Cauchy Based Matched Filter for Retinal Vessels Detection

Abstract

In this paper, a novel matched filter based on a new kernel function with Cauchy distribution is introduced to improve the accuracy ofthe automatic retinal vessel detection compared with other available matched filter‑based methods, most notably, the methods builton Gaussian distribution function. Several experiments are conducted to pick the best values of the parameters for the new designedfilter, including both Cauchy function parameters as well as the matched filter parameters such as the threshold value. Moreover,the thresholding phase is enhanced with a two‑step procedure. Experimental results employed on DRIVE retinal images databaseconfirms that the proposed method has higher accuracy compared with other available matched filter‑based methods.