RockXaminer · Surface Pore Analysis

unwrapped core side texture → pore density & size distribution

Drop a core texture here
or click to browse

Samples

Core diameter

scale: —

Pore size distribution

min —max —

Pore types & fractures

Morphology classes follow carbonate usage (Choquette–Pray / Lucia): separate vugs vs interparticle matrix. Click a row to show/hide; click a fracture to jump to it.

Measurements (selected range)

—

pores selected

—/cm²

pore density

—%

areal porosity

—µm

median d50

—µm

d90

—mm

largest pore

Surface pore openings on the unwrapped side texture, above the optical resolution floor. Scale derives from the entered core diameter (full 360° wrap = π·D).

— scroll = zoom · drag = pan · drag the bar = before/after

Load a sample or drop an image to start.

no image loaded

Porosity · depth

Method — Surface Pore Characterization from Core Scans

Image-analysis methodology note · RockXaminer photogrammetric core scanning

1 · What is measured

The quantity reported is the apparent surface pore-opening size distribution: the population of visible pore openings intersecting the outer cylindrical surface of the core or plug, expressed as equivalent-circle diameters in millimetres, together with their areal fraction (surface porosity), number density (pores/cm²), and distribution statistics (d10/d50/d90). This is petrographic image analysis (PIA) in the tradition of Anselmetti et al. (1998), applied to the unwrapped lateral surface rather than a thin section, and is methodologically analogous to vug quantification on optical borehole-image logs (Cunningham et al., USGS).

2 · Acquisition and scale calibration

The core is digitized by multi-camera photogrammetry into a textured 3D model; the lateral surface is unwrapped into an orthographic texture spanning the full 360° circumference, edge-to-edge. Because the core diameter D is entered by the operator at scan time, the pixel scale is exact and traceable, not estimated:

scale [mm/px] = π·D / image width [px]

For the 1″ plugs shown here this gives ≈19.5 µm/px (1.25″: ≈24.4 µm/px). The Nyquist-type detection floor is ~3 px; sizes are therefore reported only above ≈0.1 mm. Every reported length can be audited in-app with the 📏 measure tool, which converts pixel distance through the same single scale factor.

3 · Segmentation (deterministic, fully inspectable)

Pore openings appear as locally dark regions against the matrix. Segmentation is classical morphological image analysis — every step is deterministic, parameter-visible, and auditable (no trained black-box model):

Illumination normalisation — a Gaussian low-pass background is subtracted (black-tophat principle), making detection robust to the gentle circumferential lighting gradient of a cylindrical surface (same rationale as the moving-average baseline used on optical borehole images).
Dual detection — (a) local contrast: pixels darker than their local background by a threshold δ capture small pores; (b) strict darkness: pixels darker than a tail-anchored absolute threshold (P1 + 0.25·(median−P1) of the matrix luminance) recover the uniformly dark interiors of large vugs, where the local background dips with the pore itself.
Morphological cleanup — opening/closing and hole-filling, so vugs are measured as filled openings, not rims.
Fracture separation — elongated traces (aspect ≥ 5, length ≥ ~2 mm, mean aperture ≤ ~0.85 mm) are extracted as fractures before pore partitioning and are excluded from all porosity statistics.
Watershed partitioning — coalesced openings are split at their waists via distance-transform marker watershed before per-pore measurement.
Shape and contrast gates — components that are neither locally contrasted nor truly dark, and border-truncated components, are rejected.

4 · Measurement definitions

Quantity	Definition	Notes
Equivalent diameter	`d = 2·√(A/π)` per pore	count-weighted statistic; standard PIA descriptor
Surface porosity	Σ pore area / analysed rock area	areal fraction of visible macro-openings; lower bound (§6)
Pore density	n / analysed area [cm²]	per selected size class
Depth log	areal porosity in a 1 mm rolling window along the core axis	exported as LAS 2.0 (DEPT.MM / SPOR.PCT)
Fracture metrics	trace length; mean aperture = area/length; apparent dip = trace angle to the circumferential axis	apparent, from the unwrapped trace

5 · Pore-type classes (Choquette–Pray / Lucia framing)

Detected pores are grouped by morphology: vug/moldic (≥0.5 mm, equant), interparticle/matrix (<0.5 mm), and channel/elongated (aspect ≥ 3). This follows the spirit of Choquette & Pray (1970) pore typing and Lucia’s (1995) separate-vug vs interparticle distinction, because the two populations carry different petrophysical meaning. The assignment is morphological only — fabric-selective classification in the strict Choquette–Pray sense requires petrographic confirmation.

6 · What this measurement is — and is not

It is not mercury-injection (MICP) data: MICP measures pore-throat radii via percolation; image analysis measures pore-body openings. The two differ by construction and must not be force-matched.
It is not total porosity: openings below the optical floor (≈0.1 mm) — including the micro- and mesoporosity that often dominates carbonate pore volume — are invisible. Surface porosity here is a lower bound of visible macroporosity, expected to read below helium porosity.
It is a surface measurement: the lateral skin is not a random interior section, so stereological volume conversions are not applied. Values are reported as measured, in 2D, with no model-based 3D extrapolation.
The surface must be clean: markings, mud or saw damage will be detected as dark features (visible on the demonstration plugs, which carry depth annotations).

7 · Internal QC and recommended laboratory tie-in

Threshold stability — analyses are swept at δ ±50%; the median opening size shifts by <0.01 mm on these plugs, and results are quoted with that range.
Resolution sweep — re-analysis at 2× and 4× downsampling identifies which part of the distribution is resolution-limited rather than geological.
Manual audit — any individual opening can be measured by hand in-app and compared with its tabulated diameter.
Method-matched lab comparison (expected, not forced, agreement): helium porosimetry → closure check (image ≤ He; the gap quantifies sub-resolution porosity); micro-CT → pore-body sizes above the CT floor; MICP → throats (expect smaller modes); slabbed-core / thin-section point counts → areal porosity of comparable fabric.

8 · References

Choquette, P.W. & Pray, L.C. (1970). Geologic nomenclature and classification of porosity in sedimentary carbonates. AAPG Bulletin 54(2), 207–250.
Lucia, F.J. (1995). Rock-fabric/petrophysical classification of carbonate pore space for reservoir characterization. AAPG Bulletin 79(9), 1275–1300.
Anselmetti, F.S., Luthi, S. & Eberli, G.P. (1998). Quantitative characterization of carbonate pore systems by digital image analysis. AAPG Bulletin 82(10), 1815–1836.
Cunningham, K.J. et al. (USGS). A new method for quantification of vuggy porosity from digital optical borehole images. U.S. Geological Survey.
Delesse, A. (1847). Procédé mécanique pour déterminer la composition des roches — areal-fraction principle underlying image porosity.