Fusion of point cloud curvature and boundary information for high-precision seam feature extraction and measurement

Abstract In the design and production processes of industries such as aerospace, shipbuilding, and automotive, the precise measurement of gap and flush is critical to assembly quality and equipment performance. However, existing digital measurement technologies, limited by issues such as weak robustness in seam feature extraction, poor adaptability of measurement models, and insufficient system versatility, cannot satisfy the diverse measurement requirements of complex industrial scenarios. In response to the above challenges, this paper proposes a precise measurement method for seam gap and flush based on 3D point cloud, effectively improving measurement accuracy and system robustness. By merging curvature-based feature points with boundary feature points extracted using the angular criterion method. Three measurement models are designed, with the right method chosen according to different seam types. Experiments using five collection devices show that the proposed approach is highly robust and reliable under various interferences. It adapts well to different devices and enables quick application transfer. The gap and flush measurement models achieve excellent accuracy. The maximum flush measurement error is <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mtext>⩽</mml:mtext> </mml:mrow> </mml:math> 0.03 mm, the maximum gap measurement error is <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mtext>⩽</mml:mtext> </mml:mrow> </mml:math> 0.04 mm. The research proves that the algorithm overcomes traditional methods’ limitations in complex environments, improving measurement accuracy, and has significant engineering value for enhancing assembly quality and overall performance.