Comparison of KNN and SVM Algorithms in Facial Image Recognition Using Haar Wavelet Feature Extraction

Neneng Rachmalia Feta

doi:10.34288/jri.v5i3.224

Authors

Neneng Rachmalia Feta Indonesia Cyber University

(*) Corresponding Author

DOI:

https://doi.org/10.34288/jri.v5i3.224

Keywords:

accuracy, facial image recognition, feature extraction, Haar Wavelet, low dimension, KNN algorithm, SVM algorithm

Abstract

To process all the pixels in the face image, feature extraction can be performed using the Haar Wavelet method so that it processes identifiers with lower dimensions. However, a classification algorithm must separate the distance between classes with minimal data to classify low-dimensional facial images. KNN and SVM algorithms are classifiers that can be used for facial image recognition. When classifying images, SVM creates a hyperplane, divides the input space between classes and classifies based on which side of the hyperplane the unclassified object is placed when it is placed in the input space. KNN uses a voting system to determine which class an unclassified object belongs to, taking into account the nearest neighbor class in the decision space. When classifying, KNN will generally classify accurately, resulting in some minor misclassifications that plagued the final classified image. This study aims to compare the two algorithms on image identifiers with low dimensions resulting from haar wavelet extraction. The research results obtained are facial image classification using the haar wavelet extraction method using the SVM algorithm to obtain an accuracy of 98.8%. Whereas when using the KNN algorithm, the accuracy obtained is 96.6%. The results of this study show that the SVM algorithm produces better accuracy in facial image recognition using haar wavelet feature extraction compared to the KNN algorithm. The SVM algorithm can recognize facial images even though it uses image training data with various face poses and sizes, resulting in higher accuracy.

Downloads

Download data is not yet available.

References

Ahmad, I., Siddiqi, M. H., Fatima, I., Lee, S., & Lee, Y.-K. (2011). Weed classification based on Haar wavelet transform via k-nearest neighbor (k-NN) for real-time automatic sprayer control system. ICUIMC ’11, 1–6.

Al-Aidid, S., & Pamungkas, D. (2018). Sistem Pengenalan Wajah dengan Algoritma Haar Cascade dan Local Binary Pattern Histogram. Jurnal Rekayasa Elektrika, 14(1), 62–67.

Aminudin, A., & Cahyono, E. B. (2019). Pengukuran Performa Apache Spark dengan Library H2O Menggunakan Benchmark Hibench Berbasis Cloud Computing. Jurnal Teknologi Informasi Dan Ilmu Komputer, 6(5), 519. https://doi.org/10.25126/jtiik.2019651520

Bahri, S., & Lubis, A. (2020). Metode Klasifikasi Decision Tree Untuk Memprediksi Juara English Premier League. Jurnal Sintaksis, 2(1), 63–70.

Desylvia, S. N. (2014). Perbandingan SOM dan LVQ pada Identifikasi Citra Wajah dengan Wavelet sebagai Ekstraksi Ciri. Institute Pertanian Bogor.

Feta, N. R. (2022). Classification of Burned Peatland Using Probabilistic Neural Network Algorithm Based on High Temporal Data. Jurnal Riset Informatika, 4(2), 141–148.

Feta, N. R., & Ginanjar, A. R. (2019). Komparasi Fungsi Kernel Metode Support Vector Machine Untuk Pemodelan Klasifikasi Terhadap Comparison of the Kernel Function of Support Vector Machine Method for Modeling Classification of Soybean Plat Disease. Jurnal Ilmiah Ilmu Komputer, Sains Dan Teknologi Terapan, 1(1), 33–39.

Ginanjar, A. R., & Feta, N. R. (2019). Identifikasi Citra Wajah Menggunakan Probabilistic Neural Network dengan Ekstraksi Ciri Berbasis Wavelet. BRITech, 1(1), 24–32.

Kosasih, R. (2020). Kombinasi Metode ISOMAP Dan KNN Pada Image Processing Untuk Pengenalan Wajah. CESS (Journal of Computer Engineering, System and Science), 5(2), 166.

Nasution, M. R. A. N., & Hayaty, M. (2019). Perbandingan Akurasi dan Waktu Proses Algoritma K-NN dan SVM dalam Analisis Sentimen Twitter. Jurnal Informatika, 6(2), 226–235.

Ramdani, M. H., Wijaya, I. G. P. S., & Dwiyansaputra, R. (2022). Optimalisasi Pengenalan Wajah Berbasis Linear Discriminant Analysis Dan K-Nearest Neighbor menggunakan Particle Swarm Optimization. Teknologi Informasi, Komputer Dan Aplikasinya (JTIKA), 4(1), 40–51.

Reisenhofer, R., Bosse, S., Kutyniok, G., & Wiegand, T. (2018). A Haar wavelet-based perceptual similarity index for image quality assessment. Signal Processing: Image Communication, 61(February), 33–43.

Religia, Y. (2019). Feature Extraction Untuk Klasifikasi Pengenalan Wajah Menggunakan Support Vector Machine Dan K-Nearest Neighbor. Pelita Teknologi: Jurnal Ilmiah Informatika, Arsitektur Dan Lingkungan, 14(2), 85–92.

S.Raikwal, J., & Saxena, K. (2012). Performance Evaluation of SVM and K-Nearest Neighbor Algorithm over Medical Data set. International Journal of Computer Applications, 50(14), 35–39. https://doi.org/10.5120/7842-1055

Situmorang, G. T., Widodo, A. W., & Rahman, M. A. (2019). Penerapan Metode Gray Level Co-occurrence Matrix ( GLCM ) untuk ekstraksi ciri pada telapak tangan. Jurnal Pengembangan Teknologi Informasi Dan Ilmu Komputer, 3(5), 4710–4716.

Wijaya, Y. A., Bahtiar, A., Kaslani, & R, N. (2021). Analisa Klasifikasi menggunakan Algoritma Decision Tree pada Data Log Firewall. Jurnal Sistem Informasi Dan Manajemen, 9(3), 256–264. https://doi.org/10.47024/JS.V9I3.303

Wilhelm, B., & Mark, J. B. (2016). Digital Image Processing An Algorithmic Introduction Using Java Second Edition. In European Journal of Engineering Education (Second, Vol. 19, Issue 3). Springer.

Yohannes, Y., Sari, Y. P., & Feristyani, I. (2019). Klasifikasi Wajah Hewan Mamalia Tampak Depan Menggunakan k-Nearest Neighbor Dengan Ekstraksi Fitur HOG. Jurnal Teknik Informatika Dan Sistem Informasi, 5(1), 84–97.

Yulianti, D., Triastomoro, I., & Sa’idah, S. (2022). Identifikasi Pengenalan Wajah Untuk Sistem Presensi Menggunakan Metode Knn (K-Nearest Neighbor). Jurnal Teknik Informasi Dan Komputer (Tekinkom), 5(1), 1–10.