The worn surface contains rich information about wear evolution; in situ measurement of its surface topography is of great significance for understanding wear mechanisms and monitoring equipment conditions. Therefore, this paper aims to solve the problems of existing 3D imaging technologies that are costly and difficult for in situ measurement and propose a low-cost, in situ 3D reconstruction method.
A kaleidoscopic in situ imaging system has been proposed for image acquisition, and the DUSt3R neural network was used for 3D reconstruction of the worn surface topography.
Compared to measurements obtained by the NANOVEA PS50 3D profilometer, the average measurement errors for worn surface depth and width are 11.86% and 12.92%, respectively, and the reliability of this method was proved.
Compared with other in situ measurement methods, the proposed method has similar accuracy in the 3D reconstruction results of worn surface topography, and the system structure is simpler and the cost is lower. It provides a feasible solution for in situ monitoring of mechanical equipment wear conditions.
