数据筛选---txt分割标签筛选包含特定类别的标签和img

语义分割标签处理时，标签数据中会包含很多杂乱的标签，此时对数据作筛选，选出包含特定类别的标签会很重要。

一、txt标签文件中筛选
choice_txt.py

# encoding: utf-8

import os.path
import shutil
import re

# 遍历指定目录，显示目录下的所有文件名
def eachFile(filepath):
    pathDir = os.listdir(filepath)
    for allDir in pathDir:
        child = os.path.join('%s\%s' % (filepath, allDir))
        if os.path.isfile(child):
            readFile(child) 
            continue
        eachFile(child)

# 遍历出结果 返回文件的名字
def readFile(filenames):
    f = open(filenames, 'r')  
    line = f.readline()  # 以行的形式进行读取文件
    list1 = []
    while line:
        a = line.split()
        str = a[0]   # 读取该行的第一个数据
        if str == '0' or '1' or '9' or '11' or '12':   
            list1.append(filenames)
            shutil.copy(filenames, 'H:\mtl2fs\multyfs2single\label0')
            print("匹配到:" + filenames)
        line = f.readline()
    f.close()

if __name__ == "__main__":
    filenames = 'H:\mtl2fs\multyfs2single\label'  
    arr = []
    eachFile(filenames)

二、根据txt筛选结果挑选img
choice_img.py

# -*- coding: utf-8 -*-

import os
import shutil

# root_dir = "E:/multyfs2single/"  #根目录
# txt_dir = root_dir + "label0/"  #标签
# image_dir = root_dir + "img0/"  #结果存放地址
# origin_image_dir = "img"  #原始图片地址
txt_dir = "H:/mtl2fs/multyfs2single/label0/"
image_dir = "H:/mtl2fs/multyfs2single/img0/"
origin_image_dir = "H:/mtl2fs/multyfs2single/img"
count = 0
# train_data_list = os.listdir(images_dir)
for file in os.listdir(txt_dir):
    # count += 1
    # if file not in train_data_list:
    print(count, " ", file)
    shutil.copy(origin_image_dir + '/' + os.path.splitext(file)[0] + '.png',
                image_dir + '/' + os.path.splitext(file)[0] + '.png')