SISTEM PENGENALAN EKSPRESI CIRI WAJAH MATA DAN HIDUNG MENGGUNAKAN METODE WEIGHTLESS NEURAL NETWORK IMMEDIATE SCAN PADA MOBILE ROBOTIC

FAHALAFI, MOHAMMAD REZA and Zarkasi, Ahmad (2025) SISTEM PENGENALAN EKSPRESI CIRI WAJAH MATA DAN HIDUNG MENGGUNAKAN METODE WEIGHTLESS NEURAL NETWORK IMMEDIATE SCAN PADA MOBILE ROBOTIC. Undergraduate thesis, Sriwijaya University.

[thumbnail of RAMA_56201_09011282025088_cover.jpg]
Preview
 Image
RAMA_56201_09011282025088_cover.jpg - Accepted Version
Available under License Creative Commons Public Domain Dedication.
Download (319kB) | Preview
[thumbnail of RAMA_56201_09011282025088.pdf] Text
RAMA_56201_09011282025088.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (12MB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_TURNITIN.pdf] Text
RAMA_56201_09011282025088_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (4MB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_0225087902_01_front_ref.pdf] Text
RAMA_56201_09011282025088_0225087902_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.
Download (3MB)
[thumbnail of RAMA_56201_09011282025088_0225087902_02.pdf] Text
RAMA_56201_09011282025088_0225087902_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (1MB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_0225087902_03.pdf] Text
RAMA_56201_09011282025088_0225087902_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (1MB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_0225087902_04.pdf] Text
RAMA_56201_09011282025088_0225087902_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (3MB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_0225087902_05.pdf] Text
RAMA_56201_09011282025088_0225087902_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (195kB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_0225087902_06_ref.pdf] Text
RAMA_56201_09011282025088_0225087902_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (182kB) | Request a copy
[thumbnail of RAMA_56201_09011282025088_0225087902_07_lamp.pdf] Text
RAMA_56201_09011282025088_0225087902_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (3MB) | Request a copy

Abstract

Facial expression recognition is a branch of digital image processing widely applied in various fields, such as security, human-robot interaction, and artificial intelligence systems. However, the main challenge in implementing facial recognition systems on embedded and mobile robotic devices lies in computational power and memory constraints. Therefore, this study develops a facial expression recognition system using the Weightless Neural Network (WNN) Immediate Scan method, designed for high efficiency on resource-limited devices. The system is built using Raspberry Pi 4 and Arduino Mega 2560, with a Pi Camera as the image acquisition sensor. The image processing workflow consists of several key steps: face detection using Haar Cascade Classifier, feature extraction of the eyes and nose, image conversion into a binary representation, and pattern recognition with the Weightless Neural Network Face Recognition Algorithm (WNN-FRA). The training dataset comprises 75 images with three different expressions (neutral, happy, surprised), and system evaluation is conducted using a random split method to compare its performance. Experimental results indicate that the system achieves an average accuracy of 93,16%, with the highest performance observed for neutral expressions 97,58%, highest happy expressions at 96,62%, and highest surprised expressions at 96,97%. The implementation of the WNN Immediate Scan method enables facial expression recognition with lower power consumption and computational time than conventional methods such as Convolutional Neural Networks (CNN). This study concludes that the Weightless Neural Network Immediate Scan method can be effectively applied to facial expression recognition systems for mobile robotic devices with limited computational resources. Future research may explore hybrid approaches combining WNN with deep learning techniques, expanding the dataset with additional facial expressions, and optimizing image processing parameters to enhance recognition accuracy. Keywords: Facial Expression Recognition, Weightless Neural Network, Immediate Scan, Mobile Robotic, Raspberry Pi, Arduino Mega.

Item Type: Thesis (Undergraduate)
Subjects: Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
T Technology > T Technology (General) > T58.5-58.64 Information technology > T58.5 General works Management information systems Cf. HD30.213 Industrial management Cf. HF5549.5.C6+ Communication in personnel management Cf. TS158.6 Automatic data collection systems (Production control)
T Technology > T Technology (General) > T59.4 Mechanization
T Technology > T Technology (General) > T59.5 Automation
Divisions: 09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User: MOHAMMAD REZA FAHALAFI
Date Deposited: 21 May 2025 07:31
Last Modified: 21 May 2025 07:31
URI: http://repository.unsri.ac.id/id/eprint/173477

Actions (login required)

View Item View Item