TensorFlow Object Detection API 实践

1. Install object_detection API

pip install tensorflow
git clone https://github.com/tensorflow/models --recursive 
cd /root/myfloder/models/research
protoc object_detection/protos/*.proto --python_out=.
export PYTHONPATH=$PYTHONPATH:'/Users/terry/models/research':'/Users/terry/models/research/slim'
#解压后
python object_detection/builders/model_builder_test.py

2. Download and convent data

# From tensorflow/models/research/
wget http://www.robots.ox.ac.uk/~vgg/data/pets/data/images.tar.gz
wget http://www.robots.ox.ac.uk/~vgg/data/pets/data/annotations.tar.gz
tar -xvf annotations.tar.gz
tar -xvf images.tar.gz
python object_detection/dataset_tools/create_pet_tf_record.py \
    --label_map_path=object_detection/data/pet_label_map.pbtxt \
    --data_dir=`pwd` \
    --output_dir=`pwd`

4. config pipline file

# Faster R-CNN with Resnet-101 (v1) configured for the Oxford-IIIT Pet 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 {
  faster_rcnn {
    num_classes: 37
    image_resizer {
      keep_aspect_ratio_resizer {
        min_dimension: 600
        max_dimension: 1024
      }
    }
    feature_extractor {
      type: 'faster_rcnn_resnet101'
      first_stage_features_stride: 16
    }
    first_stage_anchor_generator {
      grid_anchor_generator {
        scales: [0.25, 0.5, 1.0, 2.0]
        aspect_ratios: [0.5, 1.0, 2.0]
        height_stride: 16
        width_stride: 16
      }
    }
    first_stage_box_predictor_conv_hyperparams {
      op: CONV
      regularizer {
        l2_regularizer {
          weight: 0.0
        }
      }
      initializer {
        truncated_normal_initializer {
          stddev: 0.01
        }
      }
    }
    first_stage_nms_score_threshold: 0.0
    first_stage_nms_iou_threshold: 0.7
    first_stage_max_proposals: 300
    first_stage_localization_loss_weight: 2.0
    first_stage_objectness_loss_weight: 1.0
    initial_crop_size: 14
    maxpool_kernel_size: 2
    maxpool_stride: 2
    second_stage_box_predictor {
      mask_rcnn_box_predictor {
        use_dropout: false
        dropout_keep_probability: 1.0
        fc_hyperparams {
          op: FC
          regularizer {
            l2_regularizer {
              weight: 0.0
            }
          }
          initializer {
            variance_scaling_initializer {
              factor: 1.0
              uniform: true
              mode: FAN_AVG
            }
          }
        }
      }
    }
    second_stage_post_processing {
      batch_non_max_suppression {
        score_threshold: 0.0
        iou_threshold: 0.6
        max_detections_per_class: 100
        max_total_detections: 300
      }
      score_converter: SOFTMAX
    }
    second_stage_localization_loss_weight: 2.0
    second_stage_classification_loss_weight: 1.0
  }
}

train_config: {
  batch_size: 1
  optimizer {
    momentum_optimizer: {
      learning_rate: {
        manual_step_learning_rate {
          initial_learning_rate: 0.0003
          schedule {
            step: 900000
            learning_rate: .00003
          }
          schedule {
            step: 1200000
            learning_rate: .000003
          }
        }
      }
      momentum_optimizer_value: 0.9
    }
    use_moving_average: false
  }
  gradient_clipping_by_norm: 10.0
  fine_tune_checkpoint: "/Users/terry/models/research/0terry_pet/model/faster_rcnn_resnet101_coco_11_06_2017/model.ckpt"
  from_detection_checkpoint: true
  load_all_detection_checkpoint_vars: true
  # 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 {
    }
  }
}

train_input_reader: {
  tf_record_input_reader {
    input_path: "/Users/terry/models/research/0terry_pet/data/train/pet_faces_train.record-?????-of-00010"
  }
  label_map_path: "/Users/terry/models/research/0terry_pet/data/pet_label_map.pbtxt"
}

eval_config: {
  metrics_set: "coco_detection_metrics"
  num_examples: 1101
}

eval_input_reader: {
  tf_record_input_reader {
    input_path: "/Users/terry/models/research/0terry_pet/data/eval/pet_faces_val.record-?????-of-00010"
  }
  label_map_path: "/Users/terry/models/research/0terry_pet/data/pet_label_map.pbtxt"
  shuffle: false
  num_readers: 1
}

