基于坐姿压力检测的人脑注意力分析研究

Research on human brain attention analysis based on sitting posture pressure detection

本文提出了一种基于坐姿压力检测的人脑注意力状态分析方法,旨在深入分析和评估人类在外界环境认知过程中的注意力状态。研究显示,坐姿不仅能反映人体健康状况,还能揭示精神状态。特别地,坐姿的变化往往与注意力的专注程度密切相关。为此,利用压力传感器采集受试者的坐姿压力数据,并借助GRU模型对其进行深度分析,以揭示坐姿压力分布及其变化,进而推断出用户的注意力集中程度。为获得准确的注意力状态标签,结合基于脑电数据的注意力分类模型和ABR检测方程,通过提取脑电数据关键通道的特征数据,使用KNN模型进行训练,并获得一个分类模型以进行注意力状态分类。同时,计算ABR值以评估注意力专注程度。在标签确定过程中,当分类模型的分类结果与ABR检测方程的判断一致时,该结果将作为最终的注意力标签;否则,相应数据将被舍弃。这种双重验证确保了注意力标签的准确性和可靠性。该方法在注意力分析领域取得了显著成效,模型准确率可达98%以上。此方法不仅成本低廉、操作简便,还具备非侵入性特点,为注意力的相关研究和应用提供了有力的支持。

This paper proposes a method for analyzing human brain attention state based on sitting posture pressure detection,aiming to deeply analyze and evaluate human attention state during the cognitive process of the external environment.Research has shown that sitting posture can not only reflect human health status,but also reveal mental state.Specifically,changes in sitting posture are often closely related to the degree of attention level.Therefore,pressure sensors are used to collect subjects'sitting posture pressure data,and the GRU model is employed for in-depth analysis to reveal the distribution and changes of sitting posture pressure,thereby inferring the degree of users'focus intensity.To obtain accurate attention state labels,a combination of EEG-based attention classification model and ABR detection equation is utilized.By extracting feature data from key channels of EEG data,the KNN model is used to obtain a classification model for attention state classification.Meanwhile,the ABR value is calculated to evaluate the degree of attention focus.During the labeling process,when the classification result of the classification model is consistent with the judgment of the ABR detection equation,the result will be taken as the final attention label;otherwise,the corresponding data will be discarded.This dual verification ensures the accuracy and reliability of the attention labels.This method has achieved remarkable results in the field of attention analysis,with a model accuracy of over 98%.This method is not only cost-effective,easy to operate,but also non-invasive,providing strong support for relevant research and applications on attention.