shallowlearn
Rasterio wrappers focused on the sentinel-2 satellite with specific Machine learning applications on water
Science Score: 54.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: mdpi.com -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (13.4%) to scientific vocabulary
Repository
Rasterio wrappers focused on the sentinel-2 satellite with specific Machine learning applications on water
Basic Info
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 1
- Releases: 0
Metadata Files
README.md
ShallowLearn: A Python Toolkit for Shallow Water Remote Sensing Analysis
ShallowLearn provides a collection of tools and utilities designed for processing, analyzing, and visualizing satellite imagery, with a particular focus on shallow water environments like coral reefs. It includes functionalities for data loading, preprocessing, feature extraction, segmentation, time-series analysis, and visualization. This repo is a work in progress with improvements to come, I'm still in the process of refactoring now that the paper is published but hopefully I'll get to it sooner rather than later.
Features
- Data Handling:
- Load satellite data from various formats (GeoTIFF, Sentinel-2 SAFE/ZIP).
- Download Sentinel-2 via APIs (
cdsetool). - Compile multi-band GeoTIFFs from raw Sentinel-2 bands.
- Handle time series and seasonal data loading.
- Image Processing:
- Radiometric normalization (PCA-based, Histogram Matching).
- Image transformations (Contrast Enhancement - LCE, BCET, Color Space Conversions - LAB, HSV).
- Cloud detection and masking (XGBoost-based).
- Image resampling operations.
- Feature Extraction:
- Calculate standard remote sensing indices.
- Extract computer vision features (LBP, Gabor, HOG, Edge Density).
- Generate comprehensive feature stacks combining spectral, index, and texture information.
- Segmentation:
- Superpixel generation using various algorithms (SLIC, Felzenszwalb, Quickshift, Watershed).
- Superpixel processing pipelines involving PCA and clustering (DBSCAN, OPTICS, GMM).
- Depth Invariant Indices (DII):
- Calculate standard band-ratio DII.
- Implement superpixel-based DII workflows for automated deep/shallow area identification.
- Time Series Analysis:
- Process and normalize image time series.
- Quick-look analysis using PCA and clustering on image thumbnails or cropped areas.
- Utilities & Visualization:
- Helper functions for date extraction, band mapping, and metadata handling.
- Plotting utilities for images, spectra, histograms, scatter plots with image overlays, and density plots.
Documentation is available here
If you've found any of the code useful please cite our paper
Owner
- Name: Zayad AlZayer
- Login: z-alzayer
- Kind: user
- Repositories: 3
- Profile: https://github.com/z-alzayer
PhD student at Imperial in the earth sciences department
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this research, please cite it as below."
preferred-citation:
type: article
authors:
- family-names: "AlZayer"
given-names: "Zayad"
- family-names: "Mason"
given-names: "Philippa"
- family-names: "Platt"
given-names: "Robert"
- family-names: "John"
given-names: "Cédric M."
title: "An Improved Machine Learning-Based Method for Unsupervised Characterisation for Coral Reef Monitoring in Earth Observation Time-Series Data"
journal: "Remote Sensing"
volume: 17
issue: 7
article-number: 1244
year: 2025
url: "https://www.mdpi.com/2072-4292/17/7/1244"
issn: "2072-4292"
GitHub Events
Total
- Watch event: 1
- Delete event: 2
- Push event: 93
- Public event: 1
- Pull request event: 8
- Create event: 5
Last Year
- Watch event: 1
- Delete event: 2
- Push event: 93
- Public event: 1
- Pull request event: 8
- Create event: 5