UFO Debris Lattice Matcher 3D

Unknown sample

Al (aluminum) 72

Fe (iron) 5

Si (silicon) 15

Ca (calcium) 8

Lattice sampling

Measurement noise 0.08

Best-match material

–

Confidence

–

Atoms sampled

–

How it works

Left lattice = the "unknown" sample: each lattice site is stochastically assigned an element according to your composition sliders, then a fraction of atoms (set by measurement noise) are randomly reassigned to mimic real detector/sampling error. Right lattice = the reference structure for whichever library material currently scores highest. Instead of comparing continuous spectrum curves (as the 2D version does), this compares discrete atom-count histograms with cosine similarity — a standard way to score compositional match when the raw data is counts, not intensities.

cos(m, r) = (m·r) / (‖m‖ ‖r‖)
confidence = cos(m, r) × 100%
m, r = realized element-fraction vectors (Al, Fe, Si, Ca)

Composition sliders — the true elemental fractions used to draw the sample lattice.
Lattice sampling (4³/6³/8³) — how many atoms are drawn; a small lattice gives a noisier realized histogram even at zero measurement noise, same as a short XRF acquisition.
Measurement noise — probability that a given atom's element is reassigned at random, simulating detector miscounts or surface contamination.
Re-sample lattice — draws a fresh random lattice from the same settings, so you can see run-to-run variance in the confidence score.

Real-world relevance: composition-histogram matching (via cosine or chi-square similarity) is the discrete-counting analogue of spectral cross-correlation, used when a lab reports element counts from EDS/SEM point analysis rather than a continuous XRF trace.

SampleBest-match reference

Al Fe Si Ca

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.