Accurately measuring the rotor’s inner diameter in industrial manufacturing often faces several challenges, including perspective distortion, edge defects, and extraneous concentric circle patterns on the rotor surface. To address these issues, this study proposes a comprehensive computer vision–based measurement method. The similar triangle–based calibration effectively compensates for perspective distortion during the rotor imaging process, thereby reducing potential measurement bias. Furthermore, a two-stage measurement strategy is introduced to robustly handle complex concentric circle patterns and edge defects, enabling highly accurate inner diameter measurements. Extensive experimental validation demonstrates that the proposed method significantly improves both the accuracy and stability of inner diameter inspection, with measurement errors controlled within 0.006 mm. Compared with existing commercial measurement systems, the proposed approach achieves superior precision and robustness, meeting the stringent requirements of high-precision rotor inspection in industrial environments.
Liu et al. (Fri,) studied this question.