This respository implements a cloud native preprocessing pipeline for sentinel 1 and sentinel 2 data, speckle filtering, texture extraction, cloud masking and NDVI derivation Processing of Sentinel datasets Overview of the project work flow

1. The project implements an end to end preprocessing and feature extract pipeline using snetinel 1 SAR and 2 optical data accessed via microsoft planetry computer API

Instalation of Libraries

• pystac (Data discovery),
• numpy (numerical operations),
• rioxarry (raster metadata),
• rasterio(raster I/O),
• scipy (numerical operations),
• scikit-image (texture extraction),
• matplotlib ( visualization).

2. Data access via API - AOI, date & time, sentinel scene collection 1 & 2

3. Sentinel preprocessing - Scene selection (VV/VH) - SPECKLE FILTERING

GLCM (SAR tecture extraction)

4. Sentinel 2 - Scene selection - Cloud masking (SCL) - visualization

5. NDVI

All data can be reproducible

Citation

Sentinel 1 and sentinel 2 from ESA Accessed via Microsoft planetary computer