p2oileen / conr_dataset Goto Github PK

import os,sys
import numpy as np
import cv2
import wget
   
input_dir = "../CoNR_Dataset_annotation_only/annotation"
output_dir = "../CoNR_Dataset_Image_only/image"
ann_file_lists = os.listdir(input_dir)
prefix = "https://cdn.donmai.us/original/"

for name in ann_file_lists:
    image_name = name.split('.')[0] + "." + name.split('.')[1]

    first = image_name[:2]
    second = image_name[2:4]
    image_url = prefix + first + "/" + second + "/" + image_name
    output_path = output_dir + "/" + image_name

    wget.download(image_url,out=output_path)