PitchHut
Log in / Sign up
sentle
45 views
Effortlessly process Sentinel data cubes at scale with powerful integrations.
Pitch

Sentle is designed for large-scale processing of Sentinel-1 and Sentinel-2 data cubes, allowing users to manage bigger-than-memory datasets seamlessly on any machine. With features like cloud detection, snow masking, and temporal composites, Sentle simplifies the complexities of remote sensing data for research and application.

Description

Sentle: Harnessing Sentinel-1 and Sentinel-2 Data at Scale

Transform your ability to work with satellite imagery with Sentle, a powerful Python package designed for handling large-scale Sentinel-1 & Sentinel-2 data cubes efficiently on any machine. Sentle offers integrated capabilities for cloud detection, snow masking, data harmonization, merging, and temporal composites, allowing you to download and process data that exceeds the memory capacity of your system.

Key Features:

  • Scalable Data Processing: Seamlessly process and manage large datasets from Sentinel satellites, tailored for tasks that exceed typical memory constraints.
  • Integrated Cloud Detection & Snow Masking: Automatically detect clouds and mask snow in your datasets to ensure data accuracy and integrity.
  • Temporal Composites: Create averaged datasets over specified time intervals to visualize trends and changes in your area of interest.

Quick Start

Initiate processing and download Sentinel data cubes with just a few lines of code:

from sentle import sentle
from rasterio.crs import CRS

sentle.process(
    zarr_store="mycube.zarr",
    target_crs=CRS.from_string("EPSG:32633"),
    bound_left=176000,
    bound_bottom=5660000,
    bound_right=216000,
    bound_top=5700000,
    datetime="2022-06-17/2023-06-17",
    target_resolution=10,
    S2_mask_snow=True,
    S2_cloud_classification=True,
    S2_cloud_classification_device="cuda",
    S1_assets=["vv", "vh"],
    S2_apply_snow_mask=True,
    S2_apply_cloud_mask=True,
    time_composite_freq="7d",
    num_workers=10,
)

This code snippet downloads and processes Sentinel-1 and Sentinel-2 data for a specified geographical area (40km x 40km) covering a year, while managing clouds and snow through efficient masking techniques. The processing leverages 10 parallel workers, optimizing speed and efficiency.

Visualization Options

Visualize your processed data easily with libraries like xarray and lexcube. Load your data and produce stunning visualizations, helping you extract insights from satellite imagery effortlessly:

import xarray as xr

da = xr.open_zarr("mycube.zarr").sentle

grasp = lexcube.Cube3DWidget(da.sel(band="B02"), vmin=0, vmax=4000)

Contributing and Support

As an open-source project, Sentle welcomes contributions from developers and researchers alike. Please report bugs or suggest enhancements by opening an issue in the GitHub repository. Together, we can enhance the capabilities of Sentle and foster a vibrant community around satellite data processing.

**Don’t miss the opportunity to revolutionize your satellite data processing with Sentle. Dive into the world of high-resolution imagery and gain unprecedented insights into environmental changes and trends!