在yolov5训练之前可能要用到的几个脚本

1.img_label对比脚本

# -*-coding:utf-8-*-
# @编译环境：windows 10 + python3.8
# @IDE：Pycharm2021.1.3
import glob

list1 = glob.glob('D:/Desktop/test/images/*g')
list2 = glob.glob('D:/Desktop/test/xml/*l')
s1 = [None] * len(list1)
s2 = [None] * len(list2)

for i in range(len(list1)):
    s1[i] = list1[i].replace('D:/Desktop/test/images\\', '')
    s1[i] = s1[i].replace('.jpg', '')

for i in range(len(list2)):
    s2[i] = list2[i].replace("D:/Desktop/test/xml\\", '')
    s2[i] = s2[i].replace('.xml', '')

print(set(s2) - set(s1))

2.判断文本文件是否为空_脚本

# -*-coding:utf-8-*-
# @编译环境：windows 10 + python3.8
# @IDE：Pycharm2021.1.3
import os, glob

list = glob.glob(r'D:\Python\pycharm project\yolov5\PCB\train\labels\*t')
print(len(list))

for i, l in enumerate(list):
    with open(l, 'r') as f:
        f.seek(0)

        if f.read() == '':
            print(l)

3.多文件夹图片标签同时重命名

# -*-coding:utf-8-*-
# @编译环境：windows 10 + python3.8
# @IDE：Pycharm2021.1.3
import os, sys  # 导入模块


def rename_subfolders(s, path1, path2, c):  # 定义函数名称
    num = 1
    old_names = os.listdir(path1)  # 取路径下的文件名，生成列表
    print(f'{c:}= ' + str(len(old_names)))
    for old_name in old_names:  # 遍历列表下的文件名
        os.rename(os.path.join(path1, old_name), os.path.join(path1, f'{c:}_' + str(num)) + ".xml")
        old_name = old_name.replace(f'D:\Desktop\{s:}labels', '')
        old_name = old_name.replace('.xml', '.jpg')
        os.rename(os.path.join(path2, old_name), os.path.join(path2, f'{c:}_' + str(num)) + ".jpg")
        num = num + 1


if __name__ == '__main__':
    fs = ['文件夹1', '文件夹2', '文件夹3', '文件夹4', '文件夹5', '文件夹6', '文件夹7', '文件夹8', '文件夹9', '文件夹10']

    c = 0
    for s in fs:
        path1 = rf'D:\Desktop\{s:}labels'
        path2 = rf'D:\Desktop\{s:}'
        rename_subfolders(s, path1, path2, str(c))
        c += 1

xml to yolo 

# -*-coding:utf-8-*-
# @编译环境：windows 10 + python3.8
# @IDE：Pycharm2021.1.3

import os
import pickle
import xml.etree.ElementTree as ET
from os import listdir, getcwd
from os.path import join
import glob

classes = ['类名1', '类名2', '类名3', '类名4', '类名5', '类名6']


def convert(size, box):
    dw = 1. / size[0]
    dh = 1. / size[1]
    x = (box[0] + box[1]) / 2.0
    y = (box[2] + box[3]) / 2.0
    w = box[1] - box[0]
    h = box[3] - box[2]
    x = x * dw
    w = w * dw
    y = y * dh
    h = h * dh
    return (x, y, w, h)


def convert_annotation(image_name):
    in_file = open('./ANNOTATIONS/' + image_name[:-3] + 'xml')
    out_file = open('./LABELS/' + image_name[:-3] + 'txt', 'w')
    tree = ET.parse(in_file)
    root = tree.getroot()
    size = root.find('size')
    w = int(size.find('width').text)
    h = int(size.find('height').text)

    for obj in root.iter('object'):
        cls = obj.find('name').text
        if cls not in classes:
            print(cls)
            continue
        cls_id = classes.index(cls)
        xmlbox = obj.find('bndbox')
        b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text),
             float(xmlbox.find('ymax').text))
        bb = convert((w, h), b)
        out_file.write(str(cls_id) + ' ' + ' '.join([str(a) for a in bb]) + '\n')


wd = getcwd()

if __name__ == '__main__':
    for image_path in glob.glob('./IMAGES/*.jpg'):
        image_name = image_path.split('\\')[-1]
        convert_annotation(image_name)