Scaling photogrammetry: a comparative evaluation and metrological assessment across small- and meso-scale domains

Abstract

Photogrammetry inspection is a Machine Vision (MV) technique intensely employed to assess the geometry of industrial assets across several measurement scales, ranging from micro-scales focusing on surface to meso- and large-scales targeting geometrical features and shape.This research endeavors to conduct a comprehensive comparative evaluation of photogrammetry across different dimensional scale domains, aiming to establish a framework for assessing performance levels in various aspects, driven by the portability of the instrumentation, measurement performance and proficiency. Central to the current methodology is employing a single camera, driven by the research’s forward-looking goal to integrate drone technology equipped with a solitary camera as the primary payload. In addition, this work presents a statistical quantitative investigation where the most relevant sources of uncertainty are taken into account. Three case studies about a small truss, a ball-bar, and a collaborative robot accompany the analysis.Finally, this study proposes a framework for assessing the expanded uncertainty and the relative uncertainty across the scales, revealing that the latter decreases with larger measurand, providing a value of 0.2% when dealing with meso-scale objects.