0
votes

I am using Tensorflow object-detection API, to be specific I am referring to detection part of this Ipython notebook (https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial.ipynb) I have input image in the test_image folder the resolution of this image is 1737x979. However when I run this code through the detection part of the Tensorflow object-detection API. I get image size as 1200x800. How do I output image with the same ratio to that of input(In this case output image should have resolution 1737x979 and not 1200x800)

IMAGE_SIZE = (12, 8) #output image size in inches


with detection_graph.as_default():
  with tf.Session(graph=detection_graph) as sess:
    # Definite input and output Tensors for detection_graph
    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.
    detection_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.
    detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')
    detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')
    num_detections = detection_graph.get_tensor_by_name('num_detections:0')

    #myFile = open('example2.csv', 'w')
    i=0
    #boxeslist=[]
    new_boxes = []
    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)
      # Actual detection.
      (boxes, scores, classes, num) = sess.run(
          [detection_boxes, detection_scores, detection_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=4)
      plt.figure(figsize=IMAGE_SIZE)
      plt.imshow(image_np)

I need to change IMAGE_SIZE such that it maintains the dimentions of the input image. Also, the output image is in form of graph with white background, how do I remove this white outline?

1

1 Answers

0
votes

It is resolved if you closely observe the output image is similar to what we get when we plot graph in matplotlib. Zoom in and you will see x and y dimensions of the image, these x and y dimensions remain same no matter you change IMAGE_SIZE variable