Design and analysis of full-scale scanning system for curved glass based on motion and 3D features

In recent years, mobile phones with glass curved screens have become more and more widely used. The irregular shape of the curved screen and the light transmittance characteristic of the glass have brought great challenges to its automatic defect detection. Aiming at the defect detection of the glass cover of the curved screen, this paper designs a full-scale scanning system by combining motion and three-dimensional (3D) features. First, a scanning system is constructed, and a geometric error modeling method is proposed to improve the accuracy of the scanning system; second, based on the point cloud of the 3D glass cover obtained by the scanning system, a point cloud registration method is presented by integrating the motion and 3D features; finally, the laser tracker is further used to calibrate the scanning system to analyze the mechanical error. Experimental results show that the introduction of straightness error and perpendicularity error can effectively solve the mismatch and fault problems of point cloud registration, and improve the accuracy of the scanning system. In addition, the registration method proposed in this paper can effectively reconstruct the complete point cloud of 3D glass cover for detection. The reconstruction accuracy of the plane part can reach 0.031 mm, and that of the curved part can reach 0.091 mm.