Abstract: The monocular structured light system composed of camera and projector plays a crucial role in the field of three-dimensional measurement. System calibration is a pivotal step in ensuring the accuracy of three-dimensional measurements. Traditional calibration algorithms,during the image acquisition process,have not considered the impact of different poses on the precision of system calibration. Addressing the calibration issue of a monocular structured light system with a planar target,two situations that should be avoided during calibration pose selection were proposed. Firstly,poses where the target plane is parallel to the imaging plane of the camera or the projection plane of the projector; secondly,poses involving only translation or rotation within the plane of the target.Building upon this,a criterion for target image selection was proposed,incorporating a threshold for the Euler angles of the target images. This criterion filters out images that negatively impact calibration precision,retaining those with advantageous target poses.Experimental results demonstrate that the selection of target image filtering contributes to ensuring system measurement accuracy,achieving high calibration precision,and meeting the accuracy requirements of contemporary industrial production and processing.

Key words: structured light measurement, system calibration, screening criteria, calibration accuracy