/
registration_callbacks.py
155 lines (130 loc) · 5.35 KB
/
registration_callbacks.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
import numpy as np
import matplotlib.pyplot as plt
import SimpleITK as sitk
from IPython.display import clear_output
from scipy import linalg
from registration_utilities import registration_errors
# Callback we associate with the StartEvent, sets up our new data.
def metric_start_plot():
global metric_values, multires_iterations
global current_iteration_number
metric_values = []
multires_iterations = []
current_iteration_number = -1
# Callback we associate with the EndEvent, do cleanup of data and figure.
def metric_end_plot():
global metric_values, multires_iterations
global current_iteration_number
del metric_values
del multires_iterations
del current_iteration_number
# Close figure, we don't want to get a duplicate of the plot latter on
plt.close()
# Callback we associate with the IterationEvent, update our data and display
# new figure.
def metric_plot_values(registration_method):
global metric_values, multires_iterations
global current_iteration_number
# Some optimizers report an iteration event for function evaluations and not
# a complete iteration, we only want to update every iteration.
if registration_method.GetOptimizerIteration() == current_iteration_number:
return
current_iteration_number = registration_method.GetOptimizerIteration()
metric_values.append(registration_method.GetMetricValue())
# Clear the output area (wait=True, to reduce flickering), and plot
# current data.
clear_output(wait=True)
# Plot the similarity metric values.
plt.plot(metric_values, "r")
plt.plot(
multires_iterations,
[metric_values[index] for index in multires_iterations],
"b*",
)
plt.xlabel("Iteration Number", fontsize=12)
plt.ylabel("Metric Value", fontsize=12)
plt.show()
# Callback we associate with the MultiResolutionIterationEvent, update the
# index into the metric_values list.
def metric_update_multires_iterations():
global metric_values, multires_iterations
multires_iterations.append(len(metric_values))
# Callback we associate with the StartEvent, sets up our new data.
def metric_and_reference_start_plot():
global metric_values, multires_iterations, reference_mean_values
global reference_min_values, reference_max_values
global current_iteration_number
metric_values = []
multires_iterations = []
reference_mean_values = []
reference_min_values = []
reference_max_values = []
current_iteration_number = -1
# Callback we associate with the EndEvent, do cleanup of data and figure.
def metric_and_reference_end_plot():
global metric_values, multires_iterations, reference_mean_values
global reference_min_values, reference_max_values
global current_iteration_number
del metric_values
del multires_iterations
del reference_mean_values
del reference_min_values
del reference_max_values
del current_iteration_number
# Close figure, we don't want to get a duplicate of the plot latter on.
plt.close()
# Callback we associate with the IterationEvent, update our data and display
# new figure.
def metric_and_reference_plot_values(registration_method, fixed_points, moving_points):
global metric_values, multires_iterations, reference_mean_values
global reference_min_values, reference_max_values
global current_iteration_number
# Some optimizers report an iteration event for function evaluations and not
# a complete iteration, we only want to update every iteration.
if registration_method.GetOptimizerIteration() == current_iteration_number:
return
current_iteration_number = registration_method.GetOptimizerIteration()
metric_values.append(registration_method.GetMetricValue())
# Compute and store TRE statistics (mean, min, max).
current_transform = sitk.CompositeTransform(
registration_method.GetInitialTransform()
)
current_transform.SetParameters(registration_method.GetOptimizerPosition())
current_transform.AddTransform(registration_method.GetMovingInitialTransform())
current_transform.AddTransform(
registration_method.GetFixedInitialTransform().GetInverse()
)
mean_error, _, min_error, max_error, _ = registration_errors(
current_transform, fixed_points, moving_points
)
reference_mean_values.append(mean_error)
reference_min_values.append(min_error)
reference_max_values.append(max_error)
# Clear the output area (wait=True, to reduce flickering), and plot current data.
clear_output(wait=True)
# Plot the similarity metric values.
plt.subplot(1, 2, 1)
plt.plot(metric_values, "r")
plt.plot(
multires_iterations,
[metric_values[index] for index in multires_iterations],
"b*",
)
plt.xlabel("Iteration Number", fontsize=12)
plt.ylabel("Metric Value", fontsize=12)
# Plot the TRE mean value and the [min-max] range.
plt.subplot(1, 2, 2)
plt.plot(reference_mean_values, color="black", label="mean")
plt.fill_between(
range(len(reference_mean_values)),
reference_min_values,
reference_max_values,
facecolor="red",
alpha=0.5,
)
plt.xlabel("Iteration Number", fontsize=12)
plt.ylabel("TRE [mm]", fontsize=12)
plt.legend()
# Adjust the spacing between subplots so that the axis labels don't overlap.
plt.tight_layout()
plt.show()