【分享】玩转谷歌物体识别
本帖最后由 wushaominkk 于 2018-5-2 10:14 编辑我在win7和ubuntu16.04下面都能运行成功。
我用的是 jupyter notebook
1.运行所需要的库：matplotlib、lxml、pillow、Cython具体参考：https://github.com/tensorflow/mo ... doc/installation.md2.object_detection包，下载地址：https://github.com/tensorflow/modelshttps://www.52pojie.cn/forum.php?mod=image&aid=1119081&size=300x300&key=1662b12c900d2066&nocache=yes&type=fixnone
3.编译好的protos文件，object_detection中的文件没有编译，我是在ubuntu下编译的，win7可以下载编译好的文件：https://github.com/1529591487/Object-Detection直接替换object_detection中的protos文件夹即可。4.代码：import numpy as np
import os
import six.moves.urllib as urllib
import sys
import tarfile
import tensorflow as tf
import zipfile
from collections import defaultdict
from io import StringIO
from matplotlib import pyplot as plt
from PIL import Image
# 这里改成你下载的object_detection包的位置
sys.path.append(r"E:\学习资料\人工智能\models-master\research")
from object_detection.utils import ops as utils_ops
if tf.__version__ < '1.4.0':
raise ImportError('Please upgrade your tensorflow installation to v1.4.* or later!')
%matplotlib inline
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util
这里会有警告，但是不影响,如果要去掉警告的话，将models-master\research\object_detection\utils\visualization_utils.py 文件中的第26行改成matplotlib.use('Agg',warn=False, force=True)
# 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_CKPT = MODEL_NAME + '/frozen_inference_graph.pb'
# 这里的路径也需要修改
PATH_TO_LABELS = os.path.join(r'E:\学习资料\人工智能\models-master\research\object_detection\data', 'mscoco_label_map.pbtxt')
NUM_CLASSES = 90
opener = urllib.request.URLopener()
opener.retrieve(DOWNLOAD_BASE + MODEL_FILE, MODEL_FILE)
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())
detection_graph = tf.Graph()
with detection_graph.as_default():
od_graph_def = tf.GraphDef()
with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
serialized_graph = fid.read()
od_graph_def.ParseFromString(serialized_graph)
tf.import_graph_def(od_graph_def, name='')
label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
category_index = label_map_util.create_category_index(categories)
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.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 = 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'], )
detection_masks = tf.squeeze(tensor_dict['detection_masks'], )
# 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'], tf.int32)
detection_boxes = tf.slice(detection_boxes, , )
detection_masks = tf.slice(detection_masks, , )
detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(
detection_masks, detection_boxes, image.shape, image.shape)
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'])
output_dict['detection_classes'] = output_dict[
'detection_classes'].astype(np.uint8)
output_dict['detection_boxes'] = output_dict['detection_boxes']
output_dict['detection_scores'] = output_dict['detection_scores']
if 'detection_masks' in output_dict:
output_dict['detection_masks'] = output_dict['detection_masks']
return output_dict
IMAGE_SIZE = (36, 24)
#这里设置图片路径
mydir=r'E:\学习资料\人工智能\models-master\research\object_detection\test_images'
# mydir = 'G:\壁纸'
for filename in os.listdir(mydir):
if os.path.splitext(filename) == '.jpg':
filepath=os.path.join(mydir, filename)
print(filepath)
image = Image.open(filepath)
# 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:
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)
fig1 = plt.gcf()
plt.figure(figsize=IMAGE_SIZE)
plt.imshow(image_np)
代码参考：https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial.ipynb
有些图片识别会失败，目前还没搞清楚，欢迎大家交流讨论
运行结果如下：
linuxprobe 发表于 2018-4-28 21:22
自从谷歌退出中国后，就没再怎么关注了。
但是人工智能还是势不可挡啊，而且github的分享很棒啊，我现在在youtube上也学到了很多有用的东西，顺便还能学英语，一举两得。 duanshifeng1994 发表于 2018-6-15 16:55
没有模块交import tensorflow as tf?我win7，python版本3.6，这安装不上怎么破？
用pip安装一下tensorflow，还有python好像要64位的 自从谷歌退出中国后，就没再怎么关注了。 好像很厉害的样子哦 764507093 发表于 2018-4-28 21:24
好像很厉害的样子哦
很简单的，改改文件路径就能实现了。 夏日已末 发表于 2018-4-28 21:33
很简单的，改改文件路径就能实现了。
我是搞Java的，没有搞懂 要我科学上网，我心累 好像很厉害的样子 谷歌就是让我永远难忘 程序好像很厉害的样子
页:
[1]
2