Automated segmentation of optic disc using statistical region merging and morphological operations
- PDF / 1,346,206 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 36 Downloads / 233 Views
SCIENTIFIC PAPER
Automated segmentation of optic disc using statistical region merging and morphological operations K. S. Nija1,2 · C. P. Anupama1,2 · Varun P. Gopi3 · V. S. Anitha2,4 Received: 5 November 2019 / Accepted: 31 May 2020 © Australasian College of Physical Scientists and Engineers in Medicine 2020
Abstract Accurate Optic Disc (OD) segmentation is vital in designing systems that aid the diagnosis and evaluation of early phases of retinal diseases. However, in many images, the OD boundary is ambiguous, which makes the automated OD segmentation process very challenging. A method to segment OD based on statistical region merging and morphological operations is proposed in this paper. The proposed method is tested on standard databases MESSIDOR, DIARETDB1, DIARETDB0, and DRIONS-DB. The average overlap ratios are found to be 91.35% for DIARETDB1 images, 88.80% for DRIONS-DB images, 86.60% for DIARETDB0 images and 89.68% for MESSIDOR images, with average accuracies of 99.68%, 99.89%, 99.69%, and 99.93% respectively. A comparison with alternative methods showed that the proposed algorithm in OD segmentation is better than existing ones. Keywords Optic disc · Statistical region merging · Diabetic retinopathy · Morphological operations · OD segmentation
Introduction The Optic Disc (OD) is an area in the human eye, in which, the retinal nerve fibers converge to form the optic nerve [1]. Understanding the morphology and location of the OD in the retina is crucial in the diagnosis of retinal diseases like Agerelated Macular Degeneration (AMD), Diabetic Maculopathy (DM), and glaucoma [2–7]. The elliptical form of major retinal blood vessels, the circular edge of the OD, and the * Varun P. Gopi [email protected] K. S. Nija [email protected] C. P. Anupama [email protected] V. S. Anitha [email protected] 1
Department of Electronics and Communication Engineering, Government Engineering College Wayanad, Wayanad, India
2
APJ Abdul Kalam Technological University, Thiruvananthapuram, Kerala, India
3
Department of Electronics and Communication Engineering, National Institute of Technology, Tiruchirappalli, India
4
Department of Computer Science and Engineering, Government Engineering College Wayanad, Wayanad, India
local darkening at the fovea are used to obtain approximate locations of the OD in the eye [3]. In optical diagnostics, the shape and position of the OD are manually segmented by a doctor from the fundus images. Manual segmentation is tedious and subjective and can be prone to errors [8]. Computerassisted diagnosis and evaluation systems can conceivably support doctors execute OD segmentation [9] with greater diagnostic accuracy and efficiency. There have been many studies in recent years to develop automated methods to segment the OD. Some of them are discussed below: Walter et al. [10] proposed a method to identify diabetic retinal lesions and detected the OD through morphological filter and watershed transformation. The algorithm was tested on a small image data set using color spa
Data Loading...