3. Move the files to a new folder 0terry_pet

+data
  -pet_label_map.pbtxt
  -train
    pet_faces_train.record-00009-of-00010 [10 files]
  -eval TFRecord file
    pet_faces_val.record-00009-of-00010 [10 files]
  -images
  -annotations
+models
  + model
    _faster_rcnn_resnet101_pets.config
    -faster_rcnn_resnet101_coco_11_06_2017
    +train
    +eval

3. Train data

# From the /Users/terry/models/research directory
PIPELINE_CONFIG_PATH=0terry_pet/model/faster_rcnn_resnet101_pets.config
MODEL_DIR=0terry_pet/model/faster_rcnn_resnet101_coco_11_06_2017
NUM_TRAIN_STEPS=500
SAMPLE_1_OF_N_EVAL_EXAMPLES=1
python object_detection/model_main.py \
   --pipeline_config_path=${PIPELINE_CONFIG_PATH} \
   --model_dir=${MODEL_DIR} \
   --num_train_steps=${NUM_TRAIN_STEPS} \
   --sample_1_of_n_eval_examples=$SAMPLE_1_OF_N_EVAL_EXAMPLES \
   --alsologtostderr

4. Startup Tensorbroad

# From the /Users/terry/models/research directory
MODEL_DIR=0terry_pet/model/faster_rcnn_resnet101_coco_11_06_2017
tensorboard --logdir=${MODEL_DIR}

观察loss不怎么变小时，可停止训练

5. create pd file

将train文件夹下的最大的数字，如下文件复制到checkpoint_best文件夹下，并去除ckpt后面的“-数字”，checkpoint文件内相应也要改

checkpoint
model.ckpt.data-00000-of-00001
model.ckpt.index
model.ckpt.meta

在项目根目录下执行：

 # From the /Users/terry/models/research directory
python object_detection/export_inference_graph.py \
--pipeline_config_path 0terry_pet/model/faster_rcnn_resnet101_pets.config \
--trained_checkpoint_prefix 0terry_pet/model/checkpoint_best/model.ckpt \
--output_directory 0terry_pet/model/pb

6. 预测新的图片

import numpy as np
import os
import six.moves.urllib as urllib
import sys
import tarfile
import tensorflow as tf
import zipfile

from distutils.version import StrictVersion
from collections import defaultdict
from io import StringIO
from matplotlib import pyplot as plt
from PIL import Image

# This is needed since the notebook is stored in the object_detection folder.
sys.path.append("..")
from object_detection.utils import ops as utils_ops

if StrictVersion(tf.__version__) < StrictVersion('1.9.0'):
  raise ImportError('Please upgrade your TensorFlow installation to v1.9.* or later!')

# This is needed to display the images.
%matplotlib inline

from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util


# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_FROZEN_GRAPH = '/Users/terry/models/research/0terry_pet/model/pb/frozen_inference_graph.pb'

# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('/Users/terry/models/research/0terry_pet/data', 'pet_label_map.pbtxt')

detection_graph = tf.Graph()
with detection_graph.as_default():
  od_graph_def = tf.GraphDef()
  with tf.gfile.GFile(PATH_TO_FROZEN_GRAPH, 'rb') as fid:
    serialized_graph = fid.read()
    od_graph_def.ParseFromString(serialized_graph)
    tf.import_graph_def(od_graph_def, name='')
    
