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
| """
ZF Infrared Belt 检测集成示例
适用于 Euro NCAP 合规的安全带误用检测
"""
import numpy as np
from typing import Tuple, List, Optional
from dataclasses import dataclass
from enum import Enum
class BeltStatus(Enum):
UNBUCKLED = 0
CORRECT = 1
MISUSE_SHOULDER = 2
MISUSE_LAP = 3
MISUSE_BOTH = 4
NOT_DETECTED = 5
@dataclass
class SeatPosition:
row: int
side: str
occupied: bool = False
class InfraredBeltDetector:
"""
红外安全带检测器
硬件要求:
- 940nm 红外摄像头 (前排+后排)
- ZF LIFETEC Infrared Belt 安全带
- 最小分辨率: 1280x720
- 帧率: ≥15fps (NCAP 最低要求)
"""
IR_THRESHOLD = 0.65
MIN_BELT_LENGTH = 80
ANGLE_TOLERANCE = 15
def __init__(self, config: dict):
self.camera_front = config.get('camera_front', 'default')
self.camera_rear = config.get('camera_rear', 'default')
self.seats = self._init_seats(config)
def _init_seats(self, config: dict) -> List[SeatPosition]:
seats = []
rows = config.get('rows', 2)
for row in range(1, rows + 1):
sides = config.get(f'row{row}_sides', ['left', 'right'])
for side in sides:
seats.append(SeatPosition(row=row, side=side))
return seats
def detect_belt_status(
self,
ir_image: np.ndarray,
seat: SeatPosition
) -> Tuple[BeltStatus, float]:
"""
检测安全带佩戴状态
Args:
ir_image: 红外图像 (H, W), uint8
seat: 座位位置
Returns:
status: BeltStatus 枚举
confidence: 置信度 0-1
"""
belt_mask = self._extract_ir_signature(ir_image)
if belt_mask.sum() < self.MIN_BELT_LENGTH:
return BeltStatus.NOT_DETECTED, 0.0
if not seat.occupied:
return BeltStatus.UNBUCKLED, 0.95
belt_path = self._trace_belt_path(belt_mask)
shoulder_ok, lap_ok = self._classify_belt_position(
belt_path, seat
)
confidence = self._calculate_confidence(belt_mask, belt_path)
if shoulder_ok and lap_ok:
return BeltStatus.CORRECT, confidence
elif not shoulder_ok and not lap_ok:
return BeltStatus.MISUSE_BOTH, confidence
elif not shoulder_ok:
return BeltStatus.MISUSE_SHOULDER, confidence
else:
return BeltStatus.MISUSE_LAP, confidence
def _extract_ir_signature(self, ir_image: np.ndarray) -> np.ndarray:
"""
从红外图像提取安全带签名
ZF 红外安全带的特征:
- 在 940nm 波段有高反射率
- 与皮肤/衣物/座椅形成高对比
"""
normalized = ir_image.astype(np.float32) / 255.0
belt_mask = normalized > self.IR_THRESHOLD
return belt_mask.astype(np.uint8)
def _trace_belt_path(self, mask: np.ndarray) -> np.ndarray:
"""追踪安全带路径"""
from scipy import ndimage
skeleton = ndimage.morphology.distance_transform_edt(mask)
path = np.argwhere(skeleton > 0)
return path
def _classify_belt_position(
self,
path: np.ndarray,
seat: SeatPosition
) -> Tuple[bool, bool]:
"""
分类肩带和腰带位置是否正确
正确佩戴标准 (Euro NCAP):
- 肩带: 从肩部跨锁骨到对侧腰部
- 腰带: 跨越盆骨（非腹部）
- 角度偏差 < 15度
"""
if len(path) < 10:
return False, False
angle = np.degrees(np.arctan2(
path[-1, 0] - path[0, 0],
path[-1, 1] - path[0, 1]
))
shoulder_ok = 25 < abs(angle) < 65
lap_ok = abs(angle) < self.ANGLE_TOLERANCE
return shoulder_ok, lap_ok
def _calculate_confidence(
self,
mask: np.ndarray,
path: np.ndarray
) -> float:
"""计算检测置信度"""
ir_strength = mask.mean()
path_length = len(path)
confidence = min(ir_strength * 1.5, 1.0)
length_factor = min(path_length / 200, 1.0)
return (confidence + length_factor) / 2
class EuroNCAPCompliance:
"""Euro NCAP 安全带检测合规检查"""
def __init__(self):
self.detector = InfraredBeltDetector({
'rows': 2,
'row1_sides': ['left', 'right'],
'row2_sides': ['left', 'center', 'right']
})
def check_all_seats(self, ir_images: dict) -> dict:
"""
检查所有座位
Args:
ir_images: {'front': ndarray, 'rear': ndarray}
Returns:
compliance_report: 每座状态 + 整体合规
"""
results = {}
for seat in self.detector.seats:
img_key = 'front' if seat.row == 1 else 'rear'
if img_key not in ir_images:
continue
status, conf = self.detector.detect_belt_status(
ir_images[img_key], seat
)
results[f"r{seat.row}_{seat.side}"] = {
'status': status.name,
'confidence': round(conf, 3),
'compliant': status in [BeltStatus.CORRECT, BeltStatus.UNBUCKLED]
}
results['overall_compliant'] = all(
r['compliant'] for r in results.values()
if isinstance(r, dict)
)
return results
if __name__ == "__main__":
compliance = EuroNCAPCompliance()
front_ir = np.random.randint(0, 100, (720, 1280), dtype=np.uint8)
rear_ir = np.random.randint(0, 100, (720, 1280), dtype=np.uint8)
front_ir[200:600, 400:420] = 200
front_ir[600:630, 300:800] = 200
report = compliance.check_all_seats({
'front': front_ir,
'rear': rear_ir
})
for seat, result in report.items():
if isinstance(result, dict):
print(f"{seat}: {result['status']} (conf={result['confidence']})")
print(f"\nOverall NCAP Compliant: {report['overall_compliant']}")
|