← All projects
Data & Test Automation2026

Sensor Radiation Analysis

Automated pipelines that turn hours of image-sensor noise and lifetime analysis into a one-minute report.

3 hr → 1 min
Analysis turnaround
−35%
Manual validation cycles

A set of analysis tools built at Teledyne DALSA to characterize image-sensor behavior (noise, quantum efficiency, and lifetime) automatically instead of by hand.

Highlights
  • Scripted pipelines to process PRNU, QE, and dark-noise metrics on low-yield BSI sensors in a Level 10 cleanroom.
  • Built a Pytest executable for sensor-lifetime estimation, cutting analysis from ~3 hours to ~1 minute.
  • Automated temperature-coefficient alignment, reducing manual calibration cycles by 35%.

01Context

Image sensors are characterized by metrics like PRNU, quantum efficiency, and dark noise. Measuring and analyzing them by hand is slow and repetitive.

02Approach

I scripted pipelines (Python/Pytest, MATLAB) that ingest raw sensor data and compute the key metrics, plus a lifetime-estimation executable and automated temperature-coefficient alignment.

03Result

Lifetime analysis dropped from ~3 hours to ~1 minute, and manual validation cycles fell 35%, freeing time for real engineering judgement.

Want to talk through how I built this?