TensorFlow Object Detection API の USBカメラ、Mjpg-streamer での利用

先ず、環境を整えます。 
1. tensorflow/models の全体を、zipファイルでダウンロードして、Windows10 の適当なディレクトリーで展開。
"<>Code" タグ  -> Clone or Download -> Download Zip 

2. Table of contents ->Setup: ->Installation に従って、環境をインストールします。
https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md
注) Windows10 なので、完全にはできません。
COCO API installation は、無視

export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim の
`pwd' は、 Winddowsなので、コマンドが無いので、models-master を展開したpath を直接記述。
`pwd` -> c:\???\models-master\research
`pwd`/slim -> c:\???\models-master\research\slim

Windows10 では、下記
set PYTHONPATH=C:\???\models-master\research;C:\???\models-master\research\slim;%PYTHONPATH%

3. c:\???\models-master\research\object_detection ディレクトリーの、object_detection_tutorial.ipynb を JupeterNotebook で開いて、実行します。

4.object_detection_tutorial.ipynb を実行すると、学習済み Model がダウンロードされるのでこれで準備完了です。
学習済み Model は、下記に幾つもあるので色々変えて試せます。

https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md

要は、COCO Dataset、Kitti dataset、Open Image dataset でそれぞれ学習させたデルが用意されているので、
自分が検出したい物をよく認識するモデル、または性能のモデルを使えば良いと言う事みたいです。
どうしてもマッチするのが無ければ、自分で一から学習させるか、学習済モデルをベースに、
転移学習(Transfer Learning)をするか。

5.インターネット上に良く紹介されている object detection のサンプルプログラムは、 C:\???\models-master\research\object_detection ディレクトリーにプログラムを作成します。
今回のプログラムも、このディレクトリーで作成します。

注) utils -> object_detection.utils の事
#from utils import label_map_util
#from utils import visualization_utils as vis_util
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_util

6.実行方法は、Dos プロンプトから
>activate tensorflow
>set PYTHONPATH=C:\???\models-master\research;C:\???\models-master\research\slim;%PYTHONPATH%
>cd C:\???\models-master\research
>python object_detection\object_detection_tutorial_cam.py

プログラムの終了は、
'q'  キーを押下 

それでは、早速ですが、コードは、こちらです。
object_detection\object_detection_tutorial_cam.py

''' Created on 2018/01/20 object_detection_tutorial_cam.py https://github.com/tensorflow/models/tree/master/research/object_detection object_detectioon_tutorial.ipynb をベースに USBカメラから入力した画像を使ってみます。 @author: nishi 実行方法 cd ....\models-master\research ディレクトリーで python object_detection/object_detection_tutorial_cam.py プログラムの終了 type key 'q' ''' # coding: utf-8 ''' Object Detection Demo Welcome to the object detection inference walkthrough! This notebook will walk you step by step through the process of using a pre-trained model to detect objects in an image. Make sure to follow the [installation instructions] (https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md) before you start. ''' # # Imports # In[1]: 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 import cv2 #if tf.__version__ < '1.4.0': # raise ImportError('Please upgrade your tensorflow installation to v1.4.* or later!') ## Env setup # In[2]: # This is needed to display the images. #get_ipython().run_line_magic('matplotlib', 'inline') # This is needed since the notebook is stored in the object_detection folder. sys.path.append("..") ''' Object detection imports Here are the imports from the object detection module. ''' # In[3]: #from utils import label_map_util #from utils import visualization_utils as vis_util from object_detection.utils import label_map_util from object_detection.utils import visualization_utils as vis_util ''' 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/research/object_detection/g3doc/detection_model_zoo.md) for a list of other models that can be run out-of-the-box with varying speeds and accuracies. ''' # In[4]: BASE_DIR='object_detection' # 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' PATH_TO_CKPT = os.path.join(BASE_DIR,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(BASE_DIR,'data', 'mscoco_label_map.pbtxt') NUM_CLASSES = 90 ''' Download Model ''' # In[5]: def model_downloader(): 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()) ''' Load a (frozen) Tensorflow model into memory. ''' # In[6]: print('PATH_TO_CKPT=',PATH_TO_CKPT) 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='') ''' Loading label map Label maps map indices to category names, so that when our convolution network predicts `5`, we know that this corresponds to `airplane`. Here we use internal utility functions, but anything that returns a dictionary mapping integers to appropriate string labels would be fine ''' # In[7]: 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 not use ''' # In[8]: 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) ''' USBカメラオープン ''' video = cv2.VideoCapture(0) # Exit if video not opened. if not video.isOpened(): print("Could not open video") sys.exit() ''' Detection ''' # In[9]: # 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(BASE_DIR,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) # In[10]: with detection_graph.as_default(): with tf.Session(graph=detection_graph) as sess: # Definite input and output Tensors for detection_graph 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. detection_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. detection_scores = detection_graph.get_tensor_by_name('detection_scores:0') detection_classes = detection_graph.get_tensor_by_name('detection_classes:0') num_detections = detection_graph.get_tensor_by_name('num_detections:0') #for image_path in TEST_IMAGE_PATHS: while True: #image = Image.open(image_path) # Read first frame. ok, image_np = video.read() if not ok: print('Cannot read video file') sys.exit() # 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. (boxes, scores, classes, num) = sess.run( [detection_boxes, detection_scores, detection_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() cv2.imshow('img',image_np) #---------------- # プログラムの終了は、 q を入力してください。 #---------------- if cv2.waitKey(1) & 0xFF == ord('q'): break video.release() cv2.destroyAllWindows()

実行した感じ?
PCのモニターの上に自分を映す様にUSBカメラを向けているですが、
モニターには、自分のお顔がクローズアップされています。
一応、Person 70% と表示されます。

序に、 USB カメラを WebCAM(Mjpeg-streamer)に変えてみます。
URL="" は、Mjpeg-streamer のアドレスを指定します。

''' USBカメラオープン ''' if False: video = cv2.VideoCapture(0) # Exit if video not opened. if not video.isOpened(): print("Could not open video") sys.exit() ''' Web CAM(Mjpeg-streamer) オープン ''' if True: URL="http://200.36.58.250/mjpg/video.mjpg?resolution=320x240" #URL="http://200.36.58.250/mjpg/video.mjpg?resolution=640x480" video = cv2.VideoCapture(URL) # Exit if video not opened. if not video.isOpened(): print("Could not open Web CAM") sys.exit()