Preparing Inputs

代碼高能預(yù)警

Tensorflow Object Detection API 在讀取數(shù)據(jù)中使用了TFRecord文件格式。API提供了兩個(gè)示例腳本烈拒，(create_pascal_tf_record.py 和 create_pet_tf_record.py)荆几。這里我們精讀一下代碼create_pascal_tf_record.py。

掌握TFRocord讀取方法的可以跳級(jí)了行拢。
先看一下License

# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# 

這個(gè)腳本的主要用處是把PASCAL數(shù)據(jù)集轉(zhuǎn)換成TFRecord舟奠。
用法是

  ./create_pascal_tf_record --data_dir=/home/user/VOCdevkit \  --year=VOC2012 \      --output_path=/home/user/pascal.record

引入各種庫

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import hashlib
import io
import logging
import os

from lxml import etree
import PIL.Image
import tensorflow as tf

from object_detection.utils import dataset_util
from object_detection.utils import label_map_util

到這一步沼瘫，程序都在引入各種各樣的庫晕鹊，沒有的裝就是了暴浦。

flags

flags = tf.app.flags
flags.DEFINE_string('data_dir', '', 'Root directory to raw PASCAL VOC dataset.')
flags.DEFINE_string('set', 'train', 'Convert training set, validation set or '
                    'merged set.')
flags.DEFINE_string('annotations_dir', 'Annotations',
                    '(Relative) path to annotations directory.')
flags.DEFINE_string('year', 'VOC2007', 'Desired challenge year.')
flags.DEFINE_string('output_path', '', 'Path to output TFRecord')
flags.DEFINE_string('label_map_path', 'data/pascal_label_map.pbtxt',
                    'Path to label map proto')
flags.DEFINE_boolean('ignore_difficult_instances', False, 'Whether to ignore '
                     'difficult instances')
FLAGS = flags.FLAGS

Tensorflow 中的flags類似于argv歌焦，基本用法是flags.DEFINE_類型('參數(shù)名稱'独撇，'默認(rèn)值'，'參數(shù)描述')卵史。進(jìn)一步了解flags用法請(qǐng)移步tensorflow 學(xué)習(xí)(三）使用flags定義命令行參數(shù) 搜立。

dict_to_tf_example

SETS = ['train', 'val', 'trainval', 'test']
YEARS = ['VOC2007', 'VOC2012', 'merged']

def dict_to_tf_example(data,
                       dataset_directory,
                       label_map_dict,
                       ignore_difficult_instances=False,
                       image_subdirectory='JPEGImages'):
 
  img_path = os.path.join(data['folder'], image_subdirectory, data['filename'])
  full_path = os.path.join(dataset_directory, img_path)
  with tf.gfile.GFile(full_path, 'rb') as fid:
    encoded_jpg = fid.read()
  encoded_jpg_io = io.BytesIO(encoded_jpg)
  image = PIL.Image.open(encoded_jpg_io)
  if image.format != 'JPEG':
    raise ValueError('Image format not JPEG')
  key = hashlib.sha256(encoded_jpg).hexdigest()

  width = int(data['size']['width'])
  height = int(data['size']['height'])

  xmin = []
  ymin = []
  xmax = []
  ymax = []
  classes = []
  classes_text = []
  truncated = []
  poses = []
  difficult_obj = []
  for obj in data['object']:
    difficult = bool(int(obj['difficult']))
    if ignore_difficult_instances and difficult:
      continue

    difficult_obj.append(int(difficult))

    xmin.append(float(obj['bndbox']['xmin']) / width)
    ymin.append(float(obj['bndbox']['ymin']) / height)
    xmax.append(float(obj['bndbox']['xmax']) / width)
    ymax.append(float(obj['bndbox']['ymax']) / height)
    classes_text.append(obj['name'].encode('utf8'))
    classes.append(label_map_dict[obj['name']])
    truncated.append(int(obj['truncated']))
    poses.append(obj['pose'].encode('utf8'))

