-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathOutput_Results.py
87 lines (70 loc) · 3.47 KB
/
Output_Results.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
"""
Visualization process
Please cited:
HANet (Chengxi Han韩承熙,Wuhan University,[email protected],https://chengxihan.github.io/)
please Cited the paper:
C. HAN, C. WU, H. GUO, M. HU, AND H. CHEN,
“HANET: A HIERARCHICAL ATTENTION NETWORK FOR CHANGE DETECTION WITH BI-TEMPORAL VERY-HIGH-RESOLUTION REMOTE SENSING IMAGES,” IEEE J. SEL. TOP. APPL. EARTH OBS. REMOTE SENS., PP. 1–17, 2023, DOI: 10.1109/JSTARS.2023.3264802.
"""
import torch.utils.data
import os
import cv2
from tqdm import tqdm
from utils.parser import get_parser_with_args
from utils.Related import get_test_loaders, initialize_metrics
from datasetHCX import HCXDataset
from torch.utils.data import DataLoader
# model setting, weighted setting and dataloder
parser, metadata = get_parser_with_args()
opt = parser.parse_args()
dev = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
os.environ["OPENCV_IO_ENABLE_OPENEXR"]="2"
# test_loader = get_test_loaders(opt, batch_size=16)
# testdataset = HCXDataset('/data/chengxi.han/data/Building change detection dataset256/test', is_training=False) #is_training=False即为取全部影像，不仅取前景
testdataset = HCXDataset(opt.dataset_dir +'test', is_training=False)
test_loader = DataLoader(testdataset, batch_size=1, shuffle=False, drop_last=False,num_workers=opt.num_workers) #batch size一定要是1！！！！
# path = 'tmp/WHU-ChangedLabelTrain-Fixed15/epoch_40.pt' # the path of the model
path = opt.weight_dir+'final_epoch99.pt'
# path = opt.weight_dir+'epoch_39.pt'
# path = opt.weight_dir+'best.pt'
model = torch.load(path)
# if not os.path.exists('Detection_Re/WHU-ChangedLabelTrain-Fixed15'):
# os.mkdir('Detection_Re/WHU-ChangedLabelTrain-Fixed15')
if not os.path.exists(opt.Output_dir):
os.mkdir(opt.Output_dir)
# test processing
model.eval()
Img_index = 0
test_metrics = initialize_metrics()
with torch.no_grad():
# Unpacking
T = tqdm(test_loader)
for Imgs1, Imgs2, labels in T:
# Transferring to the device
Imgs1 = Imgs1.float().to(dev)
Imgs2 = Imgs2.float().to(dev)
# labels = labels.long().to(dev)
# Model output
Output = model(Imgs1, Imgs2)
# print('输出影像的元组大小为：', len(Output))
Output = Output[-1]
# print('输出影像的元组大小为：', len(Output))
_, Output = torch.max(Output, 1)
# print('输出影像的元组大小为：', len(Output))
Output = Output.data.cpu().numpy()
Output = Output.squeeze() * 255
# print('输出影像的一维度大小为：',Output.shape[0],'输出影像的二维度大小为：',Output.shape[1],'输出影像的三维度大小为：',Output.shape[2])
# print('输出影像的一维度大小为：',Output.shape[0],'输出影像的二维度大小为：',Output.shape[1])
# Output=Output[0:1,:,:]
# print('输出影像的一维度大小为：', Output.shape[0], '输出影像的二维度大小为：', Output.shape[1], '输出影像的三维度大小为：', Output.shape[2])
# results saving
file_path = 'Detection_Re/S2Looking-Fixed15-Row-ColAttention-Epo50/' + str(Img_index).zfill(1)
# file_path = opt.Output_dir + str(Img_index).zfill(1)
# cv2.imshow('结果',Output)
# cv2.waitKey(0)
# cv2.destroyAllWindows()
# cv2.imwrite(file_path + '.png', Output)
cv2.imwrite(opt.Output_dir + testdataset.ids[Img_index], Output)
Img_index += 1
if __name__ == '__main__':
print('测试影像')