def testFirstColorInADE20KColorMap(self):
label = np.array([[1, 3], [10, 20]])
expected_result = np.array([[[120, 120, 120], [6, 230, 230]],
[[4, 250, 7], [204, 70, 3]]])
colored_label = get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_ade20k_name())
self.assertTrue(np.array_equal(colored_label, expected_result))
def image_inference(model_name, picure, label_color, output_name,
output_name_color):
#run single image on model
MODEL = DeepLabModel(model_name)
print('model loaded successfully!')
image = Image.open(picure)
resized_im, seg_map = MODEL.run(image)
seg_image = get_dataset_colormap.label_to_color_image(
seg_map, label_color).astype(np.uint8)
resized_im = np.array(resized_im)
color_and_mask = cv2.addWeighted(resized_im, 0.4, seg_image, 0.6, 0.0)
resized_im = Image.fromarray(resized_im, 'RGB')
seg_image = Image.fromarray(seg_image, 'RGB')
color_and_mask = Image.fromarray(color_and_mask, 'RGB')
#resized_im = cv2.cvtColor(resized_im, cv2.COLOR_RGB2BGR)
#seg_image.show()
global start_t
duration = time.time() - start_t
print(duration)
duration = format(duration, '.4f')
seg_image.save('static/' + output_name, format='PNG')
color_and_mask.save('static/' + output_name_color, format='PNG')
return duration
def save_annotation(label,
save_dir,
filename,
add_colormap=True,
colormap_type=get_dataset_colormap.get_pascal_name()):
"""Saves the given label to image on disk.
Args:
label: The numpy array to be saved. The data will be converted
to uint8 and saved as png image.
save_dir: The directory to which the results will be saved.
filename: The image filename.
add_colormap: Add color map to the label or not.
colormap_type: Colormap type for visualization.
"""
# Add colormap for visualizing the prediction.
if add_colormap:
colored_label = get_dataset_colormap.label_to_color_image(
label, colormap_type)
else:
colored_label = label
pil_image = img.fromarray(colored_label.astype(dtype=np.uint8))
with tf.gfile.Open('%s/%s.png' % (save_dir, filename), mode='w') as f:
pil_image.save(f, 'PNG')
def save_annotation(label,
save_dir,
filename,
add_colormap=True,
colormap_type=get_dataset_colormap.get_pascal_name()):
"""Saves the given label to image on disk.
Args:
label: The numpy array to be saved. The data will be converted
to uint8 and saved as png image.
save_dir: The directory to which the results will be saved.
filename: The image filename.
add_colormap: Add color map to the label or not.
colormap_type: Colormap type for visualization.
"""
# Add colormap for visualizing the prediction.
if add_colormap:
colored_label = get_dataset_colormap.label_to_color_image(
label, colormap_type)
else:
colored_label = label
pil_image = img.fromarray(colored_label.astype(dtype=np.uint8))
with tf.gfile.Open('%s/%s.png' % (save_dir, filename), mode='w') as f:
pil_image.save(f, 'PNG')
def testLabelToPASCALColorImage(self):
"""Test the value of the converted label value."""
label = np.array([[0, 16, 16], [52, 7, 52]])
expected_result = np.array([[[0, 0, 0], [0, 64, 0], [0, 64, 0]],
[[0, 64, 192], [128, 128, 128],
[0, 64, 192]]])
colored_label = get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_pascal_name())
self.assertTrue(np.array_equal(expected_result, colored_label))
def testLabelToPASCALColorImage(self):
"""Test the value of the converted label value."""
label = np.array([[0, 16, 16], [52, 7, 52]])
expected_result = np.array([
[[0, 0, 0], [0, 64, 0], [0, 64, 0]],
[[0, 64, 192], [128, 128, 128], [0, 64, 192]]
])
colored_label = get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_pascal_name())
self.assertTrue(np.array_equal(expected_result, colored_label))
def save_annotation(label,
save_dir,
filename,
add_colormap=True,
normalize_to_unit_values=False,
scale_values=False,
colormap_type=get_dataset_colormap.get_pascal_name()):
"""Saves the given label to image on disk.
Args:
label: The numpy array to be saved. The data will be converted
to uint8 and saved as png image.
save_dir: String, the directory to which the results will be saved.
filename: String, the image filename.
add_colormap: Boolean, add color map to the label or not.
normalize_to_unit_values: Boolean, normalize the input values to [0, 1].
scale_values: Boolean, scale the input values to [0, 255] for visualization.
colormap_type: String, colormap type for visualization.
