MULTI DIRECTIONAL FACE RECOGNITION DENGAN METODE CNN (CONVOLUTIONAL NEURAL NETWORK)

HIDAYATULLAH, MUH. DANI and Fachrurrozi, Muhammad and Rachmatullah, Muhammad Naufal (2023) MULTI DIRECTIONAL FACE RECOGNITION DENGAN METODE CNN (CONVOLUTIONAL NEURAL NETWORK). Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021381924152.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
	Text RAMA_55201_09021381924152_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (10MB) | Request a copy
	Text RAMA_55201_09021381924152_0222058001_0001129204_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (600kB)
	Text RAMA_55201_09021381924152_0222058001_0001129204_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (395kB) | Request a copy
	Text RAMA_55201_09021381924152_0222058001_0001129204_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (364kB) | Request a copy
	Text RAMA_55201_09021381924152_0222058001_0001129204_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (475kB) | Request a copy
	Text RAMA_55201_09021381924152_0222058001_0001129204_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021381924152_0222058001_0001129204_06_ref.pdf - Bibliography Available under License Creative Commons Public Domain Dedication. Download (136kB)
	Text RAMA_55201_09021381924152_0222058001_0001129204_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy

Abstract

Face recognition is a field that has garnered significant attention in the development of artificial intelligence systems. In this research, we focus on face recognition from multiple viewpoints using the Convolutional Neural Network (CNN) method. Conventional approaches to face recognition often overlook variations in face position and orientation, leading to unsatisfactory performance in real-world scenarios. To address this challenge, we propose an approach that utilizes a convolutional neural network, specifically the Convolutional Neural Network (CNN), which has proven successful in various complex pattern recognition tasks. The proposed method consists of several stages. First, we collect a dataset that includes various variations in face position and orientation. This dataset encompasses rotated, tilted, and scaled faces. Next, we train the CNN model using the collected dataset. The training process involves hierarchical feature extraction using convolutional and pooling layers to recognize face patterns from multiple viewpoints. After training the CNN model, we conduct testing using an unseen test dataset consisting of previously unseen faces. We evaluate the model's performance based on commonly used face recognition metrics such as accuracy, precision, and recall. We also compare our model's performance with other face recognition methods in the literature. Our research findings demonstrate that the proposed CNN method successfully recognizes faces from multiple viewpoints with high accuracy. We also discover that the CNN model has an advantage in handling variations in face position and orientation compared to conventional methods. These results highlight the potential use of CNN methods in multi-view face recognition. This research has significant implications for the development of more advanced and reliable face recognition systems. Its findings can be applied in various practical applications, including security, surveillance, and individual identification in images or videos.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Recognition, Pengenalan Wajah dari Berbagai Sisi, CNN, Real-Time
Subjects:	T Technology > T Technology (General) > T1-995 Technology (General) > T11 General works > T11.5 Translating T Technology > T Technology (General) > T10.5-11.9 Communication of technical information > T11 General works > T11.5 Translating T Technology > T Technology (General) > T10.5-11.9 Communication of technical information > T11.5 Translating
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Muh. Dani Hidayatullah
Date Deposited:	01 Aug 2023 08:17
Last Modified:	01 Aug 2023 08:17
URI:	http://repository.unsri.ac.id/id/eprint/125062

Actions (login required)

View Item