{"name": "Bathymetrix-AI", "package_name": "Bathymetrix_AI", "description": "A professional QGIS research toolkit for Satellite-Derived Bathymetry (SDB), integrating Sentinel-2 STAC imagery downloading & temporal compositing, multispectral SDB modeling, and ICESat-2 LiDAR via Auto-ML.", "about": "BATHYMETRIX-AI: ADVANCED SDB TOOLKIT & COASTAL DYNAMICS\r\n================================================================================\r\n\r\nBathymetrix-AI is a professional QGIS research toolkit for high-precision Satellite-Derived Bathymetry (SDB). The toolkit integrates multispectral satellite imagery with ICESat-2 (ATL24) LiDAR bathymetry through a modular and adaptive Machine Learning framework. It automates the end-to-end SDB processing pipeline while maintaining full control over data quality, model selection, spatial refinement, uncertainty estimation, and scientific validation.\r\n\r\n\r\n--------------------------------------------------------------------------------\r\n\ud83d\udd2c SCIENTIFIC METHODOLOGY \u2014 CORE 5-PHASE SYSTEM (SDB SINGLE MASTERFLOW)\r\n--------------------------------------------------------------------------------\r\n\r\nThe SDB Single Masterflow provides a complete end-to-end processing pipeline for a single satellite scene:\r\n\r\nPhase 01 \u2192 Phase 02 \u2192 Phase 03 \u2192 Phase 04 \u2192 Phase 05\r\n\r\n\r\nPHASE 01: ADVANCED PRE-PROCESSING\r\n\r\nPrepares satellite imagery for bathymetric modeling by isolating the aquatic domain, reducing radiometric interference, and generating depth-sensitive spectral features.\r\n\r\n- Sun-Glint Removal:\r\n  Removes surface reflections using the Hedley model, with robust handling of NaN and Inf values.\r\n\r\n- Water Segmentation:\r\n  Uses NDWI, MNDWI, and NWI with adaptive thresholding to isolate the aquatic domain.\r\n\r\n- Deep Water OSW Filtering:\r\n  Identifies and removes deep-water areas unsuitable for optical bathymetry using:\r\n\r\n  (1) Automated Knee-Point Extinction [Recommended].\r\n  (2) Turbidity-Invariant Log-Ratio Extinction using NDTI.\r\n  (3) Multi-Otsu / GMM 3D Spectral Clustering.\r\n  (4) Connected-Component Topological Cleaning.\r\n  (5) Manual Polygon ROI & Custom OSW Polygon.\r\n\r\n- OSW Boundary Extraction:\r\n  Exports Optically Shallow Water boundaries as GeoPackage vectors while preserving the source CRS.\r\n\r\n- Log-Ratio Features & Spectral Indices:\r\n  Generates depth-sensitive features, including Blue/Green Log-Ratio and additional spectral indices.\r\n\r\n\r\nPHASE 02: ROBUST FILTERING\r\n\r\nImproves training data quality by identifying and removing noisy observations, outliers, and environmental artifacts.\r\n\r\n- Noise Removal:\r\n  Supports Linear RANSAC, Least-Squares (LS) Variance Fit, and Huber Variance Fit, with optional bypass.\r\n\r\n- Dynamic Diagnostic Plotting:\r\n  Uses robust percentile-based visualization for variance and trend analysis.\r\n\r\n\r\nPHASE 03: GLOBAL AUTO-ML & FEATURE ANALYSIS\r\n\r\nAutomatically evaluates multiple machine-learning approaches instead of relying on one predefined algorithm.\r\n\r\n- Feature Analysis:\r\n  Pearson/Spearman correlation, Auto-RANSAC, and Auto-Random Forest.\r\n\r\n- Algorithm Benchmarking:\r\n  Evaluates 15+ regressors, including Random Forest, Gradient Boosting, XGBoost, LightGBM, CatBoost, SVR, and MLP.\r\n\r\n- Model Selection & Winner Stability:\r\n  Supports seven selection strategies:\r\n\r\n  (1) Winner Stability via Monte Carlo Sensitivity Simulation.\r\n  (2) SDB Composite Score.\r\n  (3) Maximum R\u00b2.\r\n  (4) Minimum RMSE.\r\n  (5) Minimum wMAPE.\r\n  (6) Minimum Absolute Bias.\r\n  (7) Minimum MAE.\r\n\r\n- Hyperparameter Optimization:\r\n  Bayesian Optimization, Grid Search, and Random Search.