gajahslot mariatogel airbet robopragma wajik77 mposlot https://siapadia.riau.go.id/plugins/pulsa-limaribu/ https://journal.parahikma.ac.id/juaraslot/ https://jurnalbening.uho.ac.id/pages/tm/ http://adj.fkg.unand.ac.id/lib/pkp/alexistoto/ http://inomet.ft.unand.ac.id/locale/bento/ https://jurnal.usk.ac.id/lib/pkp/sbobet88/ https://htn.parahikma.ac.id/sbotop/ https://jurnalbening.uho.ac.id/lib/pkp/bni4d/ https://job.uho.ac.id/lib/pkp/btn4d/ paito hk https://jpi.faterna.unand.ac.id/help/slot99/ ronaldo slot slot gacor malam ini KOMBINASI KPCA DAN EUCLIDEAN DISTANCE UNTUK PENGENALAN CITRA WAJAH | Wahyuningrum | Rekayasa

KOMBINASI KPCA DAN EUCLIDEAN DISTANCE UNTUK PENGENALAN CITRA WAJAH

Rima Tri Wahyuningrum

Abstract

Permasalahan machine learning dan pattern recognition bukan merupakan penelitian yang baru. Seiring dengan perkembangan teknologi, semakin berkembang pula teknik dan algoritma yang digunakan untuk menyelesaikan permasalahan machine learning dan pattern recognition. Pada penelitian ini telah berhasil melakukan pengenalan citra wajah menggunakan ekstraksi fitur Kernel Principal Component Analysis (KPCA) untuk menentukan karakteristik dari wajah dan Euclidean Distance sebagai metode klasifikasi berbasis statistik. Sedangkan uji coba telah dilakukan pada basis data citra wajah ORL, YALE dan BERN menggunakan kernel polynomial dan Gaussian, dengan reduksi dimensi menjadi v = 25 dan v = 50. Akurasi pengenalan citra wajah tertinggi dari ketiga basis data tersebut adalah menggunakan kernel Gaussian dan reduksi dimensi v = 50 dengan tujuh data pelatihan di setiap kelasnya. Pada basis data citra wajah ORL diperoleh akurasi pengenalan sebesar 98,50%, pada basis data citra wajah YALE diperoleh akurasi pengenalan sebesar 97,65%, dan pada basis data citra wajah BERN diperoleh akurasi pengenalan sebesar 97,95%. Dengan demikian, metode ekstraksi fitur KPCA yang dikombinasikan dengan metode klasifikasi Euclidean Distance sangat baik digunakan sebagai pengenalan citra wajah.

 

Kata kunci: Kernel Principal Component Analysis (KPCA), Euclidean Distance, kernel polynomial, kernel Gaussian

 

Abstract

Problems of machine learning and pattern recognition are not a new research. Along with the development of technology, growing techniques and algorithms used to solve the problems of machine learning and pattern recognition. In this research has been successfully performed face recognition using Kernel Principal Component Analysis (KPCA) as feature extraction to determine the characteristics of the face and Euclidean Distance as the classification method based on statistics. While the experiments have been conducted on ORL face image database, YALE and BERN using polynomial and Gaussian kernel, the dimension reduction to v = 25 and v = 50. Highest recognition accuracy of three face image database is to use the Gaussian kernel and the reduction of dimension v = 50 with seven training data in each class. In the ORL face image database obtained recognition accuracy of 98,50%, on the basis of image data obtained YALE face recognition accuracy of 97,65%, and on the basis of image data obtained BERN face recognition accuracy of 97,95%. Thus, KPCA feature extraction methods are combined with Euclidean Distance classification method is best used as a facial image recognition.

 

Key words: Kernel Principal Component Analysis (KPCA), Euclidean Distance, polynomial kernel, Gaussian kernel

Full Text:

PDF

References

Turk, M.A., Pentland, A.P., 1991. Eigenfaces for recognition. Journal of Cognitive Neuroscience 3 (1), 71–86.

Belhumeur, J.H.P.N., Kriegman, D., 1997. Eigenfaces vs. fisherfaces: Recognition using class specific linear projection, IEEE Transaction on Pattern Analysis of Machine Intelligent 19 (7), 711–720.

Scholkopf, B., Smola, A., Muller, K.R., 1998. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computing 10(5), 1299–1319.

Djakaria, I., Guritno, S., Kartiko, H.S., 2010. Visualisasi data iris menggunakan Analisis Komponen Utama dan Analisis Komponen Utama Kernel, Jurnal ILMU DASAR 11 (1), 31–38.

Wen, Y., He, L., Shi, P., 2012. Face recognition using difference vector plus KPCA, Digital Signal Processing 22 (7), 140–146.

López, M.M., Ramírez, J., Górriz, J.M., Álvarez, I., Salas-Gonzalez, D., Segovia, F., Chaves, R., 2009. SVM-based CAD system for early detection of the Alzheimer’s disease using kernel PCA and LDA, Neuroscience Letters 464 (3), 233–238.

Zhang, R., Wang, W., Mac, Y., 2010.

Approximations of the standard Principal Components Analysis and kernel PCA. Expert Systems with Applications 37 (9), 6531–6537.

Sahbi, H., 2007. Kernel PCA for similarity invariant shape recognition. Neurocomputing 70 (16-18), 3034–3045.

Ekstrom, J., 2011. Euclidean Distance beyond normal distributions. Website: http://statistics. ucla.edu/system/resources.pdf, diakses tanggal 30 Mei 2011.

Ommy, R., Rizal, A., dan Murti, M.A., 2008. Pengenalan identitas manusia melalui pola iris mata menggunakan Transformasi Wavelet dan Euclidean Distance. Konferensi Nasional Sistem dan Informatika, KNS&I08-056, Bali, 15 Nopember, 316–320.

Wang, Q., 2011. Kernel principal component analysis and its applications in face recognition and active shape models. Website: http://arxiv. org/pdf/1207.3538, diakses tanggal 1 April 2011.

DOI

https://doi.org/10.21107/rekayasa.v4i2.2343

Metrics

Refbacks

  • There are currently no refbacks.


Copyright (c) 2016 Rima Tri Wahyuningrum

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.