A Novel Method for Palmprint Feature Extraction Based on Modified Pulse-Coupled Neural Network
Most of these methods succeeded to achieve the invariance against object translation, rotation and scaling, but that could not neutralize the bright background effect and non-uniform light on the quality of the generated features. To eliminate the limit that the recent subspace learning methods for facial feature extraction are sensitive to the variations of orientation, position and illumination in capturing palmprint images, a novel palmprint feature extraction approach is proposed. Palmprint images are decomposed into a sequence of binary images using a Modified Pulse-Coupled Neural Network (M-PCNN), and then the information entropy of each binary image are calculated and regarded as features. A classifier based Support Vector Machine (SVM) is employed to implement recognition and classification. Simultaneously, it overcomes the disadvantage of standard PCNN model with high number of parameters. Theoretical and experimental results show that the proposed approach is robust to the variations of orientation, position and illumination conditions in comparison with other methods based subspace.
KeywordsPulse-Coupled Neural Network (PCNN) Palmprint recognition Feature extraction Information entropy Support vector machine (SVM)
Unable to display preview. Download preview PDF.
- 2.Yuan, W.Q., Gu, Z.H.: Research on feature selection method based on five classes of palmprint principal lines on the whole palm. Chinese Journal of Scientific Instrument 33, 942–948 (2012) (in Chinese) Google Scholar
- 4.Sun, Z., Tan, T., Wang, Y., Li, S.Z.: Ordinal palmprint represention for personal identification. In: CVPR 2005, pp. 279–284. IEEE Press, New York (2005)Google Scholar
- 6.Rava, T.H., Bettaiah, V., Ranganath, H.S.: Adaptive pulse coupled neural network parameters for image segmentation. World Acad. Sci. Eng. Technol. 73, 1046–1052 (2011)Google Scholar
- 7.Wei, S., Hong, Q., Hou, M.: Automatic image segmentation based on PCNN with adaptive threshold time constant. Neurocomputing, 1485–1491 (2011)Google Scholar
- 8.Chen, Y., Park, S.K., Ma, Y., Ala, R.: A new automatic parameters setting method of a simplified PCNN for image segmentation. IEEE Trans. Neural Network, 880–892 (2011)Google Scholar
- 9.Fu, J.C., Chen, C.C., Chai, J.W., Wong, S.T.C., Li, I.: Image segmentation by EM-based adaptive pulse coupled neural networks in brain magnetic resonance imaging. In: Wong, S. (ed.) Computerized Medical Imaging and Graphics, pp. 308–320. Elsevier (2010)Google Scholar
- 10.Forgac, R., Mokris, I.: Feature generation improving by optimized PCNN. In: 6th International Symposium on SAMI 2008, pp. 203–207. IEEE Press, New York (2008)Google Scholar
- 11.Tolba, M.F., Abdellwahab, M.S., Aboul-Ela, M., Samir, A.: Image signature improving by PCNN for Arabic sign language recognition. Machine Learning & Pattern Recognition, 1–6 (2010)Google Scholar
- 12.Weili, S., Yu, M., Zhanfang, C., Hongbiao, Z.: Research of automatic medical image segmentation based on Tsallis entropy and improved PCNN. In: Fukuda, T., et al. (eds.) ICMA 2009, pp. 1004–1008. IEEE Press, New York (2009)Google Scholar
- 13.Yide, M., Lian, L., Kun, D.: Pulse coupled neural network with digital image processing, pp. 225–237. Science Press (2006)Google Scholar