\r\n\r\n- Spatial Cross-Validation:\r\n  Independent spatial block cross-validation.\r\n\r\n- Ensemble Blending:\r\n  Standard Average, Median, Stacking, and Uncertainty-Weighted Pixel Fusion.\r\n\r\n- Memory-Efficient Prediction:\r\n  Chunk-based raster prediction for large satellite images.\r\n\r\n- Customization:\r\n  Advanced controls for model parameters and optimization settings.\r\n\r\n\r\nPHASE 04: ADAPTIVE REFINEMENT\r\n\r\nCorrects local errors remaining after global machine-learning prediction.\r\n\r\n- Decoupled Refinement Architecture:\r\n  Independent Depth Variance Correction and Spatial Residual Error Modeling.\r\n\r\n- Spatial Residual Correction:\r\n  Models spatial differences between predicted and observed depths.\r\n\r\n- Zero-Mean Centered Spatial Residuals:\r\n  Applies robust residual processing and spatial weighting to reduce local bias.\r\n\r\n- Spatial Residual Modeling:\r\n  Robust KNN, Standard KNN, and Gaussian Process / Kriging.\r\n\r\n- Spatial Cross-Validation:\r\n  Independent spatial validation of residual models.\r\n\r\n- Adaptive Re-training & Pixel Fusion:\r\n  Integrates model predictions with spatial error information.\r\n\r\n- IHO Standards Assessment:\r\n  Evaluates results against IHO S-44 Order 1a/2 Total Vertical Uncertainty criteria.\r\n\r\n\r\nPHASE 05: VALIDATION & REPORTING\r\n\r\nProvides independent scientific assessment of the final bathymetric result.\r\n\r\n- Independent Accuracy Assessment:\r\n  Evaluates model performance using unseen validation points.\r\n\r\n- Performance Metrics:\r\n  R\u00b2, RMSE, and wMAPE.\r\n\r\n- IHO S-44 Assessment:\r\n  Evaluates compliance with applicable Order 1a/2 TVU criteria.\r\n\r\n- Interactive Validation Dashboard:\r\n  Generates HTML dashboards containing model leaderboards, validation metrics, diagnostic plots, and summary information.\r\n\r\n\r\n--------------------------------------------------------------------------------\r\n\ud83d\ude80 MASTERFLOWS & STANDALONE MODULES\r\n--------------------------------------------------------------------------------\r\n\r\n1. SDB Single Masterflow:\r\n   One Scene \u2192 Pre-processing \u2192 Filtering \u2192 AI Modeling \u2192 Adaptive Refinement \u2192 Validation \u2192 One Final SDB Map.\r\n\r\n2. SDB SpatioSpectral Masterflow:\r\n   Multiple Scenes \u2192 Independent SDB Modeling \u2192 Scene Evaluation \u2192 Best Scene Selection / Weighted Median / Weighted Mean / Median / Mean \u2192 One Consolidated SDB Map.\r\n\r\n3. SDB SpatioTemporal Masterflow:\r\n   Multiple Years \u2192 Global Spatiotemporal AI Modeling \u2192 Year-Specific Adaptive Refinement \u2192 Temporal Series of SDB Maps.\r\n\r\n4. Coastal Dynamics Analysis (Module 06):\r\n   Analyzes Net Bathymetric Change, Morphological Stability Index (MSI), Statistical Level of Detection (StatCD), Erosion and Accretion, Shoreline Movement, Volumetric Sediment Tracking, and Target ROI Analytics.\r\n\r\n5. ICESat-2 Downloader:\r\n   Queries, filters, and downloads ICESat-2 ATL24 bathymetric observations directly from NSIDC.\r\n\r\n6. Tidal Datum Converter:\r\n   Converts and vertically aligns bathymetric observations between tidal reference datums using NASA GSFC GOT4.10c tidal modeling.\r\n\r\n\r\n--------------------------------------------------------------------------------\r\n\ud83d\udcca PERFORMANCE METRICS\r\n--------------------------------------------------------------------------------\r\n\r\n- R\u00b2 (Coefficient of Determination):\r\n  Measures goodness of fit between predicted and observed depths.\r\n\r\n- RMSE (Root Mean Square Error):\r\n  Measures average vertical prediction error in meters.