p–ISSN: 2723 - 6609 e-ISSN: 2745-5254

Vol. 5, No. 10, October 2024 http://jist.publikasiindonesia.id/

Indonesian Journal of Social Technology, Vol. 5, No. 10, October 2024 4253

Implementation of the YOLO Method for Detection of

Human Emotions Based on Facial Mimics

Thomas Felison

1*

, Erwin conery firtan

2

, Steven

3

4

*Correspondence

which is to recognize human facial expressions. Research on facial expression analysis

synthesis of human faces in a human-computer dialogue system, or detecting the presence

of micro-expressions of subtle movements in the face (Rosiani et al., 2018). One of the

real applications of human facial expression analysis is to recognize the facial expressions

of e-learning users, where learner interaction is a weakness that must be considered in e-

learning learning (Husdi, 2016). The process of recognizing human facial expressions is

also applied to the MOODSIC music player application, which will play music according

to the user's emotions obtained by detecting the user's facial expressions (Wijaya et al.,

2018). Emotion detection through facial expression recognition plays an important role

in daily life, such as how to properly respond to emotional expressions in social

interactions so that it can establish and build verbal or nonverbal communication with

intentions, or social communication. Face detection is the first step that must be done in

facial analysis, including facial expression recognition. Face detection aims to determine

whether or not there is a face in the image, and if there is a location of the face and the

size of each face in the image. (Budiyanta et al., 2021). In face detection, there are several

challenges such as the position of the face not facing directly to the camera, face scale,

facial expressions, face obstructed by other objects, and lighting conditions.

(Prasetyawan, 2020). Many methods can be used to carry out the face detection process,

such as Jatmoko has implemented the Viola-Jones algorithm in facial recognition, with

the results of research from all experiments obtaining an average accuracy score of 65%

(Jatmoko et al., 2020). Another research conducted by Putra and Krishna using the

only needs to look once at the input image, to predict what object is contained in the

image and where it is located. (Redmon, 2016).

Some similar studies that have been conducted before can be detailed as follows:

Author's Name

(Year)

Heading

Recency

process of recognizing human facial expressions. After that, tests will be carried out on

have the potential to be applied in various fields, such as human-computer interaction

(HCI), supervision and monitoring systems in work or education environments, as well

as the entertainment and technology industries. The implementation of the YOLO method

in facial expression detection can help improve interaction between humans and

machines, develop monitoring systems that are more responsive to human emotions, and

be used in entertainment applications that adjust content based on user expressions and

emotions. Thus, this research is expected not only to contribute to the development of

science but also to have a real impact on various technological and social sectors.

Network (CNN) to detect objects in images. This algorithm divides the image into grids

of a certain size, where each grid predicts a bounding box and a class map for the objects

it contains. If a grid predicts the existence of an object, then a bounding box will be

predicted that surrounds the object. The confidence score is calculated for each bounding

box and selection is made based on the confidence value. (Nasution & Kartika, 2022).

the research topic. To ensure the success of the research, the tools used are as follows.

Hardware Requirements

The software to design this program is:

a. Windows 10 operating system.

b. Visual Studio C# 2013

c. Microsoft SQL Server 2012.

The number of correct facial expression detections = 15 pieces and incorrect = 5 pieces

Disgust Expression

The number of correct facial expression detections = 20 pieces and incorrect = 0 pieces

Jumlah deteksi ekspresi wajah yang benar = 6 buah dan salah = 14 buah

Table 4

Bamba, I., Yashika, Singh, J., & Chawla, P. (2022). Face Recognition Techniques and

Husdi, H. (2016). Pengenalan Ekspresi Wajah Pengguna Elearning Menggunakan

Hutauruk, J. S. W., Matulatan, T., & Hayaty, N. (2020). Deteksi kendaraan secara real

Jatmoko, C., Hartanto, D., Kurniawan, A. F., Rachmawanto, E. H., Sari, C. A., &

Nasution, S. W., & Kartika, K. (2022). Eggplant disease detection using Yolo algorithm

Redmon, J. (2016). You only look once: Unified, real-time object detection. Proceedings

Rosiani, U. D., Choirina, P., Sumpeno, S., & Purnomo, M. H. (2018). Menuju Pengenalan

Wijaya, I. G. P. S., Firdaus, A. A., Dwitama, A. P. J., & Mustiari, M. (2018). Pengenalan