Abstract

In both ophthalmological and cardiovascular disease diagnosis, the accurate segmentation of the retinal vessel tree has become the prerequisite step for automatic or computer-aided diagnosis systems. Unlike typical foreground and background segmentation in normal image processing, there are three problems making the retinal vessel segmentation task even harder. First, the retinal color image tends to be red everywhere, thus having a lower contrast than normal image segmentation. Second, most retinal color images suffer from unbalanced illumination and make it harder to recognize background. Third, the symptom for retinopathy has unexpected color and shape, thus making it more difficult to separate vessel from noises. Under such circumstance, the research of the retinal blood vessel segmentation has brought much attention and been developed. In this work, the input is a retinal color image and the output will be a binary image of the vessel and non-vessel pixels. Preprocessing with CLAHE contrast enhancement, Vessel Extraction with Otsu thresholding is implemented along with performance parameters Accuracy, Sensitivity and Specificity. In the proposed work 20 images are used from the database called Digital Retinal Images for Vessel Extraction. The performance parameters Specificity, Sensitivity, Accuracy are evaluated.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.