PROJECT TITLE:

Iterative Vessel Segmentation of Fundus Images - 2015

ABSTRACT:

This paper presents a completely unique unsupervised iterative blood vessel segmentation algorithm using fundus images. 1st, a vessel enhanced image is generated by tophat reconstruction of the negative inexperienced plane image. An initial estimate of the segmented vasculature is extracted by global thresholding the vessel enhanced image. Next, new vessel pixels are identified iteratively by adaptive thresholding of the residual image generated by masking out the prevailing segmented vessel estimate from the vessel enhanced image. The new vessel pixels are, then, region grown into the existing vessel, thereby ensuing in an iterative enhancement of the segmented vessel structure. As the iterations progress, the amount of false edge pixels identified as new vessel pixels will increase compared to the number of actual vessel pixels. A key contribution of this paper may be a novel stopping criterion that terminates the iterative process leading to higher vessel segmentation accuracy. This iterative algorithm is strong to the speed of new vessel pixel addition since it achieves 93.two-95.35percent vessel segmentation accuracy with zero.9577-0.9638 area underneath ROC curve (AUC) on abnormal retinal pictures from the STARE dataset. The proposed algorithm is computationally efficient and consistent in vessel segmentation performance for retinal pictures with variations due to pathology, uneven illumination, pigmentation, and fields of view since it achieves a vessel segmentation accuracy of regarding ninety five% in a median time of 2.forty five, 3.ninety five, and eight s on images from 3 public datasets DRIVE, STARE, and CHASE_DB1, respectively. Additionally, the proposed algorithm has more than ninety% segmentation accuracy for segmenting peripapillary blood vessels in the pictures from the DRIVE and CHASE_DB1 datasets.