CAD-based object reconstruction using line photogrammetry for direct interaction between GEMS and a vision system

The paper presents an algorithm for reconstructing Geometric Element Modeli ng System (GEMS)-based industrial objects using line photogrammetry (LP) in order that there can be direct interaction between GEMS and a vision syste m. In the GEMS system, each object is constructed by Boolean set operators of constant solid geometry (CSG) primitives, each face of which is describe d by a Boundary presentation (B-pre); and straight lines, curves, planes, a nd curved surfaces are described by parametric equations. Based on this typ e of data structure and representation, the geometric elements which descri be industrial objects in GEMS, are taken as unknowns, and are directly solv ed for by matching images (2D) and objects (3D). Compared to traditional ph otogrammetric DEM products, the advantage of this algorithm is that the pro duct of reconstruction is parameters describing the industrial object, whic h can be directly fed back to the GEMS system. It is convenient to communic ate between GEMS and the vision system without any transformation. On the o ther hand, the developed mathematical model can be used not only for line-t o-line correspondences but also for line-to-surface or surface-to-surface c orrespondences. A number of simulations and practical experiments demonstra ted that this algorithm is feasible and reliable in industrial computer vis ion.