CAD-Based Photogrammetry for Reverse Engineering of Industrial Installations

When designing an industrial installation, construction engineers often make use of a library of stan- dardized CAD components. For instance, in the case of a servicing plant, such a library contains descriptions of simple components such as straight pipes, elbows, and T-junctions. A new installation is constructed by select- ing and connecting the appropriate components from the library. This article demonstrates that one can use the same approach for reverse engineering by photogramme- try. In our technique, the operator interprets images and selects the appropriate CAD component from a library. By aligning the edges of the component's wire frame to the visible edges in the images, we implicitly determine the position, orientation, and shape of the real compo- nent. For a fast object reconstruction the alignment pro- cess has been split in two parts. Initially, the operator ap- proximately aligns a component to the images. In a second step a fitting algorithm is invoked for an automatic and precise alignment. Further improvement in the efficiency of the reconstruction is obtained by imposing geometric constraints on the CAD components of adjacent object parts. the as-built situation has to be reconstructed from mea- surements because the plant design maps are no longer available, not up to date, or not accurate enough. Tradi- tionally, these measurements are done by tacheometry or classical photogrammetry (Cooper and Robson 1996). In both cases this requires the application of a large num-

[1]  Linda G. Shapiro,et al.  Computer and Robot Vision , 1991 .

[2]  Thomas C. Henderson,et al.  Feature-based reverse engineering of mechanical parts , 1999, IEEE Trans. Robotics Autom..

[3]  Stuart Robson,et al.  Chapter 2: Theory of close range photogrammetry , 1996 .

[4]  Paul Debevec,et al.  Modeling and Rendering Architecture from Photographs , 1996, SIGGRAPH 1996.

[5]  Robert B. Fisher,et al.  Object reconstruction by incorporating geometric constraints in reverse engineering , 1999, Comput. Aided Des..

[6]  Alyn P. Rockwood,et al.  Three-dimensional object reconstruction from two-dimensional images , 1997, Comput. Aided Des..

[7]  Ralph R. Martin,et al.  Reverse engineering of geometric models - an introduction , 1997, Comput. Aided Des..

[8]  George Vosselman,et al.  Mapping by Dragging and Fitting of Wire-Frame Models , 1999 .

[9]  Jitendra Malik,et al.  Modeling and Rendering Architecture from Photographs: A hybrid geometry- and image-based approach , 1996, SIGGRAPH.

[10]  Seokbae Son,et al.  Automated laser scanning system for reverse engineering and inspection , 2002 .

[11]  Sabry F. El-Hakim,et al.  SHORT-BASELINE ACTIVE TRIANGULATION FOR CAD RECONSTRUCTION OF ROOM- SIZED INDUSTRIAL ENVIRONMENTS , 2000 .

[12]  M. J. G. M. van Emmerik A Direct Manipulation Technique for Specifying 3D Object Transformations with a 2D Input Device , 1990, Comput. Graph. Forum.

[13]  Pascal Fua,et al.  Model-Based Optimization: Accurate and Consistent Site Modeling , 1996 .

[14]  Michael E. Mortenson Geometric modeling (2nd ed.) , 1997 .

[15]  Fiona Lang,et al.  3D-city modeling with a digital one-eye stereo system , 1996 .

[16]  David G. Lowe,et al.  Fitting Parameterized Three-Dimensional Models to Images , 1991, IEEE Trans. Pattern Anal. Mach. Intell..