Skip to content

Interactive lab

Real outputs you can poke at

Three demos replay recorded results; the fourth runs its algorithm live in your browser. Each one says exactly what is real and what is schematic.

01Demo

Cobot fault detection: replay a recorded cycle

Pick a fault and play the cycle. The arm replays the recorded joint angles while the torque residual and tracking error are drawn against the band of 948 normal training cycles. When the cycle ends, the PCA detector and the Random Forest give the verdicts they actually gave.

Case study

Loading 13 recorded cycles…

Real
Joint angles, residuals and detector/classifier outputs for 13 recorded voraus-AD test cycles (one normal, one per fault: the median-scoring cycle of each category).
Schematic
The arm drawing uses generic link lengths and axes 1–4 only. It shows the recorded motion, not the robot's real geometry.
Runs where
Pre-computed by the project notebook; your browser only replays and plots it. No model runs here, and nothing is uploaded.

02Demo

Defect segmentation: inspect the C++ app case by case

Drag the divider to compare the ground-truth outline with the mask the C++/TensorRT application produced. Failures are included and labelled.

Case study

17 held-out cases · 11 of 15 defects located at IoU ≥ 0.5

C++ output: Cable, bent wireGround truthC++ output
Part
Cable
Defect
bent wire
Detection
cable 0.91
Mask IoU
0.79

Verdict: located

Real
Outputs saved from the C++/TensorRT application on 17 held-out MVTec AD images, with mask IoU scored against ground truth.
Selection
One random validation image per category (seed 0) plus two random defect-free parts. Failures are included.
Runs where
Pre-recorded. The engine needs an NVIDIA GPU, so nothing is inferred in your browser.

03Demo

Infrared detection: choose an operating point

Move the threshold to see the measured trade-off between finding 3–7 pixel targets and raising false alarms, and why 0.30 was chosen.

Case study

Detection vs false alarms per frame (log) · measured

0.000.250.500.751.000.0010.010.1false alarms per frame
Detected
97.9%
False alarms
0.0473 /frame

Chosen operating point: the tracker can reject an isolated false alarm, but it cannot recover a target the detector never reports.

Real
Detection rate and false alarms measured at each threshold on 12 held-out sequences (846 full frames).
Scoring
A detection counts when its centre is within 6 px of the target (IRSTD convention).
Runs where
A lookup over measured results; no inference in the browser.

04Demo

Sub-pixel registration: run it live

This one is not pre-recorded. Set a shift to a thousandth of a pixel, add sensor noise, and the phase-correlation algorithm recovers it in your browser on a 1/100 px grid.

Case study

Wafer pattern

Referenceclean design pattern
Targetshifted by (3.427, -5.184) px, 30 dB
Correlation peak, 1/100 px gridupsampled DFT
Residual after realigning|reference − target shifted back by the estimate|
Truth
(3.427, -5.184)
Estimate
…
Error
…
Compute, your browser
…

One shift is an anecdote. Run 50 random ones at the current pattern and noise.

Real
The registration runs live in your browser: a TypeScript port of the project's NumPy library, about 40 ms per 256×256 pair on a laptop.
Checked
Same estimate as the Python library on identical inputs for logic and DRAM patterns; within 0.09 px on patterns with spectral nulls, where FFT rounding decides (mean error within 1%).
Differs from the notebook
256×256 instead of 512×512 and a browser random generator, so batch numbers will not match the notebook tables exactly. Nothing is uploaded.