This repository has been archived by the owner on Jun 21, 2022. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathchangelog
35 lines (32 loc) · 2.45 KB
/
changelog
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Release v1.4
------------
- support of STAC catalogs for Sentinel-2 on AWS and Microsoft Planetary Computer
- dockerization of the sample notebooks to allow for easy demonstration of the package
- bugfixes in the rasterization procedure
- band math functionality added for the Band class
- minor fixes in the plotting module
Release v1.3
------------
- major improvements and advances in the data model. From now on the base unit in the data model is the Band. A band holds
a 2-d numpy array, numpy masked Array, or zarr (further support for zarr will be added in the future) and has geo-localisation
information. Raster datasets consist of 0 to N Band objects; these are called `RasterCollections` henceforth.
- the second major effort of this release concerns the mapping module. The mapping module queries the satellite metadata base using
a temporal and spatial area of interest and extracts satellite data for a set of (multi-)polygons or point-like features. The
returned series is either a collection of RasterCollection objects or a GeoDataFrame in the case of point-like features
- the latest Sentinel-2 baseline changes have been included. From now on Sentinel-2 gain and offset factors are determined based
on the PDGS processing baseline and the reflectance factor values are converted automatically from UINT16 to float (scaled
using gain and offset). To avoid automatic scaling, it is possible to unset a flag in the reading methods
Release v1.2
------------
- some bugfixes in the processing of L1C data (spatial resampling and blackfill merging)
- new generic function to extract raster values for polygon features for any single- or multi-band raster
- bugfix in buffer polygon function regarding the handling of multi-polygons
- improved filtering functionality for the CREODIAS downloader (filter by cloud cover threshold)
- new features for reading Sentinel-2 and other image data into a dict-like structure supporting AgriSatPy-derived bandstacks and .SAFE datasets
- added fast and generic vegetation index calculation
- added fast and generic plotting of spectral and discrete raster band data using matplotlib
- database model has been extended to support concept of geographic regions for organizing the SAT archive by geographic extent
Release v1.1
------------
- first draft of AgriSatPy including its satellite metadata base and functions to extract pixel values from Sentinel-2 scenes
- many features still experimental and likely to undergo major changes in the (near) future