PENGENALAN OBJEK SECARA REAL TIME MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS MODEL YOLOV4-TINY

GULTOM, JOSEP and Supardi, Julian and Rizqie, M. Qurhanul (2023) PENGENALAN OBJEK SECARA REAL TIME MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORKS MODEL YOLOV4-TINY. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021281924046.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021281924046_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (6MB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_55201_09021281924046_0010077210_0203128701_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (164kB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (277kB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (694kB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (286kB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (34kB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (98kB) | Request a copy
	Text RAMA_55201_09021281924046_0010077210_0203128701_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (71kB) | Request a copy

Abstract

Object recognition is a technology related to Computer Vision which can detect and recognize objects in digital images. However, the process of detecting and recognizing objects is something that is very difficult to do by a computer due to the many factors that can affect the shape of the object in a digital image. In this study, researchers will discuss object recognition in real time using convolution neural networks. The purpose of this study is to find out how the CNN algorithm works in detecting and recognizing objects in digital images in Real Time on Mobile Phone devices and also to find out the level of accuracy and Processing Time of the CNN model implemented on Mobile phone devices. The results of this study indicate that the YOLOv4-tiny CNN model can optimally process object recognition at a threshold of 0.5 with a Precision value of 0.93, Recall of 0.93, Accuracy of 0.86, F1-Score of 0.93, and mAP of 94.27%. The average Processing Time on the software in carrying out the Real Time object recognition process is 427ms.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Pengenalan Objek, Computer Vision, CNN, YOLOv4-tiny, Real Time, Mobile Phone
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation. Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76 Computer software
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Josep Gultom
Date Deposited:	04 Sep 2023 02:39
Last Modified:	04 Sep 2023 02:39
URI:	http://repository.unsri.ac.id/id/eprint/128146

Actions (login required)

View Item