Skip to content

Latest commit

 

History

History
80 lines (69 loc) · 3.38 KB

README.md

File metadata and controls

80 lines (69 loc) · 3.38 KB
Raw

Code style: black Tests

rasterexplorer

A package for quickly viewing raster datasets on an interactive map.

This package was inspired by the method explore (from GeoDataFrame objects) implemented by the package geopandas. As geopandas focuses on vector data, this package was created focusing on raster data. It provides a function, also named explore, that allows viewing raster datasets on an interactive map created using folium.

Installation

On Windows, one of the dependencies (rasterio) must be manually installed before installing rasterexplorer. For conda users, this dependency can be easily installed using the command conda install -c conda-forge rasterio. After that, rasterexplorer can be installed with pip using the following command:

pip install git+https://github.com/lbferreira/rasterexplorer

On Linux, rasterexplorer can be installed directly using pip.

Usage examples

When working with raster datasets on a jupyter notebook, quicklying viewing the datasets can be very useful. To do it with rasterexplorer, the following examples can be followed. input_raster can be a string with path for the input raster, a DatasetReader object of the lib rasterio, or a RasterData object, which can be created from a numpy masked array.

from rasterexplorer import explorer

my_map = explorer.explore(
    input_raster='./example_data/sentinel2/area1/ndvi.tif',
    label='My example area 1',
    band=1,
    cbar_caption='NDVI',
)
my_map

A custom tile server can be provided as shown below:

tiles = "https://api.mapbox.com/v4/mapbox.satellite/{z}/{x}/{y}@2x.jpg90?access_token=" + my_token
attr = 'MapBox'
my_map = explorer.explore(
    input_raster='./example_data/sentinel2/area1/ndvi.tif',
    label='My example area 1',
    band=1,
    cbar_caption='NDVI',
    tiles=tiles,
    attr=attr,
)
my_map

When manipulating rasters in the form of numpy arrays, the interactive map can be created passing a RasterData object to the parameter input_raster of the explore function. An example is provided below.

from rasterexplorer.raster_data import RasterData

# Loading data from .tif files and extrating data as numpy masked arrays
with rasterio.open('./example_data/sentinel2/area1/b4.tif') as dataset:
    band4 = dataset.read(1, masked=True)
    band4_nodata = dataset.nodata
    band4_crs = dataset.crs
    band4_transform = dataset.transform
with rasterio.open('./example_data/sentinel2/area1/b8.tif') as dataset:
    band8 = dataset.read(1, masked=True)

# Computing NDVI
# NDVI = (NIR - RED) / (NIR + RED) where NIR = band4 nad RED = band8
ndvi = (band8 - band4) / (band8 + band4)

# Creating a RasterData object
# Note that nodata, crs and transform from band4 and band8 are the same ones.
# RasterData can be created passing both conventional and masked numpy arrays.
ndvi_raster = RasterData(data=ndvi, nodata=band4_nodata, crs=band4_crs, transform=band4_transform)

# Creating map
my_map = explorer.explore(
    input_raster=ndvi_raster,
    label='My example area 1',
    band=1,
    cbar_caption='NDVI',
)
my_map

Issues

If you find a bug, please report it as an issue.