PENERAPAN AUGMENTASI DATA DAN PRINCIPAL COMPONENT ANALYSIS PADA PENENTUAN PENYAKIT DIABETIC RETINOPATHY MENGGUNAKAN ALGORITMA RECURRENT CONVOLUTIONAL NEURAL NETWORK

UKHROWI, ZA'IM and Desiani, Anita and Irmeilyana, Irmeilyana (2021) PENERAPAN AUGMENTASI DATA DAN PRINCIPAL COMPONENT ANALYSIS PADA PENENTUAN PENYAKIT DIABETIC RETINOPATHY MENGGUNAKAN ALGORITMA RECURRENT CONVOLUTIONAL NEURAL NETWORK. Undergraduate thesis, Sriwijaya University.

	Text RAMA_44201_08011281722035.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_44201_08011281722035_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (13MB) | Request a copy
Preview	Text RAMA_44201_08011281722035_0011127702_0017057403_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (793kB) | Preview
	Text RAMA_44201_08011281722035_0011127702_0017057403_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (768kB) | Request a copy
	Text RAMA_44201_08011281722035_0011127702_0017057403_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (238kB) | Request a copy
	Text RAMA_44201_08011281722035_0011127702_0017057403_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (836kB) | Request a copy
	Text RAMA_44201_08011281722035_0011127702_0017057403_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8kB) | Request a copy
	Text RAMA_44201_08011281722035_0011127702_0017057403_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (148kB) | Request a copy

Abstract

Diabetic Retinopathy (DR) is one of the biggest health problems in the world. DR disease classification process needs to be done early to prevent the occurrence of DR. Recurrent Convolutional Neural Network (RCNN) is a deep learning-based classification algorithm that can classify image data accurately but requires a large amount of image data. Data augmentation can be applied to increase the amount of image data so that training data, validation data, and testing data are obtained as many as 15.000 images, 2.000 images, and 2.000 images. In addition, the large data dimensions make the classification performance not very efficient. Principal Component Analysis (PCA) is a method that can operate dimension reduction in image data so that the cumulative proportion of variance is approximately 99% with the number of principal components is 70. The purpose of this research is to apply data augmentation and PCA method to retinal image data for classifying DR disease using RCNN algorithm. The research methods used are data collection, image enhancement, data augmentation, implementation of PCA and RCNN algorithm, data training, data testing, evaluation, and analysis of results. Classification results were measured by calculating the values of accuracy, sensitivity, specificity, F1-Score, and Cohens Kappa with the values obtained as many as 94,55%, 89,1%, 100%, 94,23%, and 0,891. Based on the classification results obtained, it can be concluded that application of data augmentation and PCA can still provide good classification performance to classify DR disease using RCNN on retinal images.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Deep Learning, Diabetic Retinopathy
Subjects:	Q Science > QA Mathematics > QA8.9-QA10.3 Computer science. Artificial intelligence. Computational complexity. Data structures (Computer scienc. Mathematical Logic and Formal Languages
Divisions:	08-Faculty of Mathematics and Natural Science > 44201-Mathematics (S1)
Depositing User:	ZA'IM UKHROWI
Date Deposited:	07 Apr 2022 02:12
Last Modified:	07 Apr 2022 02:12
URI:	http://repository.unsri.ac.id/id/eprint/68494

Actions (login required)

View Item