Multifilters-Based Unsupervised Method for Retinal Blood Vessel Segmentation

Fundus imaging is one of the crucial methods that help ophthalmologists for diagnosing various eye diseases in modern medicine. An accurate vessel segmentation method can be a convenient tool to foresee and analyze fatal diseases, including hypertension or diabetes, which damage retinal vessel’s appearance. This work suggests an unsupervised approach vessels out images. The proposed includes multiple steps. Firstly, from colored image, green channel extracted preprocessed utilizing Contrast Limited Histogram Equalization as well Fuzzy Based contrast enhancement. To expel geometrical articles (macula, optic disk) noise, top-hat morphological operations are used. On resulted enhanced matched filter Gabor wavelet applied, outputs both added extract pixels. resulting image with now noticeable blood binarized using human visual system (HVS). A final segmented obtained by applying post-processing. suggested assessed on two public datasets (DRIVE STARE) showed comparable results regard sensitivity, specificity accuracy. we achieved respect together accuracy DRIVE database 0.7271, 0.9798 0.9573, STARE these 0.7164, 0.9760, 0.9560, respectively, less than 3.17 s average per image.