可以将voc格式中储存的图片的标注信息转化为xml文件
方便阅读以及可视化
可以二次标注
txt格式: cls、x、 y、w、h
import cv2
from lxml.etree import Element, SubElement, tostring
# 修改voc格式中数字对应的类别
classes = {0:'w',1:'b'}
# 读取图片信息 返回WHC
def Read_img(path):
image = cv2.imread(path)
list_image = (image.shape[0],image.shape[1],image.shape[2])
return list_image
# 读取与图片对应的txt文件的信息
def Read_txt(path1,path2):
xyxy_list = []
f = open(path1,'r')
for line in f.readlines():
line = line.strip('\n').split()
cls = int(line[0])
line[0] = classes[cls]
W = Read_img(path2)[0]
H = Read_img(path2)[1]
x = float(line[1])
y = float(line[2])
w = float(line[3])
h = float(line[4])
xmax = int(W * (2 * x + w)) / 2
xmin = int(W * (2 * x - w)) / 2
ymax = int(H * (2 * y + h)) / 2
ymin = int(H * (2 * y - h)) / 2
line[1] = str(xmin)
line[2] = str(ymin)
line[3] = str(xmax)
line[4] = str(ymax)
xyxy_list.append(line)
f.close()
return xyxy_list
def create_xml(list_xml,list_images,xml_path):
"""
创建xml文件,依次写入xml文件必备关键字
:param list_xml: txt文件中的box
:param list_images: 图片信息,xml中需要写入WHC
:return:
"""
node_root = Element('annotation')
node_folder = SubElement(node_root, 'folder')
node_folder.text = 'Images'
node_filename = SubElement(node_root, 'filename')
node_filename.text = str(list_images[3])
node_size = SubElement(node_root, 'size')
node_width = SubElement(node_size, 'width')
node_width.text = str(list_images[0])
node_height = SubElement(node_size, 'height')
node_height.text = str(list_images[1])
node_depth = SubElement(node_size, 'depth')
node_depth.text = str(list_images[2])
if len(list_xml)>=1:
for list_ in list_xml:
node_object = SubElement(node_root, 'object')
node_name = SubElement(node_object, 'name')
node_name.text = str(list_[0])
node_difficult = SubElement(node_object, 'difficult')
node_difficult.text = '0'
node_bndbox = SubElement(node_object, 'bndbox')
node_xmin = SubElement(node_bndbox, 'xmin')
node_xmin.text = str(list_[1])
node_ymin = SubElement(node_bndbox, 'ymin')
node_ymin.text = str(list_[2])
node_xmax = SubElement(node_bndbox, 'xmax')
node_xmax.text = str(list_[3])
node_ymax = SubElement(node_bndbox, 'ymax')
node_ymax.text = str(list_[4])
xml = tostring(node_root, pretty_print=True)
# 不是jpg格式的将 .jpg 修改为对应的格式
file_name = list_images[3].replace(".jpg",'.xml')
filename = xml_path+"/{}".format(file_name)
f = open(filename, "wb")
f.write(xml)
f.close()
if __name__ == '__main__':
import os
img_path = r"/Data/1" # 图片路径
txt_path = r"/1_txt" # xml标注保存路径
xml_path = r"/1_xml" # xml标注保存路径
filename=os.listdir(txt_path)
# if '.DS_Store' in filename:
# filename.remove('.DS_Store')
for txt in filename:
txtpath = os.path.join(txt_path,txt)
_jpg = txt.replace('.txt','.jpg')
imgpath = os.path.join(img_path,_jpg)
xyxy_list = Read_txt(txtpath,imgpath)
list_image = list(Read_img(imgpath))
list_image.append(str(_jpg))
print(list_image)
create_xml(xyxy_list, list_image, xml_path)