python解析xml+得到pascal voc xml格式用于目标检测+美化xml

1.python解析xml

    img_path='./data/001.tif'
    xml_path='./xml/001.xml'
    img=cv2.imread(img_path)
    # cv2.imshow('img', img)
    # cv2.waitKey(0)
    print(img.shape)

    try:
        xmlp = ET.XMLParser(encoding="utf-8")
        tree = ET.parse(xml_path, parser=xmlp)
        root = tree.getroot()
        print(tree)
        print(root)
    except Exception as e:
        print(e, xml_path)

    content_info={}
    for i in tree.iterfind('.//正文组/正文'):
        print("i.attrib['篇目号']", i.attrib['篇目号'])
        for j in i.findall('篇目坐标'):
            print(j.text)
            print(type(j.text))
            content_info[i.attrib['篇目号']] =j.text
    print('content_info=',content_info)

若要更改xml内容，可以

new_point='1,2,3,4'
j.text = new_point
tree.write('output_test.xml', encoding='utf-8')

2. 得到pascal voc xml格式用于目标检测

img_dir = "./images/train"
csv_dir = "./train_xml.csv"
xml_dir = "./Annotations"
if not os.path.exists(xml_dir):
    os.mkdir(xml_dir)

imgs_path_Lists = [os.path.join(img_dir,i) for i in os.listdir(img_dir)]

df=pd.read_csv(csv_dir).copy()
df_value=df.values
print(df_value.shape)
coord_h,coord_w=df_value.shape
print(df_value[:2])
for img_path_List in imgs_path_Lists:
    im = Image.open(img_path_List)
    width, height = im.size
    img_name=img_path_List.split('/')[-1]

    # write in xml file
    xml_file = open((xml_dir + '/' + img_name.split('.jpg')[0] + '.xml'), 'w')
    xml_file.write('\n')
    xml_file.write('    steel\n')
    xml_file.write('    ' + img_name + '\n')
    xml_file.write('    \n')
    xml_file.write('        ' + str(width) + '\n')
    xml_file.write('        ' + str(height) + '\n')
    xml_file.write('        3\n')
    xml_file.write('    \n')
    for i in df_value:
        if i[0].split('/')[-1]==img_name:
            xmin=i[1]
            ymin=i[2]
            xmax=i[3]
            ymax=i[4]
            class_name=i[-1]

            # write the region of image on xml file
            xml_file.write('    \n')

    xml_file.write('')

3.美化xml

原始xml如下

 	
 		22
 		male
 	 
 	 
 		22
 		female
 	 
 

代码:

def prettyXml(element, indent, newline, level = 0): # elemnt为传进来的Elment类，参数indent用于缩进，newline用于换行
    if element:  # 判断element是否有子元素
        if element.text == None or element.text.isspace(): # 如果element的text没有内容
            element.text = newline + indent * (level + 1)
        else:
            element.text = newline + indent * (level + 1) + element.text.strip() + newline + indent * (level + 1)
    #else:  # 此处两行如果把注释去掉，Element的text也会另起一行
        #element.text = newline + indent * (level + 1) + element.text.strip() + newline + indent * level
    temp = list(element) # 将elemnt转成list
    for subelement in temp:
        if temp.index(subelement) < (len(temp) - 1): # 如果不是list的最后一个元素，说明下一个行是同级别元素的起始，缩进应一致
            subelement.tail = newline + indent * (level + 1)
        else:  # 如果是list的最后一个元素， 说明下一行是母元素的结束，缩进应该少一个
            subelement.tail = newline + indent * level
        prettyXml(subelement, indent, newline, level = level + 1) # 对子元素进行递归操作
def testXML():
    from xml.etree import ElementTree as ET
    tree = ET.parse('test.xml')
    root = tree.getroot()
    print(root)
    # #查看tag和attrib
    # for person in root:
    #     print(person.tag, person.attrib)

    ZF_COORD = ET.Element('字符坐标')
    ZWZ = ET.SubElement(ZF_COORD, '正文组')
    #这里需要把001变成需要的篇目号
    ZW = ET.SubElement(ZWZ , '正文',{'篇目号':"001"})

    #栏目
    LM=ET.SubElement(ZW, '栏目')
    LM.text = '12,12,334,123;12,34,45,56'

    #引题
    YT=ET.SubElement(ZW,'引题')
    YT.text='12,12,334,123;12,34,45,56'

    # 标题
    BT = ET.SubElement(ZW, '标题')
    BT.text = '12,12,334,123;12,34,45,56'

    # 副题
    FT = ET.SubElement(ZW, '副题')
    FT.text = '12,12,334,123;12,34,45,56'

    # 小标题
    XBT = ET.SubElement(ZW, '小标题')
    XBT.text = '12,12,334,123;12,34,45,56'

    # 作者
    author = ET.SubElement(ZW, '作者')
    author.text = '12,12,334,123;12,34,45,56'

    # # 向根节点添加新的子节点 
    root.append(ZF_COORD)
    # 写入 
    tree.write('./sample.xml', encoding="utf-8",xml_declaration=True)

    #美化作用
    tree = ET.parse('./sample.xml')  # 解析test.xml这个文件，该文件内容如上文
    root = tree.getroot()  # 得到根元素，Element类
    prettyXml(root, '\t', '\n')  # 执行美化方法
    # ET.dump(root)
    tree.write('./sample.xml', encoding="utf-8",xml_declaration=True)

新生成xml如下

	
		22
		male
	
	
		22
		female
	
	<字符坐标>
		<正文组>
			<正文 篇目号="001">
				<栏目>12,12,334,123;12,34,45,56
				<引题>12,12,334,123;12,34,45,56
				<标题>12,12,334,123;12,34,45,56
				<副题>12,12,334,123;12,34,45,56
				<小标题>12,12,334,123;12,34,45,56
				<作者>12,12,334,123;12,34,45,56