Machine Learning Based Luminance Analysis of a µLED Array

Abstract

In the past years, the development of µLED arrays gained momentum since they combine the advantages of LEDs, such as high brightness and longevity, with the high resolution of a micro-scaled structure. For their development, spatially resolved measurement of luminance and color of single µLEDs and the entire light-emitting surface are usually analyzed as they quantify the visual perception. However, studying individual µLEDs is time-consuming to measure and evaluate, while examining the entire light-emitting area suffers from interference from non-functioning µLEDs. This paper presents a method to perform both analyzes with a single measurement employing unsupervised machine learning. The results suggest that a precise reconstruction of the µLEDs and a more accurate characterization µLED arrays is achieved.