Abstract
This work proposes the use of orthogonal bipolar vectors (OBV) as new target vectors for Artificial Neural Networks (ANN) of the Radial Basis Functions (RBF) type for pattern recognition problems. The network was trained and tested with three sets of biometric data: human iris, handwritten digits and signs of the Australian sign language, Auslan. The objective was to verify the network performance with the use of OBVs and to compare the results obtained with those presented for the Multilayer Perceptron (MLP) networks. Datasets used in the experiments were obtained from the CASIA Iris Image Database developed by the Chinese Academy of Sciences - Institute of Automation, Semeion Handwritten Digit of Machine Learning Repository and UCI - Machine Learning Repository. The networks were modeled using OBVs and conventional bipolar vectors for comparing the results. The classification of the patterns in the output layer was based on the Euclidean distance. The results show that the use of OBVs in the network training process improved the hit rate and reduced the amount of epochs required for convergence.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Chinese Academy of Sciences - Institute of Automation, Database of 256. Greyscale Eye Images. http://www.cbsr.ia.ac.cn/IrisDatabase.htm
Manzan, J.R.G., Nomura, S., Yamanaka, K., Carneiro, M.B.P., Veiga, A.C.P.: Improving iris recognition through new target vectors in MLP artificial neural networks. In: Artificial Neural Networks in Pattern Recognition, pp. 115–126. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-33212-8_11
Lichman, M.: UCI machine learning repository (2013). http://archive.ics.uci.edu/ml
Kadous, M.W.: Temporal classification: extending the classification paradigm to multivariate time series (Unpublished doctoral dissertation). The University of New South Wales (2002)
Leemis, L: Learning Base R. Lightning Source (2016). ISBN 978-0-9829174-8-0
Haykin, S.: Redes Neurais: Princípios e prática. Bookman, Porto Alegre RS (2001)
Silva, I.N., Spatti, D.H., Flauzino, R.A.: Redes Neurais Artificiais para engenharia e ciências aplicadas - curso prático, 1 ed., Vol. 1. ArtLiber Editora, São Paulo (2010)
Fausett, L.: Fundamentals of neural networks: architectures, algorithms, and applications (1994)
Lowe, D.: What have neural networks to offer statistical pattern processing?. In: Proceedings of SPIE 1565, Adaptive Signal Processing, 1 December 1991. https://doi.org/10.1117/12.49798. Prentice-Hall, Inc
Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification, 2nd edn (2001)
Nied, A.: Training of artificial neural networks based on systems of variable structure with adaptive learning rate. (Unpublished doctoral dissertation) – Federal University of Minas Gerais, Belo Horizonte (2007). (in Portuguese)
Manzan, J.R.G., Yamanaka, K., Nomura, S.: Orthogonal bipolar vectors as multilayer perceptron targets for biometric pattern recognition. In: International Conference on Computing, Networking and Communications - Garden Grove, CA, USA (2015)
Pereira, M.B., Veiga, A.C.P.: An application of genetic algorithms to improve the reliability of an iris recognition system. In: XXVIII National Congress of Applied and Computational Mathematics – São Paulo/ Brazil (2005). https://doi.org/10.1109/MLSP.2005.1532892
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning internal representations by error propagation. Technical report, DTIC Document (1985)
Du, K.-L., Swamy, M.N.S.: Neural Networks and Statistical Learning (2013). https://doi.org/10.1007/978-1-4471-5571-3
Nomura, S., Yamanaka, K., Katai, O., Kawakami, H., Shiose, T.: Improved mlp learning via orthogonal bipolar target vectors. JACIII 9(6), 580–589 (2005)
Nomura, S., Yamanaka, K., Katai, O., Kawwakami, H., Shiose, T.: A new approach to improving math performance of artificial neural networks (2004)
Nomura, S., Manzan, J.R.G., Yamanaka, K.: Análise experimental de novos vetores alvo na melhoria do desempenho de mlp. IX Conferência de estudos em engenharia elétrica (CBIC 2011)
Manzan, J.R.G., Yamanaka, K., Nomura, S.: Improvement in perfomance of MLP using new target vectors (in Portuguese). In: X Brazilian Congress on Computational Intelligence – Fortaleza (2011)
Manzan, J.R.G., Nomura, S., Yamanaka, K.: Mathematical evidence for target vector type influence on MLP learning improvement. In: Proceedings on the International Conference on Artificial Intelligence (ICAI), p. 1 (2012)
Manzan, J.R.G., Yamanaka, K., Nomura, S.: A mathematical discussion concerning the performance of multilayer perceptron-type artificial neural networks through use of orthogonal bipolar vectors. In: Computational and Applied mathematics 2016 (2016)
Conover, W.J.: Practical nonparametric statistics. Wiley series in probability and statistics: applied probability and statistics (1999)
Martins, G.A., Fonseca, J.S.: Curso de estatística. Atlas, 6ª Edição (2006)
Acknowledgment
The authors thank Federal University of Uberlândia and Federal Institute of Triângulo Mineiro for support in the realization of the research.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
da Cruz Santos, C., Yamanaka, K., Manzan, J.R.G., Peretta, I.S. (2019). Modified Radial Basis Function and Orthogonal Bipolar Vector for Better Performance of Pattern Recognition. In: Arai, K., Kapoor, S., Bhatia, R. (eds) Intelligent Systems and Applications. IntelliSys 2018. Advances in Intelligent Systems and Computing, vol 868. Springer, Cham. https://doi.org/10.1007/978-3-030-01054-6_31
Download citation
DOI: https://doi.org/10.1007/978-3-030-01054-6_31
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-01053-9
Online ISBN: 978-3-030-01054-6
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)