Please use this identifier to cite or link to this item:
Title: Generation of binary mask of retinal fundus image using bimodal masking
Authors: Garg M
Gupta S.
Keywords: Bimodal masking
Binary mask
Fundus image
Issue Date: 2018
Publisher: Institute of Advanced Scientific Research, Inc.
Abstract: Preprocessing is the important key in automatic diagnosis of various disorders related to retinal fundus images because it improves the localization and segmentation of various retinal image features. The processing of the noisy and surrounding areas in retinal fundus image is not necessary because lot of time is consumed at all stages of processing. Number of operations required for preprocessing can be minimized by focusing only on the retinal image feature region. To accomplish this task, binary mask of the fundus image is generated using proposed bimodal masking technique. The main parts of bimodal masking are RGB to gray conversion, histogram generation and gray to binary conversion using peaks and valleys of histogram of fundus image. The present study provides a simple and accurate method for the generation of binary mask of retinal fundus images. It can be helpful for the accurate extraction of vasculature map and other morphological attributes of the fundus image. Simulation is performed on the fundus images of DRIVE database using MATLAB software. Simulated results shows that the average Sensitivity, Specificity and Accuracy for DRIVE database comes out to be 99.46%, 99.77% and 99.56% respectively. � 2018, Institute of Advanced Scientific Research, Inc.. All rights reserved.
Appears in Collections:Journals

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.