Artificial Neural Network classification of operator workload with an assessment of time variation and noise-enhancement to increase performance.
Artificial Neural Network classification of operator workload with an assessment of time variation and noise-enhancement to increase performance.
复制标题
DOI：
10.3389/fnins.2014.00372
复制
发表时间：
2014
影响因子：
4.3
通讯作者：
Casson AJ
中科院分区：
文献类型：
--
作者：
Casson AJ
脑机分类-确定人类操作员是否处于高或低工作负荷状态，以优化其工作环境-是被动脑机接口（BCI）系统的新兴应用。实际的系统不仅要准确地检测当前的工作负载状态，还要具有良好的时间性能：需要很少的时间来设置和训练分类器，并确保报告的性能水平随着时间的推移是一致的和可预测的。本文研究了基于人工神经网络的分类系统的时间性能。对于在很少的EEG数据上训练的网络，在很短的时间内实现了良好的分类准确率（86%），但随着网络训练和实际使用之间的时间间隔增加，发现准确率大幅下降。噪声增强处理是一种潜在的技术，可以在测试信号中故意添加人为生成的噪声，以减轻这种退化，而无需使用更多数据重新训练网络。小的随机共振效应证明，从而在存在更多的噪声的分类过程变得更好。效果很小，并没有消除重新训练的需要，但它是一致的，这是第一次证明这种效果的非诱发/自由运行的EEG信号适用于被动BCI。
Workload classification—the determination of whether a human operator is in a high or low workload state to allow their working environment to be optimized—is an emerging application of passive Brain-Computer Interface (BCI) systems. Practical systems must not only accurately detect the current workload state, but also have good temporal performance: requiring little time to set up and train the classifier, and ensuring that the reported performance level is consistent and predictable over time. This paper investigates the temporal performance of an Artificial Neural Network based classification system. For networks trained on little EEG data good classification accuracies (86%) are achieved over very short time frames, but substantial decreases in accuracy are found as the time gap between the network training and the actual use is increased. Noise-enhanced processing, where artificially generated noise is deliberately added to the testing signals, is investigated as a potential technique to mitigate this degradation without requiring the network to be re-trained using more data. Small stochastic resonance effects are demonstrated whereby the classification process gets better in the presence of more noise. The effect is small and does not eliminate the need for re-training, but it is consistent, and this is the first demonstration of such effects for non-evoked/free-running EEG signals suitable for passive BCI.
登录
查看更多内容
影响因子：
8.6
作者：
Gluckman, BJ;Netoff, TI;Schiff, SJ
通讯作者：
Schiff, SJ
影响因子：
4.6
作者：
De Clercq, Wim;Vergult, Anneleen;Van Huffel, Sabine
通讯作者：
Van Huffel, Sabine
影响因子：
17.6
作者：
Chi, Yu Mike;Jung, Tzyy-Ping;Cauwenberghs, Gert
通讯作者：
Cauwenberghs, Gert
DOI：
10.1109/jstsp.2008.2011162
发表时间：
2009-02-01
影响因子：
7.5
作者：
Peng, Renbin;Chen, Hao;Varshney, Pramod K.
通讯作者：
Varshney, Pramod K.
影响因子：
4.3
作者：
Dijksterhuis C;de Waard D;Brookhuis KA;Mulder BL;de Jong R
通讯作者：
de Jong R