Article details

Title: A New Invariant Set for Pattern Recognition
Author(s): Constantin-Iulian Vizitiu               

Abstract: The choice of an appropriate feature extraction method is essential for the success of the following recognition process. This paper proposes a design method for a new pattern descriptor set belonging to the Flusser moments family, which is invariant to the elementary geometric transformations and, respectively, has an increased noise robustness. Experimental results based on a real video image database confirm the basic properties of this new descriptor set.

Keywords: feature extraction, Flusser moments, pattern recognition, neural networks.

[1] A. ABO-ZAID, O.R. HINTON, E. HORNE – About Moment Normalization and Complex Moment Descriptors, In J. Kittler (Ed.), “Pattern Recognition: 4th International Conference, Cambridge, UK, March 28-30, 1988: Proceedings”, pp. 399-409, Springer-Verlag, Berlin, Germany, 1988
[2] J. FLUSSER – On the Independence of Rotation Moment Invariants, Pattern Recognition, Vol. 33, No. 9, pp. 1405-1410, Sep. 2000
[3] B.D. RIPLEY – Pattern Recognition and Neural Networks, Cambridge University Press, Cambridge, UK, 1996
[4] M.A. RODRIGUES – Invariants for Pattern Recognition and Classification, World Scientific Publishing, Danvers, MA, 2000
[5] C.I. VIZITIU – Neural Networks Used in Pattern Recognition, Military Technical Academy Publishing House, Bucharest, Romania, 2002 (in Romanian)
[6] C.I. VIZITIU –A New Modality to Increase the Video Pattern Recognition Performances, Proc. of the 4th International Symposium on Defense Technology, pp. 416-443, ZMNDU, Budapest, Hungary, Apr. 19-20, 2006