Deteksi Emosi Wajah Menggunakan Algoritma Deep Generative Adversarial Network (DGAN) Pada Fitur Konsultasi Video Call

Attaqwa, Syukur Iman (2024) Deteksi Emosi Wajah Menggunakan Algoritma Deep Generative Adversarial Network (DGAN) Pada Fitur Konsultasi Video Call. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010026_Cover.pdf Download (1MB)
	Text (BAB 1) 20081010026_BAB 1.pdf Download (104kB)
	Other (BAB 2) 20081010026_Bab 2.PDF Restricted to Repository staff only until 19 September 2026. Download (793kB)
	Other (BAB 3) 20081010026_Bab 3.PDF Restricted to Repository staff only until 19 September 2026. Download (5MB)
	Text (BAB 5) 20081010026_BAB 5.pdf Download (85kB)
	Other (BAB 4) 20081010026_Bab 4.PDF Restricted to Repository staff only until 19 September 2026. Download (5MB)
	Text (Daftar Pustaka) 20081010026_DAFTAR PUSTAKA.pdf Download (103kB)

Abstract

Video call is one of the results of further development in the field of communication by utilizing camera capture in the form of video transmitted in real time through the internet network. Utilizing video calls as a place to conduct remote consultations or online consultations has considerable potential with 93% of adolescents interested in conducting online consultations (Mirawati, 2015). With this potential, this research will develop video calls in one of the consultation systems with technology that can help facilitate the consultation process, especially in helping to analyze emotions. The proposed technology is Facial Expression Recognition (FER) technology using the Deep Generative Adversarial Networks (DGANs) algorithm and applying the Two-branch Disentangled method in carrying out the classification process. In this study, the average accuracy result was 63.64% with a precision value of 64.37% while 62.61% for the sensitive value and 61.04% for the f1 score value. Meanwhile, overall the features of the consultation system run well in functionality testing. This can be seen from the average time value obtained of 366 milliseconds.

Item Type:

Thesis (Undergraduate)

Contributors: