Retinal vessel segmentation with Slime-Mould-Optimization based Multi-Scale-Matched-Filter

Even though a number of sensory organs are existing, eye plays a necessary role in the sensory system; which converts the incident light into meaningful visual information. If any abnormality arises in eye, then the whole sensory system gets stressed. The disease in eye is due to injury, infection and ageing and the untreated eye disease will lead to vision loss. The proposed research aims to propose a Computer-Aided-Procedure (CAP) to extract the blood-vessel section from Digital-Fundus-Image (DFI). In order to accomplish this task, a Multi-Scale-Matched-Filter (MSMF) is designed using the Slime-Mould-Optimization algorithm. In this work, the necessary test images are collected from the benchmark DRIVE and CHASE_DB1 dataset. After extracting the blood-vessel using the MSMF, an examination among extracted vessel and the Ground-Truth (GT) image is executed and the Image-Performance-Values are separately computed for each database. The attained result with this CAP confirms that the attained Jaccard, Dice and Accuracy of proposed approach is better compared to similar existing approaches in the literature.