PASCAL VOC（xml解析，xml节点删除，生成xml）

其他 2020-04-19 10:36:42 阅读次数: 0

1. xml解析

import xml.etree.ElementTree as ET

def xml_reader(filename):
    """ Parse a PASCAL VOC xml file """
    tree = ET.parse(filename)
    objects = []
    for obj in tree.findall('object'):
        obj_struct = {}
        obj_struct['name'] = obj.find('name').text
        bbox = obj.find('bndbox')
        obj_struct['bbox'] = [int(bbox.find('xmin').text),
                              int(bbox.find('ymin').text),
                              int(bbox.find('xmax').text),
                              int(bbox.find('ymax').text)]
        objects.append(obj_struct)
    return objects

if __name__ == "__main__":
    objects = xml_reader("test.xml")
    print(objects)
[{'name': 'box', 'bbox': [2524, 1218, 2955, 1431]}, {'name': 'people', 'bbox': [539, 328, 963, 858]}, {'name': 'people', 'bbox': [2164, 441, 2461, 1114]}, {'name': 'people', 'bbox': [3589, 602, 3825, 857]}, {'name': 'people', 'bbox': [2041, 624, 2168, 844]}, {'name': 'face', 'bbox': [2323, 496, 2418, 616]}, {'name': 'face', 'bbox': [629, 409, 774, 572]}]

2. xml节点删除

     读取一个xml文件，删除对应的节点，再生成新的xml文件

def xml_rm_node(filename, filename_new, node_name):
    """ 
    删除PASCAL VOC里面的节点, 删除后保存成新的xml文件
    """
    tree = ET.parse(filename)
    root = tree.getroot()

    for child in root.findall('object'):
        name = child.find('name').text
        if name == node_name:
            root.remove(child)
    
    tree.write(filename_new)

3. 根据图片，box的类别和坐标信息生成xml文件（PASCAL VOC格式）

def CreatXml(imgPath, results, xmlPath):
    img = cv2.imread(imgPath)
    imgSize = img.shape
    imgName = imgPath.split('/')[-1]

    impl = xml.dom.minidom.getDOMImplementation()
    dom = impl.createDocument(None, 'annotation', None)
    root = dom.documentElement

    folder = dom.createElement('folder')
    root.appendChild(folder)
    name_folfer = dom.createTextNode('Unknown')
    folder.appendChild(name_folfer)

    filename = dom.createElement('filename')
    root.appendChild(filename)
    name_img = dom.createTextNode(os.path.splitext(imgName)[0])
    filename.appendChild(name_img)

    filepath = dom.createElement('path')
    root.appendChild(filepath)
    path_img = dom.createTextNode(imgPath)
    filepath.appendChild(path_img)

    source = dom.createElement('source')
    root.appendChild(source)
    database = dom.createElement('database')
    database_name = dom.createTextNode('Unknown')
    database.appendChild(database_name)
    source.appendChild(database)

    img_size = dom.createElement('size')
    root.appendChild(img_size)
    width = dom.createElement('width')
    width_num = dom.createTextNode(str(int(imgSize[1])))
    width.appendChild(width_num)
    height = dom.createElement('height')
    height_num = dom.createTextNode(str(int(imgSize[0])))
    height.appendChild(height_num)
    depth = dom.createElement('depth')
    depth_num = dom.createTextNode(str(int(imgSize[2])))
    depth.appendChild(depth_num)
    img_size.appendChild(width)
    img_size.appendChild(height)
    img_size.appendChild(depth)

    segmented = dom.createElement('segmented')
    root.appendChild(segmented)
    segmented_num = dom.createTextNode('0')
    segmented.appendChild(segmented_num)

