TensorFlow Object Detection API教程——制作自己的数据集

感想

1. 前一段时间，利用tensorflow object detection跑了一些demo，然后成功的训练了自己的模型，这里我把我的方法分享出来，希望能够帮助大家。
2. tensorflow object detection api的github 开源地址为，https://github.com/tensorflow/models，这个模块比较新，有很多都在不断更新。我这里就object detection 来分享一下

1 数据集制作

我这里利用了voc格式的数据，事先要把数据集准备好，我把xml放在了 merged_xml文件夹下，把图片放在了images文件夹

我的xml文件示例为：

<?xml version="1.0" ?>
<annotation>
    <folder>VOC2007</folder>
    <filename>000009.jpg</filename>
    <size>
        <depth>3</depth>
        <width>500</width>
        <height>375</height>
    </size>
    <object>
        <name>person</name>
        <pose>Right</pose>
        <truncated>0</truncated>
        <difficult>0</difficult>
        <bndbox>
            <xmin>150</xmin>
            <ymin>141</ymin>
            <xmax>229</xmax>
            <ymax>284</ymax>
        </bndbox>
    </object>
    <object>
        <name>person</name>
        <pose>Right</pose>
        <truncated>0</truncated>
        <difficult>0</difficult>
        <bndbox>
            <xmin>285</xmin>
            <ymin>201</ymin>
            <xmax>327</xmax>
            <ymax>331</ymax>
        </bndbox>
    </object>
    <object>
        <name>person</name>
        <pose>Left</pose>
        <truncated>0</truncated>
        <difficult>0</difficult>
        <bndbox>
            <xmin>258</xmin>
            <ymin>198</ymin>
            <xmax>297</xmax>
            <ymax>329</ymax>
        </bndbox>
    </object>
</annotation>

解析也是按照这个解析的

然后利用下面的 train_test_split.py把xml数据集分为了train test validation三部分，代码如下：

import os
import random
import time
import shutil

xmlfilepath=r'merged_xml'
saveBasePath=r"./annotations"

trainval_percent=0.9
train_percent=0.85
total_xml = os.listdir(xmlfilepath)
num=len(total_xml)
list=range(num)
tv=int(num*trainval_percent)
tr=int(tv*train_percent)
trainval= random.sample(list,tv)
train=random.sample(trainval,tr)

print("train and val size",tv)
print("train size",tr)
# print(total_xml[1])
start = time.time()

# print(trainval)
# print(train)

test_num=0
val_num=0
train_num=0
# for directory in ['train','test',"val"]:
#         xml_path = os.path.join(os.getcwd(), 'annotations/{}'.format(directory))
#         if(not os.path.exists(xml_path)):
#             os.mkdir(xml_path)
#         # shutil.copyfile(filePath, newfile)
#         print(xml_path)
for i  in list:
    name=total_xml[i]
            # print(i)
    if i in trainval:  #train and val set
    # ftrainval.write(name)
        if i in train:
            # ftrain.write(name)
            # print("train")
            # print(name)
            # print("train: "+name+" "+str(train_num))
            directory="train"
            train_num+=1
            xml_path = os.path.join(os.getcwd(), 'annotations/{}'.format(directory))
            if(not os.path.exists(xml_path)):
                os.mkdir(xml_path)
            filePath=os.path.join(xmlfilepath,name)
            newfile=os.path.join(saveBasePath,os.path.join(directory,name))
            shutil.copyfile(filePath, newfile)

        else:
            # fval.write(name)
            # print("val")
            # print("val: "+name+" "+str(val_num))
            directory="validation"
            xml_path = os.path.join(os.getcwd(), 'annotations/{}'.format(directory))
            if(not os.path.exists(xml_path)):
                os.mkdir(xml_path)
            val_num+=1
            filePath=os.path.join(xmlfilepath,name)
            newfile=os.path.join(saveBasePath,os.path.join(directory,name))
            shutil.copyfile(filePath, newfile)
            # print(name)
    else:  #test set
        # ftest.write(name)
        # print("test")
        # print("test: "+name+" "+str(test_num))
        directory="test"
        xml_path = os.path.join(os.getcwd(), 'annotations/{}'.format(directory))
        if(not os.path.exists(xml_path)):
            os.mkdir(xml_path)
        test_num+=1
        filePath=os.path.join(xmlfilepath,name)
        newfile=os.path.join(saveBasePath,os.path.join(directory,name))
        shutil.copyfile(filePath, newfile)
            # print(name)

# End time
end = time.time()
seconds=end-start
print("train total : "+str(train_num))
print("validation total : "+str(val_num))
print("test total : "+str(test_num))
total_num=train_num+val_num+test_num
print("total number : "+str(total_num))
print( "Time taken : {0} seconds".format(seconds))

运行完以后，annotations文件夹下就放好了分类的xml，annotations有三个目录，分别是train，test，validation。

然后把xml转换成csv文件，我的代码文件名为xml_to_csv.py,，运行代码前，需要建一个data目录，用来放生成的csv文件，然后我的代码为：

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


def xml_to_csv(path):
    xml_list = []
    for xml_file in glob.glob(path + '/*.xml'):
        tree = ET.parse(xml_file)
        root = tree.getroot()
        # print(root)
        print(root.find('filename').text)
        for member in root.findall('object'):
            value = (root.find('filename').text,
                     int(root.find('size')[1].text),   #width
                     int(root.find('size')[2].text),   #height
                     member[0].text,
                     int(member[4][0].text),
                     int(float(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():
    for directory in ['train','test','validation']:
        xml_path = os.path.join(os.getcwd(), 'annotations/{}'.format(directory))
    # image_path = os.path.join(os.getcwd(), 'merged_xml')
        xml_df = xml_to_csv(xml_path)
        # xml_df.to_csv('whsyxt.csv', index=None)
        xml_df.to_csv('data/whsyxt_{}_labels.csv'.format(directory), index=None)
        print('Successfully converted xml to csv.')


main()

做完这一步以后，我们就来生成tfrecords文件，我的python文件名为generate_tfrecord.py,代码为：

"""
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):
    if row_label == 'car':
        return 1
    elif row_label == 'person':
        return 2
    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):
    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(_):
    writer = tf.python_io.TFRecordWriter(FLAGS.output_path)
    path = os.path.join(os.getcwd(), 'images')
    examples = pd.read_csv(FLAGS.csv_input)
    grouped = split(examples, 'filename')
    num=0
    for group in grouped:
        num+=1
        tf_example = create_tf_example(group, path)
        writer.write(tf_example.SerializeToString())
        if(num%100==0):  #每完成100个转换，打印一次
            print(num)

    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()

我运行的命令为：

python3 generate_tfrecord.py --csv_input=data/whsyxt_train_labels.csv --output_path=data/whsyxt_train.tfrecord
python3 generate_tfrecord.py --csv_input=data/whsyxt_test_labels.csv --output_path=data/whsyxt_test.tfrecord
python3 generate_tfrecord.py --csv_input=data/whsyxt_validation_labels.csv --output_path=data/whsyxt_validation.tfrecord

然后就获得了这三个训练需要的文件啦，训练方法请见我的下一篇博客

参考文献

[1].Introduction and Use - Tensorflow Object Detection API Tutorial.https://pythonprogramming.net/introduction-use-tensorflow-object-detection-api-tutorial/