基于Tensorflow训练物体、人像识别的模型

领导突发奇想一个idea，于是踏上了了解Tensorflow机器学习框架之路，踩过很多坑，做个记录。各位看官看的时候有些训练方式可能已经过时或者不对，见谅。

参考文档

安装tensorflow教程
https://www.leiphone.com/news/201606/ORlQ7uK3TIW8xVGF.html
腾讯社区
https://cloud.tencent.com/developer/article/1006123
model训练文档
http://blog.csdn.net/LiJiancheng0614/article/details/77756252

环境配置：

• ubuntu 16.0.4
• Python 2.7
• tensorflow cpu版本 1.4.1

训练过程：

1. 使用labelImg工具标注图片，生成相应的注释文件.xml，注意图片和xml不要放在同一个文件夹
2. 下载官方推荐的model

git clone https://github.com/tensorflow/models.git

cd 到models/research目录下添加路径

# From tensorflow/models/research
export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim

1. 使用xml_to_csv.python脚本将xml转换成csv文件

import os
import glob
import pandas as pd
import xml.etree.ElementTree as ET

def xml_to_csv(path):

    xml_list = []

    print("*********",glob.glob(path + '/*.xml'))

    for xml_file in glob.glob(path + '/*.xml'):

        tree = ET.parse(xml_file)

        root = tree.getroot()

        for member in root.findall('object'):

            value = (root.find('filename').text+'.JPG',    //如果生成的xml项后面没有图片格式声明，记得这里加上

                     int(root.find('size')[0].text),

                     int(root.find('size')[1].text),

                     member[0].text,

                     int(member[4][0].text),

                     int(member[4][1].text),

                     int(member[4][2].text),

                     int(member[4][3].text)

                     )

            xml_list.append(value)

    column_name = ['filename', 'width', 'height', 'class', 'xmin', 'ymin', 'xmax', 'ymax']

    xml_df = pd.DataFrame(xml_list, columns=column_name)

    return xml_df

def main():

    image_path = os.path.join(os.getcwd(), 'annotations')

    image_path = r'/mnt/hgfs/UbuntuShare/xml_hat'    //改这里的xml路径

    xml_df = xml_to_csv(image_path)

    xml_df.to_csv('hat.csv', index=None)    //生成的csv文件

    print('Successfully converted xml to csv.')

main()

1. 使用generate_tfrecord.python文件将csv转换成record文件，命令如下：

python generate_tfrecord.py --csv_input=sunglasses_test_labels.csv --output_path=sunglass_test.record

"""
Usage:
  # From tensorflow/models/
  # Create train data:
  python generate_tfrecord.py --csv_input=data/train_labels.csv  --output_path=train.record

  # Create test data:
  python generate_tfrecord.py --csv_input=data/test_labels.csv  --output_path=test.record
"""
from __future__ import division
from __future__ import print_function
from __future__ import absolute_import

import os
import io
import pandas as pd
import tensorflow as tf

from PIL import Image
from object_detection.utils import dataset_util
from collections import namedtuple, OrderedDict

flags = tf.app.flags
flags.DEFINE_string('csv_input', '', 'Path to the CSV input')
flags.DEFINE_string('output_path', '', 'Path to output TFRecord')
FLAGS = flags.FLAGS


# TO-DO replace this with label map
def class_text_to_int(row_label):
    print('********',row_label)
    if row_label == 'hat':    // 使用图片标注的label
        return 1
    else:
        None


def split(df, group):
    data = namedtuple('data', ['filename', 'object'])
    gb = df.groupby(group)
    return [data(filename, gb.get_group(x)) for filename, x in zip(gb.groups.keys(), gb.groups)]


def create_tf_example(group, path):
    print("-------------")
    with tf.gfile.GFile(os.path.join(path, '{}'.format(group.filename)), 'rb') as fid:
        encoded_jpg = fid.read()
    encoded_jpg_io = io.BytesIO(encoded_jpg)
    image = Image.open(encoded_jpg_io)
    width, height = image.size

    filename = group.filename.encode('utf8')
    image_format = b'jpg'
    xmins = []
    xmaxs = []
    ymins = []
    ymaxs = []
    classes_text = []
    classes = []

    for index, row in group.object.iterrows():
        xmins.append(row['xmin'] / width)
        xmaxs.append(row['xmax'] / width)
        ymins.append(row['ymin'] / height)
        ymaxs.append(row['ymax'] / height)
        classes_text.append(row['class'].encode('utf8'))
        classes.append(class_text_to_int(row['class']))

    tf_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(filename),
        'image/source_id': dataset_util.bytes_feature(filename),
        'image/encoded': dataset_util.bytes_feature(encoded_jpg),
        'image/format': dataset_util.bytes_feature(image_format),
        'image/object/bbox/xmin': dataset_util.float_list_feature(xmins),
        'image/object/bbox/xmax': dataset_util.float_list_feature(xmaxs),
        'image/object/bbox/ymin': dataset_util.float_list_feature(ymins),
        'image/object/bbox/ymax': dataset_util.float_list_feature(ymaxs),
        'image/object/class/text': dataset_util.bytes_list_feature(classes_text),
        'image/object/class/label': dataset_util.int64_list_feature(classes),
    }))
    return tf_example

def main(_):
    print("-------------")
    writer = tf.python_io.TFRecordWriter(FLAGS.output_path)
    path = os.path.join(os.getcwd(), 'images_hat')   //图片文件路径
    examples = pd.read_csv(FLAGS.csv_input)
    grouped = split(examples, 'filename')
    for group in grouped:
        tf_example = create_tf_example(group, path)
        writer.write(tf_example.SerializeToString())

    writer.close()
    output_path = os.path.join(os.getcwd(), FLAGS.output_path)
    print('Successfully created the TFRecords: {}'.format(output_path))

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

5.生成指定的 train.record。接下来指定标签名称，仿照models/ object_detection/data/ pet_label_map.pbtxt，重新创建一个文件，指定标签名

item {![这里写图片描述](https://img-blog.csdn.net/20180125165430032?watermark/2/text/aHR0cDovL2Jsb2cuY3Nkbi5uZXQvdTAxMzIxNDU4OA==/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70/gravity/SouthEast)
  id: 1
  name: 'dog'
}

6.打开 object_detection/samples/configs/ssd_mobilenet_v1_pets.config进行编辑修改，将num值改成自己需要训练识别的object种类个数

首先上这里下载预训练model，推荐第一个，将第一行的路径指向下载的ssd_mobilenet_v1_coco/model.ckpt，接下来圈红的改为自己的record、pbtxt路径