XML 标签读取

XML 标签读取

from xml.etree import ElementTree
copyfil

路径和文件配置

根据自己的路径配置，主要用来存储处理好的数据
import os
from shutil import copyfile

	from xml.etree import ElementTree
	a=0
	train_path='.//datasets//data//train'
	val_path='.//datasets//data//value'
	train_txt='.//datasets//data//train.txt'
	sourse='.//datasets//data//VOCdevkit//VOC2012//JPEGImages'
	path_prefix = './/datasets//data//VOCdevkit//VOC2012//ImageSets//Main//person_train.txt'
	val_txt='.//datasets//data//val.txt'
	with open(path_prefix,'r') as f:
	    filenames=f.readlines()
	train_file = open(train_txt, 'w')
	val_file = open(val_txt, 'w')
	for filename in filenames:
	    filename =filename[ :-3]
	    # 找到文件名字
	    path1='./datasets//data//VOCdevkit//VOC2012//Annotations'+'//'+filename.rstrip()+'.xml'

XML读取

用a来分训练集和验证集

# 1、XML解析根路径
tree = ElementTree.parse(path1)
root = tree.getroot()

size_tree = root.find('size')
width = float(size_tree.find('width').text)
height = float(size_tree.find('height').text)
for object_tree in root.findall('object'):
    if object_tree.find('name').text=='person':


        for bounding_box in object_tree.iter('bndbox'):
            xmin = float(bounding_box.find('xmin').text) / width
            ymin = float(bounding_box.find('ymin').text) / height
            xmax = float(bounding_box.find('xmax').text) / width
            ymax = float(bounding_box.find('ymax').text) / height
        bounding_boxes = []
        bounding_box = [(xmin+xmax)/2., (ymin+ymax)/2., xmax-xmin, ymax-ymin]
        bounding_boxes.append(bounding_box)
        class_name = object_tree.find('name').text
        # # 将类别进行one_hot编码
        # one_hot_class = self.on_hot(class_name)
        # one_hot_classes.append(one_hot_class)
        # 我现在只要人
        if a<1000:
            label_path = train_path + '//' + filename.rstrip() + '.txt'
            label_file = open(label_path, 'w')
            copyfile(sourse+'//'+filename.rstrip()+'.jpg',train_path+'//'+filename.rstrip()+'.jpg')
            print('copy done')
            label_file.write('1'+' '+' '.join([str(i) for i in bounding_box])+'\n')
            print('mask done')
            a += 1
            print(a)
            train_file.write(train_path+'//'+filename.rstrip()+'.jpg'+'\n')
        elif   a >= 1000and a <1300:
            val_label_path = val_path + '//' + filename.rstrip() + '.txt'
            val_label_file = open(val_label_path, 'w')
            copyfile(sourse + '//' + filename.rstrip() + '.jpg',val_path + '//' + filename.rstrip() + '.jpg')
            print('ssssss')
            val_label_file.write('1' + ' ' + ' '.join([str(i) for i in bounding_box]) + '\n')
            print('ssssss')
            a += 1
            print(a)
            val_file.write(val_path + '//' +filename.rstrip() + '.jpg' + '\n')
        else :
            break