Cityscapes转VOC格式，目标检测标签，内附完整代码

一次性转换Image和Target，并且按照VOC的格式存放

网上大部分cityscapes转VOC的代码都不完整，所以写了这个完整转换VOC格式的脚本，希望大家能给我点赞鼓励一下。

注意事项：

cityscapes_root需要自行修改，应设为leftImg8bit_trainvaltest和gtFine_trainvaltest所在的目录。

由于cityscapes的中并不包含VOC中的truncated和difficult信息，所以将所有object中truncated和difficult的值都设为0。

cityscapes数据集，通常将val作为test集用于验证性能，因为官方没有放出test集的标注，目的是防止有人用test集训练，在网站上刷分。

import os
import os.path
from PIL import Image
import json


def position(pos):
    # 该函数用来找出xmin,ymin,xmax,ymax即bbox包围框
    x = []
    y = []
    nums = len(pos)
    for i in range(nums):
        x.append(pos[i][0])
        y.append(pos[i][1])
    x_max = max(x)
    x_min = min(x)
    y_max = max(y)
    y_min = min(y)
    b = (float(x_min), float(y_min), float(x_max), float(y_max))
    return b


def convert_annotation(image_id):
    load_f = open(city_xml_dir + '/' + image_id + "_gtFine_polygons.json", 'r')  # 导入json标签的路径
    load_dict = json.load(load_f)
    out_file = open(city_xml_dir + '/' + '%s_leftImg8bit.txt' % (image_id), 'w')  # 输出标签的路径
    objects = load_dict['objects']
    nums = len(objects)

    cls_id = ''
    for i in range(0, nums):
        labels = objects[i]['label']
        if (labels in ['person',
                       'rider',
                       'car',
                       'truck',
                       'bus',
                       'train',
                       'motorcycle',
                       'bicycle']):  # 这里我需要用到的类别是这8类
            print(labels)
            pos = objects[i]['polygon']
            bb = position(pos)
            cls_id = labels
            out_file.write(cls_id + " " + " ".join([str(a) for a in bb]) + '\n')
    if cls_id == '':
        print('no label json:', "%s_gtFine_polygons.json" % (image_id))


def images_id(orgin_picture_dir):  # 获取训练集每个图像的名称  （orgin_picture_dir训练集图像的路径）
    a = []
    filenames = os.listdir(orgin_picture_dir)
    for filename in filenames:
        filename = filename.split('_leftImg8bit.png')[0]
        a.append(filename)
    return a


# 生成xml文件以及转换图片格式到保存路径
def transformation(orgin_picture_dir, city_xml_dir, save_dir):
    img_basenames = os.listdir(orgin_picture_dir)
    img_names = []
    for item in img_basenames:
        path = os.path.join(orgin_picture_dir, item)
        image = Image.open(path)
        save_params = {
            'format': 'JPEG',
            'pnginfo': image.info
        }

        # 分别提取图片名称和图片后缀名称（具有_leftImg8bit）
        temp1, temp2 = os.path.splitext(item)
        jpg_name = temp1 + '.jpg'

        # 将PNG图片转换为JPG格式
        image.save(os.path.join(Images_dir, jpg_name), **save_params)

        img_names.append(temp1)
        print(img_names)
    for img in img_names:  # img是没有后缀的pic名称
        im = Image.open((orgin_picture_dir + img + '.png'))
        width, height = im.size

        # open the crospronding temp2txt file
        gt = open(city_xml_dir + '/' + img + '.txt').read().splitlines()

        # write in xml file
        xml_file = open((save_dir + '/' + img + '.xml'), 'w')
        xml_file.write('<annotation>\n')
        xml_file.write('    <folder>CITYSCAPE</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(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>')


if __name__ == '__main__':

    # 添加路径
    # cityscapes_root需要自行修改，应设为leftImg8bit_trainvaltest和gtFine_trainvaltest所在的目录
    cityscapes_root = '/home/lyd/GithubProject/Cityscapes_file/'
    cityscapes_img = os.path.join(cityscapes_root, 'leftImg8bit_trainvaltest/leftImg8bit/' )
    cityscapes_xml = os.path.join(cityscapes_root, 'gtFine_trainvaltest/gtFine/')
    # cityscapes数据集，通常将val作为test集用于验证性能，因为官方没有放出test集的标注，目的是防止有人用test集训练，在网站上刷分
    datasets = ['train', 'val']

    # 待创建路径
    save_root = os.path.join(cityscapes_root, 'VOC2012/')
    Annotations_dir = os.path.join(save_root, 'Annotations')
    Images_dir = os.path.join(save_root, 'JPEGImages')
    ImageSets_dir = os.path.join(save_root, 'ImageSets/Main')

    # 创建路径
    os.makedirs(save_root, exist_ok=True)
    os.makedirs(Annotations_dir, exist_ok=True)
    os.makedirs(Images_dir, exist_ok=True)
    os.makedirs(ImageSets_dir, exist_ok=True)

    # 开始遍历数据集和城市
    for dataset in datasets:
        set = []
        xml_dataset_dir = cityscapes_xml + dataset + '/'
        img_dataset_dir = cityscapes_img + dataset + '/'
        city_names = os.listdir(xml_dataset_dir)
        print(city_names)
        for city in city_names:
            city_xml_dir = xml_dataset_dir + city + '/'  # json所在路径
            city_img_dir = img_dataset_dir + city + '/'  # cityscape的train所在路径
            for image in os.listdir(city_img_dir):
                name, png = os.path.splitext(image)
                set.append(name)

            names = images_id(city_img_dir)

            # 转换为txt格式标注
            for image_id in names:
                print(image_id)
                convert_annotation(image_id)

            # 利用txt再转换为xml
            transformation(city_img_dir, city_xml_dir, Annotations_dir)

        # 保存imageSets
        save_file = dataset + '.txt'
        with open(os.path.join(ImageSets_dir, save_file), 'w') as f:
            for item in set:
                f.write(str(item) + '\n')