-
Notifications
You must be signed in to change notification settings - Fork 1
/
show_feature.py
310 lines (227 loc) · 8 KB
/
show_feature.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
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
import torch
import torchvision.transforms as transforms
import skimage.data
import skimage.io
import skimage.transform
import numpy as np
import matplotlib
import pylab
import matplotlib.pyplot as plt
import os
import cv2
# from completion_segmentation_model import DepthCompletionFrontNet
# from completion_segmentation_model_v3_eca_attention import DepthCompletionFrontNet
import math
from models.yolo_gy import Model
import yaml
from utils.torch_utils import intersect_dicts
# https://blog.csdn.net/missyougoon/article/details/85645195
# https://blog.csdn.net/grayondream/article/details/99090247
# 定义是否使用GPU
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
# 定义数据预处理方式(将输入的类似numpy中arrary形式的数据转化为pytorch中的张量(tensor))
transform = transforms.ToTensor()
def get_picture(picture_dir, transform):
'''
该算法实现了读取图片,并将其类型转化为Tensor
'''
img = skimage.io.imread(picture_dir)
img256 = skimage.transform.resize(img, (128, 256))
img256 = np.asarray(img256)
img256 = img256.astype(np.float32)
return transform(img256)
def get_picture_rgb(picture_dir):
'''
该函数实现了显示图片的RGB三通道颜色
'''
img = skimage.io.imread(picture_dir)
img256 = skimage.transform.resize(img, (256, 256))
skimage.io.imsave('4.jpg', img256)
# 取单一通道值显示
# for i in range(3):
# img = img256[:,:,i]
# ax = plt.subplot(1, 3, i + 1)
# ax.set_title('Feature {}'.format(i))
# ax.axis('off')
# plt.imshow(img)
# r = img256.copy()
# r[:,:,0:2]=0
# ax = plt.subplot(1, 4, 1)
# ax.set_title('B Channel')
# # ax.axis('off')
# plt.imshow(r)
# g = img256.copy()
# g[:,:,0]=0
# g[:,:,2]=0
# ax = plt.subplot(1, 4, 2)
# ax.set_title('G Channel')
# # ax.axis('off')
# plt.imshow(g)
# b = img256.copy()
# b[:,:,1:3]=0
# ax = plt.subplot(1, 4, 3)
# ax.set_title('R Channel')
# # ax.axis('off')
# plt.imshow(b)
# img = img256.copy()
# ax = plt.subplot(1, 4, 4)
# ax.set_title('image')
# # ax.axis('off')
# plt.imshow(img)
img = img256.copy()
ax = plt.subplot()
ax.set_title('image')
# ax.axis('off')
plt.imshow(img)
plt.show()
def visualize_feature_map_sum(item, name, save_path):
'''
将每张子图进行相加
:param feature_batch:
:return:
'''
feature_map = item.squeeze(0)
c = item.shape[1]
print(feature_map.shape)
feature_map_combination = []
for i in range(0, c):
feature_map_split = feature_map.data.cpu().numpy()[i, :, :]
feature_map_combination.append(feature_map_split)
feature_map_sum = sum(one for one in feature_map_combination)
# feature_map = np.squeeze(feature_batch,axis=0)
plt.figure()
plt.title("combine figure")
plt.imshow(feature_map_sum)
# print(feature_map_sum.dtype) # float32
# exit()
pylab.show()
# exit()
# plt.show()
# <class 'module'>
# exit()
plt.savefig(save_path+'/feature_map_' + name + '.png') # 保存图像到本地
def visualize_feature_map_sum_1(item, name, save_path, h_img):
'''
将每张子图进行相加
:param feature_batch:
:return:
'''
feature_map = item.squeeze(0)
c = item.shape[1]
# print(type(feature_map))
print(feature_map.shape)
feature_map_sum = 0
for i in range(0, c):
feature_map_sum = feature_map.data.cpu().numpy()[i, :, :] + feature_map_sum
# heatmap = cv2.applyColorMap(heatmap, cv2.COLORMAP_JET)
# cv2.imwrite(save_path + '/cam.png', superimposed_img)
plt.figure()
plt.title("combine figure")
heatmap = cv2.resize(feature_map_sum, (h_img.shape[1], h_img.shape[0]))
Min = np.min(heatmap)
Max = np.max(heatmap)