\r\n\r\n- wMAPE (Weighted Mean Absolute Percentage Error):\r\n  Measures relative error across depth intervals.\r\n\r\n- IHO S-44 Standards:\r\n  Evaluates results against Order 1a and Order 2 TVU criteria.\r\n\r\n\r\n--------------------------------------------------------------------------------\r\n\ud83c\udfaf MASTERFLOW SELECTION GUIDE\r\n--------------------------------------------------------------------------------\r\n\r\n- SDB Single Masterflow:\r\n  One Scene \u2192 One SDB.\r\n\r\n- SDB SpatioSpectral Masterflow:\r\n  Multiple Scenes \u2192 Evaluate / Select / Aggregate \u2192 One SDB.\r\n\r\n- SDB SpatioTemporal Masterflow:\r\n  Multiple Years \u2192 Global Temporal AI \u2192 SDB for Each Year.\r\n\r\n- Coastal Dynamics Analysis:\r\n  Yearly SDB \u2192 Bathymetric Change & Morphological Analysis.\r\n\r\n\r\n--------------------------------------------------------------------------------\r\n\ud83d\udcda SCIENTIFIC REFERENCES\r\n--------------------------------------------------------------------------------\r\n\r\n- Stumpf et al. (2003):\r\n  Determination of shallow water depth with high-resolution satellite imagery.\r\n\r\n- Hedley et al. (2005):\r\n  Simple and robust removal of sun glint for high-resolution imagery.\r\n\r\n- Fischler & Bolles (1981):\r\n  Random Sample Consensus (RANSAC) for model fitting.\r\n\r\n- Zhang et al. (2021):\r\n  Adaptive variance fitting for satellite bathymetry noise reduction.\r\n\r\n- Alevizos (2020):\r\n  Spatial residual refinement and error compensation in shallow coastal waters.\r\n\r\n- Wheaton et al. (2010):\r\n  Accounting for uncertainty in DEMs from repeat topographic surveys.\r\n\r\n\r\n--------------------------------------------------------------------------------\r\n\ud83d\udcd6 SOFTWARE CITATION\r\n--------------------------------------------------------------------------------\r\n\r\nNasef M. Aly. (2026).\r\nNasef2017/Bathymetrix-AI: Bathymetrix_AI V7.9 (Version v7.9).\r\n[Computer software]. Zenodo.\r\nhttps://doi.org/10.5281/zenodo.22539469\r\n\r\n\r\n================================================================================\r\nDEVELOPMENT & ACKNOWLEDGMENT\r\n================================================================================\r\n\r\nDeveloped by Mohamed Aly Nasef for scientific research, coastal mapping, bathymetric analysis, and hydrographic applications.\r\n\r\nCode architecture, algorithms, and technical documentation significantly enhanced and optimized using Google Gemini AI.\r\n\r\n================================================================================", "homepage": "https://github.com/Nasef2017/Bathymetrix-AI", "repository": "https://github.com/Nasef2017/Bathymetrix-AI", "tracker": "https://github.com/Nasef2017/Bathymetrix-AI/issues", "author": "Mohamed Aly Nasef", "tags": ["icesat-2", "remote sensing", "machine learning", "spatial correction", "ransac", "bathymetry", "sdb", "hydrography", "python", "satellite imagery", "coastal dynamics", "volumetric change", "quantile regression", "spatiospectral", "iho-s44", "spatiotemporal"], "downloads": 7963, "latest_version": "8.0", "versions": [{"version": "8.0", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 81, "uploaded_by": "nasefmaly", "upload_datetime": "2026-10-27T07:33:13.171762"}, {"version": "7.9", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 791, "uploaded_by": "nasefmaly", "upload_datetime": "2026-09-07T01:20:02.661942"}, {"version": "7.7", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 174, "uploaded_by": "nasefmaly", "upload_datetime": "2026-09-01T11:25:10.845578"}, {"version": "7.5", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 158, "uploaded_by": "nasefmaly", "upload_datetime": "2026-08-30T07:10:30.810237"}, {"version": "7.2", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 