Foggy_driving数据集下载以及将其转换成VOC数据格式

Foggy_driving数据集官方下载地址
百度网盘下载链接：https://pan.baidu.com/s/1q4dhnlX-doxlt13Mi-uFZQ
提取码：2ap3
VOC格式的Foggy_driving数据集百度网盘下载链接：https://pan.baidu.com/s/14bIND62U0wyhXLvUy5nJFQ
提取码：ekrn

将Foggy_driving数据集转换成VOC数据格式的python代码如下：

###Foggy_driving中的gtbox标注为txt格式，现工作需要将目标检测标注转为pascal voc格式的标注格式
###Foggy_driving数据集的图片在leftImg8bit中（且分为了test:pedestrian、public和test_extra:pedestrain、web）
###目标检测的标注在bboxGt中（且分为了test:pedestrian、public和test_extra:pedestrain、web）
####所以需要将bounding_box标注放到同一个文件夹中，方便读取
import os, sys
import glob
from PIL import Image

# 图像存储位置
src_img_dir = "D:/XUEXI/CODES/Detections/DATASET/Foggy_Driving/Foggy_Driving/VOC2007/JPEGImages"
# 图像的 ground truth 的 txt 文件存放位置
src_txt_dir = "D:/XUEXI/CODES/Detections/DATASET/Foggy_Driving/bboxGt"
###生成的xml文件想要保存的位置
src_xml_dir = "D:/XUEXI/CODES/Detections/DATASET/Foggy_Driving/Foggy_Driving/VOC2007/Annotations"
### - `bboxGt`
###the bounding box annotations induced from the above semantic annotations, available for all 101 images of the dataset.
# Annotations are encoded as `txt` files, in which each line corresponds to a single object and is formatted as
#    ```
#    {class} {xmin} {ymin} {xmax} {ymax}
#    ```
#    `class` stands for the ID of the class this object belongs to, and the rest four elements encode the extent of
#    its bounding box in 1-based integer pixel coordinates. The 8 relevant classes are encoded with the following IDs:
#    - *car*: 0
#    - *person*: 1
#    - *bicycle*: 2
#    - *bus*: 3
#    - *truck*: 4
#    - *train*: 5
#    - *motorcycle*: 6
#    - *rider*: 7
classes = {'0': 'car','1': 'person','2': 'bicycle', '3': 'bus', '4': 'truck', '5': 'train', '6': 'motorcycle',
           '7': 'rider',}

img_Lists = glob.glob(src_img_dir + '/*.png')

img_basenames = []  # e.g. 100.jpg
for item in img_Lists:
    img_basenames.append(os.path.basename(item))

img_names = []  # e.g. 100
for item in img_basenames:
    temp1, temp2 = os.path.splitext(item)
    img_names.append(temp1)
    image_ids = open(r'D:\XUEXI\CODES\Detections\DATASET\Foggy_Driving\Foggy_Driving\VOC2007\ImageSets\Main\val.txt', 'a')###val.txt需要提前新建
    image_ids.write('% s\n'% (temp1))
    image_ids.close()
for img in img_names:
    im = Image.open((src_img_dir + '/' + img + '.png'))
    width, height = im.size

    # open the crospronding txt file
    ###txt的文件名：pedestrian_20161201_101324.txt  图像的文件名：pedestrian_20161201_101324_leftImg8bit.png  所以img[:-12]
    gt = open(src_txt_dir + '/' + img[:-12] + '.txt').read().splitlines()###{class} {xmin} {ymin} {xmax} {ymax} 此时的gt为List 但是length为1
    # write in xml file
    # os.mknod(src_xml_dir + '/' + img + '.xml')
    xml_file = open((src_xml_dir + '/' + img + '.xml'), 'w')
    xml_file.write('<annotation>\n')
    xml_file.write('    <folder>VOC2007</folder>\n')
    xml_file.write('    <filename>' + str(img) + '.png' + '</filename>\n')
    xml_file.write('    <size>\n')
    xml_file.write('        <width>' + str(width) + '</width>\n')
    xml_file.write('        <height>' + str(height) + '</height>\n')
    xml_file.write('        <depth>3</depth>\n')
    xml_file.write('    </size>\n')

    # write the region of image on xml file
    for img_each_label in gt:
        spt = img_each_label.split(' ')  # 这里如果txt里面是以逗号‘，’隔开的，那么就改为spt = img_each_label.split(',')。
        xml_file.write('    <object>\n')
        xml_file.write('        <name>' + str(classes[spt[0]]) + '</name>\n')
        xml_file.write('        <pose>Unspecified</pose>\n')
        xml_file.write('        <truncated>0</truncated>\n')
        xml_file.write('        <difficult>0</difficult>\n')
        xml_file.write('        <bndbox>\n')
        xml_file.write('            <xmin>' + str(spt[1]) + '</xmin>\n')
        xml_file.write('            <ymin>' + str(spt[2]) + '</ymin>\n')
        xml_file.write('            <xmax>' + str(spt[3]) + '</xmax>\n')
        xml_file.write('            <ymax>' + str(spt[4]) + '</ymax>\n')
        xml_file.write('        </bndbox>\n')
        xml_file.write('    </object>\n')

    xml_file.write('</annotation>')