Search for collections on Undip Repository

IMPLEMENTASI METODE EIGENFACE DAN INTEGRASI SUPPORT VECTOR MACHINE DALAM FACE RECOGNITION

ANNUBAHA, Chakim and Widodo, Aris Puji and Adi, Kusworo (2022) IMPLEMENTASI METODE EIGENFACE DAN INTEGRASI SUPPORT VECTOR MACHINE DALAM FACE RECOGNITION. Masters thesis, School of Postgraduate Studies.

[thumbnail of COVER.pdf] Text
COVER.pdf

Download (774kB)
[thumbnail of BAB I.pdf] Text
BAB I.pdf

Download (15kB)
[thumbnail of BAB II.pdf] Text
BAB II.pdf

Download (700kB)
[thumbnail of BAB III.pdf] Text
BAB III.pdf
Restricted to Repository staff only

Download (361kB)
[thumbnail of BAB IV.pdf] Text
BAB IV.pdf
Restricted to Repository staff only

Download (1MB)
[thumbnail of BAB V.pdf] Text
BAB V.pdf

Download (10kB)
[thumbnail of DAFTAR PUSTAKA.pdf] Text
DAFTAR PUSTAKA.pdf

Download (232kB)
[thumbnail of LAMPIRAN.pdf] Text
LAMPIRAN.pdf
Restricted to Repository staff only

Download (3MB)

Abstract

Sistem biometrika adalah sebuah teknologi pengenalan seseorang yang banyak dikembangkan akhir-akhir ini. Salah satunya adalah sistem pengenalan wajah yang berbentuk citra digital dengan algoritma eigenface. Eigenface digunakan untuk mereduksi dimensi vector wajah menjadi vector yang lebih sederhana (eigen vector). Pada penelitian ini diimplementasikan metode eigenface dan integrasi support vector machine (SVM). Eigenface menggunakan metode Principal Component Analysis (PCA) untuk mereduksi dimensi gambar wajah sehingga menghasilkan variabel yang lebih sedikit dan lebih mudah ditangani. Hasil yang diperoleh kemudian dimasukkan ke dalam pattern classifier untuk menentukan identitas pemilik wajah. Penelitian ini menggunakan 600 data wajah sebagai data uji dan data latih. Terbagi menjadi 10 kelas dengan setiap kelas 15 wajah, kemudian dilakukan simulasi data wajah dengan resolusi 500 kb, 250 kb, 125 kb, dan 75 kb. Hasil pengujian sistem menunjukkan bahwa penggunaan Eigenface dengan Support Vector Machine (SVM) sebagai classifier dapat memberikan tingkat akurasi yang cukup tinggi. Dapat dilihat dari hasil akurasi wajah dengan resolusi 75 kb memperoleh akurasi 78%, resolusi 125 kb akurasi 78%, resolusi 250 kb akurasi 83%, dan resolusi 500 kb akurasi 84%. Hal ini membuktikan semakin tinggi resolusi data yang digunakan akan meningkatkan hasil akurasi.
Kata Kunci: Face Recognition, Eigenface, Support Vector Machine (SVM)

The biometric system is a person recognition technology that has been developed recently. One of them is a face recognition system in the form of a digital image with the eigenface algorithm. Eigenface is used to reduce the dimensions of the face vector into a simpler vector (eigen vector). In this research, the eigenface method and support vector machine (SVM) integration are implemented. Eigenface uses the Principal Component Analysis (PCA) method to reduce the dimensions of the face image so that it produces fewer variables and is easier to handle. The results obtained are then entered into the pattern classifier to determine the identity of the face owner. This study uses 600 facial data as test data and training data. Divided into 10 classes with 15 faces each class, then simulated facial data with a resolution of 500 kb, 250 kb, 125 kb, and 75 kb. The system test results show that the use of Eigenface with a Support Vector Machine (SVM) as a classifier can provide a fairly high level of accuracy. It can be seen from the results of facial accuracy with 75 kb resolution, 78% accuracy, 125 kb resolution 78% accuracy, 250 kb resolution 83% accuracy, and 500 kb resolution 84% accuracy. This proves that the higher the resolution of the data used will increase the accuracy of the results
Keywords: Face Recognition, Eigenface, Support Vector Machine (SVM)

Item Type: Thesis (Masters)
Uncontrolled Keywords: Face Recognition, Eigenface, Support Vector Machine (SVM)
Subjects: Sciences and Mathemathic
Divisions: Postgraduate Program > Master Program in Information System
Depositing User: ekana listianawati
Date Deposited: 16 Nov 2022 08:25
Last Modified: 16 Nov 2022 08:25
URI: https://eprints2.undip.ac.id/id/eprint/9721

Actions (login required)

View Item View Item