The ferry filter copies data from one dimension to another, creates new dimensions or both.
The filter is guided by a list of 'from' and 'to' dimensions in the format <from>=><to>. Data from the 'from' dimension is copied to the 'to' dimension. The 'from' dimension must exist. The 'to' dimension can be pre-existing or will be created by the ferry filter.
Alternatively, the format =><to> can be used to create a new dimension without copying data from any source. The values of the 'to' dimension are default initialized.
.. embed::
.. streamable::
In this scenario, we are making copies of the X and Y dimensions into the dimensions StatePlaneX and StatePlaneY. Since the reprojection filter will modify the dimensions X and Y, this allows us to maintain both the pre-reprojection values and the post-reprojection values.
{
"pipeline":[
"uncompressed.las",
{
"type":"readers.las",
"spatialreference":"EPSG:2993",
"filename":"../las/1.2-with-color.las"
},
{
"type":"filters.ferry",
"dimensions":"X => StatePlaneX, Y=>StatePlaneY"
},
{
"type":"filters.reprojection",
"out_srs":"EPSG:4326+4326"
},
{
"type":"writers.las",
"scale_x":"0.0000001",
"scale_y":"0.0000001",
"filename":"colorized.las"
}
]
}
The ferry filter is being used to add a dimension 'Classification' to points so that the value can be set to '2' and written as a LAS file.
{
"pipeline":[
{
"type": "readers.gdal",
"filename": "somefile.tif"
},
{
"type": "filters.ferry",
"dimensions": "=>Classification"
},
{
"type": "filters.assign",
"assignment": "Classification[:]=2"
},
"out.las"
]
}
- dimensions
A list of dimensions whose values should be copied. The format of the option is <from>=><to>, <from>=><to>,... Spaces are ignored. 'from' can be left empty, in which case the 'to' dimension is created and default-initialized. 'to' dimensions will be created if necessary.
Note: the old syntax that used '=' instead of '=>' between dimension names is still supported.