#Loading label map
category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)   


def load_image_into_numpy_array(image):
  (im_width, im_height) = image.size
  return np.array(image.getdata()).reshape(
      (im_height, im_width, 3)).astype(np.uint8)
# If you want to test the code with your images, just add path to the images to the TEST_IMAGE_PATHS.
PATH_TO_TEST_IMAGES_DIR = '/Users/terry/models/research/0terry_pet/data/test_images'
TEST_IMAGE_PATHS = [ os.path.join(PATH_TO_TEST_IMAGES_DIR, 'image{}.jpg'.format(i)) for i in range(1, 3) ]

# Size, in inches, of the output images.
IMAGE_SIZE = (12, 8)

def run_inference_for_single_image(image, graph):
  with graph.as_default():
    with tf.Session() as sess:
      # Get handles to input and output tensors
      ops = tf.get_default_graph().get_operations()
      all_tensor_names = {output.name for op in ops for output in op.outputs}
      tensor_dict = {}
      for key in [
          'num_detections', 'detection_boxes', 'detection_scores',
          'detection_classes', 'detection_masks'
      ]:
        tensor_name = key + ':0'
        if tensor_name in all_tensor_names:
          tensor_dict[key] = tf.get_default_graph().get_tensor_by_name(tensor_name)
      if 'detection_masks' in tensor_dict:
        # The following processing is only for single image
        detection_boxes = tf.squeeze(tensor_dict['detection_boxes'], [0])
        detection_masks = tf.squeeze(tensor_dict['detection_masks'], [0])
        # Reframe is required to translate mask from box coordinates to image coordinates and fit the image size.
        real_num_detection = tf.cast(tensor_dict['num_detections'][0], tf.int32)
        detection_boxes = tf.slice(detection_boxes, [0, 0], [real_num_detection, -1])
        detection_masks = tf.slice(detection_masks, [0, 0, 0], [real_num_detection, -1, -1])
        detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(
            detection_masks, detection_boxes, image.shape[0], image.shape[1])
        detection_masks_reframed = tf.cast(
            tf.greater(detection_masks_reframed, 0.5), tf.uint8)
        # Follow the convention by adding back the batch dimension
        tensor_dict['detection_masks'] = tf.expand_dims(
            detection_masks_reframed, 0)
      image_tensor = tf.get_default_graph().get_tensor_by_name('image_tensor:0')

      # Run inference
      output_dict = sess.run(tensor_dict,
                             feed_dict={image_tensor: np.expand_dims(image, 0)})

      # all outputs are float32 numpy arrays, so convert types as appropriate
      output_dict['num_detections'] = int(output_dict['num_detections'][0])
      output_dict['detection_classes'] = output_dict[
          'detection_classes'][0].astype(np.uint8)
      output_dict['detection_boxes'] = output_dict['detection_boxes'][0]
      output_dict['detection_scores'] = output_dict['detection_scores'][0]
      if 'detection_masks' in output_dict:
        output_dict['detection_masks'] = output_dict['detection_masks'][0]
  return output_dict

for image_path in TEST_IMAGE_PATHS:
  image = Image.open(image_path)
  # the array based representation of the image will be used later in order to prepare the
  # result image with boxes and labels on it.
  image_np = load_image_into_numpy_array(image)
  # Expand dimensions since the model expects images to have shape: [1, None, None, 3]
  image_np_expanded = np.expand_dims(image_np, axis=0)
  # Actual detection.
  output_dict = run_inference_for_single_image(image_np, detection_graph)
  # Visualization of the results of a detection.
  vis_util.visualize_boxes_and_labels_on_image_array(
      image_np,
      output_dict['detection_boxes'],
      output_dict['detection_classes'],
      output_dict['detection_scores'],
      category_index,
      instance_masks=output_dict.get('detection_masks'),
      use_normalized_coordinates=True,
      line_thickness=8)
  plt.figure(figsize=IMAGE_SIZE)
  plt.imshow(image_np)