|
چکیده
|
Extraction of blood vessels in retinal images is helpful for ophthalmologists to screen many medical disorders. The changes in the retinal vessels due to the pathologies can be easily identified by the retinal vessels segmentation. Therefor in this paper, we propose an automatic method to extract the blood vessels from various normal and abnormal retinal images. Our proposed method uses the advantages of the optimal Gabor filter and morphological reconstruction to employ robust performance analysis to evaluate the accuracy and sensitivity. Moreover, unsharp filter is used which sharpens the edges of the vessels without increasing noise. The results illustrate the superior performance of the proposed algorithm when compared to the other existing various vessel segmentation methods. For this purpose, we use images provided from two public databases, DRIVE and STARE. Our proposed algorithm proof its better performance by achieving the greatest accuracy, sensitivity and specificity values, 95.21, 74.00 and 97.26 for the DRIVE database and 95.37, 77.66 and 96.80 for the STARE database, respectively
|