ApertureAPI — Holographic Reconstruction Studio

The Specimen

Provide the raw hologram exactly as captured by your sensor — no preprocessing required.

Uploads are secure and not shared with others.

Hologram Capture Drag & drop or click to upload the raw interference pattern

Place your hologram here

TIFF · PNG · NPY · JPG · ≤ 75 MB · ≤ 4000 px per side

Reconstruction Character

Choose the fidelity of the render and how the interference geometry should be interpreted.

Quality Mode High-Quality is recommended for most work; High Dynamic Range recovers detail in extreme-contrast scenes

Standard

Enhanced

High-Quality

High Dynamic Range

Holography Geometry Off-axis, inline, or let the pipeline detect the carrier automatically

Off-Axis

Inline

Auto-Detect

Inline Reconstruction Quality Choose between speed and twin-image suppression strength

The Optical Bench

Describe the physical rig that recorded the hologram — or begin from a curated instrument preset.

Instrument Presets

Wavelength (μm) Illumination wavelength

Pixel Pitch (μm) Sensor pixel size

Autofocus — automatically locate the optimal reconstruction distance

Min Distance (mm) Search range minimum

Max Distance (mm) Search range maximum

Propagation Method Numerical model for wavefield propagation

Beam Geometry — Off-Axis, No Reference ▼

When to use beam geometry

If you don't have a reference hologram, provide the beam angle for precise carrier extraction and tilt correction. Optionally add the beam splitter distance for spherical wave compensation (non-collimated setups).

Beam Angle (degrees) Reference beam tilt angle. Enables precise carrier frequency extraction.

Beam Splitter Distance (mm) BS-to-sensor distance. Corrects spherical reference wave.

Reference Hologram — optional, off-axis Upload a reference hologram for phase compensation. If unavailable, use beam geometry above.

Place reference hologram here

Software-Defined Calibration — optional Photograph a known reference target with your rig and upload the shot (≤1024×1024, centre-crop, don't downscale). Our digital-twin optimizer fits your camera's true pixel pitch and wavelength against the target's exact 3D model and applies them to this reconstruction. Targets: M1.6 DIN 934 hex nut (off-axis) or Eisco prepared salt crystals slide (inline).

Refinements

The defaults reflect our recommended full pipeline. Adjust only if your application demands it.

Advanced Options ▼

Background subtraction

Preserve full quality (clinical-grade; no lossy crop or quantization)

Optical aberration correction

Edge-preserving phase smoothing

Speckle pre-conditioning

Deterministic mode (reproducible audit trail)

Smoothing strength Leave blank for auto (recommended)

Detail-preservation guard Automatically revert smoothing if it removes too much detail

0.30

Processing your hologram…

Upload

Propagation

Autofocus

Mesh

Estimated time: calculating…

Reconstruction complete

Quality Report

Agreement Score

SNR Measurement

Classification

Fusion Weights

Verified Facts

Every statement in this section is backed by a committed, independently checkable artifact. Where a figure has no committed artifact, we publish no figure.

Software-Defined Calibration — live in production

Digital-twin calibration is live on the production API, including the no-account /public/reconstruct endpoint. In the committed closed-loop proof, the optimizer recovered the true pixel pitch to within 0.0005% from a nominal that was mis-specified by 6%, and the digital twin replicated the measured field to within the optimizer's convergence tolerance. This is a closed-loop synthetic validation of optimizer convergence and endpoint plumbing — not a physical-target measurement, and we label it as such. Artifact: calibration/results/sdc_end_to_end_hex_nut.json.

Deploy provenance you can query

The live /health endpoint serves a full deploy-identity block, and a hash-chained deploy ledger verifies intact from genesis. Both are registered live checks (deploy-identity, ledger-verify) — ask, and the running system answers with a live run.

Data-retention governance, fail-closed

Retention policy is class-keyed with fail-closed owner routing (unclassified data routes to a named data-integrity owner, never to a default), write-once legal-hold events with declared release conditions, and provable clock-basis divergence. Verified by the registered retention-proof check.

Conformity decisions are data, not code

Quality-threshold decisions ship as content-addressed, supersession-chained records with an enforced separation between the business decision owner and the engineering computation. The freeze gate requires at least three independent evidence lineages — a passing verdict cannot be self-manufactured by the vendor.

Benchmark discipline

Every published performance number is stamped with the committed result artifact it came from; capabilities without a committed measurement are listed as not-yet-benchmarked and carry no number. A freshness check (bench-doc) fails the build if the published document drifts from the artifacts.

Regression budget enforced at every release

A quiet-window benchmark-regression gate runs against a committed baseline with a fixed 0.80 performance budget before changes are accepted. The gate result is a registered live check (bench-regression).

Light, resolved intomeasurement