New Pattern Recognition Method Based on Wavelet De-Noising and Kernel Principal Component Analysis


Article Preview

The background noise influences the face image recognition greatly. It is crucial to remove the noise signals prior to the face image recognition processing. For this purpose, the wavelet de-noising technology has combined with the kernel principal component analysis (KPCA) to identify face images in this paper. The wavelet de-noising technology was firstly used to remove the noise signals. Then the KPCA was employed to extract useful principal components for the face image recognition. By doing so, the dimensionality of the feature space can be reduced effectively and hence the performance of the face image recognition can be enhanced. Lastly, a support vector machine (SVM) classifier was used to recognize the face images. Experimental tests have been conducted to validate and evaluate the proposed method for the face image recognition. The analysis results have showed high performance of the newly proposed method for face image identification.



Edited by:

Yuning Zhong




J. J. Zhang and L. J. Liang, "New Pattern Recognition Method Based on Wavelet De-Noising and Kernel Principal Component Analysis", Applied Mechanics and Materials, Vol. 235, pp. 74-78, 2012

Online since:

November 2012




[1] Z. Li, X. Yan, Y. Jiang, L. Qin, J. Wu, A new data mining approach for gear crack level identification based on manifold learning, Mechanika, 18 (2012) 29–34.


[2] D. Donoho, De-noising by soft-thresholding, IEEE Trans. Information Theory, 41 (1995) 613–27.


[3] Z. Li, X. Yan, C. Yuan, Z. Peng, A new intelligent fusion method of multi-dimensional sensors and its application to tribo-system fault diagnosis of marine diesel engines, Tribology Letters, 47 (2012) 1–15.


[4] Z. Li, X. Yan, C. Yuan, Z. Peng, Intelligent fault diagnosis method for marine diesel engines using instantaneous angular speed, Journal of Mechanical Science and Technology, 26 (2012) 1–11.


[5] Z. Li, X. Yan, Z. Tain, C. Yuan, Z. Peng, Blind vibration component separation and nonlinear feature extraction applied to the nonstationary vibration signals for the gearbox multi-fault diagnosis, Measurement, 2012, http: /dx. doi. org/10. 1016/j. measurement. 2012. 06. 013.


[6] Z. Li, X. Yan, C. Yuan, Z. Peng, L. L, Virtual prototype and experimental research on gear multi-faults diagnosis using wavelet-autoregressive model and principal component analysis method, Mechanical Systems and Signal Processing, 25 (2011).


[7] J. Yang, Z. Jin, J. Yang, D. Zhang, A. Frangi, Essence of Kernel Fisher Discriminant: KPCA Plus LDA, Pattern Recognition, 37 (2004) 2097–2100.