def main():
init = util.readImage('input/TestSeq/Shift0.png')
final = util.readImage('input/TestSeq/ShiftR2.png')
#final = util.readImage('input/TestSeq/ShiftR5U5.png')
#final = util.readImage('input/TestSeq/ShiftR10.png')
#final = util.readImage('input/TestSeq/ShiftR20.png')
#final = util.readImage('input/TestSeq/ShiftR40.png')
U, V = lk_flow(init, final)
#U, V = hierarchical_lk(init, final)
U = upscale(U)
V = upscale(V)
U = cv2.cvtColor(U, cv2.COLOR_GRAY2RGB)
V = cv2.cvtColor(V, cv2.COLOR_GRAY2RGB)
false_color_U = cv2.applyColorMap(U, cv2.COLORMAP_JET)
false_color_V = cv2.applyColorMap(V, cv2.COLORMAP_JET)
#namedWindow("window")
cv2.imshow("U", false_color_U)
cv2.imshow("V", false_color_V)
import time
time.sleep(30)
def main():
parser = argparse.ArgumentParser(description='Vision Simulator')
parser.add_argument(
'pixel_map',
type=str,
help='An image of what a single white pixel looks like to you.')
parser.add_argument('images',
nargs='+',
help='Images to apply your vision to.')
parser.add_argument(
'-n',
'--norm-factor',
default=1.7,
type=float,
help=
'The exponent factor used in normalizing pixel_map. You may have to adjust this value for good results.'
)
parser.add_argument('-v',
'--verbose',
action='store_true',
help='Run in verbose mode.')
args = parser.parse_args()
visionImg = util.readImage(args.pixel_map)
visionMap = util.getBrightnessMap(visionImg, args.norm_factor)
for path in args.images:
image = util.readImage(path, convertToFloat=True)
image = applyVision(visionMap, image, verbose=args.verbose)
util.writeImage(path + '.vision.png', image)
def q1b():
init = util.readImage('input/TestSeq/Shift0.png')
final = util.readImage('input/TestSeq/ShiftR10.png')
plotDisplacements(init, final, 'output/ps5-1-b-1.png')
final = util.readImage('input/TestSeq/ShiftR20.png')
plotDisplacements(init, final, 'output/ps5-1-b-2.png')
final = util.readImage('input/TestSeq/ShiftR40.png')
plotDisplacements(init, final, 'output/ps5-1-b-3.png')
def main3():
init = util.readImage('input/TestSeq/Shift0.png')
final = util.readImage('input/TestSeq/ShiftR2.png')
U, V = lk_flow(init, final)
warped = warp.warp(final, U, V)
cv2.imshow("original", init)
cv2.imshow("modified", warped)
import time
time.sleep(30)
def q3a():
image1 = util.readImage('input/DataSeq1/yos_img_01.jpg')
image2 = util.readImage('input/DataSeq1/yos_img_02.jpg')
image3 = util.readImage('input/DataSeq1/yos_img_03.jpg')
q3helper(image1, image2, 1)
q3helper(image2, image3, 5)
image1 = util.readImage('input/DataSeq2/0.png')
image2 = util.readImage('input/DataSeq2/1.png')
image3 = util.readImage('input/DataSeq2/2.png')
q3helper(image1, image2, 3)
q3helper(image2, image3, 7)
def q2b():
image = util.readImage('input/DataSeq1/yos_img_01.jpg')
gauPyr = pyramid.gaussPyramid(image, 3)
lapPyr = pyramid.laplPyramid(gauPyr)
q2_helper(lapPyr, 'output/ps5-2-b-2.png')
def q2a():
image = util.readImage('input/DataSeq1/yos_img_01.jpg')
gauPyr = pyramid.gaussPyramid(image, 3)
q2_helper(gauPyr, 'output/ps5-2-b-1.png')
def q1a():
init = util.readImage('input/TestSeq/Shift0.png')
final = util.readImage('input/TestSeq/ShiftR2.png')
final2 = util.readImage('input/TestSeq/ShiftR5U5.png')
plotDisplacements(init, final, 'output/ps5-1-a-1.png')
plotDisplacements(init, final2, 'output/ps5-1-a-2.png')
for key in metric_keys
}
for key in metric_keys:
f = f_dict[key]
f.write('name ')
for i in range(1, num_class):
f.write('OAR%d ' % (i))
f.write('average\n')
count = 0
for patient in test_list:
predict_path = os.path.join(predict_root_path, patient, 'predict.nii.gz')
label_path = os.path.join(label_root_path, patient, 'label.nii.gz')
predict = readImage(predict_path).astype(np.uint8)
label = readImage(label_path)
predict = one_hot(predict)
label = one_hot(label)
for key in metric_keys:
f = f_dict[key]
temp_list = []
temp_string = 'patient{} '.format(patient)
for i in range(1, num_class):
metric = metric_dict[key](label[..., i], predict[..., i])
temp_string += '%.4f ' % (metric)
temp_list.append(metric)
avg = average(temp_list)
temp_string += '%.4f\n' % (avg)
