Uncertainty evaluation of 3D scanning measurement systems

A binocular structured-light measurement system involves a complex model, multiple error sources, and correlation coefficients that are difficult to determine, making it hard to accurately evaluate its measurement uncertainty using the conventional Guide to the Expression of Uncertainty in Measurement(GUM). To address this issue, an improved GUM method based on grey theory is proposed. First, the grey model GM(0,2) is employed for non-statistical evaluation of Type A uncertainty components, while Type B components are assessed by the traditional approach. Then, grey relational analysis is applied to quantify the correlations among error sources and to determine the corresponding correlation coefficients. Finally, the overall measurement uncertainty is synthesized within the GUM framework. Applied to a binocular structured-light measurement system, the proposed method yields a measurement uncertainty of 0.0934, demonstrating its reliability.