Electroencephalogram (EEG) artifacts such as eyeblink, eye movement, and muscle movements widely contaminate the EEG signals. Those unwanted artifacts corrupt the information contained in the EEG signals and degrade the performance of qualitative analysis of clinical applications and as well as EEG-based brain–computer interfaces (BCIs). The applications of wavelet transform in denoising EEG signals are increasing day by day due to its capability of handling non-stationary signals. All the reported wavelet denoising techniques for EEG signals are surveyed in this paper in terms of the quality of noise removal and retrieving important information. In order to evaluate the performance of wavelet denoising techniques for EEG signals and to express the quality of reconstruction, the techniques were evaluated based on the results shown in the respective literature. We also compare certain features in the evaluation of the wavelet denoising techniques, such as the requirement of reference channel, automation, online, and performance on a single channel.

Emotion recognition using EEG has been widely studied to address the challenges associated with affective computing. Using manual feature extraction methods on EEG signals results in sub-optimal performance by the learning models. With the advancements in deep learning as a tool for automated feature engineering, in this work, a hybrid of manual and automatic feature extraction methods has been proposed. The asymmetry in different brain regions is captured in a 2D vector, termed the AsMap, from the differential entropy features of EEG signals. These AsMaps are then used to extract features automatically using a convolutional neural network model. The proposed feature extraction method has been compared with differential entropy and other feature extraction methods such as relative asymmetry, differential asymmetry and differential caudality. Experiments are conducted using the SJTU emotion EEG dataset and the DEAP dataset on different classification problems based on the number of classes. Results obtained indicate that the proposed method of feature extraction results in higher classification accuracy, outperforming the other feature extraction methods. The highest classification accuracy of 97.10% is achieved on a three-class classification problem using the SJTU emotion EEG dataset. Further, this work has also assessed the impact of window size on classification accuracy.

Emotion plays a vital role in understanding the affective state of mind of an individual. In recent years, emotion classification using electroencephalogram (EEG) has emerged as a key element of affective computing. Many researchers have prepared datasets, such as DEAP and SEED, containing EEG signals captured by the elicitation of emotion using audio–visual stimuli, and many studies have been conducted to classify emotions using these datasets. However, baseline power removal is still considered one of the trivial aspects of preprocessing in feature extraction. The most common technique that prevails is subtracting the baseline power from the trial EEG power. In this paper, a novel method called InvBase method is proposed for removing baseline power before extracting features that remain invariant irrespective of the subject. The features extracted from the baseline removed EEG data are then used for classification of two classes of emotion, i.e., valence and arousal. The proposed scheme is compared with subtractive and no-baseline-correction methods. In terms of classification accuracy, it outperforms the existing state-of-art methods in both valence and arousal classification. The InvBase method plus multilayer perceptron shows an improvement of 29% over the no-baseline-correction method and 15% over the subtractive method.

Deep learning has gained much popularity in solving challenging machine learning problems related to image, speech classification, etc. Research has been conducted to apply deep learning models in emotion classification based on physiological signals such as EEG. Most of the research works have based their model on the spatial aspects of the EEG. However, the emotion features in EEG are spread across the time domain during an emotional episode. Therefore, in this work, the emotion classification problem is modelled as a sequence classification problem. The power band frequency based features of every time segment of EEG sequences generated from 32-channel EEG data are used to train three different models of Long Short-Term Memory (LSTM1, LSTM2, and LSTM3). Four class (HVHA, HVLA, LVHA, and LVLA) classification experiments were performed based on the valence and arousal emotion models. The LSTM3 model with 128 memory cells achieved the highest classification accuracy of 90%, whereas LSTM1 (32 cells) and LSTM2 (64 cells) yielded classification accuracies of 85% and 89% respectively. Further, the impact of segment size on classification accuracy was also investigated in this work. Results obtained indicate that a smaller segment size leads to higher classification accuracy using LSTM models.