2.2.1 High-resolution fundus (HRF)

The database contains three image sets: 15 images of healthy patients, 15 images of patients with diabetic retinopathy, and 15 images of glaucomatous patients. All images are captured using a fundus camera CANON CF-60UVi equipped with CANON EOS 20D digital cameras; the capture angle is 60° (FOV). The resolution is 3504 2336 pixels. In addition, this database provided binary gold standard vessel segmentation for each image. Also the masks determining the field of view (FOV) are provided for particular datasets [25]. Figure 2 shows an example of an image from the database.

Figure 1. 3D fractal reconstruction algorithm.

Figure 2.

Example fundus image from high-resolution fundus (HRF). Database: (a) raw image 02\_h.jpg and (b) the manual segmentation of the blood vessels 02\_h.tif.
