labelme将json转换成dataset格式

这里面好多坑，有好多版本的问题，还有内部代码的问题，有的缺少info.yaml文件的输出，最后改来改去，找来找去，弄好了。只需要改B:\Anaconda\Lib\site-packages\labelme\cil。类似这个目录中的json_to_dataset.py,就可以，然后使用cmd输入B:\Anaconda\Scripts>labelme_json_to_dataset.exe A:\huang\date -o=A:\huang\date

需要对应到自己的路径，然后就可以了。

import argparse
import base64
import json
import os
import os.path as osp

import imgviz
import PIL.Image
import yaml
from labelme import logger
from labelme import utils


def main():
    logger.warning(
        "批量处理json文件。"
        "注意！文件夹中不能有其他文件存在。"
    )
    parser = argparse.ArgumentParser()
    parser.add_argument('json_file')
    parser.add_argument('-o', '--out', default=None)
    args = parser.parse_args()

    json_file = args.json_file

    # 输出路径
    if args.out is None:
        out_dir = osp.basename(json_file).replace('.', '_')
        out_dir = osp.join(osp.dirname(json_file), out_dir)
    else:
        out_dir = args.out
    if not osp.exists(out_dir):
        os.makedirs(out_dir)

    # 处理多个文件
    json_files_names = os.listdir(json_file)
    for i in range(0, len(json_files_names)):
        path = os.path.join(json_file, json_files_names[i])
        logger.info("正在处理: {}".format(path))
        if os.path.isfile(path):
            try:
                data = json.load(open(path, encoding='UTF-8'))
            except:
                data = json.load(open(path))

            if data['imageData']:
                imageData = data['imageData']
            else:
                imagePath = os.path.join(os.path.dirname(path), data['imagePath'])
                try:
                    with open(imagePath, 'rb', encoding='UTF-8') as f:
                        imageData = f.read(encoding='UTF-8')
                        imageData = base64.b64encode(imageData).decode('utf-8')
                except:
                    with open(imagePath, 'rb') as f:
                        imageData = f.read()
                        imageData = base64.b64encode(imageData).decode('utf-8')

            img = utils.img_b64_to_arr(imageData)
            label_name_to_value = {'_background_': 0}
            for shape in sorted(data["shapes"], key=lambda x: x["label"]):
                label_name = shape["label"]
                if label_name in label_name_to_value:
                    label_value = label_name_to_value[label_name]
                else:
                    label_value = len(label_name_to_value)
                    label_name_to_value[label_name] = label_value
            lbl, _ = utils.shapes_to_label(
                img.shape, data["shapes"], label_name_to_value
            )

            label_names = [None] * (max(label_name_to_value.values()) + 1)
            for name, value in label_name_to_value.items():
                label_names[value] = name

            lbl_viz = imgviz.label2rgb(
                label=lbl, image=imgviz.asgray(img), label_names=label_names, loc="rb"
            )
            # 新建路径
            save_dir = osp.join(out_dir, json_files_names[i].split('.')[0])
            if not os.path.exists(save_dir):
                os.makedirs(save_dir)

            PIL.Image.fromarray(img).save(osp.join(save_dir, 'img.png'))
            utils.lblsave(osp.join(save_dir, 'label.png'), lbl)
            PIL.Image.fromarray(lbl_viz).save(osp.join(save_dir, 'label_viz.png'))

            with open(osp.join(save_dir, 'label_names.txt'), 'w') as f:
                for lbl_name in label_names:
                    f.write(lbl_name + '\n')

            logger.info("Saved to: {}".format(save_dir))
            info = dict(label_names=label_names)
            with open(osp.join(save_dir, 'info.yaml'), 'w') as f:
                yaml.safe_dump(info, f, default_flow_style=False)

            print('Saved to: %s' % out_dir)