BAB 5 KESIMPULAN DAN SARAN
5.1
Kesimpulan
a. Jaringan RBF yang dilatih menggunakan algoritma pembelajaran Extreme Learning Machine (ELM-RBF) tidak hanya memiliki akurasi lebih tinggi melainkan juga unggul dalam hal kecepatan training bila dibandingkan dengan algoritma SCG-TLF dan BP-MLP, yaitu tingkat akurasi 92,05% dan waktu training selama 5,43 detik. Dengan kata lain, Algoritma ELM memiliki akurasi 6,95% dan 27,95% lebih tinggi dan waktu training 132,57 detik dan 10,66 detik lebih cepat masing-masing terhadap algoritma SCG-TLF dan BP-MLP. b. Akurasi klasifikasi pada ELM-RBF menggunakan fitur input hasil reduksi PCA mencapai 94,62% dan waktu training selama 4,18 detik. Hal ini menunjukkan adanya peningkatan performansi akurasi sebesar 2,79% dan reduksi waktu training selama 1,25 detik terhadap penggunaan fitur input asli c. Penggunaan jumlah hidden neuron berbanding lurus dengan waktu training namun tidak berlaku pada tingkat akurasinya, sehingga penentuan jumlah hidden neuron yang tepat turut menentukan performansi jaringan ELM-RBF.
5.2
Saran Untuk pengembangan lebih lanjut, metode klasifikasi berbasis jaringan
syaraf tiruan pada umumnya dan jaringan ELM-RBF pada khususnya dapat ditingkatkan pada level klasifikasi dengan data citra sidik jari yang lebih besar pada beberapa Database sidik jari yang tersedia di internet secara komersial seperti Database NIST.
65
(Maltoni, Maio, Jain dan salil Prabhakar, 2003) (Jain, Hong, Pankanti dan Bolle, 1997) (Li, Yau dan Wang, 2007) (Yeung, Cloete dan Shi, 2009) (Musafi, Ahmed dan Chan, 1992) (Huang, Qin-Yu-Zhu dan Siew, 2006) (Huang, Zhu dan Siew, 2006) (Huang, Zhu dan Siew, 2004) (Serrau, Marcialis, Bunke dan Roli, 2003) (Wang, Li dan Niu, 2005) (Sharma, Paliwal dan Onwubolu, 2006) (Champbell dan Meyer, 2009)
66
DAFTAR PUSTAKA Anton, H. (1987). Elementary Linear Algebra, John Wiley Songs, INC. Candela dan Grother (1995). Pcasys a pattern-level classification automation system for fingerprints, Technical report, NIST TR 5647. Champbell, S. L. dan Meyer, C. D. (2009). Generalized Inverses of Linear Transformations, Society for Industrial and Applied Mathematics, Philadelphia. Fausett, L. (1994). Fundamentals of Neural Network Architecture, Algorithms, and Application, Prentice-Hall,USA. Gray, R. M. dan Neuhoff, D. L. (1998). “quantization”, IEEE Transactions on Information Theory . Huang, G. B., Qin-Yu-Zhu dan Siew, C.-K. (2006). “Extreme Learning Machine:Theory and Applications”, Neurocomputing (70): 489–501. Huang, G. B. dan Siew, C.-K. (2004). “Extreme Learning Machine with Randomly Assigned RBF Kernels”, Proceedings of the Eight International Conference on Control, Automation, Robotics and Vision. Huang, G.-B., Zhu, Q.-Y. dan Siew, C.-K. (2004). “Extreme Learning Machine: A New Learning Scheme of Feedforward Neural Networks”, Proceedings of International Joint Conference on Neural Networks (IJCNN2004). Huang, G.-B., Zhu, Q.-Y. dan Siew, C.-K. (2006). “Extreme Learning machine:theory and applications”, Neurocomputing (70): 489–501. J¨ahne., B. (1993). Digital Image Processing - Concepts, Algorithms and Scientific Applications, Springer-Verlag. Jain, A. K., Hong, L., Pankanti, S. dan Bolle, R. (1997). “an identity authentication system using fingerprints”, Proceedings of the IEEE, Vol. 85, pp. 365–1388. Jin, C. dan Jin, P. (2009). “Fingerprint Classifikasian in DCT Domain using RBF Neural Network”, Journal of Information Science and Engineering . Komarinski, P. (2005). Automated Fingerprint Identification Systems (AFIS), Elsevier Academic Press. Li, J., Yau, W.-Y. dan Wang, H. (2007). “Combining Singular Point and Orientation Image Information for Fingerprint Classification”, The Journal of the Pattern Recognition Society . Madsen, R. E., Hansen, L. K. dan Winther, O. (2004). Singular value decomposition and principal component analysis, Technical report. 67
Maltoni, D., Maio, D., Jain, A. K. dan salil Prabhakar (2003). Handbook of Fingerprint Recognition, Springer, New York. McAndrew, A. (2004). An introduction to digital image processing with matlab, Technical report, School of Computer Science and Mathematics Victoria University of Technology. Musafi, M., Ahmed, W. dan Chan, K. (1992). “On Training of Radial Basis Function Classifiers”, Neural Networks 5(4). Park, C. H. dan Park, H. (2004). “Fingerprint Classification using Fast Fourier Transform and Nonlinear Discriminant Analysis”, Journal of the pattern recognition society . Prabhakar, S. (2001). Fingerprint Classification and Matching Using a Filterbank, PhD thesis, Michigan State University. Schott, J. R. (1997). Matrix Analysis for Statistics, John Wiley and Songs, Inc. Seemann, T. (2002). Digital Image Processing using Local Segmentation, PhD thesis, School of Computer Science and Software Engineering Faculty of Information Technology Monash University Australia. Serrau, A., Marcialis, G. L., Bunke, H. dan Roli, F. (2003). An experimental comparison of fingerprint classification methods using graphs, Technical report, Department of Electrical and Electronic Engineering University of Cagliari Italy. Sharma, A. dan Paliwal, K. K. (2007). “Fast Principal Component Analysis using Fixed-point Algorithm”, Pattern Recognition Letters (28): 1151–1155. Sharma, A., Paliwal, K. K. dan Onwubolu, G. C. (2006). “Class-dependent PCA, MDC and LDA:A combined classifier for pattern classification”, The journal of the Pattern Recognition Society (39): 1215–1229. Simon, H. (1999). Neural networks: A comprehensive foundation, New Jersey:Prentice Hall. Thompson, C. M. dan Shure, L. (1997). Image Processing Toolbox User’s Guide, The MathWorks, Inc. Wang, X., Li, J. dan Niu, Y. (2005). Advances in Neural Networks ISNN 2005, Springer Berlin / Heidelberg, chapter Fingerprint Classification Based on Curvature Sampling and RBF Neural Networks, pp. 171–176. Wuzhili (2002). Hongkong.
Fingerprint Recoqnition, Master’s thesis, Baptist University,
Yeung, D. S., Cloete, I. dan Shi, D. (2009). Sensitivity Analysis for Neural Networks, springer. 68
Zhang, R. (2005). Efficient Sequential and Batch Learning Artificial Neural Network Methods for Classification Problems, PhD thesis, Nanyang Technological University.
69
