nano pi M4+tensorflow+opencv+logi摄像头实现目标检测

 

做物体检测的网络有很多种，如faster rcnn，ssd，yolo等等，通过不同维度的对比，各个网络都有各自的优势。
毕竟nano pi M4计算能力有限，我们这里先选择专门为速度优化过最快的网络SSD，以及经典的faster-rcnn作对比，再加上能显示mask的高端网络，，，
事实上yolo v3刚出来，比SSD更快，而faster rcnn相对来说运行慢的多了，后面可以都尝试对比一下，目前先把基线系统搭建好。

环境：

          系统：rk3399-sd-friendlydesktop-bionic-4.4-arm64

                     python2.7

                    tensorflow1.8.0

                   opencv

平台：

                  nano pi M4 2G版

                   logi摄像头

                  TF卡32G

1、进入这个网站下载适合的tensorflow版本：https://github.com/lhelontra/tensorflow-on-arm/releases/download/v1.8.0这里使用tensorflow1.8.0；

      sudo pip install tensorflow-1.8.0-cp27-none-linux_aarch64.whl

      或者：在线下载安装wget https://github.com/lhelontra/tensorflow-on-arm/releases/download/v1.8.0/tensorflow-1.8.0-cp27-none-linux_aarch64.whl

                                       sudo pip install tensorflow-1.8.0-cp27-none-linux_aarch64.whl

 

2、安装matplotlib库

      sudo pip install matplotlib

3、opencv安装sudo pip install opencv-python

4、下载tensorflow提供的models API并解压，下载模型git clone https://github.com/tensorflow/models.git

5、下载用COCO训练集预训练的模型
      下载训练好的模型并放到上一步models_master下的object_detection/models目录

        

6、Protobuf 安装与配置   

protobuf是Google开发的一种混合语言数据标准，提供了一种轻便高效的结构化数据存储格式，可以用于结构化数据序列化。很适合做数据存储或 RPC 数据交换格式。可用于通讯协议、数据存储等领域的语言无关、平台无关、可扩展的序列化结构数据格式。目前提供了 C++、Java、Python 三种语言的 API。
下载地址： https://github.com/google/protobuf/releases
我们这里下载最新版本 protobuf-all-3.5.1.tar.gz

安装：

tar -xf  protobuf-all-3.5.1.tar.gz  
cd protobuf-3.5.1  
./configure   
make   
make check   ->这一步是检查编译是否正确，耗时非常长，可略过
sudo make install  
sudo ldconfig  ->更新库搜索路径，否则可能找不到库文件

 将proto格式的数据转换为python格式，从而可以在python脚本中调用，进入目录models-master/research，运行：

protoc object_detection/protos/*.proto --python_out=.

转换完毕后可以看到在object_detection/protos/目录下多了许多*.py文件。

7、测试代码

    

import numpy as np
import os
import sys
import tarfile
import tensorflow as tf
import cv2
import time
from collections import defaultdict

# 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

# What model to download.
MODEL_NAME = 'ssd_mobilenet_v1_coco_2018_01_28'
# 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'
# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('/home/pi/models/research/object_detection/data', 'mscoco_label_map.pbtxt')

model_path = "/home/pi/models/research/object_detection/models/ssd_mobilenet_v1_coco_2018_01_28/model.ckpt"
#extract the ssd_mobilenet
start = time.clock()
NUM_CLASSES = 90

end= time.clock()
print('load the model' ,(end -start))
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)

cap = cv2.VideoCapture(0)
with detection_graph.as_default():
    with tf.Session(graph=detection_graph) as sess:
        writer = tf.summary.FileWriter("logs/", sess.graph)
        sess.run(tf.global_variables_initializer())

        loader = tf.train.import_meta_graph(model_path + '.meta')
        loader.restore(sess, model_path)
        while(1):
            start = time.clock()
            ret, frame = cap.read()
            if cv2.waitKey(1) & 0xFF == ord('q'):
                break
            image_np =frame
        # the array based representation of the image will be used later in order to prepare the
        # result image with boxes and labels on it.
        # 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=6)
            end = time.clock()

            print 'One frame detect take time:' ,end - start

            cv2.imshow("capture", image_np)
            print('after cv2 show')
            cv2.waitKey(1)
cap.release()
cv2.destroyAllWindows()

保存为 test.py，到目录models-master/research/object_detection/models下。

8、运行：

进入models-master/research/object_detection/models运行命令：

sudo chmod 666 /dev/video0
python detect.py

9、测试结果：可以看到SSD模型加载花了3秒左右，识别一张图在2秒左右。

安装环境是会出现pip安装超超时等等问题，解决办法：sudo pip install --index-url https://pypi.douban.com/simple matplotlib

                                                                                       sudo apt-get install protobuf-compiler

安装opencv时 sudo apt-get upgrade  sudo apt-get install python2.7-dev   sudo apt-get install python-opencv