Predict urban growth on maps.
- Analysis of satellite images of an urban area.
This urban growth prediction model uses maps of the physical and social characteristics of an urban area to predict the growth of urbanistic characteristics in future years.
Growth attractors (such as transport, quality of life, topography and services) and limiters (such as bodies of water or protected areas) selected by the modeler according to the conditions of the urban area in question are used to determine if a pixel in the map image will exhibit urbanized features or not at the end of a designated period of time.
The predictions can be used to plan optimized urban expansion and estimate the best and worst scenarios of climate change.
Inter-American Development Bank
As inputs, the model uses monochrome images with standardized size and limits prepared from satellite images of an urban area.
These images can be physical maps, density maps or maps that denote legislative or social limits, according to the conditions of the urban area and the discretion of the modeler.
Each image should only contain information about a single characteristic, since each one will be assigned a positive or negative weight. Using these inputs, a regularized spatial logistic regression model will predict urban growth at a level of one pixel per pixel within determined limits and generate a binary Raster file.
You can see the code of this model here: