/
test.py
182 lines (157 loc) · 8.12 KB
/
test.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
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
from PhIREGAN.PhIREGANs import *
from metrics import *
from utils import *
from Interpolation.interpolation import *
import cv2
import numpy as np
from PIL import Image
class Tester:
DEFAULT_TIMESTEPS = []
COMPONENTS = {'wind': {'ua':1, 'va':1}, 'solar': {'dni':0, 'dhi':1}}
def __init__(self, timesteps=None):
self.phiregan_metrics = {'PSNR': [], 'SSIM': [], 'MSE': [], 'MAE': []}
self.cnn_metrics = {'PSNR': [], 'SSIM': [], 'MSE': [], 'MAE': []}
self.bicubic_metrics = {'PSNR': [], 'SSIM': [], 'MSE': [], 'MAE': []}
self.edsr_metrics = {'PSNR': [], 'SSIM': [], 'MSE': [], 'MAE': []}
self.timesteps = timesteps if timesteps is not None else self.DEFAULT_TIMESTEPS
def interpolate(self, HR_dir, cub):
for gt_HR in os.listdir(HR_dir):
img, _, dimension = read_image(gt_HR)
# Change Image Size
scale_percent = 20 # percent of original image size
resized_img = image_change_scale(img, dimension, scale_percent)
# Change image to original size using bicubic interpolation
cubic_img_algo = bicubic_interpolation(resized_img, dimension)
cubic_img_algo = Image.fromarray(
cubic_img_algo.astype('uint8')).convert('RGB')
# Save output
cv2.imwrite(cub+gt_HR, np.array(cubic_img_algo))
def report_metrics(self):
print("------- GANs Metrics -------")
# Calculate and print the PSNR value
gan_psnr_val = np.mean(self.gan_metrics['PSNR'])
print(f"PSNR: {gan_psnr_val}")
# Calculate and print the SSIM value
gan_ssim_val = np.mean(self.gan_metrics['SSIM'])
print(f"SSIM: {gan_ssim_val}")
# Calculate and print the MSE value
gan_mse_val = np.mean(self.gan_metrics['MSE'])
print(f"MSE: {gan_mse_val}")
# Calculate and print the MAE value
gan_mae_val = np.mean(self.gan_metrics['MAE'])
print(f"MAE: {gan_mae_val}")
print("------- SR CNN Metrics -------")
# Calculate and print the PSNR value
cnn_psnr_val = np.mean(self.cnn_metrics['PSNR'])
print(f"PSNR: {cnn_psnr_val}")
# Calculate and print the SSIM value
cnn_ssim_val = np.mean(self.cnn_metrics['SSIM'])
print(f"SSIM: {cnn_ssim_val}")
# Calculate and print the MSE value
cnn_mse_val = np.mean(self.cnn_metrics['MSE'])
print(f"MSE: {cnn_mse_val}")
# Calculate and print the MAE value
cnn_mae_val = np.mean(self.cnn_metrics['MAE'])
print(f"MAE: {cnn_mae_val}")
print("------- Bicubic Metrics -------")
# Calculate and print the PSNR value
cub_psnr_val = np.mean(self.bicubic_metrics['PSNR'])
print(f"PSNR: {cub_psnr_val}")
# Calculate and print the SSIM value
cub_ssim_val = np.mean(self.bicubic_metrics['SSIM'])
print(f"SSIM: {cub_ssim_val}")
# Calculate and print the MSE value
cub_mse_val = np.mean(self.bicubic_metrics['MSE'])
print(f"MSE: {cub_mse_val}")
# Calculate and print the MAE value
cub_mae_val = np.mean(self.bicubic_metrics['MAE'])
print(f"MAE: {cub_mae_val}")
print("------- EDSR Metrics -------")
# Calculate and print the PSNR value
edsr_psnr_val = np.mean(self.edsr_metrics['PSNR'])
print(f"PSNR: {edsr_psnr_val}")
# Calculate and print the SSIM value
edsr_ssim_val = np.mean(self.edsr_metrics['SSIM'])
print(f"SSIM: {edsr_ssim_val}")
# Calculate and print the MSE value
edsr_mse_val = np.mean(self.edsr_metrics['MSE'])
print(f"MSE: {edsr_mse_val}")
# Calculate and print the MAE value
edsr_mae_val = np.mean(self.edsr_metrics['MAE'])
print(f"MAE: {edsr_mae_val}")
print("------- ESRGAN Metrics -------")
# Calculate and print the PSNR value
esrgan_psnr_val = np.mean(self.esrgan_metrics['PSNR'])
print(f"PSNR: {esrgan_psnr_val}")
# Calculate and print the SSIM value
esrgan_ssim_val = np.mean(self.esrgan_metrics['SSIM'])
