TensorFlow：基于TensorFlow_object_detection_api实现图片识别

在配置完TensorFlow-object-detection-api的环境变量后，可以运行官方的demo。接下来就是通过自己训练的模型识别自己所需要的图片。先说明在此过程中遇到的两大巨坑：

1，在anaconda prompt下文件名中不能用空格，即空格可用_替代，这样可以为以后省下不少麻烦

2，在下载TensorFlow时，使用TensorFlow-gpu，因为cpu下实在太慢，下载gpu时，要用到cudnn和cuda，这是在网上找的教程，较为详细：

https://v.youku.com/v_show/id_XMzgwOTUzNDU3Ng==.html?x&sharefrom=android&sharekey=d9547d3a1b5e967e76a1d062b6c6446b1

下面为具体过程：

1，使用labelIMg进行数据集的收集，在image文件夹下建立train和test文件（分别包含jpg和xml文件）

因为TensorFlow默认识别一种为record的文件夹，所以要用两个脚本文件，分别实现xml_to_csv，和生成record文件，代码如下：

# -*- coding: utf-8 -*-
"""
Created on Fri Sep 14 23:58:21 2018

@author: asus
"""

# -*- coding: utf-8 -*-
"""
Created on Tue Jan 16 00:52:02 2018
@author: Xiang Guo
"""

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

os.chdir('D:\\object_detection_api\\models-master\\research\\object_detection\\image\\test')
path = 'D:\\object_detection_api\\models-master\\research\\object_detection\\image\\test'

def xml_to_csv(path):
    xml_list = []
    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,
                     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 = path
    xml_df = xml_to_csv(image_path)
    xml_df.to_csv('LYF_test.csv', index=None)
    print('Successfully converted xml to csv.')


main()

在xml转csv过程中，主要文件路径（是两个反斜杠），还有最后生成的文件名。之后可分别在test和train文件下生成LYF_test.csv和LYF_train_csv文件。之后，将这两个文件夹放入object_detection文件夹下的data文件夹内

之后，运行generate_TFR.py,生成record文件，这一步中，要特别注意在代码92行，记得路径问题，另外需要注意的是，需在D:\object_detection_api\models-master\research目录下运行此python脚本文件。此python脚本文件如下：

# -*- coding: utf-8 -*-
"""
Created on Tue Jan 16 01:04:55 2018
@author: Xiang Guo
"""

"""
Usage:
  # From tensorflow/models/
  # Create train data:
  python generate_TFR.py --csv_input=data/LYF_train.csv  --output_path=data/LYF_train.record
  # Create test data:
   python generate_TFR.py --csv_input=data/LYF_test.csv  --output_path=data/LYF_test.record
"""



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

os.chdir('D:\\object_detection_api\\models-master\\research\\object_detection\\')

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 == 'LYF':
        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):
    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(), 'image\\test\\')
    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()

至此，在data文件夹下分别有两个csv文件和record文件。

2，config文件的配置：

模型地址：https://github.com/tensorflow/models/tree/master/research/object_detection/samples/configs

以选的模型ssd_mobilenet_v1_coco.config为例，将ssd_mobilenet_v1_coco.config放在training文件夹内，并用文本编辑器notepad++打开，因为我的标识对象只有一个“LYF”，修改第九行num_classes:为1，141行的bitch_wise为4，在171行-189行分别设置训练和测试的路径，并在data文件夹下新建一个LYF.pbtxt文件夹。config文件代码和LYF.pbtxt文件代码分别如下：

# SSD with Mobilenet v1 configuration for MSCOCO Dataset.
# Users should configure the fine_tune_checkpoint field in the train config as
# well as the label_map_path and input_path fields in the train_input_reader and
# eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
# should be configured.

