NWPU VHR-10转成pascal voc的格式_github nwpu vhr-10

欢迎关注的公众号：
不定期发布各类教程、理论笔记、搬运前沿动态。
关注公众号“AIOTer”回复“ NWPU VOC ”即可获取NWPU VHR-10的pascal voc格式数据集。

csdn上好几个教程，试了一下，貌似都一个样。我好奇，原始数据集就一个txt存放的文件夹，为什么搞得这么复杂，传递地址用了4个，真费劲，居然还跑不起来，运行了一顿就输出一个xml，用labelimg查看还不能用……，本文这个用法是传递3个地址即可，转换代码在后面。
下载：NWPU VHR-10的pascal voc格式数据集
成功效果图：

from lxml.etree import Element, SubElement, tostring
from xml.dom.minidom import parseString
import xml.dom.minidom
import os
import sys
from PIL import Image

#https://blog.csdn.net/summer2day/article/details/83064727#comments

# 把txt中的内容写进xml
def deal(path):
    files = os.listdir(path)  # 列出所有文件
    for file in files:
        filename = os.path.splitext(file)[0]  # 分割出文件名
        # print(filename)
        sufix = os.path.splitext(file)[1]  # 分割出后缀
        if sufix == '.txt':
            xmins = []
            ymins = []
            xmaxs = []
            ymaxs = []
            names = []
            num, xmins, ymins, xmaxs, ymaxs, names = readtxt(file)
            # dealpath = path + "/" + filename + ".xml"
            dealpath = xmlPath + "/" + filename + ".xml"
            filename = filename + '.jpg'
            with open(dealpath, 'w') as f:
                writexml(dealpath, filename, num, xmins, ymins, xmaxs, ymaxs, names)


# 读取图片的高和宽写入xml
def dealwh(path):
    files = os.listdir(path)  # 列出所有文件
    for file in files:
        filename = os.path.splitext(file)[0]  # 分割出文件名
        sufix = os.path.splitext(file)[1]  # 分割出后缀
        if sufix == '.jpg':
            height, width = readsize(file)
            # dealpath = path + "/" + filename + ".xml"
            dealpath = xmlPath + "/" + filename + ".xml"
            gxml(dealpath, height, width)


# 读取txt文件
def readtxt(p):
    p_file = txtPath + "/" + p
    with open(p_file, 'r') as f:
        contents = f.read()
        # print(contents)
        objects = contents.split('\n')  # 分割出每个物体
        for i in range(objects.count('')):  # 去掉空格项
            objects.remove('')
        # print(objects)
        num = len(objects)  # 物体的数量
        # print(num)
        xmins = []
        ymins = []
        xmaxs = []
        ymaxs = []
        names = []
        for objecto in objects:
            # print(objecto)
            xmin = objecto.split(',')[0]
            xmin = xmin.split('(')[1]
            xmin = xmin.strip()

            ymin = objecto.split(',')[1]
            ymin = ymin.split(')')[0]
            ymin = ymin.strip()

            xmax = objecto.split(',')[2]
            xmax = xmax.split('(')[1]
            xmax = xmax.strip()

            ymax = objecto.split(',')[3]
            ymax = ymax.split(')')[0]
            ymax = ymax.strip()

            name = objecto.split(',')[4]
            name = name.strip()

            if name == "1 " or name == "1":
                name = 'airplane'
            elif name == "2 " or name == "2":
                name = 'ship'
            elif name == "3 " or name == "3":
                name = 'storage tank'
            elif name == "4 " or name == "4":
                name = 'baseball diamond'
            elif name == "5 " or name == "5":
                name = 'tennis court'
            elif name == "6 " or name == "6":
                name = 'basketball court'
            elif name == "7 " or name == "7":
                name = 'ground track field'
            elif name == "8 " or name == "8":
                name = 'habor'
            elif name == "9 " or name == "9":
                name = 'bridge'
            elif name == "10 " or name == "10":
                name = 'vehicle'
            else:
                print(txtPath)
            # print(xmin,ymin,xmax,ymax,name)
            xmins.append(xmin)
            ymins.append(ymin)
            xmaxs.append(xmax)
            ymaxs.append(ymax)
            names.append(name)
        # print(num,xmins,ymins,xmaxs,ymaxs,names)
        return num, xmins, ymins, xmaxs, ymaxs, names


# 在xml文件中添加宽和高
def gxml(path, height, width):
    dom = xml.dom.minidom.parse(path)
    root = dom.documentElement
    heights = root.getElementsByTagName('height')[0]
    heights.firstChild.data = height
    # print(height)

    widths = root.getElementsByTagName('width')[0]
    widths.firstChild.data = width
    # print(width)
    with open(path, 'w') as f:
        # with open(xmlPath, 'w') as f:
        dom.writexml(f)
    return


# 创建xml文件
def writexml(path, filename, num, xmins, ymins, xmaxs, ymaxs, names, height='256', width='256'):
    node_root = Element('annotation')

    node_folder = SubElement(node_root, 'folder')
    node_folder.text = "VOC2007"

    node_filename = SubElement(node_root, 'filename')
    node_filename.text = "%s" % filename

    node_size = SubElement(node_root, "size")
    node_width = SubElement(node_size, 'width')
    node_width.text = '%s' % width

    node_height = SubElement(node_size, 'height')
    node_height.text = '%s' % height

    node_depth = SubElement(node_size, 'depth')
    node_depth.text = '3'
    for i in range(num):
        node_object = SubElement(node_root, 'object')
        node_name = SubElement(node_object, 'name')
        node_name.text = '%s' % names[i]
        node_name = SubElement(node_object, 'pose')
        node_name.text = '%s' % "unspecified"
        node_name = SubElement(node_object, 'truncated')
        node_name.text = '%s' % "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 = '%s' % xmins[i]
        node_ymin = SubElement(node_bndbox, 'ymin')
        node_ymin.text = '%s' % ymins[i]
        node_xmax = SubElement(node_bndbox, 'xmax')
        node_xmax.text = '%s' % xmaxs[i]
        node_ymax = SubElement(node_bndbox, 'ymax')
        node_ymax.text = '%s' % ymaxs[i]

    xml = tostring(node_root, pretty_print=True)
    dom = parseString(xml)
    with open(path, 'wb') as f:
        f.write(xml)
    return


def readsize(p):
    p_file=imagePath+"/"+p
    img=Image.open(p_file)
    width = img.size[0]
    height = img.size[1]
    return height, width

if __name__ == "__main__":
    # path = ("D:/NWPU VHR-10 dataset/NWPU VHR-10 dataset/test")
    imagePath = (r"F:\NWPU VHR-10 dataset\positive image set")
    txtPath = (r"F:\NWPU VHR-10 dataset\ground truth")
    xmlPath = ("annotations")
    deal(txtPath)
    dealwh(imagePath)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190