International Journal of Pharma and Bio Sciences
ijpbs.net
editorijpbs@rediffmail.com (or) editorofijpbs@yahoo.com (or) prasmol@rediffmail.com
10.22376/ijpbs.2019.10.1.p1-12
Volume 5 Issue 3
2014 (July- September)
LOCATION OF FOVEA CENTRALIS IN DIGITAL FUNDUS IMAGES USING ADAPTIVE THRESHOLDING METHOD
In the field of biomedical, one of the most significant research area is pathology identification in human eyes. In the fundus image analysis, fovea detection plays a vital role in screening Diabetic Retinopathy. Blood vessels that originate from the optic disc may leak fluid which leads to microaneurysms and deposit in the fovea region. Detecting the fovea center helps in identifying the presence of microaneurysms, exudates, and hemorrhages which causes diabetic retinopathy. Exhaustive survey of literature shows that Wellner's adaptive thresholding method has not been applied to retinal images to detect Diabetic Retinopathy. In this paper, Wellner's adaptive thresholding method is used to locate the center of fovea for early detection of Diabetic Retinopathy. This method uses local threshold values that separate the foreground from the background region which helps in identifying the fovea region more clearly than other method. Both the healthy and pathological images were analysed for the detection of fovea center. The experiment was carried out with 1200 fundus images from MESSIDOR database. In which 660 images were pathological and 540 images were healthy retina. It is found that the proposed method has an improvement of 3.1 percent and an accuracy of 99.25 percent in detecting the fovea center compared to the existing methods.
E. NANDHINI AND S.R. MALATHI
Diabetic Retinopathy, Optic Disc, Exudates, Microaneurysms, Fovea, Macula.
590-600