model {
  ssd {
    num_classes: 1
    box_coder {
      faster_rcnn_box_coder {
        y_scale: 10.0
        x_scale: 10.0
        height_scale: 5.0
        width_scale: 5.0
      }
    }
    matcher {
      argmax_matcher {
        matched_threshold: 0.5
        unmatched_threshold: 0.5
        ignore_thresholds: false
        negatives_lower_than_unmatched: true
        force_match_for_each_row: true
      }
    }
    similarity_calculator {
      iou_similarity {
      }
    }
    anchor_generator {
      ssd_anchor_generator {
        num_layers: 6
        min_scale: 0.2
        max_scale: 0.95
        aspect_ratios: 1.0
        aspect_ratios: 2.0
        aspect_ratios: 0.5
        aspect_ratios: 3.0
        aspect_ratios: 0.3333
      }
    }
    image_resizer {
      fixed_shape_resizer {
        height: 300
        width: 300
      }
    }
    box_predictor {
      convolutional_box_predictor {
        min_depth: 0
        max_depth: 0
        num_layers_before_predictor: 0
        use_dropout: false
        dropout_keep_probability: 0.8
        kernel_size: 1
        box_code_size: 4
        apply_sigmoid_to_scores: false
        conv_hyperparams {
          activation: RELU_6,
          regularizer {
            l2_regularizer {
              weight: 0.00004
            }
          }
          initializer {
            truncated_normal_initializer {
              stddev: 0.03
              mean: 0.0
            }
          }
          batch_norm {
            train: true,
            scale: true,
            center: true,
            decay: 0.9997,
            epsilon: 0.001,
          }
        }
      }
    }
    feature_extractor {
      type: 'ssd_mobilenet_v1'
      min_depth: 16
      depth_multiplier: 1.0
      conv_hyperparams {
        activation: RELU_6,
        regularizer {
          l2_regularizer {
            weight: 0.00004
          }
        }
        initializer {
          truncated_normal_initializer {
            stddev: 0.03
            mean: 0.0
          }
        }
        batch_norm {
          train: true,
          scale: true,
          center: true,
          decay: 0.9997,
          epsilon: 0.001,
        }
      }
    }
    loss {
      classification_loss {
        weighted_sigmoid {
        }
      }
      localization_loss {
        weighted_smooth_l1 {
        }
      }
      hard_example_miner {
        num_hard_examples: 3000
        iou_threshold: 0.99
        loss_type: CLASSIFICATION
        max_negatives_per_positive: 3
        min_negatives_per_image: 0
      }
      classification_weight: 1.0
      localization_weight: 1.0
    }
    normalize_loss_by_num_matches: true
    post_processing {
      batch_non_max_suppression {
        score_threshold: 1e-8
        iou_threshold: 0.6
        max_detections_per_class: 100
        max_total_detections: 100
      }
      score_converter: SIGMOID
    }
  }
}

train_config: {
  batch_size: 4
  optimizer {
    rms_prop_optimizer: {
      learning_rate: {
        exponential_decay_learning_rate {
          initial_learning_rate: 0.004
          decay_steps: 800720
          decay_factor: 0.95
        }
      }
      momentum_optimizer_value: 0.9
      decay: 0.9
      epsilon: 1.0
    }
  }
  # Note: The below line limits the training process to 200K steps, which we
  # empirically found to be sufficient enough to train the pets dataset. This
  # effectively bypasses the learning rate schedule (the learning rate will
  # never decay). Remove the below line to train indefinitely.
  num_steps: 200000
  data_augmentation_options {
    random_horizontal_flip {
    }
  }
  data_augmentation_options {
    ssd_random_crop {
    }
  }
}

train_input_reader: {
  tf_record_input_reader {
    input_path: "data/LYF_train.record"
  }
  label_map_path: "data/LYF.pbtxt"
}

eval_config: {
  num_examples: 8000
  # Note: The below line limits the evaluation process to 10 evaluations.
  # Remove the below line to evaluate indefinitely.
  max_evals: 10
}

eval_input_reader: {
  tf_record_input_reader {
    input_path: "data/LYF_test.record"
  }
  label_map_path: "data/LYF.pbtxt"
  shuffle: false
  num_readers: 1
}

item {
  id: 1
  name: "LYF"
}

3，训练模型

注意：！！！！！！为避免GPU显存被占满，在训练命令执行前，先执行命令

Windows SET CUDA_VISIBLE_DEVICES=0
Linux export CUDA_VISIBLE_DEVICES=0

之后执行训练命令：

python legacy/train.py --pipeline_config_path=D:/object_detection_api/models-master/research/object_detection/training/ssd_mobilenet_v1_coco.config --train_dir=D:/object_detection_api/models-master/research/object_detection/training --num_train_steps=100 --num_eval_steps=5 --alsologtostderr

之后，可以看到训练开始了，如图所示是我训练八个小时后的结果：

打开tensorboard可视化训练过程：

tensorboard --logdir=training

4，导出模型

在object_detection文件夹下运行export_inference_graph.py文件，运行时，需要使用以下命令行：

python export_inference_graph.py --input_type image_tensor --pipeline_config_path training/ssd_inception_v3_pets.config --trained_checkpoint_prefix training/model.ckpt-15441 --output_directory person_detection

ckpt后的数字根据自己上一步training文件夹下最大的训练步数判断，LYF_detection是我在object_detection文件夹下新创建的，运行成功后，会在文件夹下生成如下文件：

其中，.pb文件就是下一步需要测试的模型。

测试结果及源码见：http://localhost:8888/notebooks/object_detection_tutorial-Copy1.ipynb

部分测试结果如下：

可见，由于数据集过小，测试结果不尽人意，但终于走通了第一步！

整个过程自己走了很过弯路，也不算是弯路吧，想成功一点终究要付出辛苦的，路漫漫其修远兮，吾将上下而求索！