Face recognition using DCT coefficients selection

ABSTRACT

This paper presents a face recognition method based on Discrete Cosine Transform (DCT) coefficient's selection. Without a normalization phase, the proposed method uses, in its feature selection stage, a technique based only on the DCT coefficients amplitudes. Three coefficient selection criterions were analyzed: the first one is the average of the coefficients' amplitudes; the second one is based on counting the occurrence of each coefficient, which are stored in a set of lists containing the most significant coefficients; finally, the third criterion is based on the average position of the coefficients in a list of coefficients ordered by amplitude. Experimental tests on the ORL Face Database [1] achieved 99.00% of recognition accuracy using only 50 DCT coefficients, with low computational cost. Additionally, the method achieved 100.00% of recognition accuracy when the correct face is within a range of eight returned faces.

References

1. AT&T Laboratories, Cambridge, UK. "The ORL Database of Faces" (now AT&T "The Database of Faces"). Available {Online}: http://www.cl.cam.ac.uk/Research/DTG/attarchive/pub/data/att_faces.zip {15/September/2007}.
2. Batista, L. V., Carvalho, L. C. and Melcher, E. U. K. Compression of ECG Signals Base on Optimum Quantization of Discrete Cosine Transform Coefficients and Golomb-Rice Coding. In Proceeding of the 25th Annual International Conference on the IEEE EMBS. Mexico, Sep. 17--21, 2003.
3. Bicego, M., Castellani, U. and Murino V. Using HMM and Wavelets for Face Recognition. Proceedings of the 12th International Conference on Image Analysis and Processing. 2003 IEEE.
4. Chai, Douglas and Wong, Kok Wai. Facial Image Processing: An Overview. Proceeding of the 2004 IEEE Conference on Cybernetics and Intelligent Systems, p. 307--311. Singapore, 1--3 December, 2004.
5. Duda, R. O.; Hart, P. E. and Stork, D. G. Pattern Classification. Second Edition. Wiley-Interscience. 2000.
6. Fáundez-Zanuy, Marcos. Face Recognition in a Transformed Domain. In Proceedings of IEEE 37th Annual 2003 International Carnahan Conference on Security Technology, 2003, p. 290--297.
7. Hafed, Ziad M and Levine, Marin D. Face Recognition Using Discrete Cosine Transform. International Journal of Computer Vision, v. 43(3), p. 167--188. 2001.
8. Haykin, Simon. Redes Neurais - Princípios e práticas. Tradução de 2a Edição. Bookman. 2001.
9. Ivancevic, V.; Kaine, A. K.; McLindin, B. A. and Sunde, J. Factor Analysis of Essencial Facial Features. 25th Int. Conf. Information Technology Interface ITI 2003, p. 187--191. June 16--19, Croatia. 2003.
10. Podilchuk, C. and Zhang, X. 1996. Face recognition using DCT-based feature vectors. In Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE international Conference - Volume 04 (May 07 - 10, 1996). ICASSP. IEEE Computer Society, Washington, DC, 2144--2147.
11. Rao, K. R. and Yip, P. Discrete Cosine Transform: Algorithms, Advantages, Applications. Academic Press, Inc. 1990.
12. Samra, Ahmed Shabann; Allah, Salah. E. T. G and Ibrahim, Rehab Mahmound. Face Recognition Using Wavelet Transform, Fast Fourier Transform and Discrete Cosine Transform. In Proceedings of the 46th IEEE International Midwest Symposium on Circuits and Systems, 2003 {MWSCAS '03}, v. 1, p. 272--275.
13. Zhao, W.; Chellappa, R.; Phillips, P. J. and Rosenfeld, A. Face Recognition: A Literature Survey. ACM Computing Surveys, v. 35, n. 4, p. 399--458, 2003.

Index Terms

• Computing methodologies

  • Artificial intelligence

    • Computer vision

      • Computer vision problems

        • Object recognition

  • Machine learning

    • Learning paradigms

      • Supervised learning

        • Supervised learning by classification

    • Machine learning approaches

      • Classification and regression trees