    for i in range(len(results)):
        img_object = dom.createElement('object')
        root.appendChild(img_object)
        label_name = dom.createElement('name')
        namecls = dom.createTextNode(results[i]['name'])
        label_name.appendChild(namecls)
        pose = dom.createElement('pose')
        pose_name = dom.createTextNode('Unspecified')
        pose.appendChild(pose_name)
        truncated = dom.createElement('truncated')
        truncated_num = dom.createTextNode('0')
        truncated.appendChild(truncated_num)
        difficult = dom.createElement('difficult')
        difficult_num = dom.createTextNode('0')
        difficult.appendChild(difficult_num)
        bndbox = dom.createElement('bndbox')
        xmin = dom.createElement('xmin')
        xmin_num = dom.createTextNode(str(int(results[i]['bbox'][0])))
        xmin.appendChild(xmin_num)
        ymin = dom.createElement('ymin')
        ymin_num = dom.createTextNode(str(int(results[i]['bbox'][1])))
        ymin.appendChild(ymin_num)
        xmax = dom.createElement('xmax')
        xmax_num = dom.createTextNode(str(int(results[i]['bbox'][2])))
        xmax.appendChild(xmax_num)
        ymax = dom.createElement('ymax')
        ymax_num = dom.createTextNode(str(int(results[i]['bbox'][3])))
        ymax.appendChild(ymax_num)
        bndbox.appendChild(xmin)
        bndbox.appendChild(ymin)
        bndbox.appendChild(xmax)
        bndbox.appendChild(ymax)
        img_object.appendChild(label_name)
        img_object.appendChild(pose)
        img_object.appendChild(truncated)
        img_object.appendChild(difficult)
        img_object.appendChild(bndbox)

    f = open(xmlPath, 'w')
    dom.writexml(f, addindent='  ', newl='\n')
    f.close()

整体代码如下：

import xml.etree.ElementTree as ET
import xml.dom.minidom
import cv2
import os

def xml_reader(filename):
    """ Parse a PASCAL VOC xml file """
    tree = ET.parse(filename)
    objects = []
    for obj in tree.findall('object'):
        obj_struct = {}
        obj_struct['name'] = obj.find('name').text
        bbox = obj.find('bndbox')
        obj_struct['bbox'] = [int(bbox.find('xmin').text),
                              int(bbox.find('ymin').text),
                              int(bbox.find('xmax').text),
                              int(bbox.find('ymax').text)]
        objects.append(obj_struct)
    return objects


def xml_rm_node(filename, filename_new, node_name):
    """ 
    删除PASCAL VOC里面的节点, 删除后保存成新的xml文件
    """
    tree = ET.parse(filename)
    root = tree.getroot()

    for child in root.findall('object'):
        name = child.find('name').text
        if name == node_name:
            root.remove(child)
    
    tree.write(filename_new)
 

def CreatXml(imgPath, results, xmlPath):
    img = cv2.imread(imgPath)
    imgSize = img.shape
    imgName = imgPath.split('/')[-1]

    impl = xml.dom.minidom.getDOMImplementation()
    dom = impl.createDocument(None, 'annotation', None)
    root = dom.documentElement

    folder = dom.createElement('folder')
    root.appendChild(folder)
    name_folfer = dom.createTextNode('Unknown')
    folder.appendChild(name_folfer)

    filename = dom.createElement('filename')
    root.appendChild(filename)
    name_img = dom.createTextNode(os.path.splitext(imgName)[0])
    filename.appendChild(name_img)

    filepath = dom.createElement('path')
    root.appendChild(filepath)
    path_img = dom.createTextNode(imgPath)
    filepath.appendChild(path_img)

    source = dom.createElement('source')
    root.appendChild(source)
    database = dom.createElement('database')
    database_name = dom.createTextNode('Unknown')
    database.appendChild(database_name)
    source.appendChild(database)

    img_size = dom.createElement('size')
    root.appendChild(img_size)
    width = dom.createElement('width')
    width_num = dom.createTextNode(str(int(imgSize[1])))
    width.appendChild(width_num)
    height = dom.createElement('height')
    height_num = dom.createTextNode(str(int(imgSize[0])))
    height.appendChild(height_num)
    depth = dom.createElement('depth')
    depth_num = dom.createTextNode(str(int(imgSize[2])))
    depth.appendChild(depth_num)
    img_size.appendChild(width)
    img_size.appendChild(height)
    img_size.appendChild(depth)

