Setiawan, Haris (2022) MULTI BRANCH CONVOLUTION NEURAL NETWORK UNTUK IDENTIFIKASI JENIS KELAMIN DAN USIA DENGAN MENGGUNAKAN MULTICLASS CLASSIFICATION DAN FACENET MODEL. S2 thesis, Universitas Mercu Buana Jakarta-Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (Cover)
55420120003-Haris Setiawan-01 Cover - Haris Setiawan.pdf Download (532kB) |
|
Text (Abstrak)
55420120003-Haris Setiawan-02 Abstrak - Haris Setiawan.pdf Download (92kB) |
|
Text (Bab 1)
55420120003-Haris Setiawan-03 Bab 1 - Haris Setiawan.pdf Restricted to Registered users only Download (145kB) |
|
Text (Bab 2)
55420120003-Haris Setiawan-04 Bab 2 - Haris Setiawan.pdf Restricted to Registered users only Download (520kB) |
|
Text (Bab 3)
55420120003-Haris Setiawan-05 Bab 3 - Haris Setiawan.pdf Restricted to Registered users only Download (358kB) |
|
Text (Bab 4)
55420120003-Haris Setiawan-06 Bab 4 - Haris Setiawan.pdf Restricted to Registered users only Download (386kB) |
|
Text (Bab 5)
55420120003-Haris Setiawan-07 Bab 5 - Haris Setiawan.pdf Restricted to Registered users only Download (100kB) |
|
Text (Daftar Pustaka)
55420120003-Haris Setiawan-08 Daftar Pustaka - Haris Setiawan.pdf Restricted to Registered users only Download (106kB) |
|
Text (Surat Pernyataan)
55420120003-Haris Setiawan-09 Hasil Scan Formulir Pernyataan Keabsahan dan Persetujuan Publikasi Tugas Akhir - Haris Setiawan.pdf Restricted to Registered users only Download (137kB) |
Abstract
Wajah manusia memberikan banyak informasi mengenai jenis kelamin, usia, etnis dan juga emosi. Jenis kelamin dan usia dianggap sebagai biometrik penting sebagai atribut untuk proses identifikasi dan dapat di kembangkan kedalam Human Computer Interaction (HCI) dimana memilki area pengembangan yang potensial untuk aplikasi forensic, law enforcement dan security control. Namun identifikasi jenis kelamin dan usia ini dipengaruhi oleh banyak faktor dinamis yang dapat berubah dari waktu ke waktu seperti penuaan, gaya rambut dan ekspresi sehingga proses identifikasi sering mengalami kendala dalam akurasi dan memperbesar nilai loss. Beberapa metodologi face recognition telah dicoba untuk mengatasi permasalahan faktor dinamis tersebut seperti multibranch convolutional neural network, dimana beberapa penelitian sebelumnya menggunakan metode tersebut untuk menanggani masalah overfitting dan backpropagation, namun masih diperlukan metode pendukung lainnya untuk meningkatkan nilai akurasi. Penelitian ini akan membahas multibranch convolutional neural network untuk identifikasi jenis kelamin dan usia dengan menggunakan metode multiclass classification dan facenet model. Dari metode usulan dalam penelitian ini di dapatkan nilai akurasi jenis kelamin senilai 90.75% dan mean absolute error maximum 4.485 dan minimum 0.072. Multi branch convolutional neural network (CNN) digunakan untuk mengoptimalkan backpropagation dan mengurangi tingkat kesalahan dengan mengatur bobot berdasarkan perbedaan keluaran dan target yang diinginkan. Multiclass classification digunakan untuk melakukan pengelompokan data berdasarkan umur, sedangkan FaceNet digunakan untuk menyelesaikan permasalahan di terkait face verification dan overfitting Kata Kunci : Face Recognition, Multibranch convolutional neural network, multiclass classification, FaceNe The human face provides a wealth of information regarding gender, age, ethnicity and emotions. Gender and age are considered as important biometrics and attributes for the identification process, it can be developed into Human Computer Interaction (HCI) which has potential development areas for forensic, law enforcement and security control applications. However, the identification of gender and age is influenced by many dynamic factors that can change over time such as aging, hairstyles and expressions, so the identification process have a problems in accuracy and increases the loss. Several face recognition methodologies have been tried to overcome these dynamic factor problems, one of them is multibranch convolutional neural networks. The previous studies used these methods to deal with overfitting and backpropagation, but other supporting methods are still needed to increase the accuracy. This experiments will propose a multibranch convolutional neural network for gender and age identification using the multiclass classification method and the facenet model. The proposed method has gender accuracy 90.75% and maximum mean absolute error (mae) 4.485 and the minimum value 0.072 . Multi branch convolutional neural network (CNN) is used to optimize backpropagation and reduce the error rate by adjusting the weights based on the difference between the output and the desired target. Multiclass classification is used to group data based on age, while FaceNet is used to solve problems related to face verification and overfitting. Keywords: Face Recognition, Multibranch convolutional neural network, multiclass classification, FaceNet
| Item Type: | Thesis (S2) |
|---|---|
| NIM/NIDN Creators: | 55420120003 |
| Uncontrolled Keywords: | Face Recognition, Multibranch convolutional neural network, multiclass classification, FaceNe Face Recognition, Multibranch convolutional neural network, multiclass classification, FaceNet |
| Subjects: | 600 Technology/Teknologi > 620 Engineering and Applied Operations/Ilmu Teknik dan operasi Terapan > 621 Applied Physics/Fisika terapan |
| Divisions: | Pascasarjana > Magister Teknik Elektro |
| Depositing User: | ALFINA DHEA NOVA |
| Date Deposited: | 12 Jan 2023 07:44 |
| Last Modified: | 12 Jan 2023 07:44 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/73362 |
Actions (login required)
 |
View Item |
