Skip to main content

A convolution-based approach to detect urban extents.

Project description

PyPI version fury.io Build Status Coverage Status GitHub license Binder

Urban footprinter

A reusable convolution-based approach to detect urban extents from raster datasets.

LULC Convolution result Computed urban extent
LULC Convolution result Urban extent

The approach is built upon the methods used in the Atlas of Urban Expansion. The main idea is that a pixel is considered part of the urban extent depending on the proportion of built-up pixels that surround it. See the notebook overview or this blog post for a more detailed description of the procedure.

Installation and usage

To install use pip:

$ pip install urban-footprinter

Or clone the repo:

$ git clone https://github.com/martibosch/urban-footprinter.git
$ python setup.py install

Then use it as:

import urban_footprinter as ufp

# Or use `ufp.urban_footprint_mask_shp` to obtain the urban extent as a 
# shapely geometry
urban_mask = ufp.urban_footprint_mask("path/to/raster.tif",
                                      kernel_radius,
                                      urban_threshold,
                                      urban_classes=urban_classes)

where

help(ufp.urban_footprint_mask)

Help on function urban_footprint_mask in module urban_footprinter:

urban_footprint_mask(raster, kernel_radius, urban_threshold, urban_classes=None, largest_patch_only=True, buffer_dist=None, res=None)
    Computes a boolean mask of the urban footprint of a given raster.
    
    Parameters
    ----------
    raster : ndarray or str, file object or pathlib.Path object
        Land use/land cover (LULC) raster. If passing a ndarray (instead of the
        path to a geotiff), the resolution (in meters) must be passed to the
        `res` keyword argument.
    kernel_radius : numeric
        The radius (in meters) of the circular kernel used in the convolution.
    urban_threshold : float from 0 to 1
        Proportion of neighboring (within the kernel) urban pixels after which
        a given pixel is considered urban.
    urban_classes : int or list-like, optional
        Code or codes of the LULC classes that must be considered urban. Not
        needed if `raster` is already a boolean array of urban/non-urban LULC
        classes.
    largest_patch_only : boolean, default True
        Whether the returned urban/non-urban mask should feature only the
        largest urban patch.
    buffer_dist : numeric, optional
        Distance to be buffered around the urban/non-urban mask. If no value is
        provided, no buffer is applied.
    res : numeric, optional
        Resolution of the `raster` (assumes square pixels). Ignored if `raster`
        is a path to a geotiff.
    
    Returns
    -------
    urban_mask : ndarray

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for urban-footprinter, version 0.2.0
Filename, size File type Python version Upload date Hashes
Filename, size urban-footprinter-0.2.0.tar.gz (17.3 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page