config = InferenceConfig()
config.display()
#####
result_path = DEFAULT_IMAGE_DIR + '/result'
cropped_path = DEFAULT_IMAGE_DIR + '/cropped'
if not os.path.isdir(result_path):
os.mkdir(DEFAULT_IMAGE_DIR + '/result')
os.mkdir(DEFAULT_IMAGE_DIR + '/cropped')
################# LOAD MODELS #######################
#### MASK-R-CNN
mrcnn_model = modellib.MaskRCNN(mode="inference",
config=config,
model_dir=DEFAULT_LOGS_DIR)
weights_path = DEFAULT_MRCNN_MODEL_DIR
mrcnn_model.load_weights(weights_path, by_name=True)
##### EFFICIENTNET
efficient_net = EfficientNetB0(weights='imagenet',
input_shape=(32, 32, 3),
include_top=False,
pooling='max')
eff_model = Sequential()
eff_model.add(efficient_net)
eff_model.add(Dense(units=120, activation='relu'))
eff_model.add(Dense(units=120, activation='relu'))
brush_result_path = DEFAULT_IMAGE_DIR + '/result_brush'
crack_result_path = DEFAULT_IMAGE_DIR + '/result_crack'
cropped_path = DEFAULT_IMAGE_DIR + '/cropped'
if not os.path.isdir(brush_result_path):
os.mkdir(DEFAULT_IMAGE_DIR + '/result_brush')
if not os.path.isdir(crack_result_path):
os.mkdir(DEFAULT_IMAGE_DIR + '/result_crack')
if not os.path.isdir(cropped_path):
os.mkdir(DEFAULT_IMAGE_DIR + '/cropped')
brush_model = modellib.MaskRCNN(mode="inference",
config=brush_config,
model_dir=DEFAULT_LOGS_DIR)
brush_weights_path = DEFAULT_BRUSH_DIR
#### MASK-R-CNN brush
with tf.device('/gpu:0'):
brush_model.load_weights(brush_weights_path, by_name=True)
crack_model = modellib.MaskRCNN(mode="inference",
config=crack_config,
model_dir=DEFAULT_LOGS_DIR)
crack_weights_path = DEFAULT_CRACK_DIR
#### MASK-R-CNN crack
with tf.device('/gpu:1'):
crack_model.load_weights(crack_weights_path, by_name=True)