Abstract
This paper proposes a new supervised method for blood vessel segmentation using Zernike moment-based shape descriptors. The method implements a pixel wise classification by computing a 11-D feature vector comprising of both statistical (gray-level) features and shape-based (Zernike moment) features. Also the feature set contains optimal coefficients of the Zernike Moments which were derived based on the maximum differentiability between the blood vessel and background pixels. A manually selected training points obtained from the training set of the DRIVE dataset, covering all possible manifestations were used for training the ANN-based binary classifier. The method was evaluated on unknown test samples of DRIVE and STARE databases and returned accuracies of 0.945 and 0.9486 respectively, outperforming other existing supervised learning methods. Further, the segmented outputs were able to cover thinner blood vessels better than previous methods, aiding in early detection of pathologies.
Highlights
Digital Eye fundus imaging in ophthalmology plays a significant role in the medical diagnosis of cardiovascular diseases and other pathological conditions such as diabetes and hypertension [1]
For detecting thin blood vessels, unsupervised Deep Neural Networks (DNN) were used by Maji et al [27], which employed a fusion of deep and ensemble learning for vessel classification via denoising auto-encoder proposed by Roy et al [28] used on just the trained retinal vascular patches
We explore a computationally simpler but more effective realization of blood vessel segmentation, inspired by the work of [35]
Summary
Digital Eye fundus imaging in ophthalmology plays a significant role in the medical diagnosis of cardiovascular diseases and other pathological conditions such as diabetes and hypertension [1]. For detecting thin blood vessels, unsupervised Deep Neural Networks (DNN) were used by Maji et al [27], which employed a fusion of deep and ensemble learning for vessel classification via denoising auto-encoder proposed by Roy et al [28] used on just the trained retinal vascular patches This approach achieved an accuracy of 93.27%. The novelty lines in; (a) use of a superior pre-processing technique to sharpen the blood vessels which were normally performed using a combination of Gaussian and Mean filter, (b) utilization of computationally inexpensive Higher Order, Orthogonal Zernike Moments which can accurately distinguish thinner blood vessel pixels from the background and (c) usage of efficient training set which takes into account all the possible manifestations of the blood vessel, covering the entire training set of the DRIVE database. The intense experimentation required for achieving this demonstrates the rigor of this work
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: 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.
