GeoTIFF (Library)¶

When rasterio is installed, inspect_file on a .tif or .tiff automatically checks whether the file carries a coordinate reference system (CRS). If it does, the result gets enriched with geospatial fields. If it doesn't, you get a plain TiffInspectResult with no extra keys and no error.

This enrichment happens transparently inside inspect_file. There is no separate function to call and no flag to pass.

Install¶

pip install voxelkit[geo]

Without rasterio, .tif and .tiff files are still fully supported. All the normal inspect, report, and preview functions work fine. You just won't get the geo fields.

Enriched inspect result¶

from voxelkit import inspect_file

result = inspect_file("scene.tif")

print(result.get("is_geotiff"))  # True if rasterio is installed and a CRS was found
print(result.get("crs"))         # "EPSG:32633"
print(result.get("bounds"))      # [300000.0, 5790000.0, 310240.0, 5800240.0]
print(result.get("resolution"))  # [10.0, 10.0]
print(result.get("band_count"))  # 4

The full result for a georeferenced file looks like:

{
  "filename": "scene.tif",
  "format": "tiff",
  "shape": [4, 1024, 1024],
  "ndim": 3,
  "dtype": "uint16",
  "page_count": 4,
  "axes": "CYX",
  "is_geotiff": true,
  "crs": "EPSG:32633",
  "bounds": [300000.0, 5790000.0, 310240.0, 5800240.0],
  "resolution": [10.0, 10.0],
  "band_count": 4
}

Geo fields¶

Field	Type	Description
`is_geotiff`	`bool`	Present and `True` when geo metadata was found
`crs`	`str`	Coordinate reference system string, e.g. `"EPSG:4326"`
`bounds`	`list[float]`	`[left, bottom, right, top]` in the CRS unit
`resolution`	`list[float]`	Pixel size as `[x_res, y_res]` in the CRS unit
`band_count`	`int`	Number of raster bands

Using report_file with GeoTIFFs¶

report_file works exactly the same as for any TIFF. Stats are computed over the full pixel array and the result is a standard FileReportResult without the geo fields. If you need geo metadata alongside QA stats, call both functions and combine:

from voxelkit import inspect_file, report_file

meta = inspect_file("scene.tif")
report = report_file("scene.tif")

geo_keys = ("crs", "bounds", "resolution", "band_count")
combined = {**report, **{k: meta[k] for k in geo_keys if k in meta}}

Batch workflows¶

report_batch picks up .tif and .tiff files and each file gets inspect enrichment when rasterio is available. That means CRS, bounds, and resolution appear alongside QA stats in one pass:

from voxelkit import report_batch
import json

results = report_batch("./satellite_archive/")

with open("inventory.json", "w") as f:
    json.dump(results, f, indent=2)

geo_files = [r for r in results if r.get("is_geotiff")]
print(f"{len(geo_files)} GeoTIFF files in archive")

Notes¶

Files without a CRS (plain scientific TIFFs, z-stacks, microscopy images) come back as a normal TiffInspectResult. No geo fields, no error.
rasterio is an optional dependency. If it is not installed, the inspect result is the same as if the file had no geo data.
Multi-band GeoTIFFs are supported. The shape follows tifffile's axis ordering (bands first for CYX arrays).
The GeoTiffInspectResult type is importable from voxelkit.core.types if you need type annotations in your own code.