Smartphone scene generator for efficient characterization of visible imaging detectors

Abstract

Full characterization of imaging detectors involves subjecting them to spatially and temporally varying illumination patterns over a large dynamic range. Here we present a scene generator that fulfills many of these functions. Based on a modern smartphone, it has a number of good features, including the ability to generate nearly arbitrary optical scenes, high spatial resolution (13 um), high dynamic range (~10^4), near-Poisson limited illumination stability over time periods from 100 ms to many days, and no background noise. The system does not require any moving parts and may be constructed at modest cost. We present the optical, mechanical, and software design, test data validating the performance, and application examples.

14 pages. This version includes code, available here: https://github.com/Leo-Neat