3-D imaging and modeling – Part II: integration and duplication

Summary One of the key modules in a typical reverse engineering system is the integration of multiple range images of three-dimensional objects with complex geometry. In the Part I of this article [1], we have presented a self- made optical digitizer that is employed for quickly acquiring range data of the objects from multiple views, and introduced a visualized registration procedure that can provide a reliable estimate for the fine registration of multiple range images in a unified coordinate system. Here we propose a new approach for merging multiple range images to build up a complete geometric model of the objects in 3-D space. This approach is based on the segmentation of a pair of range images using the concepts of “double stitching lines”, as well as a constrained local triangulation process. Experiment results are presented to show the effectiveness of this approach in 3-D imaging and modeling for the applications of reverse engineering.