Skip to main content
. 2016 Sep 29;2016:6838976. doi: 10.1155/2016/6838976

Table 1.

Databases for fundus images.

Name Image acquisition Number of images Resolution Uses
DRIVE [8] 3-CCD camera with 45-fold view 20 color fundus testing images.
20 color fundus training images.
768 × 584 Exudates, hemorrhages, microaneurysms, and abnormal blood vessels detection

Image-Ret (DIARETB0, DIARETB1) [9] 50-fold view DIARETB0: total 130 images in which 20 images are normal and 110 with DR.
DIARETB1: total 89 images: five images are normal and 84 images with DR.
1500 × 1152 Exudates, hemorrhages, microaneurysms, and abnormal blood vessels detection

Messidor [10] 3CCD camera at 45-fold view 1200 images. 1440 × 960, 2240 × 1488, and 2304 × 1536 Exudates, hemorrhages, microaneurysms, and abnormal blood vessels detection

Retinopathy Online Challenge [11] Canon CR5-45-NM camera at 45-fold view 100 digital fundus images. 768 × 576, 1058 × 1061, and 1389 × 1383 Microaneurysms detection

e-ophtha_EX [12] OPHDIAT© Tele-medical network It contains 47 images with exudates and 35 images with no lesion. 2048 × 1360 Exudates detection

e-ophtha_MA [12] OPHDIAT© Tele-medical network It contains 148 images with microaneurysms or small hemorrhages and 233 images with no lesion. 2048 × 1360 Microaneurysms detection

STARE [13] Top con TRV 50 fundus camera at 35-fold view 400 images. 605 × 700 Exudates, hemorrhages, microaneurysms, and abnormal blood vessels detection

HEI-MED [7] Zeiss Visucam PRO fundus camera at 45-fold view 169 images in which 115 images are abnormal and 54 images are healthy. 2196 × 1958 Exudates detection