ESA PUTRA SAKTIAWAN, . (2021) PENGUJIAN PERFORMA REAL TIME SIDE FACE RECOGNITION DENGAN METODE LOCAL BINARY PATTERN HISTOGRAM (LBPH) DAN LIBRARY OPENCV. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.
![[img]](http://repository.unj.ac.id/style/images/fileicons/text.png) |
Text
COVER.pdf Download (1MB) |
|
Text
BAB I.pdf Download (342kB) |
|
Text
BAB II.pdf Restricted to Registered users only Download (952kB) |
|
Text
BAB III.pdf Restricted to Registered users only Download (3MB) |
|
Text
BAB IV.pdf Restricted to Registered users only Download (1MB) |
|
Text
BAB V.pdf Restricted to Registered users only Download (229kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (253kB) |
|
Text
LAMPIRAN.pdf Restricted to Registered users only Download (5MB) |
Abstract
Pengaplikasian visi komputer yang mana menggunakan kamera pada proses image acquisition, dapat dimanfaatkan pada identifikasi biometrik dalam pengenalan wajah secara real time. Dalam metode pengenalan wajah, metode Local Binary Pattern Histogram (LBPH) adalah salah satu metode pengenalan wajah yang memiliki tingkat kesalahan terkecil. Pada proses pengenalan wajah, proses awal yang perlu dilakukan dalam visi komputer adalah proses deteksi wajah. Dalam proses deteksi wajah, algoritma Haar Cascade Classifier atau dikenal dengan Viola Jones merupakan metode yang paling umum digunakan dalam proses deteksi citra karena dapat memproses citra dengan cepat dan memiliki keakurasian deteksi yang tinggi. Pada pengaplikasiannya, metode Viola Jones juga memiliki batasan yaitu hanya terbatas pada deteksi wajah tampak depan. Hal tersebut tentu akan menjadi masalah pada pendeteksian wajah dalam sudut kemiringan tertentu. Pada library OpenCV terdapat model classifier untuk mendeteksi sisi samping wajah dan dapat menjadi solusi dalam mendeteksi sisi samping wajah. Setelah wajah dapat dideteksi, barulah citra wajah dapat diolah menjadi data latih yang kemudian dijadikan model classifier dengan metode Local Binary Pattern Histogram (LBPH). Pada pengujian, wajah dikenali secara real time dengan dideteksi kembali wajahnya dengan algoritma Haar Cascade Classifier kemudian bagian wajah yang terdeteksi diolah dengan metode Local Binary Pattern Histogram (LBPH) dan dibandingkan nilai histogram pengujian dengan histogram pada model classifier dari data latih. Nilai terkecil dari histogram pengujian dan model classifier akan menjadi nilai hasil pengenalan wajah yang kemudian dianalisis untuk didapatkan hasil performa dari seluruh subjek penelitian. Pada penelitian ini dilakukan pengujian performa yaitu akurasi, precision dan recall terhadap pengenalan sisi samping wajah pada 10 subjek penelitian. Berdasarkan hasil pembahasan, performa pengenalan sisi kanan wajah memiliki akurasi sebesar 90%, precision sebesar 50%, dan recall sebesar 50% serta performa pengenalan sisi kiri wajah memiliki akurasi sebesar 95%, precision sebesar 75%, dan recall sebesar 75%. Applications of computer vision which use the camera in the process of image acquisition, can be used for biometric identification in real time face recognition. For the face recognition method, the Local Binary Pattern Histogram (LBPH) is a facial recognition method that has the lowest error rate. For the face recognition process on computer vision, the initial process is the face detection. In the face detection process, Haar Cascade Classifier or known as Viola Jones method is the most commonly used method in the image detection process because it can process images quickly and has high detection accuracy. In its application, the Viola Jones method also has a limitation, which is only limited to detect a frontal face. This will certainly be a problem in detecting faces at a certain angle of inclination. In the OpenCV library there is a classifier model to detect the sides of the face and can be a solution in detecting the sides of the face. After the face can be detected, then the face image can be processed into training data which is then used as a classifier model using the Local Binary Pattern Histogram (LBPH) method. In the test, faces are recognized in real time by detecting their faces with the Haar Cascade Classifier algorithm then the detected facial parts are processed using the Local Binary Pattern Histogram (LBPH) method and the test histogram values with the histogram on the classifier model from the training data are compared. The smallest value from the test histogram and the classifier model will be the face recognition result value which is then analyzed to obtain the performance results of all research subjects. This research discusses the performance testing was carried out namely accuracy, precision and recall on the recognition of the side of the face in 10 research subjects. Based on the results of the discussion, the recognition performance of the right side face has an accuracy of 90%, precision of 50%, and recall of 50% and the left face recognition performance has an accuracy of 95%, precision by 75%, and recall by 75%
| Item Type: | Thesis (Sarjana) |
|---|---|
| Additional Information: | 1) Drs. Mulyono, M. Kom. 2)Muhammad Eka Suryana, M.Kom. |
| Subjects: | Sains > Matematika > Ilmu Komputer |
| Divisions: | FMIPA > S1 Ilmu Komputer |
| Depositing User: | Users 14629 not found. |
| Date Deposited: | 13 Jul 2022 03:32 |
| Last Modified: | 19 Jan 2024 05:28 |
| URI: | http://repository.unj.ac.id/id/eprint/31728 |
Actions (login required)
 |
View Item |
