University of Maryland, Marco Millones, Geography, University of Mary Washington, Albert Decatur, Hachure, Hichem Omrani, Luxembourg Institute of Socio-Economic Research, Giuseppe Amatulli, Yale School of Forestry and Environmental Studies, Tijana Jovanovic, National Socio-Environmental Synthesis Center, University of Maryland, Neeti Neeti, Natural Resources, TERI University
Topics: Land Use and Land Cover Change, Spatial Analysis & Modeling, Geographic Information Science and Systems
Keywords: Deep Learning, GIS, land cover change modeling
Session Type: Paper
Day: 4/3/2019
Start / End Time: 4:30 PM / 6:10 PM
Room: Virginia B, Marriott, Lobby Level
Presentation File: No File Uploaded

Deep learning is becoming ubiquitous in many fields and may be useful in improving modeling in land change sciences. In this research, we model land cover change in one of fastest growing urban region in the US, Houston, using deep learning with the Keras framework. We use National Land Cover Database dataset to calibrate the models using urban change from the 2001-2006 time period. Base on the implemented models we predict urban change over the 2006-2011 time period. Using covariates, we compare the performance of deep learning methods to more widely used methods in the land change modeling field (e.g. logistic model). We also examine to what extend information on relationship between covariates and predictor can be extracted from the deep learning framework.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login