更改xml文件脚本

很久没有更新博客了，今天更新点更改xml文件的脚本，也是为了自己能保存下来吧。
１.一个很全的可以更改xml文件中任意一行节点的内容，按照自己的需要选择注释掉什么，解注释哪些。可成批修改。

import os
import re

xml_dir="./Annotations1/"
xml_filenames=os.listdir(xml_dir)


for xml_name in xml_filenames:
    print(xml_dir+xml_name)
    xml_name_seperate=xml_name.split('.')[0]
    f=open(xml_dir+xml_name,'r+')
    xml_data=f.read()
    # xml_data = re.sub('\<folder>(.*?)\</folder>','<folder>JPEGImages</folder>', xml_data)
    # xml_data = re.sub('\<filename>(.*?)\</filename>', '<filename>'+xml_name_seperate+'</filename>', xml_data)
    # xml_data = re.sub('\<path>(.*?)\</path>',
    #                   '<path>/home/tsn/workspace/py-faster-rcnn/data/VOCdevkit2007/VOC2007/JPEGImages/' +
    #                   xml_name_seperate + '.jpg</path>', xml_data)
    #
    # xml_data = re.sub('\<name>(.*?)\</name>','<name>normal</name>', xml_data)
    xml_data = re.sub('\<width>(.*?)</width>','<width>1920</width>', xml_data)
    xml_data = re.sub('\<height>(.*?)</height>','<height>1080</height>', xml_data)
    # xml_data = re.sub('\<truncated>1</truncated>', '<truncated>0</truncated>', xml_data)
    f.seek(0) # set point to 0
    f.truncate() # clear all
    f.write(xml_data) # write all data
    f.close()

2.修改单个节点，修改哪个换成哪个。

#修改xml文件中的目标的名字，
import os, sys
import glob
from xml.etree import ElementTree as ET

# 批量读取Annotations下的xml文件
# per=ET.parse(r'C:\Users\rockhuang\Desktop\Annotations\000003.xml')
xml_dir = r'./VOC2018/Annotations/'
xml_list = glob.glob(xml_dir + '/*.xml')
for xml in xml_list:
    print(xml)
    per = ET.parse(xml)
    p = per.findall('/size')   #输入要更改的节点的名字

    for oneper in p:  # 找出person节点
        child = oneper.getchildren()[0]  # 找出person节点的子节点
        if child.text == '960':   #需要修改的名字
            child.text = '1152'    #修改成什么名字
   
    per.write(xml)
    print(child.tag, ':', child.text)

# coding=utf-8
import os
import os.path
import xml.dom.minidom

#获得文件夹中所有文件
FindPath = './Annotations1/'
FileNames = os.listdir(FindPath)
s = []
xml_path = './Annotations1/'
for file_name in FileNames:
    if not os.path.isdir(file_name):  # 判断是否是文件夹,不是文件夹才打开
        print (file_name)

    #读取xml文件
    dom = xml.dom.minidom.parse(os.path.join(FindPath,file_name))

    root = dom.documentElement

    # 获取标签对name之间的值
    name = root.getElementsByTagName('name')
    for i in range(len(name)):
        print(len(name))
        print(name[i].firstChild.data),
        if name[i] .firstChild.data== 'normal_bolt':
            name[i].firstChild.data = 'norbolt'
            print ('修改后的 name'),
            print (name[i].firstChild.data)
    #将修改后的xml文件保存
    with open(os.path.join(xml_path, file_name), 'w') as fh:
        dom.writexml(fh)
        print('write name/pose OK!')

import os
import re

# 设置为自己Annotations保存路径
_dir = "./VOC2018/Annotations/"
xmlList = os.listdir(_dir)
n = 0
for xml in xmlList:
    # f = open(_dir + xml, "r")
    f = open(_dir + xml, "r", encoding='utf-8')
    xmldata = f.read()
    # 设置为希望修改的path即可
    xmldata = re.sub('\<path>(.*?)\</path>',
                     '/home/tsn/workspace/py-faster-rcnn/data/VOCdevkit2007/VOC2007/JPEGImages/' + str(n).zfill(6) + '.jpg</path>', xmldata)
    f.close()
    f = open(_dir + xml, "w")
    f.write(xmldata)
    f.close()
    n += 1

以上三个都可以单个节点进行修改，可以任选其一使用。

３.对已经标注好的图片进行修改了size之后，对它的标注文件里的目标框的坐标也要进行修改。下面的代码就是按图片resize的比例成批修改x，y的值的。该例子中原大小为１１５２×８６４，更改成１９２０×１０８０

# -*- coding: utf-8 -*-
from xml.etree.ElementTree import ElementTree, Element

import xml.etree.ElementTree as ET
import os
import re

xml_dir="./Annotations/"
xml_dir1="./Annotations1/"
xml_filenames=os.listdir(xml_dir)


for xml_name in xml_filenames:
    tree = ET.parse(xml_dir+xml_name)
# tree = ET.parse('D:\python\venv1\output.xml')
    root = tree.getroot()
    print(root.text)
    # 遍历文件所有的tag 为目标的值得标签
    for elem in root.iter('xmin'):
        new_elem =int (int(elem.text) / 1152*1920)
        elem.text = str(new_elem)
    for elem in root.iter('ymin'):
        new_elem = int(int(elem.text) / 864 *1080)
        elem.text = str(new_elem)
    for elem in root.iter('xmax'):
        new_elem = int(int(elem.text) / 1152*1920)
        elem.text = str(new_elem)
    for elem in root.iter('ymax'):
        new_elem = int(int(elem.text) / 864 *1080)
        elem.text = str(new_elem)
    tree.write(xml_dir1+xml_name)