-
Notifications
You must be signed in to change notification settings - Fork 0
/
4.py
294 lines (218 loc) · 7.41 KB
/
4.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
from transformers import YolosFeatureExtractor, YolosForObjectDetection
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = YolosFeatureExtractor.from_pretrained('hustvl/yolos-tiny')
model = YolosForObjectDetection.from_pretrained('hustvl/yolos-tiny')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
# model predicts bounding boxes and corresponding COCO classes
logits = outputs.logits
bboxes = outputs.pred_boxes
import torch
from transformers import AutoImageProcessor
image_processor = AutoImageProcessor.from_pretrained("hustvl/yolos-tiny")
#==========
# convert outputs (bounding boxes and class logits) to COCO API
target_sizes = torch.tensor([image.size[::-1]])
results = image_processor.post_process_object_detection(outputs, threshold=0.9, target_sizes=target_sizes)[0]
for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
box = [round(i, 2) for i in box.tolist()]
print(
f"Detected {model.config.id2label[label.item()]} with confidence "
f"{round(score.item(), 3)} at location {box}"
)
#===========画图.
import numpy as np
import matplotlib
matplotlib.use('agg')
import matplotlib.pyplot as plt
#========根据box, 画图片可视化结果. 网络输出出来的结果是, box 的中心点x,y 和 w ,h 的百分比.
# lx, ly , rx ,ry image.size box_hat
outbox=[]
box_hat=results["boxes"]
for i in box_hat:
# lx=(i[0]-i[2]/2).clamp(0,1) #==========一定要根据我的写法理解宽高和矩阵里面反过来.这里面要看好0,1 哪个是宽,哪个是高.
# rx=(i[0]+i[2]/2).clamp(0,1)
# ly=(i[1]-i[3]/2).clamp(0,1)
# ry=(i[1]+i[3]/2).clamp(0,1)
lx=i[0].item()
ly=i[1].item()
rx=i[2].item()
ry=i[3].item()
outbox.append((lx,rx,ly,ry))#========这个坐标很乱, 自己对应一下.我也调了半天.
# print(outbox)
from PIL import ImageDraw
image = image # 打开一张图片
draw = ImageDraw.Draw(image) # 在上面画画
for dex,i in enumerate(outbox): # 注意画图需要的坐标顺序!!!!!!!!!!!!
draw.rectangle([i[0],i[2],i[1],i[3]], outline=(255,0,0)) # [左上角x，左上角y，右下角x，右下角y]，outline边框颜色
draw.text((i[1], i[2]),model.config.id2label[results["labels"][dex].item()]+" "+str(results["scores"][dex].item()), fill=(255, 0, 0))
print('图片保存为tmp.png')
image.save("tmp.png")
if 0:
print(outputs)
#===========下面我们画出来这个图.
#========完全就是detr的配置.
logits = outputs.logits
bboxes = outputs.pred_boxes
print(logits,bboxes)
usedex=logits.max(2)[1]!=91
usedex2=logits.max(2)[0]>0.7
usedex3=logits[(logits.max(2)[1]!=91) & (logits.max(2)[0]>0.7)]
box=bboxes[(logits.max(2)[1]!=91) & (logits.max(2)[0]>0.7)]
raise
VOC_DATA = {
"NUM": 20,
"CLASSES": [
"life",
"name",
"idn",
"front",
"back",
],
}
#重新自动刷新数量
VOC_DATA["NUM"]=len(VOC_DATA["CLASSES"])
numbeijing=VOC_DATA["NUM"]
id2label= {
"0": "life",
"1": "name",
"2": "idn",
"3": "front",
"4": "back",}
usedex=logits.max(2)[1]!=numbeijing # ========91是背景分类, 表示空物体.
usedex2=logits.max(2)[0]>yuzhi
logits_hat=logits[(logits.max(2)[1]!=numbeijing) & (logits.max(2)[0]>yuzhi)]
box_hat=bboxes[(logits.max(2)[1]!=numbeijing) & (logits.max(2)[0]>yuzhi)]
classify_hat=logits_hat.argmax(-1) # box的分类结果
classify_hat=[id2label[str(int(i))] for i in classify_hat]
gailv =logits_hat.softmax(-1).max(-1)[0].tolist()
print("识别到的物体是",classify_hat)
print("概率是",gailv)
import numpy as np
import matplotlib
matplotlib.use('agg')
import matplotlib.pyplot as plt
#========根据box, 画图片可视化结果. 网络输出出来的结果是, box 的中心点x,y 和 w ,h 的百分比.
