Как решить Raspberry Pi 2 / Tensorflow Распределение 5031936 превышает 10% ошибки системной памяти? - PullRequest
Новая
       29

Как решить Raspberry Pi 2 / Tensorflow Распределение 5031936 превышает 10% ошибки системной памяти?

0 голосов
/ 07 апреля 2019

Я пытаюсь обнаружить людей в видеокадре, используя api обнаружения объекта tenorflow и raspberry pi 2. Видеовход от ip-камеры.Когда я запускаю свой код, я получаю следующую ошибку: 

2019-04-07 13:57:17.996374: W 
tensorflow/core/framework/allocator.cc:124] Allocation of 5660928 
exceeds 10% of system memory.
2019-04-07 13:57:18.096145: W 
tensorflow/core/framework/allocator.cc:124] Allocation of 25159680 
exceeds 10% of system memory.
2019-04-07 13:57:18.208663: W 
tensorflow/core/framework/allocator.cc:124] Allocation of 10063872 
exceeds 10% of system memory.
2019-04-07 13:57:18.260229: W 
tensorflow/core/framework/allocator.cc:124] Allocation of 5031936 
exceeds 10% of system memory.
2019-04-07 13:57:18.293027: W 
tensorflow/core/framework/allocator.cc:124] Allocation of 5031936 
exceeds 10% of system memory.
Killed

Я следовал этому руководству, чтобы установить tenorflow и opencv на свой пи

https://www.youtube.com/watch?v=npZ-8Nj1YwY

Этокод, который я использую 

import numpy as np
import tensorflow as tf
import cv2
import time


class DetectorAPI:
def __init__(self, path_to_ckpt):
    self.path_to_ckpt = path_to_ckpt


    self.detection_graph = tf.Graph()
    with self.detection_graph.as_default():
        od_graph_def = tf.GraphDef()
        with tf.gfile.GFile(self.path_to_ckpt, 'rb') as fid:
            serialized_graph = fid.read()
            od_graph_def.ParseFromString(serialized_graph)
            tf.import_graph_def(od_graph_def, name='')

    self.default_graph = self.detection_graph.as_default()
    self.sess = tf.Session(graph=self.detection_graph)

    # Definite input and output Tensors for detection_graph
    self.image_tensor = self.detection_graph.get_tensor_by_name('image_tensor:0')
    # Each box represents a part of the image where a particular object was detected.
    self.detection_boxes = self.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.
    self.detection_scores = self.detection_graph.get_tensor_by_name('detection_scores:0')
    self.detection_classes = self.detection_graph.get_tensor_by_name('detection_classes:0')
    self.num_detections = self.detection_graph.get_tensor_by_name('num_detections:0')

def processFrame(self, image):
    # Expand dimensions since the trained_model expects images to have shape: [1, None, None, 3]
    image_np_expanded = np.expand_dims(image, axis=0)
    # Actual detection.
    start_time = time.time()
    (boxes, scores, classes, num) = self.sess.run(
        [self.detection_boxes, self.detection_scores, self.detection_classes, self.num_detections],
        feed_dict={self.image_tensor: image_np_expanded})
    end_time = time.time()

    print("Elapsed Time:", end_time-start_time)

    im_height, im_width,_ = image.shape
    boxes_list = [None for i in range(boxes.shape[1])]
    for i in range(boxes.shape[1]):
        boxes_list[i] = (int(boxes[0,i,0] * im_height),
                    int(boxes[0,i,1]*im_width),
                    int(boxes[0,i,2] * im_height),
                    int(boxes[0,i,3]*im_width))

    return boxes_list, scores[0].tolist(), [int(x) for x in classes[0].tolist()], int(num[0])
def close(self):
    self.sess.close()
    self.default_graph.close()

if __name__ == "__main__":
model_path ='/home/pi/tensorflow1/models/research/object_detection/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb'
odapi = DetectorAPI(path_to_ckpt=model_path)
threshold = 0.7
cap = cv2.VideoCapture('rtsp://vamshi@123:vikas123@192.168.0.5:88/videoMain')

while True:
    r, img = cap.read()
    img = cv2.resize(img, (500, 400))
    cv2.line(img,(0,400 // 2),(500 ,400 // 2),(0,255,255),2)
    cv2.line(img,(500 // 2,0),(500 // 2,400),(0,255,255),2)
    boxes, scores, classes, num = odapi.processFrame(img)

    # Visualization of the results of a detection.

    for i in range(len(boxes)):
        # Class 1 represents human
        if classes[i] == 1 and scores[i] > threshold:
            box = boxes[i]
            cv2.rectangle(img,(box[1],box[0]),(box[3],box[2]),(255,255,0),2)
            cX=int((box[1] + box[3]) / 2.0)
            cY=int((box[0] + box[2]) / 2.0)
            cv2.circle(img,(cX,cY),4,(0,255,0),-1)
            if cX < 500 // 2 and cY < 400 //2:
                print ('Quadrant 1')
            elif cX < 500 // 2 and cY > 400 //2:
                print ('Quadrant 3')
            elif cX > 500 // 2 and cY < 400 //2:
                print ('Quadrant 2')
            else: 
                print ('Quadrant 4')

    cv2.imshow("Human Detection", img)
    key = cv2.waitKey(1)
    if key & 0xFF == ord('q'):
        break

Версия Tensorflow: 1.13.1

Версия OpenCV: 3.4.4

...