vibespatial.raster.geokeys¶

GeoTIFF GeoKey parsing for nvImageCodec geo metadata.

Converts raw GeoKey dicts (MODEL_PIXEL_SCALE, MODEL_TIEPOINT, EPSG codes) into the affine 6-tuple and pyproj.CRS that vibespatial-raster uses.

No dependency on nvImageCodec, CuPy, or rasterio.

Functions¶

`geokeys_to_affine`(→ tuple[float, float, float, float, ...)	Convert GeoTIFF geo-referencing tags to a GDAL-style affine 6-tuple.
`geokeys_to_crs`(→ pyproj.CRS \| None)	Build a `pyproj.CRS` from a GeoTIFF EPSG code.
`parse_nvimgcodec_geo_metadata`(→ tuple[tuple[float, ...)	Parse an nvImageCodec geo-metadata dict into an affine tuple and CRS.

Module Contents¶

vibespatial.raster.geokeys.geokeys_to_affine(pixel_scale: list[float] | None = None, tiepoint: list[float] | None = None, model_transformation: list[float] | None = None) → tuple[float, float, float, float, float, float] | None¶

Convert GeoTIFF geo-referencing tags to a GDAL-style affine 6-tuple.

The affine tuple (a, b, c, d, e, f) maps pixel coordinates to world coordinates:

world_x = a * col + b * row + c world_y = d * col + e * row + f

Parameters¶

pixel_scalelist[float] or None: MODEL_PIXEL_SCALE — [scaleX, scaleY, scaleZ].
tiepointlist[float] or None: MODEL_TIEPOINT — [i, j, k, x, y, z].
model_transformationlist[float] or None: MODEL_TRANSFORMATION — 16-element row-major 4x4 matrix.

Returns¶

tuple[float, float, float, float, float, float] or None: (a, b, c, d, e, f) affine, or None if insufficient data.

vibespatial.raster.geokeys.geokeys_to_crs(epsg_code: int | None, *, model_type: int | None = None) → pyproj.CRS | None¶

Build a pyproj.CRS from a GeoTIFF EPSG code.

Parameters¶

epsg_codeint or None: EPSG code (e.g. 4326, 32633). None means no CRS.
model_typeint or None: GT_MODEL_TYPE (1 = Projected, 2 = Geographic). Currently informational only; the EPSG code already encodes this distinction.

Returns¶

CRS or None: A pyproj.CRS object, or None if the code is missing or unrecognised.

vibespatial.raster.geokeys.parse_nvimgcodec_geo_metadata(geo_metadata: dict) → tuple[tuple[float, float, float, float, float, float] | None, pyproj.CRS | None]¶

Parse an nvImageCodec geo-metadata dict into an affine tuple and CRS.

This is the main entry point for converting the raw dict that nvImageCodec attaches to decoded GeoTIFF images.

Parameters¶

geo_metadatadict: Raw geo-metadata dict with keys such as MODEL_PIXEL_SCALE, MODEL_TIEPOINT, MODEL_TRANSFORMATION, PROJECTED_CRS, GEODETIC_CRS, and GT_MODEL_TYPE.

Returns¶

tuple[tuple | None, CRS | None]: (affine, crs) — either value may be None when the corresponding metadata is missing or unparseable.