Meshes

A mesh is a surface made of triangles — each triangle defined by three vertex positions. Meshes are the right representation when you need a continuous surface, e.g., for rendering, leaf-area calculation, or as a target for cloud-to-mesh registration.

Where meshes come from

There are four ways to bring a mesh into Phytograph:

1. Import an existing .obj, .ply, or .stl file. A textured .obj with a sibling .mtl and image files is imported with its textures applied (see Textures). A .ply is recognized as a mesh when it contains faces, and as a point cloud when it does not — see File formats: PLY.
2. Triangulate a point cloud — see Triangulate a mesh.
3. Generate a plant — every procedurally generated plant arrives as a mesh of stems, branches, and leaves. See Generate a plant.
4. Generate a DEM / DSM / CHM — a gridded terrain surface reconstructed from a cloud, stored as a heightmap mesh coloured by elevation and exportable as a GIS raster. See Terrain surfaces below and Generate a DEM / DSM / CHM.

Terrain surfaces: DTM, DSM, CHM

The DEM tool builds three related gridded surfaces from a cloud. Each is a regular grid of one value per cell, reconstructed as a heightmap mesh and exportable as a GeoTIFF / ESRI ASCII raster:

• DTM (Digital Terrain Model) — the bare-earth ground surface. Built from the cloud's ground points (from ground segmentation, or auto-detected), taking a low per-cell percentile so residual low vegetation and noise don't lift the terrain. This is the classic "DEM".
• DSM (Digital Surface Model) — the top-of-canopy surface: the highest return in each cell (a high per-cell percentile over first returns). It includes vegetation and structures, and does not need ground classification.
• CHM (Canopy Height Model) — DSM − DTM: vegetation height above the bare earth. The DTM and DSM are gridded on one aligned grid and subtracted, so each cell reports the canopy height above the ground beneath it. Values are floored at zero (canopy height is never negative) and a first-pass pit-fill removes isolated within-canopy dips. The CHM is a core forestry product — tree-height and canopy-structure metrics derive directly from it. The exported raster holds the canopy height (referenced to the ground), while the displayed CHM surface is draped on the terrain (drawn at ground elevation + height) so it sits where the canopy actually is rather than floating from zero.

First returns for the DSM are read from the cloud's multi-return target_index (0 = first return); a single-return cloud treats every point as a first return.

DTM layers: one surface, many bands

A DTM is a single surface that carries several scalar layers on its grid — you colour the one terrain mesh by any of them (the Color by dropdown) and export any of them as a raster. This mirrors how a point cloud carries multiple scalar fields. All layers are computed automatically with the DTM and share its grid:

• Elevation — the ground height (the default band).
• Point density — points per cell; Return density — laser pulses (first returns) per cell. Gridded directly from the points (empty cells stay void).
• Intensity — mean return intensity per cell (when the cloud carries intensity).
• Hillshade / Slope / Aspect — derived from the elevation grid (not from points): hillshade is shaded relief with a fixed sun (azimuth 315°, altitude 45°); slope is the grid's steepness from horizontal; aspect its downslope compass bearing. Grid-based, matching GIS conventions.

Switching the layer recaptions the colorbar; Export raster writes the value of whichever band(s) you tick. DSM and CHM stay single-value surfaces (their geometry is the value).

Triangulation methods

Method	Best for	Speed
Delaunay	Quick previews, mostly-flat surfaces	Fast
Ball Pivot	Branch surfaces with consistent point density	Medium
Poisson	Watertight surfaces from dense clouds	Slow
Alpha Shape	Wrapping concave shapes tightly	Medium
Helios	Multi-scan TLS data with known scanner positions	Medium

All five live in one Triangulation Setup modal (pick the method from its dropdown). The Helios method is unique: it uses the scan geometry (where the scanner was, and the angular sweep) to triangulate only the rays that actually returned, producing accurate branch surfaces without the "shrink-wrap" artifacts that come from cloud-only methods. See Triangulate a mesh: Helios.

What you can do with a mesh

Operation	Workflow
Filter triangles (Lmax / aspect)	Expand a Helios mesh row → Filter — drops long or mis-shapen triangles live. Helios-only: cloud methods (ball pivot, Poisson, alpha shape, Delaunay) apply their length scale during reconstruction, so there's nothing left to filter
Plot leaf angles	Expand the mesh row → Leaf angles… — inclination PDF + azimuth rose + de Wit fit from the triangle normals; works on any triangulation method (details)
Translate / rotate / scale	Click the ⤢ (double-arrow) button on the mesh's row in the Meshes panel to open its Transform panel
Scan it into a point cloud	Synthetic LiDAR scan
Align two meshes (ICP)	Register & compare: M2M
Compare a cloud to a mesh	Register & compare: C2M
Export	Import & export

Display options

• Solid — flat-shaded triangles in the mesh's color.
• Wireframe — only the edges, useful for inspecting topology. Toggled once for all meshes from the Mesh Settings footer of the Meshes panel.
• Opacity — set per mesh. Expand a mesh row (the chevron at its left) and drag the Opacity slider to make that surface semi-transparent, e.g. to see an underlying point cloud through a triangulation. Each mesh keeps its own value. The slider is only shown for meshes where blending is meaningful — solid and vertex-colored surfaces. It is hidden for generated plants and other textured meshes (see Textures).

Textures

Meshes that carry image textures are rendered with them automatically — there is no toggle to turn on. A mesh is textured when it has UV coordinates and at least one material that references an image; otherwise it falls back to its vertex colors or solid color.

Two sources produce textured meshes:

• Generated plants. Leaves and bark use the textures from the Helios plant library. Each organ samples the correct part of its leaf-image atlas, and the leaf silhouette is cut out using the image's transparency. Stems, branches, and flowers that have no texture render with their organ colors.
• Imported .obj files. When an .obj references a .mtl and the images it names sit next to the file on disk, Phytograph loads the textures and applies them. Faces whose material has no image fall back to that material's diffuse color.

Wireframe applies to textured meshes too. Opacity does not: textured plants draw their leaf textures as crisp alpha cut-outs that ignore a blend factor, so the per-mesh Opacity slider is hidden for them — it would have no visible effect. Use opacity on solid or vertex-colored surfaces instead.

Triangle counts and performance

A scan-triangulated mesh of a single mature tree typically has 1–5 million triangles. Phytograph draws these without simplification but performance degrades above ~10M triangles. For very large meshes, consider exporting and re-importing a decimated version (in MeshLab, Blender, or similar) before further analysis.