Concatenate feature extraction for robust 3D elliptic object localization

ABSTRACT

Developing an efficient object localization system for complicated industrial objects is an important, yet difficult robotic task. To tackle this problem, we have developed a system consisting first of a vision model acquisition editor, where the object salient features are acquired through a human-in-the-loop approach. Subsequently, two feature extraction algorithms, region-growing and edge-grouping, are applied to the object scene. Finally, by Kalman filter estimation of a proper ellipse representation, our object localization system successfully generates ellipse hypotheses by grouping edge fragments in the scene. The proposed system is validated by experiments using actual industrial objects.

REFERENCES

• 1: R C Bolles , P Horaud, 3DPO: A three-dimensional part orientation system, International Journal of Robotics Research, v.5 n.3, p.3-26, Fall 1986 [doi>10.1177/027836498600500301]
• 2: Chen, C. H. and Kak, A. C. Robot vision system for recognizing 3D objeccts in low-order polynomial time Systems, IEEE Transactions on Man and Cybernetics, Vol. 19, No. 6, 1989. 1535--1563.
• 3: Chatterjee, C. and Chong, E. K. P. Efficient algorithms for finding the centers of conics and quadrics in noisy data, Pattern Recognition, Vol. 30, No. 5, 1997, 673--684.
• 4: J Canny, A computational approach to edge detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.8 n.6, p.679-698, Nov. 1986 [doi>10.1109/TPAMI.1986.4767851]
• 5: Olivier Faugeras, Three-dimensional computer vision: a geometric viewpoint, MIT Press, Cambridge, MA, 1993
• 6: Andrew Fitzgibbon , Maurizio Pilu , Robert B. Fisher, Direct Least Square Fitting of Ellipses, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.21 n.5, p.476-480, May 1999 [doi>10.1109/34.765658]
• 7: Frédéric Guichard Jean-Philippe Tarel, Curve Finder Combining Perceptual Grouping and a Kalman Like Fitting, Proceedings of the International Conference on Computer Vision-Volume 2, p.1003, September 20-25, 1999
• 8: Steven L. Horowitz , Theodosios Pavlidis, Picture Segmentation by a Tree Traversal Algorithm, Journal of the ACM (JACM), v.23 n.2, p.368-388, April 1976 [doi>10.1145/321941.321956]
• 9: Hough, P. V. C. Method and means for recognizing complex patterns, US Patent, No. 3 069 654, 1962.
• 10: Daniel P. Huttenlocher , Peter C. Wayner, Finding convex edge groupings in an image, International Journal of Computer Vision, v.8 n.1, p.7-27, July 1992 [doi>10.1007/BF00126398]
• 11: David W. Jacobs, Robust and Efficient Detection of Salient Convex Groups, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.18 n.1, p.23-37, January 1996 [doi>10.1109/34.476008]
• 12: Akio Kosaka , Avinash C. Kak, Fast vision-guided mobile robot navigation using model-based reasoning and prediction of uncertainties, CVGIP: Image Understanding, v.56 n.3, p.271-329, Nov. 1992 [doi>10.1016/1049-9660(92)90045-5]
• 13: Kosaka A. and Kak, A. C. Stereo vision for industrial applications, Handbook of Industrial Robotics, edit. S. Y. Nof, John Wiley & Sons, Inc., 1999. 269--294.
• 14: Motai, Y. and Kak, A.C. An interactive framework for acquiring vision models of 3D objects, IEEE Trans. of Systems, Man, and Cybernetics - Part B: Cybernetics, in press, 2003.
• 15: Motai, Y. and Kosaka, A. SmartView: Hand-eye robotic calibration for active viewpoint generation and object grasping, Proc. of IEEE Int. Conference of Robotics and Automation, 2001. 2183--2190.
• 16: Olson, C. F. Improving the generalized Hough transform through imperfect grouping, Image and Vision Computing, Vol. 16, 1998. 627--634.
• 17: J. Porrill, Fitting ellipses and predicting confidence envelopes using a bias corrected Kalman filter, Image and Vision Computing, v.8 n.1, p.37-41, February 1990 [doi>10.1016/0262-8856(90)90054-9]
• 18: Klaus Voss , Herbert Suesse, Invariant Fitting of Planar Objects by Primitives, IEEE Transactions on Pattern Analysis and Machine Intelligence, v.19 n.1, p.80-84, January 1997 [doi>10.1109/34.566815]
• 19: Yoshimi, B. H. and Allen, P. K. Active uncalibrated visual servoing, Proceeding of IEEE International Conference on Robotics and Automation, Vol. 1, 1994. 156--161.

INDEX TERMS

General Terms:
Algorithms, Experimentation, Human Factors, Measurement, Performance, Reliability

Keywords:
3D robot vision system, Kalman filter estimation, elliptic edge grouping, human-in-the-loop segmentation, salient feature extraction