一种面向嵌入式平台的脑电信号注意力检测方法

Attention detection method for EEG signal based on embedded platform

近年来,基于脑机接口的注意力检测技术在教育和娱乐领域得到了广泛应用。本文针对有限资源计算平台上的注意力检测方法存在检测精度较低的问题,提出了一种适用于嵌入式计算平台的脑电信号注意力检测方法,通过对高精度的优化复杂网络算法进行范式简化和参数优选,显著降低了参数计算所需的时间和训练数据量,并在STM32F407单片机上进行了实现。对6名被试者的在线试验结果表明,相比于嵌入式平台最常用的注意力计量算法,本文提出的方法取得了更高的注意力检测精度,有望提升教育和娱乐领域相关产品的用户体验。

In recent years,attention detection technology based on brain-computer interface has been widely used in education and entertainment fields.Aiming at the problem of low detection accuracy of attention detection methods on limited-resource computing platforms,an EEG signal attention detection method suitable for embedded computing platforms is proposed.Through the paradigm simplification and parameter optimization of the high-precision optimized complex network method,the time required for parameter calculation and the amount of training data are significantly reduced,and it is implemented on the STM32F407 microcontroller.The online test results of 6 subjects show that compared with the most commonly used attention meter method on embedded platforms,the method proposed in this paper has achieved higher attention detection accuracy and is expected to improve the user experience of related products in the entertainment and education field.