環境
Windows7 x64
conda 4.3.30
1、TensorFlow安裝
首先在conda中建立TensorFlow環境
conda create -n tensorflow python=3.6.2
激活tensorflow環境
activate tensorflow
安裝tensorflow
pip install tensorflow==1.12.0
安裝完成後會在anaconda安裝路徑envs檔案下自動建立tensorflow目錄
2、tensorflow object detection API安裝
下載下傳安裝包：https://github.com/xizhonghuai/tensorflow_object_detection_demo
下載下傳完成後解壓，将tensorflow_models檔案夾拷貝到不含中文任意路徑下,如：
在anaconda安裝路徑下找到剛才新建立的tensorflow環境目錄
在\Lib\site-packages下建立tensorflow_model.pth檔案，内容如下：
E:\python\workspace\tensorflow_models\research
E:\python\workspace\tensorflow_models\research\slim
進入E:\python\workspace\tensorflow_models\research\slim檔案夾下分别執行一下指令：
python setup.py build
python setup.py install
運作後如果出現error: could not create ‘build’，請删掉\slim檔案夾下BUILD檔案，在重新執行。
進入E:\python\workspace\tensorflow_models\research檔案夾下分别執行一下指令：
python setup.py build
python setup.py install
這個過程時間稍長，若出現報錯情況，請仔細檢視報錯資訊是否提示缺少相關依賴，手動通過pip安裝。
測試tensorflow object detection API是否安裝成功，在E:\python\workspace\tensorflow_models\research下執行
object_detection/builders/model_builder_test.py
無報錯表示安裝成功。
說明：
報錯資訊提示找不到相關子產品，請解除安裝tensorflow，重新安裝其他版本。
報錯資訊提示DLL相關錯誤，請解除安裝tensorflow，在以下連結中安裝 https://github.com/fo40225/tensorflow-windows-wheel
3、測試
下載下傳官方已訓練好的模型到本地，這裡我們使用ssd_mobilenet_v1_coco(模型包含90種常見物體的識别)
https://github.com/xizhonghuai/tensorflow_object_detection_demo/blob/master/tensorflow_models/research/object_detection/g3doc/detection_model_zoo.md
點選下載下傳ssd_mobilenet_v1_coco模型你解壓到本地。
打開iead，建立python工程，建立com包，在com包下分别建立，data、test_images、model檔案夾
将剛才解壓模型包中的frozen_inference_graph.pb檔案拷貝到model檔案下。
在E:\python\workspace\tensorflow_models\research\object_detection\data_back_up檔案夾下拷貝mscoco_label_map.pbtxt檔案到data檔案下。
在網上随便找一張圖檔重命名為image1.jpg存放到test_images下
項目結構如下：
在com包下建立py代碼：
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
## Env setup
# This is needed to display the images.
# This is needed since the notebook is stored in the object_detection folder.
sys.path.append("..")
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util
os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"
# Model preparation
## Variables
#Any model exported using the `export_inference_graph.py` tool can be loaded here #simply by changing `PATH_TO_CKPT` to point to a new .pb file.
#By default we use an "SSD with Mobilenet" model here. See the [detection model zoo](https://github.com/tensorflow/models/blob/master/object_detection/g3doc/detection_mo#del_zoo.md) for a list of other models that can be run out-of-the-box with varying #speeds and accuracies.
# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_CKPT = 'model/frozen_inference_graph.pb'
PATH_TO_LABELS = os.path.join('data', 'mscoco_label_map.pbtxt')
NUM_CLASSES = 1
## Load a (frozen) Tensorflow model into memory.
detection_graph = tf.Graph()
with detection_graph.as_default():
od_graph_def = tf.GraphDef()
with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
print(PATH_TO_CKPT)
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)
## Helper code
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)
# Detection
# 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, 2) ]
#TEST_IMAGE_PATHS = ['test_images']
# Size, in inches, of the output images.
IMAGE_SIZE = (12, 8)
with detection_graph.as_default():
with tf.Session(graph=detection_graph) as sess:
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)
image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')
# Each box represents a part of the image where a particular object was detected.
boxes = detection_graph.get_tensor_by_name('detection_boxes:0')
# Each score represent how level of confidence for each of the objects.
# Score is shown on the result image, together with the class label.
scores = detection_graph.get_tensor_by_name('detection_scores:0')
classes = detection_graph.get_tensor_by_name('detection_classes:0')
num_detections = detection_graph.get_tensor_by_name('num_detections:0')
# Actual detection.
(boxes, scores, classes, num_detections) = sess.run(
[boxes, scores, classes, num_detections],
feed_dict={image_tensor: image_np_expanded})
# Visualization of the results of a detection.
vis_util.visualize_boxes_and_labels_on_image_array(
image_np,
np.squeeze(boxes),
np.squeeze(classes).astype(np.int32),
np.squeeze(scores),
category_index,
use_normalized_coordinates=True,
line_thickness=8)
# plt.figure(figsize=IMAGE_SIZE) # plt.imshow(image_np)
# plt.show()
im = Image.fromarray(image_np)
im.save("ret.jpeg")
print("OK")
測試：
執行python代碼，運作成功後将生成識别結果并儲存圖檔到com包下。