Deep neural networks (DNNs) are observed to be successful in pattern classification. However, the high classification performances of DNNs are related to their large training sets. Unfortunately, in the literature, the datasets used to classify motor imagery (MI) electroencephalogram (EEG) signals contain a small number of samples. To achieve high performances with small-sized datasets, most of the studies have employed a transformation such as the common spatial patterns (CSP) before the classification process. However, the CSP is dependent on subjects and introduces computational load in real time applications. It is observed in the literature that data augmentation is not applied for increasing the classification performance of EEG signals. In this study, we have investigated the effect of the augmentation process on the classification performance of MI EEG signals instead of using a preceding transformation such as the CSP, and we have demonstrated that the augmentation process is able to compete with the CSP by generating high success rates for the classification of MI EEGs. In addition to the augmentation process, we have modified the DNN structure to increase the classification performance, to decrease the number of nodes in the structure, and to use less number of hyper parameters. A minimum distance network following the last layer of the convolutional neural network (CNN) was used as the classifier instead of a fully connected neural network (FCNN). By augmenting the EEG dataset and focusing solely on CNN’s training, the training algorithm of the proposed structure is strengthened without applying any transformation. We tested these improvements on brain–computer interface (BCI) competitions 2005 and 2008 databases with two and four classes, and the positive effects of the augmentation on the average accuracies are demonstrated.

据观察，深度神经网络 (DNN) 在模式分类方面取得了成功。然而，DNN 的高分类性能与其庞大的训练集有关。不幸的是，在文献中，用于对运动图像 (MI) 脑电图 (EEG) 信号进行分类的数据集包含少量样本。为了在小数据集上实现高性能，大多数研究在分类过程之前采用了诸如公共空间模式（CSP）之类的转换。然而，CSP 依赖于主题并在实时应用中引入计算负载。在文献中观察到，数据增强并未用于提高 EEG 信号的分类性能。在这项研究中，我们已经研究了增强过程对 MI EEG 信号分类性能的影响，而不是使用前面的转换，如 CSP，并且我们已经证明，增强过程能够通过产生高成功率来与 CSP 竞争MI EEG 的分类。除了增强过程之外，我们还修改了 DNN 结构以提高分类性能，减少结构中的节点数量，并使用更少的超参数。使用卷积神经网络 (CNN) 最后一层之后的最小距离网络代替全连接神经网络 (FCNN) 作为分类器。通过扩充 EEG 数据集并专注于 CNN 的训练，所提出结构的训练算法在不应用任何变换的情况下得到加强。我们在 2005 年和 2008 年的脑机接口 (BCI) 竞赛数据库中测试了这些改进的两个和四个类别，并证明了增强对平均准确度的积极影响。