    segmented = dom.createElement('segmented')
    root.appendChild(segmented)
    segmented_num = dom.createTextNode('0')
    segmented.appendChild(segmented_num)

    for i in range(len(results)):
        img_object = dom.createElement('object')
        root.appendChild(img_object)
        label_name = dom.createElement('name')
        namecls = dom.createTextNode(results[i]['name'])
        label_name.appendChild(namecls)
        pose = dom.createElement('pose')
        pose_name = dom.createTextNode('Unspecified')
        pose.appendChild(pose_name)
        truncated = dom.createElement('truncated')
        truncated_num = dom.createTextNode('0')
        truncated.appendChild(truncated_num)
        difficult = dom.createElement('difficult')
        difficult_num = dom.createTextNode('0')
        difficult.appendChild(difficult_num)
        bndbox = dom.createElement('bndbox')
        xmin = dom.createElement('xmin')
        xmin_num = dom.createTextNode(str(int(results[i]['bbox'][0])))
        xmin.appendChild(xmin_num)
        ymin = dom.createElement('ymin')
        ymin_num = dom.createTextNode(str(int(results[i]['bbox'][1])))
        ymin.appendChild(ymin_num)
        xmax = dom.createElement('xmax')
        xmax_num = dom.createTextNode(str(int(results[i]['bbox'][2])))
        xmax.appendChild(xmax_num)
        ymax = dom.createElement('ymax')
        ymax_num = dom.createTextNode(str(int(results[i]['bbox'][3])))
        ymax.appendChild(ymax_num)
        bndbox.appendChild(xmin)
        bndbox.appendChild(ymin)
        bndbox.appendChild(xmax)
        bndbox.appendChild(ymax)
        img_object.appendChild(label_name)
        img_object.appendChild(pose)
        img_object.appendChild(truncated)
        img_object.appendChild(difficult)
        img_object.appendChild(bndbox)

    f = open(xmlPath, 'w')
    dom.writexml(f, addindent='  ', newl='\n')
    f.close()



if __name__ == "__main__":
    filename = "test.xml"
    objects = xml_reader(filename)
    print(objects)

    xml_rm_node(filename, "new.xml", "box")

    CreatXml("test.jpg", objects, "creat.xml")
CV-deeplearning
发布了10 篇原创文章 · 获赞 14 · 访问量 1319
私信 关注

猜你喜欢

转载自blog.csdn.net/Guo_Python/article/details/105582897
今日推荐
TIOBE 5 月榜单：Fortran “复活”进入 Top 10
GCC 14.1 发布
面壁智能发布 Eurux-8x22B 开源大模型 —— 堪称「理科状元」
开源日报 | 谷歌扶持鸿蒙上位；开源Rabbit R1；Docker加持的安卓手机；微软的焦虑和野心；海尔电器把开放平台关了
中国码农的“35岁魔咒”
蘭雅 CorelDRAW 插件 2024.5.1 国际劳动节版，免费下载
Arc Browser for Windows 1.0 正式 GA
90后程序员开发视频搬运软件、不到一年获利超 700 万，结局很刑！
周排行
基本数据类型封装类比较 Java源码解读(一) 8种基本类型对应的封装类型
JS实现无缝滚动上
深入解析HashMap原理（基于JDK1.8）
mysql的连接池
关于.htc
linux下的ubuntu12.04图形界面
【数论】好推不好记的扩展欧几里德
设备树详解
cscope + tags 简单设置
xml学习
每日归档
更多
2024-05-09(35)
2024-05-08(42)
2024-05-07(14)
2024-05-06(40)
2024-05-05(0)
2024-05-04(7)
2024-05-03(19)
2024-05-02(0)
2024-05-01(4)
2024-04-30(1)

代码天地 © 2018 codetd.com

境外服务器   最新电视剧2022