Abstract
In the analysis and processing of image recognition, extracting useful and valuable data from the original dataset has become a problem. Since the data to be processed often presents high dimensional and nonlinear feature, reasonable dimensionality reduction is an necessary method for improving the accuracy of data analysis. One of the dimensionality reduction methods Kernel Principal Component Analysis (KPCA) has certain advantages in dealing with nonlinear data, but it also has defects when facing the dataset in which data owe highly complex relationship. The other linear dimensional reduction method Linear Discriminant Analysis (LDA) has supervisory characteristics, which can reduce the dimensionality of dataset in which data owe highly complex relationship. However it can only handle linear data. So we propose a hybrid method which is the combination of the above two methods called KPCA-LDA. By it the new dataset obtained by the step of dimensionality reduction is beneficial to be processed in next step for classification. We combine KPCA-LDA with the Back Propagation Neural Network (BPNN) method to achieve the classification of handwritten numbers. The experimental results show that the classification accuracy of the proposed KPCA-LDA-BPNN model can reach 98.67%, which is about 3%-5% higher than the original method using K Nearest Neighbor (KNN) and Support Vector Machine (SVM).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Huang, D., Zhang, D., Liu, Y., Zhang, S., Zhu, W.: A KPCA based fault detection approach for feed water treatment process of coal-fired power plant. In: Proceeding of the 11th World Congress on Intelligent Control and Automation, pp. 3222–3227. IEEE, Shenyang (2014). https://doi.org/10.1109/WCICA.2014.7053247
Wang, X., Huang, L., Zhang, Y.: Modeling and monitoring of nonlinear multi-mode processes based on similarity measure-KPCA. J. Central S. Univ. 24(3), 665–674 (2017). https://doi.org/10.1007/s11771-017-3467-z
Ghassabeh, Y.A., Rudzicz, F., Moghaddam, H.A.: Fast incremental LDA feature extraction. Pattern Recogn. 48(6), 1999–2012 (2015). https://doi.org/10.1016/j.patcog.2014.12.012
Fu, R., Qin, B., Liu, T.: Open-categorical text classification based on multi-LDA models. Soft. Comput. 19(1), 29–38 (2015)
Yin, S., Jing, C., Hou, J., Kaynak, O., Gao, H.: PCA and KPCA integrated support vector machine for multi-fault classification. In: IECON 2016–42nd Annual Conference of the IEEE Industrial Electronics Society, pp. 7215–7220. IEEE, Florence (2016). https://doi.org/10.1109/IECON.2016.7793188
Ibrahim, A.M., Baharudin, B.: Classification of mammogram images using shearlet transform and kernel principal component analysis. In: 2016 3rd International Conference on Computer and Information Sciences (ICCOINS), pp. 340–344. IEEE, Kuala Lumpur (2016). https://doi.org/10.1109/ICCOINS.2016.7783238
Chen, X., Wang, S., Ruan, X.: Recognition of partially occluded face by error detection with logarithmic operator and KPCA. In: 2016 9th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)), pp. 460–464. IEEEE, Datong (2016). https://doi.org/10.1109/CISP-BMEI.2016.7852755
Jha, S.K., Josheski, F., Marina, N., Hayashi, K.: GC-MS characterization of body odour for identification using artificial neural network classifiers fusion. Int. J. Mass Spectrom. 406, 35–47 (2016). https://doi.org/10.1016/j.ijms.2016.06.002
Qian, X., Chen, J.-P., Xiang, L.-J., Zhang, W., Niu, C.-C.: A novel hybrid KPCA and SVM with PSO model for identifying debris flow hazard degree: a case study in Southwest China. Environ. Earth Sci. 75(11), 991 (2016). https://doi.org/10.1007/s12665-016-5774-3