  example = tf.train.Example(features=tf.train.Features(feature={
      'image/height': dataset_util.int64_feature(height),
      'image/width': dataset_util.int64_feature(width),
      'image/filename': dataset_util.bytes_feature(
          data['filename'].encode('utf8')),
      'image/source_id': dataset_util.bytes_feature(
          data['filename'].encode('utf8')),
      'image/key/sha256': dataset_util.bytes_feature(key.encode('utf8')),
      'image/encoded': dataset_util.bytes_feature(encoded_jpg),
      'image/format': dataset_util.bytes_feature('jpeg'.encode('utf8')),
      'image/object/bbox/xmin': dataset_util.float_list_feature(xmin),
      'image/object/bbox/xmax': dataset_util.float_list_feature(xmax),
      'image/object/bbox/ymin': dataset_util.float_list_feature(ymin),
      'image/object/bbox/ymax': dataset_util.float_list_feature(ymax),
      'image/object/class/text': dataset_util.bytes_list_feature(classes_text),
      'image/object/class/label': dataset_util.int64_list_feature(classes),
      'image/object/difficult': dataset_util.int64_list_feature(difficult_obj),
      'image/object/truncated': dataset_util.int64_list_feature(truncated),
      'image/object/view': dataset_util.bytes_list_feature(poses),
  }))
  return example

這段主要定義了一個(gè)函數(shù)dict_to_tf_example的函數(shù)，用以將PASCAL數(shù)據(jù)集中的XML標(biāo)注文件轉(zhuǎn)換為tf.Example.
輸入?yún)?shù)為：

• data: 包含標(biāo)注信息的XML文件址晕。PASCAL數(shù)據(jù)集中顿锰，每張圖片的標(biāo)注信息存放于對(duì)應(yīng)的XML文件中。在main函數(shù)中刘陶，data是通過dataset_util.recursive_parse_xml_to_dict的方法將XML中信息導(dǎo)入為字典獲取的易核；
• dataset_directory: 你懂得浪默；
• label_map_dict： 為每一個(gè)類別賦予一個(gè)id纳决；由默認(rèn)路徑下已有文本給出碰逸；
• ignore_difficult_instances: 是否忽略數(shù)據(jù)集中的difficult_instances。 保持默認(rèn)即可阔加；
• image_subdirectory： 包含Images的PASCAL數(shù)據(jù)集的子文件夾饵史，同樣保持默認(rèn)即可。

在得到圖片的絕對(duì)路徑后(full_path)胜榔，通過GFile實(shí)現(xiàn)對(duì)圖片的讀取胳喷，并用PIL打開成為我們喜聞樂見的[c,h,w]格式。

而后夭织，將data傳過來的信息轉(zhuǎn)化為規(guī)范化的格式（x/width,y/height）添加到列表中吭露。說到這里就不得不夸一下dataset_util.recursive_parse_xml_to_dict這個(gè)配件了，from XML to dict尊惰，很方便的讲竿。

再然后定義了一個(gè)tf.train.Example 實(shí)例example弄屡，將獲得的信息全加進(jìn)去题禀，最后返回example。

def main(_):
  if FLAGS.set not in SETS:
    raise ValueError('set must be in : {}'.format(SETS))
  if FLAGS.year not in YEARS:
    raise ValueError('year must be in : {}'.format(YEARS))

  data_dir = FLAGS.data_dir
  years = ['VOC2007', 'VOC2012']
  if FLAGS.year != 'merged':
    years = [FLAGS.year]

  writer = tf.python_io.TFRecordWriter(FLAGS.output_path)

  label_map_dict = label_map_util.get_label_map_dict(FLAGS.label_map_path)

  for year in years:
    logging.info('Reading from PASCAL %s dataset.', year)
    examples_path = os.path.join(data_dir, year, 'ImageSets', 'Main',
                                 'aeroplane_' + FLAGS.set + '.txt')
    annotations_dir = os.path.join(data_dir, year, FLAGS.annotations_dir)
    examples_list = dataset_util.read_examples_list(examples_path)
    for idx, example in enumerate(examples_list):
      if idx % 100 == 0:
        logging.info('On image %d of %d', idx, len(examples_list))
      path = os.path.join(annotations_dir, example + '.xml')
      with tf.gfile.GFile(path, 'r') as fid:
        xml_str = fid.read()
      xml = etree.fromstring(xml_str)
      data = dataset_util.recursive_parse_xml_to_dict(xml)['annotation']

      tf_example = dict_to_tf_example(data, FLAGS.data_dir, label_map_dict,
                                      FLAGS.ignore_difficult_instances)
      writer.write(tf_example.SerializeToString())

  writer.close()


if __name__ == '__main__':
  tf.app.run()

把example保存為TFRecord格式膀捷。