基于谷歌开源的Object-Detection API实现视频目标检测（tensorflow+opencv+anaconda3）

之前在做实时监控中人脸识别、人体姿态识别等项目，可以说一直在与视频打交道，今日心血来潮，顺便帮助师妹快速了解目标检测，特意选择了谷歌开源的Object-Detection API实现基于视频的目标检测。

测试环境：Win7、Anaconda3、tensorflow、opencv、CPU

一、Anaconda3下安装tensorflow和opencv

1、创建anaconda虚拟环境

conda create -n tf_object python=3.6.7

其中tf_object为虚拟环境名称，可以根据自己喜好起名。

激活虚拟环境 activate tf_object  若退出可执行deactivate

2、安装tensorflow

打开Anaconda中的Anaconda Navigator

点击环境，然后选择虚拟环境tf_object

然后选择All，再搜索tensorflow再点击Apply进行安装；opencv的安装按照同样的方式进行安装，具体操作可以搜索相关CSDN博客。

进行验证，是否安装成功！！

打开cmd，再激活tf_object环境，然后输入python，再输入

import tensorflow
import cv2

不报错则安装成功！！

二、protoc安装

什么是Protoc？Protoc是用来编译.Proto文件，Protocol Buffers (ProtocolBuffer/ protobuf )是Google公司开发的一种数据描述语言，类似于XML能够将结构化数据序列化，可用于数据存储、通信协议等方面。现阶段支持C++、JAVA、Python等三种编程语言。

Protoc用于编译相关程序运行文件，进入Protoc下载页，下载类似下图中带win32的压缩包。

然后解压这个文件，并记住bin文件夹路径，最好不要出现中文。

三、Git安装

git +网址是目前主流的在线下载指令，在官网找到Windows下载安装，按步骤操作就行，记得选择windows的命令框

四、安装其他包

pip install pillow
pip install lxml
pip install jupyter
pip install matplotlib
pip install requests
pip install moviepy

注意这些需要先激活虚拟环境下再安装这些包

五、下载模型并编译

打开cmd输入

git clone http://github.com/tensorflow/models.git

下载后放在某个文件夹内，然后在cmd中进入models/research下，再进行编译

E:\protoc\bin\protoc object_detection\protos\*.proto --python_out=.

其中E:\protoc\bin\protoc表示你解压的protoc路径；object_detection\protos\*.proto --python_out=.是进行编译object_detection\protos\下的所有proto文件，运行成功，会编译生成py文件。

六、运行notebook demo

打开cmd 进入models/research再输入

jupyter-notebook

浏览器自动打开如下

然后新建python3程序，输入以下代码（对之前原始代码进行了些许改进）：

import os
import cv2
import time
import argparse
import multiprocessing
import numpy as np
import tensorflow as tf
from matplotlib import pyplot as plt
%matplotlib inline

import six.moves.urllib as urllib
import sys
import tarfile
import zipfile

from collections import defaultdict
from io import StringIO
from PIL import Image


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


CWD_PATH = os.getcwd()
# Path to frozen detection graph. This is the actual model that is used for the object detection.
MODEL_NAME = 'ssd_mobilenet_v1_coco_2017_11_17'
PATH_TO_CKPT = os.path.join(CWD_PATH, 'object_detection', 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(CWD_PATH, 'object_detection', 'data', 'mscoco_label_map.pbtxt')


NUM_CLASSES = 90
# Loading label map
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 detect_objects(image_np, sess, detection_graph):
    # 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)
    return image_np


# First test on images
PATH_TO_TEST_IMAGES_DIR = 'object_detection/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 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)




from PIL import Image
for image_path in TEST_IMAGE_PATHS:
    image = Image.open(image_path)
    image_np = load_image_into_numpy_array(image)
    plt.imshow(image_np)
    print(image.size, image_np.shape)

运行之后出来结果

继续输入代码：

#Load a frozen TF model 
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='')



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)
            image_np = load_image_into_numpy_array(image)
            image_process = detect_objects(image_np, sess, detection_graph)
            print(image_process.shape)
            plt.figure(figsize=IMAGE_SIZE)
            plt.imshow(image_process)

得到图片检测结果如下所示

下面部分是对视频进行检测，继续输入代码

# Import everything needed to edit/save/watch video clips
import imageio
imageio.plugins.ffmpeg.download()

from moviepy.editor import VideoFileClip
from IPython.display import HTML



def process_image(image):
    # NOTE: The output you return should be a color image (3 channel) for processing video below
    # you should return the final output (image with lines are drawn on lanes)
    with detection_graph.as_default():
        with tf.Session(graph=detection_graph) as sess:
            image_process = detect_objects(image, sess, detection_graph)
            return image_process



white_output = 'video1_out.mp4'
clip1 = VideoFileClip("video1.mp4").subclip(0,2)
white_clip = clip1.fl_image(process_image) #NOTE: this function expects color images!!s
%time white_clip.write_videofile(white_output, audio=False)

结果如下

输出视频video1_out.mp4保存到了代码所在文件目录中。

可以查看，输入下面代码

HTML("""
<video width="960" height="540" controls>
  <source src="{0}">
</video>
""".format(white_output))