def preprocess(trainingSamples):
global MS, IS
XS = l.create_partial_XS(trainingSamples)
MS = np.zeros((c.get_nb_teeth(), c.get_nb_dim()))
IS = np.zeros((c.get_nb_teeth(), c.get_nb_dim()))
for j in range(c.get_nb_teeth()):
S = XS[j, :, :]
M, Y = pa.PA(S)
MS[j, :] = M
mtx = mty = ms = mtheta = 0
n = S.shape[0]
for i in range(n):
tx, ty, s, theta = mu.full_align_params(
M, fu.original_to_cropped(S[i, :]))
mtx += tx
mty += ty
ms += s
mtheta += theta
n = float(n)
mtx /= n
mty /= n
ms /= n
mtheta /= n
IS[j, :] = mu.full_align(M, mtx, mty, ms, mtheta)
def show_feedback(M, P_before, P_after):
'''
Shows feedback between iterations
@param M: the model for the tooth
@param P_before: the current points for the target tooth before validation
in the image coordinate frame
@param P_after: the current points for the target tooth after validation
in the image coordinate frame
'''
tx_old, ty_old, s_old, theta_old = mu.full_align_params(P_before, M)
tx, ty, s, theta = mu.full_align_params(P_after, M)
xm_old, ym_old = mu.get_center_of_gravity(P_before)
xm, ym = mu.get_center_of_gravity(P_after)
print(
str((xm_old - xm)) + ' # ' + str((ym_old - ym)) + ' # ' +
str((s_old - s)) + ' # ' + str((theta_old - theta)))
def validate(img, tooth_index, P_before, nb_it, show=False):
'''
Validates the current points P for the target tooth corresponding to the given
tooth index.
@param img: the image
@param P_before: the current points for the target tooth before validation
in the image coordinate frame
@param tooth_index: the index of the the target tooth (used in MS, EWS, fs)
@param nb_it: the number of this iteration
@param show: must the intermediate results (after each iteration) be displayed
'''
MU = MS[tooth_index]
E, W = EWS[tooth_index]
xm, ym = mu.get_center_of_gravity(P_before)
tx, ty, s, theta = mu.full_align_params(P_before, MU)
PY_before = mu.full_align(P_before, tx, ty, s, theta)
bs = pca.project(W, PY_before, MU)
bs = np.maximum(np.minimum(bs, tolerable_deviation * E),
-tolerable_deviation * E)
PY_after = pca.reconstruct(W, bs, MU)
P_after = mu.full_align(PY_after, xm, ym, 1.0 / s, -theta)
if (show):
fu.show_validation(MU, nb_it, PY_before, PY_after)
fu.show_iteration(np.copy(img), nb_it, P_before, P_after)
cv2.waitKey(0)
pyplot.close()
return P_after
def create_average_models(trainingSamples, method=''):
XS = l.create_partial_XS(trainingSamples)
MS = np.zeros((c.get_nb_teeth(), c.get_nb_dim()))
IS = np.zeros((c.get_nb_teeth(), c.get_nb_dim()))
for j in range(c.get_nb_teeth()):
S = XS[j,:,:]
M, Y = pa.PA(S)
MS[j,:] = M
mtx = mty = ms = mtheta = 0
n = S.shape[0]
for i in range(n):
tx, ty, s, theta = mu.full_align_params(M, fu.original_to_cropped(S[i,:]))
mtx += tx
mty += ty
ms += s
mtheta += theta
n = float(n)
IS[j,:] = mu.full_align(M, (mtx / n), (mty / n), (ms / n), (mtheta / n))