Filter ground returns using the approach outlined in [Mongus2012].


Our implmentation of Mongus is in an alpha state. We’d love to have you kick the tires and provide feedback, but do not plan on using this in production.

The current implementation of filters.mongus differs slightly from the original paper. We weren’t too happy with the criteria for how control points at the current level are compared against the TPS at the previous scale and were exploring some alternate metrics.

Some warts about the current implementation:

  • It writes a bunch of intermediate/debugging outputs to the current directory while processing. This should be made optional and then eventually go away.
  • We require specification of a max level, whereas the original paper automatically determined an appropriate max level.
[Mongus2012]Mongus, D., Zalik, B., 2012. Parameter-free ground filtering of LiDAR data for automatic DTM generation. ISPRS J. Photogramm. Remote Sens. 67, 1–12.

Default Embedded Stage

This stage is enabled by default


The sample pipeline below uses filters.mongus to segment ground and non-ground returns, writing only the ground returns to the output file.



Cell size. [Default: 1.0]
Apply classification labels (i.e., ground = 2)? [Default: true]
Extract ground returns (non-ground returns are cropped)? [Default: false]
Standard deviation multiplier to be used when thresholding values. [Default: 3.0]
Maximum level in the hierarchical decomposition. [Default: 8]