Open Access Open Access  Restricted Access Subscription or Fee Access

Recognition of Occluded Objects Using Wavelet Transform

V. Subbaroyan, Dr.S. Karthik

Abstract


This paper presents an approach for the recognition of multiple view objects based on wavelet transform, even when come portion of the object is occluded. Object recognition is achieved by extracting the energy computed from all the sub-bands of Discrete Wavelet Transform (DWT) fused with the colour moments. Therefore the proposed method considers not only the frequency relationship but also the spatial relationship of pixels in the objects. The extracted features are used as an input to the K Nearest Neighbor (K-NN) for classification. The evaluation of the system is carried on using COIL database and the performance of the proposed system is studied by varying the training set sizes. The experiment was extended further to cover objects that were partly occluded. Experimental results show that the proposed method produces more accurate classification rate, even with partial occlusion.

Keywords


Wavelet Transform, Colour Moments, KNN Classifier, Object Recognition, Occlusion

Full Text:

PDF

References


Bicchi A. and Kumar V., “Robotic grasping and contact: A review,” in Proceedings of the IEEE International Conference on Robotics and Automation, ICRA’00, 2000, pp. 348–353.

Nelson R. and Selinger A., “A cubist approach to object recognition,” in ICCV’98, 1998, pp.614–621.

Bjorkman M. and Kragic D., “Combination of foveal and peripheral vision for object recognition and pose estimation,” Proceedings. IEEE International Conference on Robotics and Automation, ICRA’04, vol. 5,pp. 5135 – 5140, 2004.

Kaiser M. and Dillman R., “Building elementary robot skills from human demonstration,” Proceedings of the IEEE International Conference on Robotics and Automation, v. 3, pp. 2700–2705, 1996.

Chen J. and Zelinsky A., “Programming by demonstration: removing suboptimal actions in a partially known configuration space,”Proceedings of the IEEE Intl. Conf. on Robotics and Automation (ICRA ’01), vol. 4, pp. 4096–4103, 2001.

Ekvall S. and Kragic D., “Interactive grasp learning based on human demonstration,” in Proc. IEEE/RSJ International Conference on Robotics

Petersson L., Jensfelt P., Tell D., Strandberg M., Kragic D., and Christensen H. I., “Systems integration for real-world manipulation tasks,”in IEEE International Conference on Robotics and Automation, ICRA 2002, vol. 3, 2002, pp. 2500 – 2505.

Chaumette F., “Image moments: a general and useful set of features for visual servoing,” IEEE Trans. on Robotics, vol. 20 (4), 2004.

Taylor G. and Kleeman L., “Grasping unknown objects with a humanoid robot,” Australiasian Conference on Robotics and Automation, 2002.

Ekvall F. H. S. and Kragic D., “Object recognition and pose estimation for robotic manipulation using color co-occurrence histograms,” in Proc. IEEE/RSJ International Conference Intelligent Robots and Systems, IROS’ 03, 2003.

Selinger A. and Nelson R, “A perceptual grouping hierarchy for appearance - based 3d object recognition,” CVIU, vol. 76, no. 1, pp. 83– 92, October 1999.

http://www.nada.kth.se/~caputo/cognitive-nipsw03.html.

Arbter K., Snyder W. E., Burkhardt H., and Hirzinger G., “Application of affine-invariant Fourier descriptors to recognition of 3-d objects.” IEEE Trans. Pattern Anal. Mach. Intell., vol.12, no. 7, pp. 640–647, 1990.

Singh R. and Papanikolopoulos N., “Planar shape recognition by shape morphing.” Pattern Recognition, vol. 33, no. 10, pp. 1683–1699, 2000.

Gdalyahu Y. and Weinshall D., “Flexible syntactic matching of curves and its application to automatic hierarchical classification of silhouettes.” IEEE Trans. Pattern Anal. Mach. Intell., vol. 21, no.12, pp. 1312–1328, 1999.

Christopoulos Vassilios N. and Schrater Paul,” Handling shape and contact location uncertainty in grasping two-dimensional planar objects” Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, San Diego, CA, USA, Oct 29 - Nov 2, 2007.


Refbacks

  • There are currently no refbacks.