A novel method for analyzing the influence of relative attitude of laser profile scanner on 3D measurement system

Scanning measurement systems are a pivotal solution to enable the rapid collection of 3D data from physical entities and provide further information for quality control. However, some external factors besides the scanner itself can also influence the measurement accuracy of the scanning system. In this paper, we propose a framework for analyzing the relative attitude influence of scanners on measurement uncertainty in the X and Z directions, respectively, containing the measured uncertainty calculation by the proposed optimal edge points detection and plane segmentation algorithm and the measurement uncertainty prediction model by machine learning. Experiments demonstrate that the distribution feature of measurement uncertainty differs in the X and Z directions and can provide predictions of the measurement uncertainty with about 0.2 % averaged MRE value. Our analysis advances the understanding of measurement uncertainty of scanning systems in different directions while it can be further applied to refine the calibration and measurement process.