Technique automatic for detection and segmentation of optic disc area in retinal image

Erwin, Erwin (2020) Technique automatic for detection and segmentation of optic disc area in retinal image. In: Journal of Physics: Conference Series: FIRST 2019.

[thumbnail of Erwin_2020_J._Phys.__Conf._Ser._1500_012089.pdf]
Preview
Text
Erwin_2020_J._Phys.__Conf._Ser._1500_012089.pdf

Download (901kB) | Preview

Abstract

Optic disc (OD) is a part that must be observed in diagnosing retinal disease. In this paper an approach will be proposed for segmentation and detection of the OD area. Segmentation and detection are important steps in detecting a retinal disease automatically. Segmentation in this study was carried out using the Luminance Channel and Active Contours method, and for OD area detection was carried out using the Averaging Filter method and Image Histogram. Before segmenting and detecting the OD area begins with pre-processing as an increase in image quality and eliminating blood vessels. The detection results obtained an accuracy of 90% and 97.50% using the STARE and DRIVE dataset, while segmentation uses the DRIVE dataset with the acquisition of PSNR of 19.705 dB, accuracy of 86.45%, sensitivity 100% and specificity is 85.175%. The results of the research obtained can be used as a further step in diagnosing retinal disease.

Item Type: Conference or Workshop Item (Paper)
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: Dr Erwin Erwin
Date Deposited: 08 Dec 2020 04:11
Last Modified: 08 Dec 2020 04:11
URI: http://repository.unsri.ac.id/id/eprint/38393

Actions (login required)

View Item View Item