print(f"SSIM: {esrgan_ssim_val}")
# Calculate and print the MSE value
esrgan_mse_val = np.mean(self.esrgan_metrics['MSE'])
print(f"MSE: {esrgan_mse_val}")
# Calculate and print the MAE value
esrgan_mae_val = np.mean(self.esrgan_metrics['MAE'])
print(f"MAE: {esrgan_mae_val}")
def compare_output_helper(self, data_type, component, timestep, i):
gt_HR = "output/{data_type} test/{data_type} images/{data_type}/HR/{component}_{timestep}_{i}.png".format(data_type=data_type, component=component, timestep=timestep, i=i)
phiregan = "output/{data_type} test/phiregan images/phiregan_{component}_{timestep}_{i}.png".format(data_type=data_type, component=component, timestep=timestep, i=i)
cub = "output/{data_type} test/bicubic/{component}_{timestep}_{i}.png".format(data_type=data_type, component=component, timestep=timestep, i=i)
edsr = "output/{data_type} test/edsr/sr_output/{component}_{timestep}_{i}.png".format(data_type=data_type, component=component, timestep=timestep, i=i)
esrgan = "output/{data_type} test/esrgan/inference_result/{component}_{timestep}_{i}.png".format(data_type=data_type, component=component, timestep=timestep, i=i)
cnn = "output/{data_type} test/cnn images/cnn_{component}_{timestep}_{i}.png".format(data_type=data_type, component=component, timestep=timestep, i=i)
imgA, _, _ = read_image(gt_HR)
imgB, _, _ = read_image(phiregan)
imgC, _, _ = read_image(cnn)
imgD, _, _ = read_image(cub)
imgE, _, _ = read_image(edsr)
imgF, _, _ = read_image(esrgan)
phiregan_psnr_val = psnr(imgA, imgB)
phiregan_ssim_val = ssim(imgA, imgB)
phiregan_mse_val = mse(imgA, imgB)
phiregan_mae_val = mae(imgA, imgB)
self.phiregan_metrics['PSNR'].append(phiregan_psnr_val)
self.phiregan_metrics['SSIM'].append(phiregan_ssim_val)
self.phiregan_metrics['MSE'].append(phiregan_mse_val)
self.phiregan_metrics['MAE'].append(phiregan_mae_val)
cnn_psnr_val = psnr(imgA, imgC)
cnn_ssim_val = ssim(imgA, imgC)
cnn_mse_val = mse(imgA, imgC)
cnn_mae_val = mae(imgA, imgC)
self.cnn_metrics['PSNR'].append(cnn_psnr_val)
self.cnn_metrics['SSIM'].append(cnn_ssim_val)
self.cnn_metrics['MSE'].append(cnn_mse_val)
self.cnn_metrics['MAE'].append(cnn_mae_val)
cub_psnr_val = psnr(imgA, imgD)
cub_ssim_val = ssim(imgA, imgD)
cub_mse_val = mse(imgA, imgD)
cub_mae_val = mae(imgA, imgD)
self.bicubic_metrics['PSNR'].append(cub_psnr_val)
self.bicubic_metrics['SSIM'].append(cub_ssim_val)
self.bicubic_metrics['MSE'].append(cub_mse_val)
self.bicubic_metrics['MAE'].append(cub_mae_val)
edsr_psnr_val = psnr(imgA, imgE)
edsr_ssim_val = ssim(imgA, imgE)
edsr_mse_val = mse(imgA, imgE)
edsr_mae_val = mae(imgA, imgE)
self.edsr_metrics['PSNR'].append(edsr_psnr_val)
self.edsr_metrics['SSIM'].append(edsr_ssim_val)
self.edsr_metrics['MSE'].append(edsr_mse_val)
self.edsr_metrics['MAE'].append(edsr_mae_val)
esrgan_psnr_val = psnr(imgA, imgF)
esrgan_ssim_val = ssim(imgA, imgF)
esrgan_mse_val = mse(imgA, imgF)
esrgan_mae_val = mae(imgA, imgF)
self.esrgan_metrics['PSNR'].append(esrgan_psnr_val)
self.esrgan_metrics['SSIM'].append(esrgan_ssim_val)
self.esrgan_metrics['MSE'].append(esrgan_mse_val)
self.esrgan_metrics['MAE'].append(esrgan_mae_val)
def compare_output(self, data_type, component):
for timestep in self.timesteps:
for i in range(256):
self.compare_output_helper(data_type, component, timestep, i)
self.report_metrics()
if __name__ == '__main__':
test_wind_timesteps = [3461]
test = Tester(test_wind_timesteps)
test.interpolate("output/{data_type} test/{data_type} images/{data_type}/HR/", "output/{data_type} test/bicubic/")
test.compare_output(data_type='wind', component=None)