为什么我的TensorFlow object_detection模型在BW图像上训练不正确？

Question

我想在黑白图像上训练我的TensorFlow object_detection模型，因为我要检测的对象不需要颜色。但是，当我在黑白图像上训练模型时，每当尝试使用适应的图像测试脚本对其进行测试时，都会出现以下错误：     ValueError：无法为张量为'（？，？，？，3）'的张量'image_tensor：0'输入形状（1，1080，1920）的值

我不能说太多，但是我的模型应该基于ssd_mobilenet_v1_coco_2018_01_28来检测管道上的变形。大多数变形是通过改变形状从视觉上检测到的，这就是为什么颜色只会妨碍物体检测过程的原因。 （至少我认为是这样，对我来说，仅检测形状会提高准确性。）

TensorBoard模型图

我尝试将图像重塑为（1，1080，1920，3），1080x1920是我的图像分辨率。

代码如下：

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)


# # Detection

# In[18]:


# 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 = FLAGS.image_path
TEST_IMAGE_PATHS = [ os.path.join(PATH_TO_TEST_IMAGES_DIR, 'img ({}).jpeg'.format(i)) for i in range(1, len(os.listdir(FLAGS.image_path))) ]

# Size, in inches, of the output images.
IMAGE_SIZE = (24, 16)


# In[ ]:





# In[19]:


def run_inference_for_single_image(image, graph):
  with graph.as_default():
    with tf.Session() as sess:
      # Get handles to input and output tensors
      ops = tf.get_default_graph().get_operations()
      all_tensor_names = {output.name for op in ops for output in op.outputs}
      tensor_dict = {}
      for key in [
          'num_detections', 'detection_boxes', 'detection_scores',
          'detection_classes', 'detection_masks'
      ]:
        tensor_name = key + ':0'
        if tensor_name in all_tensor_names:
          tensor_dict[key] = tf.get_default_graph().get_tensor_by_name(
              tensor_name)
      if 'detection_masks' in tensor_dict:
        # The following processing is only for single image
        detection_boxes = tf.squeeze(tensor_dict['detection_boxes'], [0])
        detection_masks = tf.squeeze(tensor_dict['detection_masks'], [0])
        # Reframe is required to translate mask from box coordinates to image coordinates and fit the image size.
        real_num_detection = tf.cast(tensor_dict['num_detections'][0], tf.int32)
        detection_boxes = tf.slice(detection_boxes, [0, 0], [real_num_detection, -1])
        detection_masks = tf.slice(detection_masks, [0, 0, 0], [real_num_detection, -1, -1])
        detection_masks_reframed = utils_ops.reframe_box_masks_to_image_masks(
            detection_masks, detection_boxes, image.shape[1], image.shape[2])
        detection_masks_reframed = tf.cast(
            tf.greater(detection_masks_reframed, 0.5), tf.uint8)
        # Follow the convention by adding back the batch dimension
        tensor_dict['detection_masks'] = tf.expand_dims(
            detection_masks_reframed, 0)
      image_tensor = tf.get_default_graph().get_tensor_by_name('image_tensor:0')

      # Run inference
      output_dict = sess.run(tensor_dict,
                             feed_dict={image_tensor: image})

      # all outputs are float32 numpy arrays, so convert types as appropriate
      output_dict['num_detections'] = int(output_dict['num_detections'][0])
      output_dict['detection_classes'] = output_dict[
          'detection_classes'][0].astype(np.int64)
      output_dict['detection_boxes'] = output_dict['detection_boxes'][0]
      output_dict['detection_scores'] = output_dict['detection_scores'][0]
      if 'detection_masks' in output_dict:
        output_dict['detection_masks'] = output_dict['detection_masks'][0]
  return output_dict


# In[20]:


for image_path in TEST_IMAGE_PATHS:
  print('Showing image')
  #image = Image.open(image_path)
  image_np = cv2.imread(image_path)
  image_np = cv2.cvtColor(image_np, cv2.COLOR_BGR2RGB) #This line and the one above make it work with color images
  #image_np = cv2.imread(image_path, cv2.COLOR_BGR2GRAY) #This line makes it break
  # 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.
  output_dict = run_inference_for_single_image(image_np_expanded, detection_graph)
  # Visualization of the results of a detection.
  vis_util.visualize_boxes_and_labels_on_image_array(
      image_np,
      output_dict['detection_boxes'],
      output_dict['detection_classes'],
      output_dict['detection_scores'],
      category_index,
      instance_masks=output_dict.get('detection_masks'),
      use_normalized_coordinates=True,
      line_thickness=8)
  #plt.figure(figsize=IMAGE_SIZE)
  #plt.imshow(image_np)

  cv2.imshow('Image', image_np)
  cv2.waitKey(0)

这是整个错误：

Traceback (most recent call last):
  File "image.py", line 212, in <module>
    output_dict = run_inference_for_single_image(image_np_expanded, detection_graph)
  File "image.py", line 185, in run_inference_for_single_image
    feed_dict={image_tensor: image})
  File "C:\Users\Charles.averill\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 950, in run
    run_metadata_ptr)
  File "C:\Users\Charles.averill\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\client\session.py", line 1149, in _run
    str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (1, 1080, 1920) for Tensor 'image_tensor:0', which has shape '(?, ?, ?, 3)'

为什么会这样？我似乎无法从互联网上找到答案。

Answer 1

您的输入图像是灰度的，因此在运行模型时应进行适当的更改。看一下这段代码：

for image_path in TEST_IMAGE_PATHS:
  print('Showing image')
  #image = Image.open(image_path)
  image_np = cv2.imread(image_path)
  image_np = cv2.cvtColor(image_np, cv2.COLOR_BGR2RGB) #This line and the one above make it work with color images

您输入的图片不是BGR格式，因此您无法像在此处的代码中那样将其转换为RGB：

image_np = cv2.cvtColor(image_np, cv2.COLOR_BGR2RGB)

通过

替换上面的代码

for image_path in TEST_IMAGE_PATHS:
  print('Showing image')
  #image = Image.open(image_path)
  image_np = cv2.imread(image_path)
  image_np = cv2.cvtColor(image_np, cv2.COLOR_GRAY2RGB)