{"name": "Terranova", "package_name": "terranova", "version": "1.0.3", "experimental": false, "qgis_min": "4.0.0", "qgis_max": "4.99.0", "downloads": 113, "uploaded_by": null, "upload_datetime": "2026-06-04T15:25:11.532194", "changelog": "1.0.3 (2026-06)\n- Officially target QGIS 4.0+ only (qgisMinimumVersion raised to\n4.0). The plugin is built on PyQt6 and uses QtWebEngine for its\nmap dock; the QGIS 3.x / PyQt5 LTR series is not supported.\n1.0.2 (2026-06)\n- Declare the plugin's external Python dependencies and how to\ninstall them in the About metadata (per plugins.qgis.org\nguidelines).  No functional change.\n1.0.1 (2026-06)\n- Beta-features toggle: time-series, SAM, foundation models and CDSE\nare hidden by default behind a header switch.\n- Catalogue composite is now discoverable via a checkbox under the\ndate range, with a max-images cap.\n- Accuracy: validation-vector vs random-points modes; interactive\npoint-by-point labelling pad with class-name persistence;\nmetrics shown inline (OA as a percentage) and exported to Excel.\n- Faster, lighter dropdowns on lower-spec / HiDPI machines.\n- Fixes: AOI-rectangle persistence, percentage formatting in PDF\nand Excel reports, plus packaging cleanups.\n1.0.0 (2026-05)\nFirst public release.\n- STAC catalogue search (Planetary Computer / Earth Search / CDSE) with\nembedded interactive Leaflet map for AOI picking and overlap-aware\nscene-footprint inspection.  Multi-select batch download as COG;\noptional mean/median composite from up to N images.\n- Supervised classification: Random Forest, Extra Trees, Gradient\nBoosting, LightGBM, XGBoost, KNN, Logistic Regression, MLP \u2014 each\nwith a 2-4 sentence pros/cons hint in the panel.\n- Unsupervised classification: K-Means + ISODATA, fit on a random\nsubsample with output as a labelled COG.\n- Accuracy assessment: validation-vector mode (sample raster at every\npixel covered by a vector) AND random-points mode (generate\nrandom / stratified / equalized-stratified points; step through\neach one in a labelling pad with auto-pan + zoom-to-pixel +\nclass-name persistence; compute confusion matrix straight from\nthe labelled points file).  PDF + Excel report with OA / kappa /\nper-class UA / PA / F1; OA shown as a percentage.\n- Beta tabs (off by default): time-series + per-pixel change detection\n(CuSum + BFAST/LandTrendr); SAM segmentation; Prithvi / Clay\nfoundation-model fine-tune scaffolding; CDSE OAuth sign-in.\n- Dual UI: native Qt dialogs for compatibility, React-in-QWebEngine\ndock with embedded OpenStreetMap / Esri satellite basemap for\nthe primary surface.\n- NDVI Processing algorithm; CI on Linux/Windows/macOS.", "external_deps": null, "download_url": "https://plugins.qgis.org/plugins/terranova/version/1.0.3/download/"}