python提取脑电功率谱密度特征
时间: 2023-05-31 16:06:43 浏览: 166
脑电功率谱密度是一种对脑电信号进行频谱分析的方法,可以用于提取脑电信号的频率特征。以下是一个示例代码,用于提取脑电功率谱密度特征:
```python
import numpy as np
from scipy import signal
from scipy.integrate import simps
# 定义函数求取功率谱密度特征
def get_psd_features(data, fs):
freqs, psd = signal.welch(data, fs=fs, nperseg=512, scaling='density')
# 计算总功率
total_power = simps(psd, freqs)
# 计算alpha波带的功率
alpha_power = simps(psd[np.logical_and(freqs >= 8, freqs <= 12)], freqs)
# 计算beta波带的功率
beta_power = simps(psd[np.logical_and(freqs >= 13, freqs <= 30)], freqs)
# 计算theta波带的功率
theta_power = simps(psd[np.logical_and(freqs >= 4, freqs <= 7)], freqs)
# 计算delta波带的功率
delta_power = simps(psd[np.logical_and(freqs >= 1, freqs <= 4)], freqs)
# 计算gamma波带的功率
gamma_power = simps(psd[np.logical_and(freqs >= 30, freqs <= 100)], freqs)
# 计算normalized alpha功率
norm_alpha_power = alpha_power / (total_power - delta_power)
# 计算normalized beta功率
norm_beta_power = beta_power / (total_power - delta_power)
# 计算normalized theta功率
norm_theta_power = theta_power / (total_power - delta_power)
# 计算normalized gamma功率
norm_gamma_power = gamma_power / (total_power - delta_power)
return [total_power, alpha_power, beta_power, theta_power, delta_power, gamma_power, norm_alpha_power, norm_beta_power, norm_theta_power, norm_gamma_power]
# 示例使用
data = np.random.rand(10000)
fs = 1000
psd_features = get_psd_features(data, fs)
print(psd_features)
```
上述代码中,我们使用了Scipy库中的signal.welch函数对输入的脑电信号进行功率谱密度估计,然后计算了不同频率带的功率和归一化功率。其中,total_power代表总功率,alpha_power、beta_power、theta_power、delta_power、gamma_power分别代表不同频率带的功率,norm_alpha_power、norm_beta_power、norm_theta_power、norm_gamma_power分别代表不同频率带的归一化功率。这些特征可以用于脑机接口和其他脑电信号分析任务。