@misc{bc3c690a7a1c471c84339198a288d4b5,
title = "Improvements in Optical Surface Measurement Using Reflected Computer Vision Targets",
abstract = "Since 2021, NREL has been developing a system to measure heliostats by measuring the deflection of printed computer vision targets, called the Reflected Target Non-intrusive Assessment (ReTNA) [6], [7]. While this system will have lower resolution than a fringe deflectometry system, it has several important advantages that make it a complimentary technology: 2D surface slope measurement can be generated from a single image, it can operate in ambient lighting, target points can be directly located in 3D space with photogrammetry allowing for a non-flat target, and it's well suited to using a smaller target, and multiple images to measure larger optical surfaces. ReTNA has undergone several significant changes and improvements, described below. This talk will summarize new system layouts designed for commercial use, new computer vision algorithms used to automate the analysis process and validation campaigns for the ReTNA software.",
keywords = "computer vision, concentrated solar, deflectometry, heliostat, metrology, optics, photogrammetry, reflection",
author = "Devon Kesseli and Daniel Tsvankin and Tucker Farrell and Rebecca Mitchell and Milo Davis and Guangdong Zhu",
year = "2024",
language = "American English",
series = "Presented at the 30th SolarPACES Conference, 8-11 October 2024, Rome, Italy",
type = "Other",
}