class VirtualDMSSensor:
    """
    DMS传感器数字孪生模型
    
    模拟三星ISOCELL Auto传感器特性
    用于Synopsys/Ansys虚拟验证流程
    """
    
    def __init__(self, sensor_config: dict):
        self.qe_curve = sensor_config.get('qe_curve', 
                                           {400: 0.45, 550: 0.55, 650: 0.50, 850: 0.30, 940: 0.25})
        self.dark_current = sensor_config.get('dark_current', 2.0)  # e-/s
        self.read_noise = sensor_config.get('read_noise', 1.5)  # e-
        self.full_well = sensor_config.get('full_well', 15000)  # e-
        self.bit_depth = sensor_config.get('bit_depth', 12)
        self.exposure_time = sensor_config.get('exposure_time', 0.033)  # s
    
    def simulate_capture(self, scene_radiance: np.ndarray,
                         wavelength: float = 550,
                         temperature: float = 25) -> np.ndarray:
        """
        模拟传感器捕获图像
        
        Args:
            scene_radiance: (H, W) 场景辐射度（来自光学仿真）
            wavelength: 波长(nm)
            temperature: 传感器温度(°C)
        
        Returns:
            sensor_output: (H, W) 模拟的传感器输出
        """
        # 1. 量子效率
        qe = self._interp_qe(wavelength)
        
        # 2. 光子数 = 辐射度 × 曝光时间 × QE
        photons = scene_radiance * self.exposure_time * qe
        
        # 3. 泊松噪声（光子散粒噪声）
        signal = np.random.poisson(photons).astype(np.float64)
        
        # 4. 暗电流（温度相关）
        dark = self.dark_current * (2 ** ((temperature - 25) / 8))
        dark_electrons = np.random.poisson(dark * self.exposure_time,
                                            size=scene_radiance.shape)
        
        # 5. 读出噪声
        read = np.random.normal(0, self.read_noise,
                                size=scene_radiance.shape)
        
        # 6. 总信号
        total = signal + dark_electrons + read
        
        # 7. 满阱限制
        total = np.clip(total, 0, self.full_well)
        
        # 8. ADC量化
        digital = (total / self.full_well * 
                   (2 ** self.bit_depth - 1)).astype(np.uint16)
        
        return digital
    
    def _interp_qe(self, wavelength: float) -> float:
        """插值获取指定波长的量子效率"""
        wavelengths = sorted(self.qe_curve.keys())
        return np.interp(wavelength, wavelengths,
                        [self.qe_curve[w] for w in wavelengths])


class VirtualDMSValidation:
    """
    虚拟DMS验证流程
    
    1. 从Ansys光学仿真获取场景辐射度
    2. 通过传感器数字孪生生成模拟图像
    3. 用DMS算法处理模拟图像
    4. 验证算法在模拟数据上的表现
    """
    
    def __init__(self, sensor_config: dict):
        self.sensor = VirtualDMSSensor(sensor_config)
        self.test_scenarios = []
    
    def add_scenario(self, name: str, radiance: np.ndarray,
                     expected_result: dict):
        """添加测试场景"""
        self.test_scenarios.append({
            'name': name,
            'radiance': radiance,
            'expected': expected_result
        })
    
    def run_validation(self, dms_algorithm) -> dict:
        """
        运行虚拟验证
        
        Args:
            dms_algorithm: DMS处理函数 (image) -> result
        
        Returns:
            validation_report: 验证报告
        """
        results = []
        
        for scenario in self.test_scenarios:
            # 模拟传感器捕获
            simulated_image = self.sensor.simulate_capture(
                scenario['radiance']
            )
            
            # 运行DMS算法
            dms_result = dms_algorithm(simulated_image)
            
            # 对比预期结果
            passed = self._check_result(dms_result, scenario['expected'])
            
            results.append({
                'scenario': scenario['name'],
                'passed': passed,
                'result': dms_result,
                'expected': scenario['expected']
            })
        
        pass_rate = sum(r['passed'] for r in results) / len(results)
        
        return {
            'total_scenarios': len(results),
            'passed': sum(r['passed'] for r in results),
            'pass_rate': pass_rate,
            'results': results
        }
    
    def _check_result(self, result, expected):
        """检查结果是否符合预期"""
        for key, value in expected.items():
            if result.get(key) != value:
                return False
        return True


import numpy as np

# 测试
if __name__ == "__main__":
    sensor_config = {
        'qe_curve': {400: 0.45, 550: 0.55, 650: 0.50, 850: 0.30, 940: 0.25},
        'dark_current': 2.0,
        'read_noise': 1.5,
        'full_well': 15000,
        'bit_depth': 12,
        'exposure_time': 0.033
    }
    
    validator = VirtualDMSValidation(sensor_config)
    
    # 模拟测试场景
    H, W = 100, 100
    
    # 场景1: 正常光照，驾驶员睁眼
    validator.add_scenario(
        "正常-睁眼",
        np.ones((H, W)) * 5000,
        {'face_detected': True}
    )
    
    # 场景2: 低光照
    validator.add_scenario(
        "低照度",
        np.ones((H, W)) * 500,
        {'face_detected': True}  # IR补光后应可检测
    )
    
    # 场景3: 高动态范围（阳光直射+阴影）
    radiance = np.ones((H, W)) * 200
    radiance[:H//2] = 10000  # 上半阳光
    validator.add_scenario(
        "HDR场景",
        radiance,
        {'face_detected': True}
    )
    
    # 模拟DMS算法（简化）
    def mock_dms(image):
        return {'face_detected': image.mean() > 300}
    
    # 运行验证
    report = validator.run_validation(mock_dms)
    
    print(f"虚拟验证报告:")
    print(f"  总场景: {report['total_scenarios']}")
    print(f"  通过: {report['passed']}")
    print(f"  通过率: {report['pass_rate']:.1%}")
    for r in report['results']:
        print(f"  {r['scenario']:15s}: {'✅' if r['passed'] else '❌'}")