# lx, ly , rx ,ry image.size box_hat
outbox=[]
for i in box_hat:
lx=(i[0]-i[2]/2).clamp(0,1)*image.size[0] #==========一定要根据我的写法理解宽高和矩阵里面反过来.这里面要看好0,1 哪个是宽,哪个是高.
rx=(i[0]+i[2]/2).clamp(0,1)*image.size[0]
ly=(i[1]-i[3]/2).clamp(0,1)*image.size[1]
ry=(i[1]+i[3]/2).clamp(0,1)*image.size[1]
lx=lx.item()
rx=rx.item()
ly=ly.item()
ry=ry.item()
outbox.append((lx,rx,ly,ry))
# print(outbox)
from PIL import ImageDraw
image = image # 打开一张图片
draw = ImageDraw.Draw(image) # 在上面画画
for dex,i in enumerate(outbox): # 注意画图需要的坐标顺序!!!!!!!!!!!!
draw.rectangle([i[0],i[2],i[1],i[3]], outline=(255,0,0)) # [左上角x，左上角y，右下角x，右下角y]，outline边框颜色
draw.text((i[1], i[2]), classify_hat[dex], fill=(255, 0, 0))
image.save("tmp.png")
# image.show()
#=========根据box_hat画出来坐标
#=========解析box (center_x, center_y, width, height)
# logits.max(2) 看分类.
# 分类表:
"""
"id2label": {
"0": "N/A",
"1": "person",
"2": "bicycle",
"3": "car",
"4": "motorcycle",
"5": "airplane",
"6": "bus",
"7": "train",
"8": "truck",
"9": "boat",
"10": "traffic light",
"11": "fire hydrant",
"12": "N/A",
"13": "stop sign",
"14": "parking meter",
"15": "bench",
"16": "bird",
"17": "cat",
"18": "dog",
"19": "horse",
"20": "sheep",
"21": "cow",
"22": "elephant",
"23": "bear",
"24": "zebra",
"25": "giraffe",
"26": "N/A",
"27": "backpack",
"28": "umbrella",
"29": "N/A",
"30": "N/A",
"31": "handbag",
"32": "tie",
"33": "suitcase",
"34": "frisbee",
"35": "skis",
"36": "snowboard",
"37": "sports ball",
"38": "kite",
"39": "baseball bat",
"40": "baseball glove",
"41": "skateboard",
"42": "surfboard",
"43": "tennis racket",
"44": "bottle",
"45": "N/A",
"46": "wine glass",
"47": "cup",
"48": "fork",
"49": "knife",
"50": "spoon",
"51": "bowl",
"52": "banana",
"53": "apple",
"54": "sandwich",
"55": "orange",
"56": "broccoli",
"57": "carrot",
"58": "hot dog",
"59": "pizza",
"60": "donut",
"61": "cake",
"62": "chair",
"63": "couch",
"64": "potted plant",
"65": "bed",
"66": "N/A",
"67": "dining table",
"68": "N/A",
"69": "N/A",
"70": "toilet",
"71": "N/A",
"72": "tv",
"73": "laptop",
"74": "mouse",
"75": "remote",
"76": "keyboard",
"77": "cell phone",
"78": "microwave",
"79": "oven",
"80": "toaster",
"81": "sink",
"82": "refrigerator",
"83": "N/A",
"84": "book",
"85": "clock",
"86": "vase",
"87": "scissors",
"88": "teddy bear",
"89": "hair drier",
"90": "toothbrush"
},
"""