参考自:
https://www.cnblogs.com/leviatan/p/10740105.html
https://www.cnblogs.com/gezhuangzhuang/p/10613468.html
关于如何安装tensorflow-gpu参考我这篇博客
https://blog.csdn.net/ourkix/article/details/103577082
目录
2. 把xml转换成csv文件,xml_to_csv.py 将生成的csv文件放在 object_detection/data/
3. 生成tfrecord文件,在research目录下建立generate_tfrecord.py
1. 在object_detection/data文件夹下创建标签分类的配置文件(labelmap.pbtxt),需要检测几种目标,将创建几个id,代码如下:
2. 配置管道配置文件,找到object_detection/samples/config/ssd_mobilenet_v1_coco.config,复制到data文件夹下。修改后的代码如下:
3.下载预训练模型(用我上传的文件的话,已经在object_detection/ssd_model/ssd_mobilenet目录下了)
4. 开始训练(这个train.py 文件可能就在object_detection目录下 也可能在object_detection/legacy下)
6.测试模型(在object_detection目录下创建文件seahorse_ssd_detect.py)
下载文件
下载地址: https://github.com/tensorflow/models
也可以使用我上传的里面有数据集和预训练文件 和 测试图片,文件有点大分卷下载了,要都下载下来一起解压
下载地址:https://download.csdn.net/download/ourkix/12068490
下载地址:https://download.csdn.net/download/ourkix/12068504
下载后得到一个 models-master.zip 文件,解压后移动到 (关于如何在文件查看其中看到隐藏的文件 Ctrl + H 快捷键)
/home/nvidia/.local/lib/python3.6/site-packages/tensorflow
文件夹下,并重命名为 models
如果用我上传的,下载解压后是个models文件夹,里面还有个models,进去吧里面的models复制到
/home/nvidia/.local/lib/python3.6/site-packages/tensorflow
文件夹下
依赖安装 如按之前博客来的话 应该已经安装好了
python3 -m pip install pillow --user
python3 -m pip install lxml --user
python3 -m pip install matplotlib --user
python3 -m pip install pandas --user
这里查看自己是否有安装 protobuf
protoc --version
出现
libprotoc 3.0.0
代表有安装
如没安装
sudo apt-get install -y python3-protobuf
#也可以用pip
python3 -m pip install protobuf --user
进入 models/research/
目录,并编译 protobuf (这里可能会报错 没有pandas 库 安装就是了)
cd /home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research
protoc object_detection/protos/*.proto --python_out=.
安装 object_detection 库
python3 setup.py build
python3 setup.py install
设置 PYTHONPATH
编辑 .bashrc文件
sudo gedit ~/.bashrc
最后加上
export PYTHONPATH=$PYTHONPATH:/home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research
export PYTHONPATH=$PYTHONPATH:/home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research/slim
保存,使环境生效
source ~/.bashrc
测试 object_detection 库是否安装成功
cd /home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research
python3 object_detection/builders/model_builder_test.py
运行测试目标检测脚本测试 在object_detection目录下有个 object-detection-turorial.ipynb 这里不用jupyter-notebook,改用python,更方便。
新建一个文件 object-detection-turorial.py
touch object-detection-turorial.py
编辑,写入
import numpy as np
import os
import six.moves.urllib as urllib
import sys
import tarfile
import tensorflow as tf
import zipfile
import matplotlib
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!')
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!')
from utils import label_map_util
from utils import visualization_utils as vis_util
global output_num
global output_img_dic
matplotlib.use('TkAgg')