heatmap = (heatmap - Min) / (Max - Min)
heatmap = heatmap * 150
# heatmap = np.uint8(255 * heatmap)
# superimposed_img = np.sum(heatmap * 0.4,h_img)
# print(h_img.shape[2])
# exit()
h_img[:,:,0] = h_img[:,:,0] + heatmap
plt.imshow(h_img)
plt.axis('off')
plt.savefig(save_path + '/thermal_night_fam.png') # 保存图像到本地
def get_feature():
# 输入数据
# root_path = '/home/ubuntu/dataset/kaist dataset/gy_copy/test/'
# pic_dir = 'set06_V003_lwir_I02799.png'
# pc_path = root_path + 'thermal/' + pic_dir
# rgb_path = root_path + 'visible/' + pic_dir
root_path = '/home/ubuntu/dataset/kaist_pixel_level/preparing_data/train/'
pic_dir = 'set00_V004_I01225.png'
pic_dir_1 = 'set00_V004_I01225.png'
pic_dir_2 = 'set00_V004_I01225.png'
pc_path = root_path + 'images/' + pic_dir_1
rgb_path = root_path + 'RGB/' + pic_dir
th_path = root_path + 'images/' + pic_dir_2
save_path = '/home/ubuntu/code/yolov5-master-copy-copy/feature_map/nosaliency/'
dir_name = pic_dir.split('.')[0]
save_path = save_path+dir_name
if not os.path.exists(save_path):
os.mkdir(save_path)
img_rgb = get_picture(rgb_path, transform)
# 插入维度
img_rgb = img_rgb.unsqueeze(0)
img_rgb = img_rgb.to(device)
img_pc = get_picture(pc_path, transform)
# 插入维度
img_pc = img_pc.unsqueeze(0)
img_pc = img_pc.to(device)
# 加载模型
ckpt = torch.load('/home/ubuntu/code/yolov5-master-copy-copy/runs/train/early_train/RGB+thermal+all-train+0.23779-0.24695-0.24365-0.27173*MBM/weights/best.pt')
# ckpt = torch.load('/home/ubuntu/code/yolov5-master-copy-copy/runs/train/early_train/small_data/RGB+thermal+MBM-0.237-0.246-0.243-0.271-+saliency/weights/best.pt')
cfg = ckpt['model'].yaml
hyp = 'data/hyp.scratch.yaml'
with open(hyp) as f:
hyp = yaml.load(f, Loader=yaml.SafeLoader)
resume = False
model = Model(cfg, ch=3, nc=1, anchors=hyp.get('anchors')).to(device) # create
exclude = ['anchor'] if (cfg or hyp.get('anchors')) and not resume else [] # exclude keys
state_dict = ckpt['model'].float().state_dict() # to FP32
state_dict = intersect_dicts(state_dict, model.state_dict(), exclude=exclude) # intersect
model.load_state_dict(state_dict, strict=False) # load
# print(model)
# myexactor = FeatureExtractor(model, exact_list)
img1 = {
'thermal_img': img_pc, 'imgs': img_rgb
}
# print(img1['pc'])
# x = myexactor(img1)
result,gy_thermal_dict,gy_rgb_dict,gy_fused_dict = model(img1)
outputs = []
x_rgb = []
x_thermal = []
x_fused = []
for item in gy_rgb_dict:
# print(item)
# continue
x_rgb.append(gy_rgb_dict[item])
for item in gy_thermal_dict:
# print(item)
# continue
x_thermal.append(gy_thermal_dict[item])
for item in gy_fused_dict:
# print(item)
# continue
x_fused.append(gy_fused_dict[item])
# 特征输出可视化
outputs = [x_rgb,x_thermal,x_fused]
namex = ['rgb','thermal','fused']
# h_img = cv2.imread(rgb_path)
# h_img = cv2.imread(pc_path)
h_img = cv2.imread(th_path)
for i in range(len(namex)):
k = 0
xname = namex[i]
for item in outputs[i]:
# c = item.shape[1]
plt.figure()
name = "{}{}".format(xname,k)
name = str(name)
plt.suptitle(name)
item = x_fused[0]
# for i in range(c):
# wid = math.ceil(math.sqrt(c))
# ax = plt.subplot(wid, wid, i + 1)
# ax.set_title('Feature {}'.format(i))
# ax.axis('off')
# figure_map = item.data.cpu().numpy()[0, i, :, :]
# plt.imshow(figure_map, cmap='jet')
# plt.savefig('/home/ubuntu/code/yolov5-master-copy-copy/feature_map/feature_map_' + name + '.png') # 保存图像到本地
visualize_feature_map_sum_1(item, name,save_path,h_img)
k = k + 1
exit()
# plt.show()
# 训练
if __name__ == "__main__":
# get_picture_rgb(pic_dir)
get_feature()