-
Notifications
You must be signed in to change notification settings - Fork 0
/
test.py
220 lines (195 loc) · 8.24 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
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
import time
from detectron2.config import get_cfg
import cv2
from detectron2.utils.visualizer import Visualizer
from detectron2.utils.visualizer import ColorMode
from detectron2.engine import DefaultPredictor
from detectron2.data import MetadataCatalog
from detectron2.data.datasets import register_coco_instances
import numpy as np
from PIL import ImageFont, ImageDraw, Image # 한글 폰트 사용하기 위해
from Acheck_TR import Acheck_TR
cfg = get_cfg()
yaml = "./configs/COCO-InstanceSegmentation/mask_rcnn_R_101_C4_3x.yaml"
weight = './output/model_final_cascade_with_hair/mask_rcnn_R_101_C4_3x/model_0029999.pth'
cfg.merge_from_file(
yaml
)
cfg.DATALOADER.NUM_WORKERS = 2
# cfg.MODEL.WEIGHTS = "detectron2://COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x/137849600/model_final_f10217.pkl" # initialize from model zoo
## (load pretrained weights)
cfg.MODEL.ROI_HEADS.NUM_CLASSES = 3 # 3 classes (Kong, lee, Huh)
register_coco_instances("Acheck", {}, "./Acheck_hair.json", "./img_hair")
MetadataCatalog.get("Acheck").thing_classes = ["Kong", "Lee" , "Huh"]
Acheck_metadata = MetadataCatalog.get("Acheck")
cfg.MODEL.WEIGHTS = weight
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.8 # set the testing threshold for this model
cfg.DATASETS.TEST = ("Acheck", )
predictor = DefaultPredictor(cfg)
def image_test(start,end):
for i in range(start,end):
k = "./test_images/{}.jpg".format(i)
im = cv2.imread(k)
outputs = predictor(im)
v = Visualizer(im[:, :, ::-1],
metadata=Acheck_metadata,
scale=0.8,
instance_mode=ColorMode.IMAGE_BW)
v = v.draw_instance_predictions(outputs["instances"].to("cpu"))
cv2.imshow("d",v.get_image()[:, :, ::-1])
cv2.waitKey()
class Person(object):
def __init__(self,SN,major,name):
self.SN= SN
self.major = major
self.name = name
self.appear = 0
self.percent = 0
self.result = 0
person1 = Person(20151739,'기계공학과','공대현(Kong)')
person2 = Person(20151476,'전자공학과','이원석(Lee)')
person3 =Person(20141713,'화생공학과','허 찬(Huh)')
person4 = Person(20161739, '방송연예학', '사 나(Sana)')
person_list = [person1, person2, person3, person4]
anum=[ 0 , 0 , 0 ,0]
aboard = {0:'Kong',1:'Lee',2:'Huh',3:'Sana'}
b , g , r , a= 0 , 255 , 0 , 0
fontpath="./fonts/gulim.ttf"
font = ImageFont.truetype( fontpath , 20 )
fourcc = cv2.VideoWriter_fourcc(*'MJPG')
def video_test(video_file,frame_term): #frame_term must be integer
stream = cv2.VideoCapture(video_file)
frame = -1
start = time.time()
while True :
frame = frame + 1
rr , im=stream.read()
if rr == False:
break
if frame%frame_term != 0:
pass
else:
#im = cv2.flip(im , 0) # 상하반전
outputs=predictor( im )
v = Visualizer(im[:, :, ::-1],
metadata=Acheck_metadata,
scale=0.8,
instance_mode=ColorMode.IMAGE_BW)
v = v.draw_instance_predictions(outputs["instances"].to("cpu"))
img = v.get_image()[:, :, ::-1]
#####################################################################
classes=outputs[ 'instances' ].__dict__.get( '_fields' ).get( 'pred_classes' )
classes_list=list( set( classes.tolist() ) )
l=len( classes_list )
for i in range(l):
anum[ classes_list[ i ] ]=anum[ classes_list[ i ] ] + 1
for i in range(len( anum )) :
if anum[ i ] :
img_pil=Image.fromarray( img )
draw=ImageDraw.Draw( img_pil )
str = "{}:{} frame".format(aboard.get(i),anum[i])
draw.text( (60 , 30+i*30) , str , font=font , fill=(b , g , r , a) )
img=np.array( img_pil )
#####################################################################
if frame/frame_term == 8 :
print( "지각검사 체크" )
print('when frame is 8 :',anum)
for i in range(len(anum)):
if anum[i] <= 1:
person_list[i].result = '지각(late)'
if frame/frame_term == 12 :
print( "결석검사 체크" )
