python从文件夹中取一定数量图片

前面转自：https://www.jianshu.com/p/2777a74c7985

按比例从文件夹中取一定数量图片

# -*- coding: utf-8 -*- import os, random, shutil def moveFile(fileDir): pathDir = os.listdir(fileDir) # 取图片的原始路径 filenumber = len(pathDir) rate = 0.75 # 自定义抽取图片的比例 picknumber = int(filenumber * rate) # 按照rate比例从文件夹中取一定数量图片 sample = random.sample(pathDir, picknumber) # 随机选取picknumber数量的样本图片 print (sample) for name in sample: shutil.move(fileDir + name, tarDir + name) return if __name__ == '__main__': fileDir = "/mnt/NIvsCG/datasets/prcg_images/" # 源图片文件夹路径 tarDir = '/mnt/NIvsCG/datasets/train/' # 移动到新的文件夹路径 moveFile(fileDir)

随机选取一定数量的样本图片

import os, random, shutil data_base_dir = "/mnt/NIvsCG/temp/train/" # 源图片文件夹路径 tarDir = "/mnt/NIvsCG/random/train/" # 移动到新的文件夹路径 for file in os.listdir(data_base_dir): fileDir = data_base_dir + file + '/' # 取图片的原始路径 pathDir = os.listdir(fileDir) filenumber = len(pathDir) if filenumber > 30: picknumber = 30 # 所取图片数量 sample = random.sample(pathDir, picknumber) # 随机选取picknumber数量的样本图片 print(sample) for name in sample: shutil.copy(fileDir + name, tarDir + name)

按比例从文件夹中取一定数量图片，并在移动到新文件夹之前重命名选取的图片

import os, random, shutil def moveFile(fileDir): pathDir = os.listdir(fileDir) # 取图片的原始路径 filenumber = len(pathDir) rate = 0.75 # 自定义抽取图片的比例 picknumber = int(1000 * rate) # 按照rate比例从文件夹中取一定数量图片 sample = random.sample(pathDir, picknumber) # 随机选取picknumber数量的样本图片 count = 1 for namein sample: shutil.move(fileDir + name, tarDir + name) os.rename(os.path.join(tarDir, name), os.path.join(tarDir, str(count) + ".jpg")) count += 1 return if __name__ == '__main__': fileDir = "/mnt/NIvsCG/datasets/mobilephone_img/" # 源图片文件夹路径 tarDir = "/mnt/NIvsCG/datasets/train/" # 移动到新的文件夹路径 moveFile(fileDir)

最后按照上述方法写了自己想要的，按比例抽取文件夹内不同分辨率的图片，并移动到新的文件夹中

# -*- coding: utf-8 -*- """ Created on Wed Oct 30 16:16:02 2019 @author: Administrator """ import math import os, random, shutil import os.path as osp from scipy import misc import collections data_base_dir = "D:/Cargo" # 源图片文件夹路径 tarDir = "D:/Cargo_randomselect" # 移动到新的文件夹路径 if not osp.exists(tarDir): os.mkdir(tarDir) resolution_list=[] resolution=[] filelist = os.listdir(data_base_dir) for file in filelist: if file.endswith('tif'): fileDir = data_base_dir + '/' + file # 取图片的路径 picshape = misc.imread(fileDir).shape resolution_list.append(str(picshape[0])+'-'+file) resolution.append(picshape[0]) statistical_result = collections.Counter(resolution) for _, res in enumerate(statistical_result.keys()): temp=[] for tif in resolution_list: if tif.split('-')[0] == str(res): temp.append(tif.split('-')[1]) if res< 30: sample = temp elif res>30 and res<42: picknumber = math.ceil(0.6 * statistical_result[res]) sample = random.sample(temp, picknumber) elif res>42: picknumber = math.ceil(0.25 * statistical_result[res]) sample = random.sample(temp, picknumber) for name in sample: src = data_base_dir + '/' + name#待复制文件路径 dst = tarDir+ '/' + name#目标路径 shutil.copy(src,dst) print('selecting resolution {} picture'.format(res))