DETEKSI OPTIK DISK PADA CITRA RETINA MENGGUNAKAN FASTER RCNN DENGAN BACKBONE RESNTE-50

RAVI, FAHRUL and Erwin, Erwin (2022) DETEKSI OPTIK DISK PADA CITRA RETINA MENGGUNAKAN FASTER RCNN DENGAN BACKBONE RESNTE-50. Undergraduate thesis, Sriwijaya University.

[thumbnail of RAMA_56201_09011281722053.pdf] Text
RAMA_56201_09011281722053.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (13MB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_0029017101_01_front_ref.pdf]
Preview
Text
RAMA_56201_09011281722053_0029017101_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.

Download (2MB) | Preview
[thumbnail of RAMA_56201_09011281722053_0029017101_02.pdf] Text
RAMA_56201_09011281722053_0029017101_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_0029017101_03.pdf] Text
RAMA_56201_09011281722053_0029017101_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (4MB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_0029017101_04.pdf] Text
RAMA_56201_09011281722053_0029017101_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (4MB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_0029017101_05.pdf] Text
RAMA_56201_09011281722053_0029017101_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (296kB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_0029017101_06_ref.pdf] Text
RAMA_56201_09011281722053_0029017101_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (340kB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_0029017101_07_lamp.pdf] Text
RAMA_56201_09011281722053_0029017101_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (2MB) | Request a copy
[thumbnail of RAMA_56201_09011281722053_TURNITIN.pdf] Text
RAMA_56201_09011281722053_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (2MB) | Request a copy

Abstract

The Optical Disc (OD) is the starting point of the optic nerve, the place where the optical fibers meet and convey information to the center and also carry more than 1 million neurons from the eye to the brain. Optical Disc (OD) Usually appears in bright yellowish areas. Optical Disk (OD) is detected using Deep Learning with Faster-RCNN. Faster R-CNN is one method that is often used in object detection. Faster R-CNN consists of a combination of the Fast R-CNN and the Region Proposal Network (RPN). In this study, there are 2 different data, namely original data and data after augmentation which will be used as comparisons both in terms of results and accuracy. This study focuses on the accuracy results issued by the Faster RCNN with Backbone Resnet-50 with the object of detection, namely Optical Disk (OD). Of the 2 models tested, the best model is obtained using data that has been augmented with Backbone Resnet-50 with batch size 4,learning rate 0.001, epoch 1000. The result of the mean average precision (MAP) obtained with the best model is 87,129%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Optik Disk(OD), Faster RCNN
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: Fahrul Ravi
Date Deposited: 09 Aug 2022 08:02
Last Modified: 09 Aug 2022 08:02
URI: http://repository.unsri.ac.id/id/eprint/76950

Actions (login required)

View Item View Item