Abstract
The rapid development of digitization and 3D printing is creating an ever-increasing demand for methods for the automated generation of 3D models from real components. Thanks to the progress and widespread use of computer vision, it is now possible to merge classical engineering tasks with image processing techniques. Computer-aided design can therefore be automated using information from image data. In this study, we present a novel method for automated digitization of 3D structures using AprilTag fiducial system and Solid Geometry Library. The proposed design process is implemented in matlab. AprilTags are used to realize 3D coordinate measurements to digitally capture the 3D dimensions of real components. Based on these data, 3D replica models are generated with the Solid Geometry Library toolbox, which enables the automated design of 3D surface models in matlab. The mathematical background of this procedure is described. The capability of the proposed method is demonstrated on 3D structures composed of components with fixed cross sections and fundamental 3D structures such as prisms, cylinders, and spheres. Further improvements in the coordinate measurement process using AprilTag and further implementation in matlab can extend the functionality for the digitization of more complex 3D structures.