Paper Details

we describe an automated method to locate and outline blood vessels in images of the ocular fundus. Such atool should prove useful to eye care specialist for purpose of patient screening, treatment evaluation, and clinical study.Our method differs from previously known methods in that it use local and global vessel feature cooperatively to segmentthe vessel network. We evaluate our method using hand labeled ground truth segmentations of 20 images. A plot of theoperating characteristic shows that our method reduces false positives by as much as 15 times over basic thresholding ofa matched filter response (MFR), at up to a 75% true positive rate. For a baseline, we also compared the ground truthagainst a second hand – labeling, yielding a 90% true positive and a 4% false positive detection rate, on average. Thesenumbers suggest there there is still room for a 15% true positive rate improvement, with the same false positive rate,over our method. We are making all our images and hand labeling publicly available for interested researchers to use inevaluating related methods

Dhanashri Dilip Dere, Dr.V.N.Nitnaware, “Locating Blood vessel in Retinal Images by Piecewise Threshold Probing of a Matched Filter Response”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 7, pp. 233-241, July 2017.