PENGEMBANGAN MODEL SEGMENTASI DAN ALIGMENT CITRA PRA KANKER SERVIKS MENGGUNAKAN ORIENTED BOUNDING BOX DAN YOU ONLY LOOK ONCE

SUSANTI, DWI and Nurmaini, Siti (2025) PENGEMBANGAN MODEL SEGMENTASI DAN ALIGMENT CITRA PRA KANKER SERVIKS MENGGUNAKAN ORIENTED BOUNDING BOX DAN YOU ONLY LOOK ONCE. Undergraduate thesis, Sriwijaya University.

	Image RAMA_56201_09011482326006_cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (11MB)
	Text RAMA_56201_09011482326006.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_56201_09011482326006_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (13MB) | Request a copy
	Text RAMA_56201_09011482326006_0002085908_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_56201_09011482326006_0002085908_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (729kB) | Request a copy
	Text RAMA_56201_09011482326006_0002085908_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011482326006_0002085908_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_09011482326006_0002085908_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (240kB) | Request a copy
	Text RAMA_56201_09011482326006_0002085908_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (241kB) | Request a copy
	Text RAMA_56201_09011482326006_0002085908_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (740kB) | Request a copy

Abstract

Cervical cancer is one of the leading causes of death in women worldwide, so early detection of precancerous lesions is crucial to reduce mortality. Conventional screening methods such as VIA and Pap Smear still have limited accuracy and risk producing false positives and false negatives. This study aims to develop a deep learning-based cervical image segmentation model using the You Only Look Once (YOLO) architecture, YOLOv8 and YOLOv11 versions. In addition, an image alignment approach using Oriented Bounding Box (OBB) was also applied to improve accuracy through rotation based on object orientation. Based on the evaluation results, the YOLOv11m-Seg model with rotation based on the cervical area class showed the best performance with an Intersection over Union (IoU) value of 0.532, a dice coefficient of 0.449, and a pixel accuracy of 93.4%. This model has the potential to support a more accurate and efficient cervical cancer screening system.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	YOLOv8, YOLOv11, Segmentasi Citra, Kanker Serviks, Deep Learning, Oriented Bounding Box, Alignment Citra
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:	Dwi Susanti
Date Deposited:	23 Jul 2025 04:49
Last Modified:	23 Jul 2025 04:49
URI:	http://repository.unsri.ac.id/id/eprint/180095

Actions (login required)

View Item