PENGENALAN WAJAH PADA PLATFORM EMBEDDED MELALUI IDENTIFIKASI MATA DAN HIDUNG MENGGUNAKAN METODE WEIGHTLESS NEURAL NETWORK

FITRIYANTO, MEGI and Zarkasi, Ahmad (2024) PENGENALAN WAJAH PADA PLATFORM EMBEDDED MELALUI IDENTIFIKASI MATA DAN HIDUNG MENGGUNAKAN METODE WEIGHTLESS NEURAL NETWORK. Undergraduate thesis, Sriwijaya University.

	Text RAMA_56201_09011281924056.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_56201_09011281924056_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (12MB) | Request a copy
	Text RAMA_56201_09011281924056_0225087902_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (830kB)
	Text RAMA_56201_09011281924056_0225087902_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (740kB) | Request a copy
	Text RAMA_56201_09011281924056_0225087902_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (798kB) | Request a copy
	Text RAMA_56201_09011281924056_0225087902_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011281924056_0225087902_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (64kB) | Request a copy
	Text RAMA_56201_09011281924056_0225087902_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (82kB) | Request a copy
	Text RAMA_56201_09011281924056_0225087902_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

Technological advances in the field of Computer Vision are increasingly complex, especially in research and industrial needs. Computer Vision allows computers to process and recognize images with a level of accuracy close to human capabilities. The purpose of this research is to develop a face recognition system that previously ran on a microcomputer, so that it can run on a microcontroller with limited memory. With this development, face recognition can be implemented in embedded systems. Weightless Neural Networks (WNN) is the method used in face recognition in this research. This method uses face data at a binary level and for binary recognition. Moreover, the sample face data in binary form is compared with the primary face data obtained from a particular camera or image. The dataset that will be created is 10 photos of the author's own face with a frame width of 110 x 110 to 90 x 90. Furthermore, each face photo will be processed by taking the eye and nose area and saving it into an image file. When the camera captures the image in real time, the Viola-Jones algorithm will perform preprocessing. When the face is detected, the size and position of the face frame will be calculated and the DC motor will move to adjust the face position of the frame. The face frame will be detected for both eyes and nose. Then, both images will be converted into binary format. Binary data will be sent from Raspberry Pi to Arduino Mega via serial to continue the recognition process. Of the 10 faces belonging to researchers who were tested, the decision resulted in 8 recognizable faces and 2 faces that failed to be recognized because the highest eye similarity was only worth 83.08% and 84.09%. Then of the 10 faces belonging to the researcher's friend tested the system produces a decision not to be recognized. because it only produces a similarity level for the eyes around 70% even though the results of the similarity level for the nose are around 85%.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Pengenalan Wajah, Platform Embedded
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Megi Fitriyanto
Date Deposited:	20 Jan 2024 02:26
Last Modified:	20 Jan 2024 02:26
URI:	http://repository.unsri.ac.id/id/eprint/139027

Actions (login required)

View Item