National conference on Digital Image and Signal Processing |
Foundation of Computer Science USA |
DISP2015 - Number 1 |
April 2015 |
Authors: Deepali D. Rathod, Ramesh R. Manza, Yogesh M. Rajput |
bb0a8204-fd64-4dec-a8bb-c5bb62e6702b |
Deepali D. Rathod, Ramesh R. Manza, Yogesh M. Rajput . Retinal Blood Vessels Extraction using Matched Filter on High Resolution Fundus Image Database. National conference on Digital Image and Signal Processing. DISP2015, 1 (April 2015), 15-22.
In this paper we have extracted the retinal blood vessels using 2D Matched filter. The proposed algorithm is consisting of some preprocessing steps on RGB image like extraction of green channel; contrast limited adaptive histogram equalization, and morphological opening operation. After extraction of retinal blood vessels using matched filter, we have calculated fractal dimension for finding the architectural distortion. Accuracy is calculated using person correlation coefficient, and achieved 98. 3 % accuracy. Operations are done on the 45 images of HRF database. 15 images are of Healthy, 15 images are of Diabetic retinopathy, and 15 are of Glaucoma.