Alam, S., Kwon, G.-R.: The Alzheimer’s disease neuroimaging initiative: Alzheimer disease classification using KPCA, LDA, and multi-kernel learning SVM. Int. J. Imaging Syst. Technol. 27(2), 133–143 (2017). https://doi.org/10.1002/ima.22217
Mei, C., Yang, M., Shu, D., Hui, J., Liu, G.: Monitoring wheat straw fermentation process using electronic nose with pattern recognition methods. Anal. Methods 7(13), 6006–6011 (2015)
Urmila, K., Chen, Q., Li, H., Zhao, J., Hui, Z.: Quantifying of total volatile basic nitrogen (TVB-N) content in chicken using a colorimetric sensor array and nonlinear regression tool. Anal. Methods 7, 5682–5688 (2015)
Mika, S., Ratsch, G., Weston, J., Scholkopf, B., Mullers, K.R.: Fisher discriminant analysis with kernels. In: Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No. 98TH8468), pp. 41–48. IEEE, Madison (1999). https://doi.org/10.1109/NNSP.1999.788121
Deng, M., Chen, X., Chen, T.X., Wang, H.R., Lu, H.-X.: Improved kernel principal component analysis based on a clustering algorithm. CAAI Trans. Intell. Syst. (2010)
Han, Z.S., Li, Y., Zhang, Y.N.: A comparative study on face recognition using LDA-based algorithm. Microelectron. Comput. (2005)
Shermina, J.: Illumination invariant face recognition using discrete cosine transform and principal component analysis. In: 2011 International Conference on Emerging Trends in Electrical and Computer Technology, Nagercoil, pp. 826–830. IEEE, Nagercoil (2011). https://doi.org/10.1109/ICETECT.2011.5760233
Xie, Y.L.: LDA algorithm and its application to face recognition. Comput. Eng. Appl. (2010)
Jenssen, R.: Kernel entropy component analysis. IEEE Trans. Pattern Anal. Mach. Intell. 32(5), 847–860 (2010). https://doi.org/10.1109/TPAMI.2009.100
El Haimoudi, K., Issati, I., Daanoun, A.: The particularities of the counter propagation neural network application in pattern recognition tasks. In: Ezziyyani, M., Bahaj, M., Khoukhi, F. (eds.) AIT2S 2017. LNNS, vol. 25, pp. 474–487. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-69137-4_42
Jiang, P., Ge, Y., Wang, C.: Research and application of a hybrid forecasting model based on simulated annealing algorithm: a case study of wind speed forecasting. J. Renew. Sustain. Energy 8(1), 015501 (2016). https://doi.org/10.1063/1.4940408
Ashiquzzaman, A., Tushar, A.K.: Handwritten Arabic numeral recognition using deep learning neural networks. In: 2017 IEEE International Conference on Imaging, Vision Pattern Recognition (icIVPR), pp. 1–4. IEEE, Dhaka (2017). https://doi.org/10.1109/ICIVPR.2017.7890866
Acknowledgement
The research is supported by Natural Science Foundation of China under Grant No. 61662054, 61262082, Inner Mongolia Science and Technology Innovation Team of Cloud Computing and Software Engineering and Inner Mongolia Application Technology Research and Development Funding Project “Mutual Creation Service Platform Research and Development Based on Service Optimizing and Operation Integrating” under Grant 201702168, Inner Mongolia Engineering Lab of Cloud Computing and Service Software and Inner Mongolia Engineering Lab of Big Data Analysis Technology.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zhao, Z., Zhou, J., Xing, H. (2019). A High Accuracy Nonlinear Dimensionality Reduction Optimization Method. In: Sun, Y., Lu, T., Yu, Z., Fan, H., Gao, L. (eds) Computer Supported Cooperative Work and Social Computing. ChineseCSCW 2019. Communications in Computer and Information Science, vol 1042. Springer, Singapore. https://doi.org/10.1007/978-981-15-1377-0_55
Download citation
DOI: https://doi.org/10.1007/978-981-15-1377-0_55
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-1376-3
Online ISBN: 978-981-15-1377-0
eBook Packages: Computer ScienceComputer Science (R0)