#images = glob.glob('./run_track1/*.jpg')
images = ['./examples/input_salient_track2_6.jpg']
total = len(images)
# Initial call to print 0% progress
printProgressBar(0, total, prefix='Progress:', suffix='Complete', length=50)
batch_print = 100
for i, path in enumerate(images):
image_array = mpimg.imread(path)
yuv = cv2.cvtColor(image_array, cv2.COLOR_BGR2YUV)
y, u, v = cv2.split(yuv)
y = y.reshape(160, 320, -1)
filename = path.split('/')[-1]
img_predicted = model.predict(y[None, :, :, :], batch_size=1)
img_predicted = img_predicted.reshape(img_predicted.shape[1:])
img = ((img_predicted - img_predicted.min()) * 255 /
(img_predicted.max() - img_predicted.min())).astype(np.uint8)
new = [[[j if j <= 80 else 0, j if j > 80 else 0, 0] for j in i]
for i in img]
dt = np.dtype('uint8')
new = np.array(new, dtype=dt)
cval=0.,
horizontal_flip=True,
vertical_flip=False,
rescale=None)
test_datagen = ImageDataGenerator(featurewise_center=False,
samplewise_center=True,
featurewise_std_normalization=False,
samplewise_std_normalization=True)
if preview_augmentation:
# Preview resultant images:
img = load_img(train_data_dir +
'/mountain/land132.jpg') # this is a PIL image
x = img_to_array(img) # this is a Numpy array with shape (3, 150, 150)
x = x.reshape(
(1, ) + x.shape) # this is a Numpy array with shape (1, 3, 150, 150)
i = 0
for batch in train_datagen.flow(x,
batch_size=1,
save_to_dir='preview',
save_prefix='image',
save_format='jpeg'):
i += 1
if i > 20:
break # otherwise the generator would loop indefinitely
train_generator = train_datagen.flow_from_directory(train_data_dir,
target_size=(img_width,
img_height),
batch_size=batch_size,
class_mode='categorical')
