res = 5.0 Contents

• Terrain Meshes
• Lidar
• Rasters
• 3D Rendering

Terrain Meshes

This tutorial covers how to generate meshes from point clouds and raster elevation data in Rhino using the RhinoTerrain plugin. The 3D modeling program Rhino has plugins for terrain modeling, 3D rendering, and computer aided manufacturing (CAM). With Rhino and its ecosystem of plugins you can model, visualize, and digitally fabricate landscapes and cities. To learn the basics of Rhino watch this overview of the user interface and read the user guide. The RhinoTerrain plugin can import geospatial data including point clouds, rasters, and vector datasets and can efficiently compute triangulated irregular networks (TIN), e.g. terrain meshes, from this data. See the tutorials on 3D printing a city and computer numerical control (CNC) surface milling to learn about digitally fabricating terrain models.

Lidar

Model a terrain mesh for Governor’s Island, NYC from the 2014 airborne lidar survey. Terrain data is often captured by airborne lidar or terrestrial laser scanning. RhinoTerrain can import .las lidar point clouds and generate terrain meshes from them. Note that older versions of RhinoTerrain may not be able to read newer .las specifications. Search for lidar data from the 2014 survey of Governor’s Island at https://orthos.dhses.ny.gov/. Download the tiles 18TWL820030.las and 18TWL820045.las.

Either run the command RtImportLas in the command line or in the RhinoTerrain menu select Import / Export and then Import las. Import 18TWL820030.las and then 18TWL820045.las. Points will be imported onto new layers by class. Delete or turn off the Las Low Point layer with noisy data. Then create a terrain mesh from both point clouds. Either run the command RtMeshTerrainCreate or in the RhinoTerrain menu select Mesh then Create terrain mesh. Once the command finishes, check the command line and accept the result. Turn off the layers with points clouds and set each of the viewports’ display mode to Rendered. If necessary, zoom to the data with the command Zoom and the options All viewports and then Extents.

RtImportLas
RtMeshTerrainCreate

If the terrain mesh takes too long to compute, has too wide an extent, or is too noisy, then edit the point cloud. To reduce the size of the point cloud and the resulting mesh, decimate the point cloud with the command RtPointCloudDecimate. To manually remove points from the point cloud use the command RtPointCloudPointsRemoveByWindow. Drag a window selection to remove points from the point cloud and then delete them. Once the point cloud has been edited, generate the terrain mesh again with RtMeshTerrainCreate.

Rasters

Use an elevation raster to generate a terrain mesh for part of Manhattan. RhinoTerrain can import elevation rasters in the GeoTIFF format. This is a .tif raster embedded with georeferencing data. While elevation rasters can be imported directly into RhinoTerrain, Rhino may run out of memory for rasters larger than about 300 MB depending on your hardware. If an elevation raster is too large, then preprocess the raster in GIS before importing it into Rhino with RhinoTerrain.

Elevation rasters from the 2017 topobathymetric lidar survey of New York City – including digital elevation and surface models – can be downloaded either from https://orthos.dhses.ny.gov/. or via file transfer protocol (FTP) from ftp://ftp.gis.ny.gov/. Note that FTP support is being disabled or removed from web browsers like Chome and Firefox due to security concerns. Either use a program such as FileZilla, the command line with curl, or enable FTP in your browser. In Chrome go to chrome://flags/ and turn on enable support for FTP URLs. Download hh_NYC_020.tif, the digital surface model for south Manhattan and Governor’s Island. Also download and extract the borough boundaries shapefiles for New York City from NYC Open Data. In the web mapping interface select Export, Download, and then Original to download nybb_21a.zip

Use GIS to define and mask the southern tip of Manhattan as the study region. Start GRASS GIS and create a new location from georeferenced data. When creating the new location read the projection and datum terms from the digital surface model hh_NYC_020.tif. First import the digital surface model hh_NYC_020.tif with r.in.gdal and then import the borough boundaries shapefile nybb.shp with v.in.ogr.

Extract Manhattan from the vector map of boroughs using v.extract with the SQL statement "BoroName = 'Manhattan'". Then set the region to the digital surface model with g.region and then create a vector area from the region with v.in.region. Find the intersection between the extracted borough and the vector region using v.overlay with the and operator. Remove the columns a_Shape_Leng and a_Shape_Area from the database of the resulting vector map with v.db.dropcolumn as these may cause the map not export properly. Set the region and mask to this vector map with g.region and r.mask. Then export the vector map with v.out.ogr and the masked raster map with r.out.gdal.

If the raster is still too large to process with RhinoTerrain, then use g.region to change the raster resolution from 1 x 1 foot cells to 2 x 2 foot cells with g.region res=2 and export again.

r.in.gdal input=hh_NYC_020.tif output=surface_2017
v.in.ogr input=nybb.shp output=boroughs
v.extract input=boroughs where="BoroName = 'Manhattan'" output=manhattan
g.region raster=surface_2017
v.in.region output=region
v.overlay ainput=manhattan binput=region operator=and output=manhattan_region
v.extract --overwrite input=manhattan_region cats=1 output=manhattan
v.db.dropcolumn map=manhattan columns=a_Shape_Leng,a_Shape_Area
g.remove -f type=vector name=region,manhattan_region
g.region vector=manhattan res=1
r.mask vector=manhattan
v.out.ogr input=manhattan output=manhattan.shp format=ESRI_Shapefile
r.out.gdal input=surface_2017 output=manhattan.tif format=GTiff

Start Rhino with the RhinoTerrain plugin. To import the digital surface model of Manhattan as a point cloud either run RtImportElevation in the command line or in the RhinoTerrain menu under Import / Export select Import elevation raster. Then import the vector area either by running RtImportVector in the command line or by selecting Import Shapefile vector in the RhinoTerrain menu under Import / Export.

Then use the command RtMeshTerrainCreate to generate a mesh from the point cloud using the vector area as a boundary. In the command line parameters set the boundary style to user defined and select the vector area, set input to Delete, and then click Accept. Set the viewport mode to Rendered or Artic to better visualize the mesh.

RtImportElevation
RtImportVector
RtMeshTerrainCreate

3D Rendering

Use a render engine to generate better visualizations of the terrain model. Render engines with plugins for Rhino include Thea Render, Vray, Maxwell, OctaneRender, and Enscape. Render engines with bridges to Rhino include Maverick Studio, Keyshot, and Lumion. This tutorial will use the Thea for Rhino plugin to render the model of Manhattan. Thea Render has real-time and production render modes that use the CPU and GPU simultaneously.

Start Rhino. From the Thea Render menu open the Render Settings panel. In the Render Settings tab turn on ambient occlusion and under devices select the CPU and GPUs to use for rendering. In the Environment tab turn on soft shadows and enable uniform illumination. Maximize the perspective viewport and then in the Camera tab sync with the Rhino viewport.

First adjust the lighting using Presto in interactive mode in the viewport. In the Thea Render toolbar select start interactive rendering in viewport. For lighting run the command Sun in Rhino’s command line and adjust the settings. When satisfied with the lighting stop the interactive rendering in the viewport.

Open Thea Darkroom, set the render engine to Presto and the mode to Presentation Render, and start rendering. When it finishes adjust the display settings. In Thea Darkroom’s Display Settings panel set tone mapping to filmic, shadows to 36, sharpness to 50, burn to 5, chroma to 50, and enable denoising. Save the rendering as a .png portable networks graphic file.

Edit the image in Photoshop. In the Image menu under Adjustments select Levels. On the input levels slider move the black point from 0 to 85 to increase the tonal range in the image. Save.