The DiaRetDB1 is a public database for evaluating and benchmarking diabetic retinopathy detection algorithms. The database contains digital images of eye fundus and expert annotated ground truth for several well-known diabetic fundus lesions (hard exudates, soft exudates, microaneurysms and hemorrhages). The original images and the raw ground truth are both available. In addition to the data we also provide Matlab functionality (M-files) to read data (XML-files), fuse data of several experts and to evaluate detection methods.
This database is related to ImageRet project and the ground truth was collected using our ImgAnnoTool image annotation tool (contact Lasse Lensu for more information). For a more detailed description, see our documentation, please.
Authors The following authors have significantly contributed to the actual work of establishing and collecting the data and implementing the methods for the database: Tomi Kauppi, Valentina Kalesnykiene, Iiris Sorri, Asta Raninen, Raija Voutilainen, Joni Kamarainen, Lasse Lensu and Hannu Uusitalo.
|来源机构||芬兰拉普兰塔理工大学（Lappeenranta University of Technology，LUT）机器视觉与模式识别实验室|