> For the complete documentation index, see [llms.txt](https://docs-7.phoenixlidar.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs-7.phoenixlidar.com/lidarmill-desktop/workflow-kk/classification/find-ground-ptd.md).

# Find Ground (PTD)

Classifies ground points using a progressive tin densification algorithm.

Select input class(s) and output class., then Click Run Filter.

After processing has finished click **Save Classification to Cloud** to save changes, or click **Cancel Fitler** to cancel changes. 

{% hint style="info" %}
It is recommended to use a **Ground Thickness** value equal to, or slightly greater than the ground thickness of your lidar pointcloud
{% endhint %}

![](/files/-M_XeCfuWgP5DEHA3Siz)

This ground filter is a Progressive TIN Densification (PTD) algorithm, where TIN stands for Triangulated Irregular Network. The algorithm classifies points as ground by the following steps:

1. First, a uniform planar grid is spanned across whole dataset and each cell is assigned the lowest point falling into its bounds. 
2. This provides initial samples for the TIN creation. 
3. TIN mesh created using delaunay triangulation, with respect to XZ plane and then use it as a base for iterative classification.
4. In each iteration, points are tested for incidence angle distance to the nearest triangle of the TIN, and assigned ground class, should they pass. 
5. At the end of the iteration, these points are added to the TIN, refining it and allowing more points to be classified as ground. 

### Parameters

* **Cell Size** - Cell size for initial TIN creation. Should reflect max building size within project area
* **Feature size** - Finer cell size for initial TIN creation. Smaller values increase memory consumption, but improve the algorithm's ability to climb slopes
* **Iteration Angle** - Angle to plane as described above
* **Iteration Distance** - Distance to plane per iteration 
* **Maximum iterations** - Maximum number of densification iterations
* **Ground Thickness** - How much algorithm should thicken the final ground. Should reflect expected ground thickness.