# What model to download.
MODEL_NAME = 'ssd_mobilenet_v1_coco_2017_11_17'
MODEL_FILE = MODEL_NAME + '.tar.gz'
DOWNLOAD_BASE = 'http://download.tensorflow.org/models/object_detection/'
# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_FROZEN_GRAPH = MODEL_NAME + '/frozen_inference_graph.pb'
# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('data', 'mscoco_label_map.pbtxt')
print(PATH_TO_LABELS)
# For the sake of simplicity we will use only 2 images:
# image1.jpg
# image2.jpg
# 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 = '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)
output_num = 1
output_img_dic = r'\output_images'
opener = urllib.request.URLopener()
print("--\n")
opener.retrieve(DOWNLOAD_BASE + MODEL_FILE, MODEL_FILE)
print("--\n")
tar_file = tarfile.open(MODEL_FILE)
for file in tar_file.getmembers():
file_name = os.path.basename(file.name)
if 'frozen_inference_graph.pb' in file_name:
tar_file.extract(file, os.getcwd())
print("--\n")
detection_graph = tf.Graph()
with detection_graph.as_default():
od_graph_def = tf.compat.v1.GraphDef()
with tf.io.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='')
print("--\n")
category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)
print("--\n")
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)
def run_inference_for_single_image(image, graph):
with graph.as_default():
with tf.compat.v1.Session() as sess:
# Get handles to input and output tensors
ops = tf.compat.v1.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.compat.v1.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)
print(1,image_np)
plt.imshow(image_np)
plt.show()
if not os.path.exists(output_img_dic):
os.mkdir(output_img_dic)
output_img_path = os.path.join(output_img_dic,str(output_num)+".png")
plt.savefig(output_img_path)
保存,运行
python3 object-detection-turorial.py
等待运行,nano运行比较久,要下载文件什么的,等个2-3分钟。
训练自己的数据集
生成tfrecord文件
VOC数据集目录结构是这样的
我在object_detection目录下建立了ssd_model目录,里面放了VOCdeckit,我会提供整个models文件夹内容(包括预训练模型,海马数据集,测试数据),你们可以按我的来
|--VOCdevkit
|--VOC2007
|--Annotations
|--ImageSets
|--Layout
|--Main
|--Segmentation
|--JPEGImages
1. 在自己的voc数据格式文件夹内,新建 train_test_split.py 把xml文件数据集分为了train、test、validation三部分,并存储在Annotations文件夹中,训练验证集占80%,测试集占20%。训练集占训练验证集的80%。代码如下:
import os
import random
import time
import shutil
xmlfilepath=r'./Annotations'
saveBasePath=r"./Annotations"
trainval_percent=0.8
train_percent=0.8
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))
2. 把xml转换成csv文件,xml_to_csv.py 将生成的csv文件放在 object_detection/data/
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('whsyxt.csv', index=None)
xml_df.to_csv('/home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research/object_detection/data/seahorse_{}_labels.csv'.format(directory), index=None)
print('Successfully converted xml to csv.')
main()
3. 生成tfrecord文件,在research目录下建立generate_tfrecord.py
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
#Usage:
# From tensorflow/models/
# Create train data:
#python generate_tfrecord.py --csv_input=data/tv_vehicle_labels.csv --output_path=train.record
# Create test data:
#python generate_tfrecord.py --csv_input=data/test_labels.csv --output_path=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('/home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research/')
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):
# 你的所有类别
if row_label == 'seahorse':
return 1
else:
return 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(), 'object_detection/ssd_model/VOCdevkit/VOC2007/JPEGImages/')
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()
主要是在 row_label 这里要添加上你标注的类别,字符串 row_label 应于labelImg中标注的名称相同。同样 path 为图片的路径。
执行生成前要去改一下cvs文件,把3个文件里面的jpeg改成jpg,这里是我图片后缀问题,不改会报错。
cd /home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research
python3 generate_tfrecord.py --csv_input=object_detection/data/seahorse_train_labels.csv --output_path=object_detection/data/seahorse_train.tfrecord
generate_tfrecord.py 需要在research目录下,也就是object_detection的上级目录,因为在脚本中使用了 object_detection.utils,如果在 object_detection 下执行命令会报错(No module named object_detection)。
类似的,我们可以输入如下命令,将验证集和测试集也转换为tfrecord格式。
python3 generate_tfrecord.py --csv_input=object_detection/data/seahorse_validation_labels.csv --output_path=object_detection/data/seahorse_validation.tfrecord
python3 generate_tfrecord.py --csv_input=object_detection/data/seahorse_test_labels.csv --output_path=object_detection/data/seahorse_test.tfrecord
训练
1. 在object_detection/data文件夹下创建标签分类的配置文件(labelmap.pbtxt),需要检测几种目标,将创建几个id,代码如下:
item {
id: 1 # id 从1开始编号
name: 'seahorse'
}
2. 配置管道配置文件,找到object_detection/samples/config/ssd_mobilenet_v1_coco.config,复制到data文件夹下。修改后的代码如下:
# 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 {
#修改,分类的总数
num_classes: 2
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: {
#修改,批次大小,nano的话在图形界面下跑4会出现卡顿OOM,内存不足,2的话勉强可以跑。可以在不启动图形界面跑会好些
batch_size: 2
optimizer {
rms_prop_optimizer: {
learning_rate: {
exponential_decay_learning_rate {
#修改,初始学习率
initial_learning_rate: 0.0001
decay_steps: 800720
decay_factor: 0.95
}
}
momentum_optimizer_value: 0.9
decay: 0.9
epsilon: 1.0
}
}
#修改,预训练模型
fine_tune_checkpoint: "ssd_model/ssd_mobilenet/model.ckpt"
from_detection_checkpoint: 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.
