PENERAPAN METODE ENSEMBLE LEARNING PADA HASIL SEGNET DAN RESNET MENGGUNAKAN TEKNIK WEIGHTED VOTING UNTUK SEGMENTASI CITRA RETINA

SUEDARMIN, MUHAMMAD and Suprihatin, Bambang and Desiani, Anita (2023) PENERAPAN METODE ENSEMBLE LEARNING PADA HASIL SEGNET DAN RESNET MENGGUNAKAN TEKNIK WEIGHTED VOTING UNTUK SEGMENTASI CITRA RETINA. Undergraduate thesis, Sriwijaya University.

	Text RAMA_44201_08011381924105.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011381924105_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (7MB) | Request a copy
	Text RAMA_44201_08011381924105_0026017102_0011127702_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_44201_08011381924105_0026017102_0011127702_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (882kB) | Request a copy
	Text RAMA_44201_08011381924105_0026017102_0011127702_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (252kB) | Request a copy
	Text RAMA_44201_08011381924105_0026017102_0011127702_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011381924105_0026017102_0011127702_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (183kB) | Request a copy
	Text RAMA_44201_08011381924105_0026017102_0011127702_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (314kB) | Request a copy

Abstract

Detection of exudate on retinal images in an effort to diagnose Diabetic Retinopathy can be done by image segmentation using the Convolutional Neural Network (CNN). The CNN method has been developed in the medical world. Architectures that are often used to perform image segmentation include SegNet and ResNet. Several studies have proposed the Ensemble Learning method to combine the performance results of the several basic models used. In this study, the implementation of Ensemble Learning was carried out on the segmentation results of the SegNet and ResNet architectures using the Weighted Voting technique on retinal image datasets in detecting Diabetic Retinopathy. The stages of the research carried out included data collection, pre-processing, training of each basic model, implementing the Ensemble Learning Weighted Voting technique by selecting the largest weight in the predictions of the predictions of each model, evaluation, analysis and interpretation of the results, and drawing conclusions. The results of the study using retinal image data obtained an accuracy value of 99.06%, a sensitivity of 70.63%, a specificity of 99.71%, an F1- Score of 77.09% and an IoU of 62.73%. Based on the results obtained, the implementation of Ensemble Learning on the results of SegNet and ResNet segmentation using the Weighted Voting technique is able to properly segment exudate on retinal images.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	SegNet, ResNet, Ensemble Learning, Weighted Voting, Citra Retina
Subjects:	Q Science > QA Mathematics > QA1-939 Mathematics > QA1.T553 Mathematics--Periodicals. Computer science--Periodicals. Computer science.
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	Muhammad Suedarmin
Date Deposited:	21 Aug 2023 02:38
Last Modified:	21 Aug 2023 02:38
URI:	http://repository.unsri.ac.id/id/eprint/127515

Actions (login required)

View Item