270, "uploaded_by": "nasefmaly", "upload_datetime": "2026-08-19T22:25:35.391308"}, {"version": "7.1", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 145, "uploaded_by": "nasefmaly", "upload_datetime": "2026-08-18T06:57:49.328943"}, {"version": "7.0", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 158, "uploaded_by": "nasefmaly", "upload_datetime": "2026-08-14T04:35:26.071826"}, {"version": "6.5", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 309, "uploaded_by": "nasefmaly", "upload_datetime": "2026-08-06T14:14:17.227002"}, {"version": "6.4", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 204, "uploaded_by": "nasefmaly", "upload_datetime": "2026-08-02T12:24:23.951318"}, {"version": "6.3", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 436, "uploaded_by": "nasefmaly", "upload_datetime": "2026-07-16T06:33:37.071600"}, {"version": "6.1", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 406, "uploaded_by": "nasefmaly", "upload_datetime": "2026-07-10T10:12:44.772453"}, {"version": "6.0", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 194, "uploaded_by": "nasefmaly", "upload_datetime": "2026-07-07T10:38:19.664382"}, {"version": "5.2", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 192, "uploaded_by": "nasefmaly", "upload_datetime": "2026-07-02T03:43:33.163448"}, {"version": "5.1", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 276, "uploaded_by": "nasefmaly", "upload_datetime": "2026-06-23T13:21:19.150571"}, {"version": "5.0", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 216, "uploaded_by": "nasefmaly", "upload_datetime": "2026-06-21T03:44:36.916673"}, {"version": "4.8", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 465, "uploaded_by": "nasefmaly", "upload_datetime": "2026-05-25T17:49:03.248788"}, {"version": "4.7", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "4.99.0", "downloads": 381, "uploaded_by": "nasefmaly", "upload_datetime": "2026-05-09T06:24:06.990974"}, {"version": "4.6", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 322, "uploaded_by": "nasefmaly", "upload_datetime": "2026-04-23T07:20:30.150107"}, {"version": "4.5", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 164, "uploaded_by": "nasefmaly", "upload_datetime": "2026-04-22T04:58:17.735258"}, {"version": "4.4", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 146, "uploaded_by": "nasefmaly", "upload_datetime": "2026-04-22T03:00:50.708322"}, {"version": "4.3", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 313, "uploaded_by": "nasefmaly", "upload_datetime": "2026-04-07T00:01:56.962288"}, {"version": "4.2", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 148, "uploaded_by": "nasefmaly", "upload_datetime": "2026-04-02T13:40:52.819255"}, {"version": "4.1", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 611, "uploaded_by": "nasefmaly", "upload_datetime": "2026-02-22T01:51:22.218820"}, {"version": "4.0", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 480, "uploaded_by": "nasefmaly", "upload_datetime": "2026-01-29T12:50:52.978416"}, {"version": "3.3", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 240, "uploaded_by": "nasefmaly", "upload_datetime": "2026-01-25T13:22:28.905958"}, {"version": "3.2", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 259, "uploaded_by": "nasefmaly", "upload_datetime": "2026-01-11T23:41:05.860406"}, {"version": "3.1", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 168, "uploaded_by": "nasefmaly", "upload_datetime": "2026-01-10T23:43:49.931487"}, {"version": "3.0", "experimental": false, "qgis_min": "3.22.0", "qgis_max": "3.99.0", "downloads": 255, "uploaded_by": "nasefmaly", "upload_datetime": "2025-12-30T09:52:02.168442"}]}