Rasterization: Difference between revisions
No edit summary |
No edit summary |
||
| Line 2: | Line 2: | ||
The rasterization model is a part of the [[Simulation Core]], that is tasked with converting attributes and polygons from features of various layers into raster values. | The rasterization model is a part of the [[Simulation Core]], that is tasked with converting attributes and polygons from features of various layers into raster values. | ||
* Most geospatial data is represented as polygons, such as the 2D layout of a building or woodlands. When you select | * Most geospatial data is represented as polygons, such as the 2D layout of a building or woodlands. When you select [[Compute Cloud|Update Now]], the {{software}} automatically begins rasterizing this data into the required raster grid at the resolution you specify. | ||
* The internal Rasterizer is specifically designed for small grid cells (<1m), leveraging the power of cloud computing to enable multiple cores to work on the conversions simultaneously. In many cases, you may have multiple simulations running, which are also processed in parallel in real-time. | * The internal Rasterizer is specifically designed for small grid cells (<1m), leveraging the power of cloud computing to enable multiple cores to work on the conversions simultaneously. In many cases, you may have multiple simulations running, which are also processed in parallel in real-time. | ||
Revision as of 09:58, 3 August 2026
The rasterization model is a part of the Simulation Core, that is tasked with converting attributes and polygons from features of various layers into raster values.
- Most geospatial data is represented as polygons, such as the 2D layout of a building or woodlands. When you select Update Now, the Tygron Platform automatically begins rasterizing this data into the required raster grid at the resolution you specify.
- The internal Rasterizer is specifically designed for small grid cells (<1m), leveraging the power of cloud computing to enable multiple cores to work on the conversions simultaneously. In many cases, you may have multiple simulations running, which are also processed in parallel in real-time.
- Internal caching further optimizes this process, ensuring that when you hit “Recalculate,” the rasterization occurs even more quickly.
Layers
The layer rasterization sequence of the polygon features is:
Methods
Converting polygons to grid cells can be done in different ways. A common approach is by checking the intersection between the polygon outline and the cell. As displayed below:
However this will create a small overshoot resulting in larger cell area than the original polygon. To correct for this one could also rasterize based on the center point. As displayed below:
While more accurate, this method may fail for objects smaller than the grid cell area. If the cell center point misses a small object, it won't be rasterized and remains as traversable space in the simulation.
To compensate for this situation the default option is combined center + outline. This method uses center-point rasterization for most objects and outline rasterization for small ones to ensure coverage.
Via the editor Calculation panel the rasterization method can be selected.
- Combined Center for large and Outline for small polygons (default).
- Outline Overshoot results in more cells than polygon area.
- Center Fastest but missing small polygons.
Note: Since version 2025 combined is the default, before that outline was used.
