One of the most common products from a LIDAR project are digital elevation models (DEM) and, to a lesser extent, topographic contours. Back in the bad old days of prodding about with an RTK rover (or even more ancient, rod leveling), data points were extremely sparse relative to the desired density of the elevation model.
These sparse points work fine on areas that do not deviate too much from a planar surface (even if that surface were tilted) but could leave out important detail where sharp transitions in elevation occurred; e.g. a sharp ridgeline, roads carved into slopes and so forth.
The common approach to dealing with this issue was to collect supplemental lines of points along these breaks and inject them into the elevation model. These injected lines (they can also be points and polygons) are termed “breaklines.”
We find that traditional surveyors who transition to drone-based LIDAR systems are often concerned with how breaklines can be added to models in our EVO post-processing software. The good news is that True View EVO contains extensive tools for both adding and enforcing breaklines in topographical modeling. The even better news is that breaklines are seldom needed in the high density data that is collected with True View sensors. Recall that breaklines are primarily used to supplement sparsely collected elevation models. A typical True View 3D Imaging Sensor (3DIS®) data collect contains, at a minimum, over 50 points/m2 so sparse data are not an issue!
That said, there are occasions where breaklines are needed. One classic example are still water bodies such as lakes and ponds. Since the water is not moving. It must be flat. An elevation model that includes elevations over water bodies (some elevation models simply declare these regions as “voids”) should reflect this flat surface. Water body flattening is one of the more basic breakline tools in True View EVO.
Figure 1 shows an area containing holding ponds that was collected with a True View 410 LIDAR/Imaging (3DIS) system. The data are not an ortho but rather 3D LIDAR points colorized by the True View 410’s co-registered cameras. We will focus on the center pond for this discussion.
Figure 1: Holding ponds I applied EVO’s automatic ground classification (without taking any time to tweak settings) and displayed contours with a 30 cm (~1 ft) interval (see Figure 2). Actually, these contours look pretty good in the center pond area. You can clearly see two high (or low) points indicated by red arrows but overall, it looks pretty good for a fully automated process!
Figure 2: Automatic Ground Classification With Contours AppliedThe steps for really cleaning this up are:
- Digitize a polygonal boundary for the pond using EVO’s Feature Edit tools
- Set the elevation of the polygon to the water surface elevation (which can be determined in a variety of ways), again using EVO’s Feature Edit tools
- Classify all points interior to the polygon as something other than Ground (usually Water). This is done with EVO’s “Classify by Geometry” tools
- Declare the polygon to be a breakline
- Generate the output elevation model and/or contours using EVO’s Export Wizard
Figure 3 shows the result of the above process. The shoreline polygon is shown in blue. It was declared a “breakline” using EVO’s breakline tools (a subset of the Feature Edit tools) and the breakline enforcement in the Map View was enabled. Notice the perfectly flat surface and the fact that no contours cross the breakline boundary. If I were to apply Export Wizard to this project to create a DEM, the water body interior points would all have a constant elevation.
Figure 3: Shoreline Polygon Used As A Water Flattening Polygon This is just another example of the extensive product-oriented tools you will find within True View EVO. For many production scenarios, you will have all the tools you need to move data from raw sensor output to finished products such as hydrographically correct elevation models. As you can see, there is tremendous value in partnering with a systems provider who builds an entire solution, not just a sensor.