Sage Journals: Discover world-class research

Abstract

Achieving inspection automation is desirable in a production line in order to reduce labour cost and ensure consistent product quality. This paper presents an automated inspection system for a three-dimensional complex surface according to the specification of surface profile tolerancing. A range scanner is used to measure the surface profile of an object and an algorithm is designed to decide upon the acceptability of the part. The pyramidal generalized Hough transform is used to obtain a rough estimate of the object's pose in the three-dimensional world. This pose estimate is then refined over a linearized optimization process to achieve an optimal solution from which shape acceptability can be determined on the basis of a statistical interpretation. Experiments showed that the current inspection system is capable of registering objects having arbitrary deformations and that the system is capable of inspecting objects of complex shape.

Keywords

visual inspection surface profile tolerance range image registration Hough transform Poisson distribution

References

Bolles

R. C.

Horaud

3DPO: A three-dimensional part orientation system. Int. J. Robotics Res., 1986, 5(3), 3–26.

Krishnapuram

Casasent

Determination of three-dimensional object location and orientation from range images. IEEE Trans. Pattern Analysis and Mach. Intell., November 1989, 11(11), 1158–1167.

Park

H. D.

Mitchell

O. R.

CAD based planning and execution of inspection. IEEE Computer Vision and Pattern Recognition, 1988, 858–863.

Besl

P. J.

McKay

N. D.

A method for registration of 3-D shapes. IEEE Trans. Pattern Analysis and Mach. Intell., February 1992, 14(2), 239–256.

Gunnarsson

K. T.

Prinz

F. B.

CAD model-based localization of parts in manufacturing. IEEE Computer, August 1987, 66–74.

Menq

C. H.

Yau

H. T.

Lai

G. Y.

Automated precision measurement of surface profile in CAD-directed inspection. IEEE Trans. Robotics and Automn, April 1992, 8(2), 268–278.

Merriman

A Text-Book on the Method of Least Squares, 8th edition, 1909, Ch. III (John Wiley, New York).

Australian Standard, Technical Drawing, Part 101: General Principles, AS 1100. 101- 1992, Standards Australia, New South Wales, 1992.

Arman

Aggarwal

J. K.

Model-based object recognition in dense-range images–a review. ACM Computing Surveys, March 1993, 25(1), 5–43.

10.

Besl

P. J.

Jain

R. C.

Three-dimensional object recognition. ACM Computing Surveys, 1985, 17, 75–145.

11.

Pope

A. R.

Model-based object recognition–a survey of recent research. Technical Report 94–04, January 1994.

12.

Lau

K. H.

Automated Surface Profile Inspection. MEngSc thesis, Monash University, Clayton, Victoria, Australia, August 1997.

13.

Edelsbrunner

Algorithms in Combinatorial Geometry, 1987, Ch. 1 (Springer-Verlag, Berlin).

14.

Breuel

T. M.

Fast recognition using adaptive subdivisions of transformation space. IEEE Computer Vision and Pattern Recognition, 1992, 445–451.

15.

Encyclopedia of Artificial Intelligence (Ed. Shapiro

S.C.

), 2nd edition, 1992, Vol. 2, pp. 1468–1472 (John Wiley, New York).

16.

Lavin

M. A.

LeMaster

R. J.

Fast Hough transform. In Proceedings of 3rd International Workshop on Computer Vision: Representation and Control, October 1985, pp. 75–83.

17.

Conover

W. J.

Practical Nonparametric Statistics, 2nd edition, 1980, pp. 117–121 (John Wiley, New York).

18.

3D Shape Measurement System User's Manual–Incorporating SHAPEâ© Software User's Manual, Monash University, Clayton, Victoria, Australia, 1989.

19.

Metricor 3D Precision Shape Measurement System (ver. 1.0) User Manual, Montech Pty Limited, Melbourne, Victoria, Australia, 1996.

Automated visual inspection for surface profile tolerancing

Abstract

Abstract

Keywords

References