print( 'when frame is 12 :' , anum )
for i in range(len(anum)):
if person_list[i].result != '지각(late)':
continue
if anum[i] <= 1 :
person_list[i].result = '결석(absence)'
#####################################################################
cv2.imshow("video_test,frame term:{}".format(frame_term), img)
if cv2.waitKey( 1 ) & 0xFF == ord( 'q' ) :
break
total_time = time.time()-start
frame = frame/frame_term
atime=[ 0 , 0 , 0 ,0]
Acheck_TR(anum,atime,person_list,frame,total_time)
def ipcam_test():
stream = cv2.VideoCapture(0)
start=time.time()
frame = 0
while True :
frame += 1
rr , im=stream.read()
outputs=predictor( im )
v=Visualizer( im[ : , : , : :-1 ] , metadata=Acheck_metadata , scale=0.8 , instance_mode=ColorMode.IMAGE_BW )
v=v.draw_instance_predictions( outputs[ "instances" ].to( "cpu" ) )
img=v.get_image()[ : , : , : :-1 ]
#####################################################################
classes=outputs[ 'instances' ].__dict__.get( '_fields' ).get( 'pred_classes' )
classes_list=list( set( classes.tolist() ) )
l=len( classes_list )
for i in range( l ) :
anum[ classes_list[ i ] ]=anum[ classes_list[ i ] ] + 1
for i in range( len( anum ) ) :
if anum[ i ] :
img_pil=Image.fromarray( img )
draw=ImageDraw.Draw( img_pil )
str="{}:{} frame".format( aboard.get( i ) , anum[ i ] )
draw.text( (60 , 30+i*30) , str , font=font , fill=(b , g , r , a) )
img=np.array( img_pil )
#####################################################################
if frame == 8 :
print( "지각검사 체크" )
print( 'when frame is 8 :' , anum )
for i in range( len( anum ) ) :
if anum[ i ] <= 1 :
person_list[ i ].result='지각(late)'
if frame == 12 :
print( "결석검사 체크" )
print( 'when frame is 12 :' , anum )
for i in range( len( anum ) ) :
if person_list[ i ].result != '지각(late)' :
continue
if anum[ i ] <= 3 :
person_list[ i ].result='결석(absence)'
#####################################################################
cv2.imshow( "ipcam_test" , img )
if cv2.waitKey( 1 ) & 0xFF == ord( 'q' ) :
break
total_time = time.time()-start
frame = frame
atime=[ 0 , 0 , 0 ,0]
Acheck_TR(anum,atime,person_list,frame,total_time)
def mAPtest(yaml,weight):
from detectron2.engine import DefaultTrainer
from detectron2.config import get_cfg
cfg=get_cfg()
cfg.merge_from_file( yaml)
cfg.DATALOADER.NUM_WORKERS=2
# cfg.MODEL.WEIGHTS = "detectron2://COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x/137849600/model_final_f10217.pkl" # initialize from model zoo
cfg.MODEL.ROI_HEADS.NUM_CLASSES=3 # 3 classes (Kong, lee, Huh)
from detectron2.data.datasets import register_coco_instances
register_coco_instances( "Acheck_test" , { } , "./Acheck_hair_test.json" , "./img_hair_test" )
from detectron2.data import MetadataCatalog
MetadataCatalog.get( "Acheck_test" ).thing_classes=[ "Kong" , "Lee" , "Huh" ]
Acheck_metadata=MetadataCatalog.get( "Acheck_test" )
from detectron2.data import DatasetCatalog
dataset_dicts=DatasetCatalog.get( "Acheck_test" )
cfg.DATASETS.TRAIN=("Acheck_test" ,)
from detectron2.engine import DefaultPredictor
cfg.MODEL.WEIGHTS= weight
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST=0.8 # set the testing threshold for this model
cfg.DATASETS.TEST=("Acheck_test" ,)
predictor=DefaultPredictor( cfg )
trainer=DefaultTrainer( cfg )
trainer.resume_or_load( resume=False )
from detectron2.evaluation import COCOEvaluator , inference_on_dataset
from detectron2.data import build_detection_test_loader
evaluator=COCOEvaluator( "Acheck_test" , cfg , False , "./output/" )
val_loader=build_detection_test_loader( cfg , "Acheck_test" )
inference_on_dataset( trainer.model , val_loader , evaluator )
###################################################################################################
#ipcam_test()
video_test('./test_videos/13.mp4',10)
#image_test(1,20)
#mAPtest(yaml,weight)