Classification from  Lidar DSM

Specifically for lidar

Remove the points in a database from a grid: From the grid's map menu, Edit, Mark as missing, Points from database, and pick an open point database.

Replace voids in the classification grid with mode of the surrounding points:  From the grid's map menu, Edit, DEM holes, Replace with mode (8 or 24). 

Remote "orphan" points in the classification:

Streetlight, find distance to nearest neighbor

Database to grid:

Edit options on the grid

Combine two grids:

Combine a Geometry grid from the DEM and the hyperspectral

Others to use

Use in conjunction with supervised classification on imagery.

Access from the training set database, which must be opened from the image.

Pick Stats, Supervised classification, Grid statistics.

Pick the grid you want.

The three colored symbols are the min, mean, and max for the training points.  The black line can be picked on the Analysis tab of the options form:

  • One standard deviation on either side of the mean.
  • Quantiles 25 and 75.

last revision 3/1/2013