odc.geo.overlap.compute_output_geobox
- odc.geo.overlap.compute_output_geobox(gbox, crs, *, resolution='auto', shape=None, tight=False, anchor='default', tol=0.01, round_resolution=None)[source]
Compute output
GeoBox
.Find best fitting, axis aligned GeoBox in a different coordinate reference given source
GeoBox
on input.- Parameters:
crs (
Union
[str
,int
,CRS
,CRS
,Dict
[str
,Any
]]) – Desired CRS of the outputresolution (
Union
[float
,int
,Resolution
,Literal
['auto'
,'fit'
,'same'
]]) –“same” use exactly the same resolution as src
”fit” use center pixel to determine scale change between the two
- ”auto” is to use the same resolution on the output if CRS units arethe same between the source and destination and otherwise use “fit”
Ignored if
shape=
is suppliedElse resolution in the units of the output crs
shape (
Union
[Tuple
[int
,int
],XY
[int
],Shape2d
,Index2d
,int
,None
]) – Span that many pixels, if it’s a single number then span that many pixels along the longest dimension, other dimension will be computed to maintain roughly square pixels. Takes precedence overresolution=
parameter.tight (
bool
) – By default output pixel grid is adjusted to align pixel edges to X/Y axis, supplingtight=True
produces unaligned geobox on the output.anchor (
Union
[AnchorEnum
,XY
[float
],float
,Literal
['center'
],Literal
['centre'
],Literal
['edge'
],Literal
['floating'
],Literal
['default'
]]) – Control pixel snapping, default is to snap pixel edge toX=0,Y=0
. Ignored whentight=True
is supplied.tol (
float
) – Fraction of the output pixel that can be ignored, defaults to 1/100. Bounding box of the output geobox is allowed to be smaller by that amount than transformed footprint of the original.round_resolution (
Union
[None
,bool
,Callable
[[float
,str
],float
]]) –round_resolution(res: float, units: str) -> float
- Return type:
- Returns:
Similar resolution, axis aligned geobox that fully encloses source one but in a different projection.