#修改,迭代总次数
num_steps: 5000
data_augmentation_options {
random_horizontal_flip {
}
}
data_augmentation_options {
ssd_random_crop {
}
}
}
train_input_reader: {
tf_record_input_reader {
#修改,训练数据 按理这里是seahorse_train.tfrecord
input_path: "data/seahorse.tfrecord"
}
#修改,labelmap路径
label_map_path: "data/labelmap.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 {
#修改,训练验证数据
input_path: "data/seahorse_validation.tfrecord"
}
#修改,labelmap路径
label_map_path: "data/labelmap.pbtxt"
shuffle: false
num_readers: 1
}
3.下载预训练模型(用我上传的文件的话,已经在object_detection/ssd_model/ssd_mobilenet目录下了)
下载 ssd_mobilenet 至 ssd_model/
目录下,解压并重命名为 ssd_mobilenet
ssd_mobilenet: http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_coco_11_06_2017.tar.gz
tar zxvf ssd_mobilenet_v1_coco_11_06_2017.tar.gz
mv ssd_mobilenet_v1_coco_11_06_2017 ssd_mobilenet
将 ssd_mobilenet_v1_coco.config 中 fine_tune_checkpoint 修改为如下格式的路径(上面已经改好)
fine_tune_checkpoint: "ssd_model/ssd_mobilenet/model.ckpt"
关闭图形界面,训练时再关闭(看你的平台情况而定,训练不了就关闭)ps:我nano在图形界面勉强可以训练
# ubuntu关闭图形用户界面
sudo systemctl set-default multi-user.target
sudo reboot
# ubuntu启用图形用户界面
sudo systemctl set-default graphical.target
4. 开始训练(这个train.py 文件可能就在object_detection目录下 也可能在object_detection/legacy下)
python3 legacy/train.py --logtostderr --train_dir=training/ --pipeline_config_path=data/ssd_mobilenet_v1_coco.config
5.训练完成后,运行 export_inference_graph.py 脚本将训练出的模型固化成 TensorFlow 的 .pb 模型,其中 trained_checkpoint_prefix 要设置成 model.ckpt-[step],其中 step 要与训练迭代次数相同
python3 ./object_detection/export_inference_graph.py --input_type image_tensor --pipeline_config_path ./object_detection/ssd_model/ssd_mobilenet_v1_coco.config --trained_checkpoint_prefix ./object_detection/training/model.ckpt-5000 --output_directory ./object_detection/ssd_model/model/
转换后生成的 .pb 模型位于 object_detection/ssd_model/model/
目录下
6.测试模型(在object_detection目录下创建文件seahorse_ssd_detect.py)
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
import cv2
if StrictVersion(tf.__version__) < StrictVersion('1.9.0'):
raise ImportError('Please upgrade your TensorFlow installation to v1.9.* or later!')
from utils import label_map_util
from utils import visualization_utils as vis_util
global output_num
global output_img_dic
matplotlib.use('TkAgg')
# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_FROZEN_GRAPH = 'ssd_model/model/frozen_inference_graph.pb'
# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('data', 'labelmap.pbtxt')
print(PATH_TO_LABELS)
# For the sake of simplicity we will use only 2 images:
# image1.jpg
# image2.jpg
# 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 = 'test_images'
TEST_IMAGE_PATHS = [ os.path.join(PATH_TO_TEST_IMAGES_DIR, 'image{}.jpg'.format(i)) for i in range(3, 7) ]
# Size, in inches, of the output images.
IMAGE_SIZE = (12, 8)
output_num = 1
output_img_dic = r'\output_images'
detection_graph = tf.Graph()
with detection_graph.as_default():
od_graph_def = tf.compat.v1.GraphDef()
with tf.io.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='')
print("--\n")
category_index = label_map_util.create_category_index_from_labelmap(PATH_TO_LABELS, use_display_name=True)
print("--\n")
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)
def run_inference_for_single_image(image, graph):
with graph.as_default():
with tf.compat.v1.Session() as sess:
# Get handles to input and output tensors
ops = tf.compat.v1.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.compat.v1.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
def detect(imgfile):
#origimg = cv2.imread(imgfile)
image = Image.open(imgfile)
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)
print(1,image_np)
cv2.imshow("SSD", image_np)
k = cv2.waitKey(0) & 0xff
#Exit if ESC pressed
if k == 27 : return False
return True
test_dir = "/home/nvidia/.local/lib/python3.6/site-packages/tensorflow/models/research/object_detection/seahorseImages"
for f in os.listdir(test_dir):
if detect(test_dir + "/" + f) == False:
break
# if not os.path.exists(output_img_dic):
# os.mkdir(output_img_dic)
# output_img_path = os.path.join(output_img_dic,str(output_num)+".png")
# plt.savefig(output_img_path)
测试(任意键下一张图,ESC退出)
python3 seahorse_ssd_detect.py