EKSTRAKSI CITRA RETINA UNTUK MENDAPATKAN CITRA BINER PEMBULUH DARAH RETINA MENGGUNAKAN DYNAMIC THRESHOLD DENGAN SAUVOLA THRESHOLD DAN FILTER PENINGKATAN CITRA

KIYATMOKO, TOMI and Erwin, Erwin (2019) EKSTRAKSI CITRA RETINA UNTUK MENDAPATKAN CITRA BINER PEMBULUH DARAH RETINA MENGGUNAKAN DYNAMIC THRESHOLD DENGAN SAUVOLA THRESHOLD DAN FILTER PENINGKATAN CITRA. Undergraduate thesis, Sriwijaya University.

Preview	Text RAMA_56201_09011381520076_0029017101_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB) | Preview
	Text RAMA_56201_09011381520076_0029017101_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (568kB) | Request a copy
	Text RAMA_56201_09011381520076_0029017101_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (511kB) | Request a copy
	Text RAMA_56201_09011381520076_0029017101_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011381520076_0029017101_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (214kB) | Request a copy
	Text RAMA_56201_09011381520076_0029017101_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (349kB) | Request a copy
	Text RAMA_56201_09011381520076_0029017101_07_lamp.pdf - Supplemental Material Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (680kB) | Request a copy

Abstract

Retinal blood reversal becomes important in determining subsequent processes such as detection, determining and classification. Therefore, the extraction of the retinal image is a key step in the process of computerized diagnosis of shapes and sizes. For this reason, retinal image extraction to obtain a binary image of blood in this final project uses Dynamic Threshold with Sauvola Threshold also supported by Image Enhancement Filters. In the process, gamma correction is performed, then it is improved by using Contrast Limited Adaptive Histogram Equalization, followed by Image Enhancement Filters as refiners and sharpeners, then retinal image segmentation using Dynamic Threshold with Sauvola Threshold is used with morphology closing, and median filter. Dynamic Threshold achieves performance results on DRIVE with an accuracy of 74,94% and STARE with an accuracy of 61,66 %. Sauvola Threshold achieves performance results on DRIVE with an accuracy of 83,54 % and STARE with an accuracy of 71,56 %. Pembuluh darah retina menjadi peran penting dalam menentukan keakuratan proses selanjutnya seperti deteksi, identifikasi dan klasifikasi. Oleh karena itu ekstraksi citra retina merupakan langkah kunci untuk proses diagnosis terkomputerisasi dari bentuk dan ukuran. Untuk itu ektraksi citra retina untuk mendapatkan citra biner pembuluh darah pada tugas akhir ini menggunakan Dynamic Threshold dengan Sauvola Threshold serta didukung oleh Filter Peningkatan Citra. Dalam prosesnya dilakukan Koreksi gama, lalu ditingkatkan kontrasnya menggunakan Contrast Limited Adaptive Histogram Equalization, diikuti Filter Peningkatan Citra sebagai penghalus dan penajam citra kemudian segmentasi citra retina menggunakan Dynamic Threshsold dengan Sauvola Threshold diikuti dengan morfologi closing, dan median filter. Dynamic Threshold mencapai Performa hasil pada DRIVE dengan akurasi 74,94 % dan STARE dengan akurasi 61,66 %. Sauvola Threshold mencapai Performa hasil pada DRIVE dengan akurasi 83,54 % dan STARE dengan akurasi 71,56 %

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Segmentation, Retinal Blood Vessels, Local Threshold, Dynamic Threshold, Image Enhancement Filter.
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation. R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Users 2824 not found.
Date Deposited:	31 Oct 2019 02:16
Last Modified:	31 Oct 2019 02:16
URI:	http://repository.unsri.ac.id/id/eprint/14059

Actions (login required)

View Item