Bug fixes. Improved Imperial Unit measurement.

The App Store listing is now available in six languages. Minor internal improvements.

Drone photogrammetry: from planning to model, without leaving EasyLidar. NEW — Flight planner • Draw the area on the map, pick your drone and camera from the catalogue, set the GSD you want, and EasyLidar works out the mission: height, passes, overlaps and photo count. • Terrain following: it uses your own DTM to keep height above ground constant. Over broken ground that means a uniform GSD across the whole flight, instead of photos with uneven detail. • Mark the take-off point on the map and check elevations and distances before you fly. • Export the mission as KMZ/WPML and load it into the drone. A built-in guide lists which models accept it. NEW — Photogrammetric processing • Import your photos with their EXIF data and check the real coverage on the map. • Build a point cloud and an orthophoto with the native macOS engine — nothing to install. Prefer another? You can connect NodeODM. • The result joins your project as one more layer: measure it, compare it against the LiDAR, and work it with all the usual tools. Much sharper views • Fixed a flaw that ruined detail on some tiles: SLRM and combined openness set their contrast tile by tile, so a tile with steep banks flattened its micro-relief while the flat one next door came out perfect. The scale is now physical and shared across the mosaic. Regenerate your views to see it. • The 2D viewer no longer has a resolution ceiling: zooming in now composes at your data's real resolution, not the mosaic's. • Finer resolution steps when building the model, so tiles of similar density no longer end up with very different quality. IGN downloads • A new switch in the CNIG download window works around the blocks the IGN turns on from time to time. If a download fails, change method and carry on.

Minor bug fixes. Since 1.7, EasyLidar has grown from a LiDAR relief viewer into a full multi-source GIS. Here is everything new in 2.0. OPTICAL AND MULTISPECTRAL IMAGERY — NEW Bring in aerial or satellite optical imagery — RGB, multiband, 16-bit, Sentinel-2 and similar — as its own data world alongside LiDAR and vectors. Compute NDVI, NDWI, NDRE and SAVI and use them directly in zonal statistics or the raster calculator. Optical Studio builds a non-destructive stack of visual transformations — stretch, sharpen, contrast, band composites, pseudocolor — with a live preview. CLASSIFICATION: MACHINE LEARNING AND DEEP LEARNING — NEW Classify optical imagery by drawing training samples, then vectorize the result with an automatic legend and a full confusion matrix. Deep Learning classification runs on Core ML and ships with FLAIR, a ready-to-use land-cover model from IGN France — or bring your own. Train a Random Forest on deep features extracted from your own samples, no external tools required. FIND TERRAIN BY EXAMPLE — NEW Draw one example of what you're looking for — a mound, a wall, a hollow — and the morphological radar finds the rest across the whole mosaic. Automatic landform discovery maps peaks, ridges, valleys and flats with no training at all, now about five times faster, running in parallel with a progress bar and ETA. KRIGING AND KERNEL DENSITY — NEW Ordinary Kriging with an automatically fitted variogram and a prediction-error map, kernel density heatmaps, and a live low-resolution preview while you tune parameters. VECTOR GEOMETRY — NEW Voronoi/Thiessen diagrams, Delaunay triangulation, convex hull and simplify, straight from any point or polygon layer. RAIN SIMULATION — NEW Simulate rainfall runoff across the whole terrain, with a directional animation in 3D. CARTOGRAPHIC OUTPUT — NEW A print composer lays out a finished map with legend, scale bar, north arrow and title block, ready to export as PDF. Contour lines now carry elevation labels, and GeoPackage export and field photos on markers are new. BRING IN ANY FILE — NEW A universal importer now reads practically any geographic format, LiDAR included, without you telling it what it is first. A MANUAL INSIDE THE APP — NEW The complete user manual now lives inside EasyLidar, with a navigable index and screenshots, in six languages. AN AI ASSISTANT THAT KEEPS GROWING Every feature above ships with its own AI tools: Claude, Codex or any MCP client can run a classification, launch the morphological radar, build a Kriging map or compose a print layout from a plain-language request. ALSO FASTER AND MORE RELIABLE Elevation models now build up to six times faster, with identical results. Undo in the layer tools now steps back one operation at a time instead of removing the whole layer. The AI connection is local-only, and destructive operations stay blocked until you allow them in Settings. Built exclusively for Apple Silicon.

Since 1.7, EasyLidar has grown from a LiDAR relief viewer into a full multi-source GIS. Here is everything new in 2.0. OPTICAL AND MULTISPECTRAL IMAGERY — NEW Bring in aerial or satellite optical imagery — RGB, multiband, 16-bit, Sentinel-2 and similar — as its own data world alongside LiDAR and vectors. Compute NDVI, NDWI, NDRE and SAVI and use them directly in zonal statistics or the raster calculator. Optical Studio builds a non-destructive stack of visual transformations — stretch, sharpen, contrast, band composites, pseudocolor — with a live preview. CLASSIFICATION: MACHINE LEARNING AND DEEP LEARNING — NEW Classify optical imagery by drawing training samples, then vectorize the result with an automatic legend and a full confusion matrix. Deep Learning classification runs on Core ML and ships with FLAIR, a ready-to-use land-cover model from IGN France — or bring your own. Train a Random Forest on deep features extracted from your own samples, no external tools required. FIND TERRAIN BY EXAMPLE — NEW Draw one example of what you're looking for — a mound, a wall, a hollow — and the morphological radar finds the rest across the whole mosaic. Automatic landform discovery maps peaks, ridges, valleys and flats with no training at all, now about five times faster, running in parallel with a progress bar and ETA. KRIGING AND KERNEL DENSITY — NEW Ordinary Kriging with an automatically fitted variogram and a prediction-error map, kernel density heatmaps, and a live low-resolution preview while you tune parameters. VECTOR GEOMETRY — NEW Voronoi/Thiessen diagrams, Delaunay triangulation, convex hull and simplify, straight from any point or polygon layer. RAIN SIMULATION — NEW Simulate rainfall runoff across the whole terrain, with a directional animation in 3D. CARTOGRAPHIC OUTPUT — NEW A print composer lays out a finished map with legend, scale bar, north arrow and title block, ready to export as PDF. Contour lines now carry elevation labels, and GeoPackage export and field photos on markers are new. BRING IN ANY FILE — NEW A universal importer now reads practically any geographic format, LiDAR included, without you telling it what it is first. A MANUAL INSIDE THE APP — NEW The complete user manual now lives inside EasyLidar, with a navigable index and screenshots, in six languages. AN AI ASSISTANT THAT KEEPS GROWING Every feature above ships with its own AI tools: Claude, Codex or any MCP client can run a classification, launch the morphological radar, build a Kriging map or compose a print layout from a plain-language request. ALSO FASTER AND MORE RELIABLE Elevation models now build up to six times faster, with identical results. Undo in the layer tools now steps back one operation at a time instead of removing the whole layer. The AI connection is local-only, and destructive operations stay blocked until you allow them in Settings. Built exclusively for Apple Silicon. Thank you for using EasyLidar. Tell us what you would like to see next.

• A more powerful and stable AI assistant: long tasks are split into subtasks, project memory persists across sessions, a shared working folder, and more robust networking. • New "How to use AI with EasyLidar" guide, plus one-click connection to Claude Desktop, Codex and Antigravity. • Performance improvements and stability fixes.

Fixes an issue that prevented LiDAR filtering and the conversion of vector and raster files on some macOS installations. Also improves handling of vector layers in coordinate systems other than WGS84 (UTM, ETRS89).