report op• Data kinds: image2d → scalar
• Call: import fullseye as fs; fs.ledger.data_range_of(*arrays, data_range=None) (to call the implementation directly, import imgmetrics; imgmetrics.data_range_of(*arrays, data_range=None); from the registry, opsimgmetrics.get("data_range_of"))
• GPU: none (CPU reference implementation only)
Decide the span the pixel values may take. It does not guess.
Integer dtypes (uint8 / uint16 / uint32 / int8 / int16) are uniquely determined by the dtype (other integer dtypes are not guessed, so give data_range explicitly). Float is treated as 1.0 only when it lies within [0, 1] and otherwise an explicit data_range is required.
Measuring PSNR of a [0, 1] float as if it were 255 is off by 48.13 dB, yet no exception is raised -- only a plausible-looking number comes out, so it is stopped here.
*arrays : ndarray
Arrays that should share the same pixel-value convention (usually 2 images).
data_range : float, optional
The range when given explicitly. ValueError unless it is a positive finite value.
float
• image_difference_metrics family guide
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• image_quality_metrics — py -3.11 examples/image_quality_metrics.py
scalar as input)luma_limited_u8 · rank_data · tid2013_index
report)compare_images · measure_with · metrics_table
*Provenance: imgmetrics.py — IMGMETRICS operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.