"""
# Add colormap for visualizing the prediction.
if add_colormap:
colored_label = get_dataset_colormap.label_to_color_image(
label, colormap_type)
else:
colored_label = label
if normalize_to_unit_values:
min_value = np.amin(colored_label)
max_value = np.amax(colored_label)
range_value = max_value - min_value
if range_value != 0:
colored_label = (colored_label - min_value) / range_value
if scale_values:
colored_label = 255. * colored_label
pil_image = img.fromarray(colored_label.astype(dtype=np.uint8))
# with tf.gfile.Open('%s/%s.png' % (save_dir, filename), mode='w') as f:
# pil_image.save(f, 'PNG')
pil_image.save('{}{}.png'.format(save_dir, filename))
def testUnExpectedLabelDimensionForLabelToADE20KColorImage(self):
label = np.array([250])
with self.assertRaises(ValueError):
get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_ade20k_name())
def testUnExpectedLabelDimensionForLabelToPASCALColorImage(self):
"""Raise ValueError if input dimension is not correct."""
label = np.array([120])
with self.assertRaises(ValueError):
get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_pascal_name())
def testUnExpectedLabelValueForLabelToPASCALColorImage(self):
"""Raise ValueError when input value exceeds range."""
label = np.array([[120], [300]])
with self.assertRaises(ValueError):
get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_pascal_name())
def main(_):
output_dir = os.path.expanduser(FLAGS.output_dir)
if not os.path.exists(output_dir):
os.makedirs(output_dir)
start = time.time()
host, port = FLAGS.server.split(':')
channel = implementations.insecure_channel(host, int(port))
stub = prediction_service_pb2.beta_create_PredictionService_stub(channel)
start = time.time()
#data, w, h = cvimread(FLAGS.image)
#print('img load: ',time.time()-start)
image_paths = []
if FLAGS.input_dir != '':
files = FLAGS.input_dir + '/*.jpg'
for image_path in glob.glob(files):
image_paths.append(image_path)
else:
image_paths.append(FLAGS.image)
img_length = len(image_paths)
batch_shape = (img_length, ) + (380, 513, 3)
#batch_shape = (img_length,)+(INPUT_SIZE, INPUT_SIZE, 3)
#X_batch = np.zeros(batch_shape, dtype=np.float32)
X_batch = np.zeros(batch_shape, dtype=np.uint8)
print(X_batch.shape)
i = 0
for image_path in image_paths:
#print(w,h)
#print(data.shape)
print(image_path)
filename = os.path.splitext(os.path.split(image_path)[1])[0]
output_filename = os.path.join(output_dir, filename + '.pkl')
#if os.path.isfile(output_filename):
# continue
data, (w, h), resized = load_input(image_path)
print(w, h)
#X_batch[i]=data
i = i + 1
request = predict_pb2.PredictRequest()
request.model_spec.name = FLAGS.model
request.model_spec.signature_name = 'predict_seg'
request.inputs['img'].CopyFrom(
tf.contrib.util.make_tensor_proto(data, shape=(1, h, w, 3)))
#tf.contrib.util.make_tensor_proto(X_batch, shape=X_batch.shape))
start = time.time()
out = stub.Predict(request, 30.0) # 5 secs timeout
print('network: ', time.time() - start)
print(out.outputs['seg'].tensor_shape)
#import sys
#sys.exit(1)
shape = []
shape.append(out.outputs['seg'].tensor_shape.dim[1].size)
shape.append(out.outputs['seg'].tensor_shape.dim[2].size)
seg = np.asarray(out.outputs['seg'].int64_val, np.int64)
seg = seg.reshape(shape)
pickle.dump(seg, gzip.open(output_filename, 'wb'))
#segMap = pickle.load(gzip.open(output_filename,'rb'))
seg_image = get_dataset_colormap.label_to_color_image(
seg, get_dataset_colormap.get_pascal_name()).astype(np.uint8)
cv2.imwrite('deeplab/seg_map.jpg', seg_image)
img = np.clip(resized, 0, 255).astype('uint8')
img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
overlay = cv2.addWeighted(img, 0.3, seg_image, 0.7, 0)
cv2.imwrite('deeplab/overlay.jpg', overlay)
def testUnExpectedLabelDimensionForLabelToPASCALColorImage(self):
"""Raise ValueError if input dimension is not correct."""
label = np.array([120])
with self.assertRaises(ValueError):
get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_pascal_name())
def testUnExpectedLabelValueForLabelToPASCALColorImage(self):
"""Raise ValueError when input value exceeds range."""
label = np.array([[120], [300]])
with self.assertRaises(ValueError):
get_dataset_colormap.label_to_color_image(
label, get_dataset_colormap.get_pascal_name())