資料集處理

一、資料標註

　　使用labelme,如下：

　　資料圖片和標註json檔案放到同一個目錄下

二、影象標註後的資料轉換

（1）訓練資料集生成標籤圖

python labelme2voc.py F:\blackbord\deeplabv3\image --labels labels.txt

__ignore__
_background_
blackboard
screen

（2）程式碼如下

#!/usr/bin/env python

from __future__ import print_function

 
import argparse
import glob
import json
import os
import os.path as osp
import sys

import numpy as np
import PIL.Image

import labelme


def main():
    parser = argparse.ArgumentParser(
        formatter_class=argparse.ArgumentDefaultsHelpFormatter
    )
    parser.add_argument('--input_dir', default= r"
 
F:\blackbord\deeplabv3\image",help='input annotated directory')
    parser.add_argument('--output_dir', default= r"F:\blackbord\deeplabv3\masks",help='output dataset directory')
    parser.add_argument('--labels', default = r"F:\blackbord\deeplabv3\class_label.txt",help='labels file', )
    args = parser.parse_args()

    
 
if osp.exists(args.output_dir):
        print('Output directory already exists:', args.output_dir)
        sys.exit(1)
    os.makedirs(args.output_dir)
    os.makedirs(osp.join(args.output_dir, 'JPEGImages'))
    os.makedirs(osp.join(args.output_dir, 'SegmentationClass'))
    os.makedirs(osp.join(args.output_dir, 'SegmentationClassPNG'))
    os.makedirs(osp.join(args.output_dir, 'SegmentationClassVisualization'))
    print('Creating dataset:', args.output_dir)

    class_names = []
    class_name_to_id = {}
    for i, line in enumerate(open(args.labels).readlines()):
        class_id = i - 1  # starts with -1
        class_name = line.strip()
        class_name_to_id[class_name] = class_id
        if class_id == -1:
            assert class_name == '__ignore__'
            continue
        elif class_id == 0:
            assert class_name == '_background_'
        class_names.append(class_name)
    class_names = tuple(class_names)
    print('class_names:', class_names)
    out_class_names_file = osp.join(args.output_dir, 'class_names.txt')
    with open(out_class_names_file, 'w') as f:
        f.writelines('\n'.join(class_names))
    print('Saved class_names:', out_class_names_file)

    colormap = labelme.utils.label_colormap(255)

    for label_file in glob.glob(osp.join(args.input_dir, '*.json')):
        print('Generating dataset from:', label_file)
        with open(label_file,"r",encoding="utf-8") as f:
            base = osp.splitext(osp.basename(label_file))[0]
            out_img_file = osp.join(
                args.output_dir, 'JPEGImages', base + '.jpg')
            out_lbl_file = osp.join(
                args.output_dir, 'SegmentationClass', base + '.npy')
            out_png_file = osp.join(
                args.output_dir, 'SegmentationClassPNG', base + '.png')
            out_viz_file = osp.join(
                args.output_dir,
                'SegmentationClassVisualization',
                base + '.jpg',
            )

            data = json.load(f)

            label_file = label_file.rstrip(".json")
            print(label_file)
            # img_file = osp.join(osp.dirname(label_file), data['imagePath'])
            img_file =label_file +".jpg"
            print(img_file)
            img = np.asarray(PIL.Image.open(img_file))
            PIL.Image.fromarray(img).save(out_img_file)

            lbl = labelme.utils.shapes_to_label(
                img_shape=img.shape,
                shapes=data['shapes'],
                label_name_to_value=class_name_to_id,
            )
            labelme.utils.lblsave(out_png_file, lbl)

            np.save(out_lbl_file, lbl)

            viz = labelme.utils.draw_label(
                lbl, img, class_names, colormap=colormap)
            PIL.Image.fromarray(viz).save(out_viz_file)


if __name__ == '__main__':
    main()

(3) mask灰度值的轉換:

python datasets/remove_gt_colormap.py --original_gt_folder /lwh/models/research/deeplab/datasets/blackboard/png --output_dir /lwh/models/research/deeplab/datasets/blackboard/mask

(4)製作指引檔案,為生成tfrecord資料格式做準備

import os,shutil
from PIL import Image
 


train_path = r'F:\blackbord\deeplabv3\masks\train'
filelist_train = sorted(os.listdir(train_path))
val_path = r'F:\blackbord\deeplabv3\masks\val'
filelist_val = sorted(os.listdir(val_path))
index_path = r'F:\blackbord\deeplabv3\masks\index'

VOC_file_dir = index_path


VOC_train_file = open(os.path.join(VOC_file_dir, "train.txt"), 'w')
VOC_test_file = open(os.path.join(VOC_file_dir, "val.txt"), 'w')
VOC_train_file.close()
VOC_test_file.close()

VOC_train_file = open(os.path.join(VOC_file_dir, "train.txt"), 'a')
VOC_test_file = open(os.path.join(VOC_file_dir, "val.txt"), 'a')

for eachfile in filelist_train:
    (temp_name,temp_extention) = os.path.splitext(eachfile)
    img_name = temp_name
    VOC_train_file.write(img_name + '\n')

for eachfile in filelist_val:
    (temp_name, temp_extention) = os.path.splitext(eachfile)
    img_name = temp_name
    VOC_test_file.write(img_name + '\n')

VOC_train_file.close()
VOC_test_file.close()

（4）製作tfrecord檔案

需要四個檔案路徑

image存放原始的訓練圖片，index存放指引檔案，mask存放去除水霧的label圖片，tfrecord為儲存訓練資料，執行下面指令碼命令，生成訓練資料

python build_voc2012_data.py --image_folder="/lwh/models/research/deeplab/datasets/CamVid/image" 
--semantic_segmentation_folder="/lwh/models/research/deeplab/datasets/CamVid/mask" 
--list_folder="/lwh/models/research/deeplab/datasets/CamVid/index" --image_format="png" --label_format="png" 
--output_dir="/lwh/models/research/deeplab/datasets/CamVid/tfrecord"