Earth Observation for Python. A collection of helper functions and classes to search, download and process earth observation data.
Here's an example of some basic contrast enhancement (BCET) and decorrelation stretching (DDS) techniques applied to a Landsat-8 false colour composite.
This library also includes the SFIM pansharpenning method, here's an example using a Landsat-8 RGB image where the 30 m multispectral bands are pansharpened using the 15 m panchromatic band.
Here's an example of searching for imagery in the cloud, Landsat-8 and Sentinel-2 are currently supported.
from datetime import datetime, timedelta
from shapely.geometry import Point
from eopy.cloud import Searcher
from eopy.geometry import GeoPolygon
latitude, longitude = 51.507351, -0.127758
search_boundary = GeoPolygon(Point((longitude, latitude)).buffer(0.1), epsg=4326)
searcher = Searcher()
scenes = searcher.search(
search_boundary,
start=datetime.now() - timedelta(days=7),
end=datetime.now()
)
print(scenes[0])
>>> <Scene: S2A_39GWH_20191122_0 | Cloud: 15.57 | Date: 2019-11-22>from eopy.cloud import Downloader
from eopy.image import Loader
downloader = Downloader("data")
downloader.download(scene=scene, bands=['B8'])
loader = Loader()
image = loader.load('data/LC8202024019319.tif')
print(image.shape)
>>> (15801, 15601)It's not always convenient to have to download a whole image if we are only interested in a specific section. Streaming a scene at a particular boundary is much faster than downloading the whole image and allows us to read it directly into memory.
import maptlotlib.pyplot as plt
image = downloader.stream(scene, bands=['B8'], boundary=search_boundary.transform(scene.epsg))
plt.imshow(image.pixels, cmap='Greys')
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent