Deformation visual inspection of industrial parts with image sequence

Abstract.A new approach to reconstructing and inspecting the deformation of industrial parts, especially sheetmetal parts based on CAD-designed data and hybrid point-line bundle adjustment with image sequence, is proposed. Nonmetric image sequence and CAD-designed data of parts are used as the source of information. The strategy of our approach is to reconstruct and inspect deformations of parts automatically with image points and line segments extracted from the images. Basic error equation of line photogrammetry and its modified form are addressed in detail. It is shown that when a certain proper weight is selected, adjustment by condition equations and adjustment by observation equations are equivalent for line photogrammetry. A novel hybrid point-line bundle adjustment algorithm is used to reconstruct industrial parts. The proposed hybrid adjustment model can be used in various 3D reconstruction applications of objects mainly composed of points and lines. The developed inspection system is tested with true image data acquired by a CCD camera, and the results are very satisfying.

[1]  Yongjun Zhang,et al.  MULTI-VIEW 3 D CITY MODEL GENERATION WITH IMAGE SEQUENCES , 2003 .

[2]  A. Gruen.  ADAPTIVE LEAST SQUARES CORRELATION: A POWERFUL IMAGE MATCHING TECHNIQUE , 1985 .

[3]  George Vosselman.  Semi-automatic CAD based Reconstruction of Industrial Installations , 2001, Künstliche Intell..

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

[5]  M. Kasser,et al.  Digital photogrammetry , 2001 .

[6]  Robert Sablatnig,et al.  Flexible automatic visual inspection based on the separation of detection and analysis , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[7]  Theodora Varvarigou,et al.  Automated inspection of gaps on the automobile production line through stereo vision and specular reflection , 2001, Comput. Ind..

[8]  Jianqing Zhang,et al.  Camera calibration technique with planar scenes , 2003, IS&T/SPIE Electronic Imaging.

[9]  D. A. Tait,et al.  Book Review: Photogrammetry, vol. 1. Fundamentals and standard processes. Karl Kraus. Ferdinand Dümmels Verlag, Bonn, 1993, 402 pp., DM 58.00, ISBN 3-427-78681-1 , 1994 .

[10]  Wilfried Linder,et al.  Digital Photogrammetry , 2003 .

[11]  Urs Hirschberg,et al.  INTEGRATION OF DIGITAL PHOTOGRAMMETRY AND CAAD: CONSTRAINT-BASED MODELLING AND SEMI-AUTOMATIC MEASUREMENT , 1995 .

[12]  Cihan H. Dagli,et al.  Automatic PCB Inspection Algorithms: A Survey , 1996, Comput. Vis. Image Underst..

[13]  Cordelia Schmid,et al.  AUTOMATIC LINE MATCHING AND 3D RECONSTRUCTION OF BUILDINGS FROM MULTIPLE VIEWS , 1999 .

[14]  Frank A. van den Heuvel,et al.  Line-photogrammetric mathematical model for the reconstruction of polyhedral objects , 1998 .