Bathymetrix-AI
Plugin ID: 4513
An advanced Machine Learning pipeline for Satellite-Derived Bathymetry (SDB). Features ICESat-2 integration, RANSAC filtering, and Spatial Residual Stacking.
(4) votes
Download latest
<h3>Bathymetrix-AI: Advanced SDB Modeling & Spatial Refinement</h3>
<p><b>Bathymetrix-AI</b> is a specialized QGIS research toolkit designed to derive high-precision bathymetry from corrected multispectral satellite imagery (e.g., Sentinel-2 L2A). It systematically integrates physics-based corrections with data-driven Machine Learning to overcome traditional SDB limitations.</p>
<p><b>Core Workflow (The 4-Phase System):</b></p>
<ul>
<li><b>1. Automated Pre-processing:</b> Sun-glint removal (Hedley), dynamic water masking (Otsu), and physics-based feature generation (Log-Ratios).</li>
<li><b>2. Robust Altimetry Filtering:</b> Uses the <b>RANSAC algorithm</b> to clean ICESat-2 (ATL24) data by identifying high-confidence inliers and removing environmental noise.</li>
<li><b>3. Global Auto-ML Modeling:</b> Competitive benchmarking of <b>11 ML algorithms</b> with Randomized Hyperparameter Optimization to find the optimal global depth function.</li>
<li><b>4. Spatial Residual Stacking:</b> Enhances accuracy by analyzing prediction residuals and re-training the model with a <b>Stacked Error Surface</b> to correct local biases.</li>
</ul>
<p><b>Key References:</b><br>
- <b>Stumpf et al. (2003):</b> Log-Ratio Algorithm for SDB inversion.<br>
- <b>Hedley et al. (2005):</b> Physics-based sun-glint correction.<br>
- <b>Fischler & Bolles (1981):</b> RANSAC algorithm for robust estimation.<br>
- <b>Alevizos (2020):</b> Residual analysis and spatial refinement in shallow waters.<br>
- <b>Bergstra & Bengio (2012):</b> Randomized search for hyperparameter optimization.<br>
- <b>Parrish et al. (2025):</b> Analysis and assessment of global ICESat-2 bathymetry.</p>
<p><i>Developed for scientific research and hydrographic applications. Creating the Codes and documentation optimized using <b>Google Gemini AI</b>.</i></p>
