1. 部署环境：
在PC上安装CUDA10和对应cuDNN，网上教程很多，这里不再累赘， 推荐使用conda集成环境，1. 新建python环境，2. 安装tensorflow-gpu=1.13, TensorFlow对象检测API需要使用其GitHub存储库中提供的特定目录结构, 所以第三步：从GitHub下载TensorFlow对象检测API存储库(下载TF V1.13版本，这里要与我们Python tensorflow对应 wget
安装依赖
sudo apt-get install protobuf-compiler python-pil python-lxml python-tk pip install --user Cython pip install --user contextlib2 pip install --user jupyter pip install --user matplotlib
安装 COCO API
下载 cocoapi ，然后复制 pycocotools 文件夹到 tensorflow/models/research 文件夹。默认使用基于 Pascal VOC 的评价指标； 如果你对使用 COCO 评价指标感兴趣：使用 COCO 目标检测（object detection）指标，请添加metrics_set: "coco_detection_metrics"到配置文件eval_config消息中；使用 COCO 实例分割（instance segmentation）指标，请添加metrics_set: "coco_mask_metrics"到配置文件eval_config消息中。
git clone https://github.com/cocodataset/cocoapi.git cd cocoapi/PythonAPI make cp -r pycocotools <path_to_tensorflow>/models/research/
编译Protobuf和将库添加进 PYTHONPATH
Tensorflow Object Detection API 使用 Protobufs 来控制模型与训练参数。在使用框架之前，Protobuf 库必须被编译。这可以在 tensorflow/models/research/ 文件夹下运行命令：
./bin/protoc object_detection/protos/*.proto --python_out=.
当在本地运行时，tensorflow/models/research/ 和 slim 文件夹需要加入 PYTHONPATH 。这可以在 tensorflow/models/research/ 文件夹下运行下列命令来完成：
export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim
注意： 如果这里添加路径之后还提示找不到"nets"文件，可以直接复制slim文件到research目录下
测试是否成功
python object_detection/builders/model_builder_test.py
返回OK则OK
2. 制作数据集
使用label-image标注工具对样本进行标注，得到VOC格式数据。将所有的图片放入images/文件夹，标注得到的xml文件保存到merged_xml/文件夹内，并新建文件夹Annotations/
训练集划分与配置文件修改
新建train_test_split.py把xml数据集分为了train 、test、 validation三部分，并存储在Annotations文件夹中，train为训练集占76.5%，test为测试集10%，validation为验证集13.5%，train_test_split.py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) start = time.time() test_num = 0 val_num = 0 train_num = 0 for i in list: name = total_xml[i] if i in trainval: # train and val set if i in train: 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: 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) else: 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) 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))
xml文件转换为csv文件
新建csvdata/目录存放生成的csv文件，代码如下：
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.find('filename').text) for member in root.findall('object'): value = (root.find('filename').text, int(root.find('size')[0].text), #width int(root.find('size')[1].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)) xml_df = xml_to_csv(xml_path) xml_df.to_csv('csvdata/tf_{}.csv'.format(directory), index=None) print('Successfully converted xml to csv.') main()
在csvdata/文件夹下生成训练、验证和测试的csv格式文件：
csv格式数据生成tf record格式数据
建generate_tfrecord.py脚本，并新建tfdata/文件夹，代码如下：
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 object_detection.utils import label_map_util from collections import namedtuple flags = tf.app.flags flags.DEFINE_string('csv_input', '', 'Path to the CSV input') flags.DEFINE_string('images_input', '', 'Path to the images input') flags.DEFINE_string('output_path', '', 'Path to output TFRecord') flags.DEFINE_string('label_map_path', '', 'Path to label map proto') FLAGS = flags.FLAGS 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, label_map_dict, images_path): with tf.gfile.GFile(os.path.join( images_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(label_map_dict[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) label_map_dict = label_map_util.get_label_map_dict(FLAGS.label_map_path) images_path = FLAGS.images_input examples = pd.read_csv(FLAGS.csv_input) grouped = split(examples, 'filename') for group in grouped: tf_example = create_tf_example(group, label_map_dict, images_path) writer.write(tf_example.SerializeToString()) writer.close() output_path = FLAGS.output_path print('Successfully created the TFRecords: {}'.format(output_path)) if __name__ == '__main__': tf.app.run()
用法：
python generate_tfrecord.py \ --csv_input=./csvdata/tf_train.csv \ --images_input=images \ --output_path=./tfdata/train.record \ --label_map_path=./label_map.pbtxt
类似的依次生成训练、验证和测试数据集：
类别文件
创建label_map.pbtxt文件， 根据自己训练的类别进行修改， 有几个类别就做几个itme！
item { name: "face" id: 1 display_name: "face" } item { name: "telephone" id: 2 display_name: "telephone" } item { name: "cigarette" id: 3 display_name: "cigarette" }
配置pipeline.config
到models/research/object_detection/samples/configs/文件夹下将ssd_mobilenet_v2_coco.config拷贝到训练文件夹下，修改内容主要是：
①总类别数
②tfrecord文件的路径，包括训练集、验证集等路径
③label_map的路径
④预训练模型路径，如果没有则注释掉。也可以设置网络的各种学习参数，如：batch_size，学习率和退化率，训练的总步数等。
●num_classes:3
●fine_tune_checkpoint:“ssd_mobilenet_v1_coco_11_06_2017/model.ckpt” # 预训练模型位置
●num_steps:30000 # 训练步数设置，根据自己数据量来设置，默认为200000
●train_input_reader/input_path:“train.record” # 注意修改成自己的路径位置
●train_input_reader/label_map_path:“label_map.pbtxt” # 类别文件位置，注意修改成自己的路径位置
●num_examples:78 # test数据集的数量
●num_visualizations:78
●#max_evals:10 ＃注释这个变量，避免一些错误，个人习惯，之前因为这个遇到过错误
●eval_input_reader/inputpath:“test.record” # 注意修改成自己的位置
●eval_input_reader/label_map_path: “label_map.pbtxt” # 注意修改成自己的路径位置
训练模型与导出模型
首先在legacy文件夹中复制一份train.py到object_detection文件夹下，然后运行以下指令(ckpt模型训练后的输出位置)
python train.py --logtostderr --train_dir=training/ --pipeline_config_path=ssd_mobilenet_v2_coco.config
训练结果
生成一堆models.ckpt-xxx的文件，不同数字代表不同训练步数下保存的模型文件
