graviti
产品服务
解决方案
知识库
公开数据集
关于我们
Resized 2015 & 2019 Blindness Detection Images
许可协议: CC-BY-SA 4.0

Overview

This dataset contains images from the 2015 Diabetic Retinopathy Detection and APTOS 2019 Blindness Detection competitions. Each image has been resized and cropped to have a maximum size of 1024px. The code for resizing & cropping can be found here, and is based off of ilovescience's code.

For the 2015 images, a left and right field is provided for every subject. Images are labeled with a subject id as well as either left or right (e.g. 1_left.jpeg is the left eye of patient id 1). The 2019 images do not specify a subject or eye.

A clinician has rated each image for the severity of diabetic retinopathy on a scale of 0 to 4:

> 0 - No DR

> 1 - Mild

> 2 - Moderate

> 3 - Severe

> 4 - Proliferative DR

Like any real-world data set, you will encounter noise in both the images and labels. Images may contain artifacts, be out of focus, underexposed, or overexposed. The images were gathered from multiple clinics using a variety of cameras over an extended period of time, which will introduce further variation.

Files

labels.zip - contains the 2015 train and test labels, the 2019 train labels, and the 2019 public test image names

resized train 15.zip - the 2015 training set images, resized and cropped

resized test 15.zip - the 2015 test set images, resized and cropped

resized train 19.zip - the 2019 training set images, resized and cropped

resized test 19.zip - the 2019 public test set images, resized and cropped

数据概要
数据格式
image,
数据量
94.296K
文件大小
2.17GB
发布方
Benjamin Warner
| 数据量 94.296K | 大小 2.17GB
Resized 2015 & 2019 Blindness Detection Images
许可协议: CC-BY-SA 4.0

Overview

This dataset contains images from the 2015 Diabetic Retinopathy Detection and APTOS 2019 Blindness Detection competitions. Each image has been resized and cropped to have a maximum size of 1024px. The code for resizing & cropping can be found here, and is based off of ilovescience's code.

For the 2015 images, a left and right field is provided for every subject. Images are labeled with a subject id as well as either left or right (e.g. 1_left.jpeg is the left eye of patient id 1). The 2019 images do not specify a subject or eye.

A clinician has rated each image for the severity of diabetic retinopathy on a scale of 0 to 4:

> 0 - No DR

> 1 - Mild

> 2 - Moderate

> 3 - Severe

> 4 - Proliferative DR

Like any real-world data set, you will encounter noise in both the images and labels. Images may contain artifacts, be out of focus, underexposed, or overexposed. The images were gathered from multiple clinics using a variety of cameras over an extended period of time, which will introduce further variation.

Files

labels.zip - contains the 2015 train and test labels, the 2019 train labels, and the 2019 public test image names

resized train 15.zip - the 2015 training set images, resized and cropped

resized test 15.zip - the 2015 test set images, resized and cropped

resized train 19.zip - the 2019 training set images, resized and cropped

resized test 19.zip - the 2019 public test set images, resized and cropped

0
立即开始构建AI
graviti
wechat-QR
长按保存识别二维码,关注Graviti公众号

Copyright@Graviti
沪ICP备19019574号
沪公网安备 31011002004865号