SISTEM PENDETEKSIAN EMOSI MULTI-WAJAH MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN)

RISHAD, AHMAD EMERY and Fachrurrozi, Muhammad (2025) SISTEM PENDETEKSIAN EMOSI MULTI-WAJAH MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN). Undergraduate thesis, Sriwijaya University.

Preview	Image RAMA_55201_09021382126168_cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (94kB) | Preview
	Text RAMA_55201_09021382126168.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_55201_09021382126168_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021382126168_0222058001_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (659kB)
	Text RAMA_55201_09021382126168_0222058001_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (410kB) | Request a copy
	Text RAMA_55201_09021382126168_0222058001_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (387kB) | Request a copy
	Text RAMA_55201_09021382126168_0222058001_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (554kB) | Request a copy
	Text RAMA_55201_09021382126168_0222058001_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (444kB) | Request a copy
	Text RAMA_55201_09021382126168_0222058001_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (198kB) | Request a copy
	Text RAMA_55201_09021382126168_0222058001_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy

Abstract

This research aims to develop and implement a real-time multi-face detection and emotion recognition system. The Haar Cascade method is used in the face detection stage to identify the position of faces in the image, while the Convolutional Neural Network (CNN) is used in the emotion classification stage based on the detected faces. The datasets used include FDDB for face detection testing and FER-2013 for emotion classification testing. The results of multi-face detection testing using Haar Cascade on a subset of FDDB showed an accuracy of 63%. Emotion classification testing using CNN on the FER-2013 dataset resulted in an accuracy of 74%. The overall success rate of the application was calculated, resulting in a value of 47%. Based on these results, the system is capable of performing real-time multi-face detection and emotion classification with a moderate success rate and still has room for improvement, especially in the multi-face detection stage

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Deteksi Wajah, Deteksi Emosi, Haar Cascade, Convolutional Neural Network (CNN), Multi-Wajah, FER-2013, Sistem Real-Time.
Subjects:	Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76 Computer software
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Ahmad Emery Rishad
Date Deposited:	26 Aug 2025 04:06
Last Modified:	26 Aug 2025 04:06
URI:	http://repository.unsri.ac.id/id/eprint/183263

Actions (login required)

View Item