Question

我正在执行https://github.com/tensorflow/tensorflow这个检测图像中对象的例子。

我希望得到检测到的对象的数量，这是代码，它为我提供了在图像中绘制的检测到的对象。但是我无法计算检测到的物体。

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=1)
      plt.figure(figsize=IMAGE_SIZE)
      plt.imshow(image_np)

这是提供实际物体检测的代码块，如下图所示：

如何获取对象数？

Answer 1

解决它只是打印boxes.shape的长度

print(len(boxes.shape))

Answer 2

您可以使用 TensorFlow Object Counting API ，它是在TensorFlow之上构建的开源框架，可以轻松开发对象计数系统以计数任何对象！

示例项目1是使用TensorFlow Object Counting API开发的“行人计数”：

示例项目2是使用 TensorFlow Object Counting API 开发的“车辆计数”：

示例项目3是使用 TensorFlow Object Counting API 开发的“实时对象计数”：

有关详细信息，请参见TensorFlow Object Counting API，如果发现有用的话，请给星号that回购，以表示对开源社区的支持！

Answer 3

重要的是要注意盒子的数量总是100个。

如果你查看实际绘制框的代码，即vis_util.visualize_boxes_and_labels_on_image_array函数，你会发现他们正在定义一个阈值 - min_score_thresh=.5 - 到将绘制的框限制为只有那些分数> 1的检测结果。 0.5。您可以将此视为仅精确检测概率> 50％的绘图框。您可以向上或向下调整此阈值以增加绘制的框数。但是，如果将它降低得太低，则会出现很多不准确的框。

Answer 4

您应该检查分数并将对象计为手动。代码在这里：

#code to test image start

    (boxes, scores, classes, num) = sess.run(
        [detection_boxes, detection_scores, detection_classes, num_detections],
        feed_dict={image_tensor: image_np_expanded})

#code to test image finish

#add this part to count objects

    final_score = np.squeeze(scores)    
        count = 0
        for i in range(100):
            if scores is None or final_score[i] > 0.5:
                    count = count + 1

#count is the number of objects detected

Answer 5

添加此部分以计数对象

final_score = np.squeeze(scores)    
    count = 0
    for i in range(100):
        if scores is None or final_score[i] > 0.5:
                count = count + 1

count是检测到的对象数

此部分将打印计数，但将以连续方式打印，可以像最终计数=某个值那样仅打印一次，而不必重复打印

如何在Tensorflow Object Detection API中计算对象

5 个答案: