The DR HAGIS database has been created to aid the development of vessel extraction algorithms suitable for retinal screening programmes. Contrary to most available databases, it takes into account the variety of images that have to be processed across different screening centres.
This database therefore consists of 39 high-resolution images divided into four comorbidity subgroups (AMD, DR, glaucoma, and hypertension) and reflects the range of image resolutions, digital cameras and fundus cameras used in clinics (Topcon TRC-NW6s, Topcon TRC-NW8 or a Canon CR DGi; the images are 4752x3168 pixels, 3456x2304 pixels, 3126x2136 pixels, 2896x1944 pixels or 2816x1880 pixels in size).
The fundus images are saved as compressed JPEG files with 8 bits per colour plane. The ground truth and mask images are saved as binary PNG files.
The database is available at: https://personalpages.manchester.ac.uk/staff/niall.p.mcloughlin/
Researchers are encouraged to test their segmentation algorithms using this database.