Abstract:Traditional brain-computer systems are complex and expensive, and emotion classification algorithms lack repre-sentations of the intrinsic relationships between different channels of electroencephalogram (EEG) signals. There is still room for improvement in accuracy. To lower the research barrier for EEG and harness the rich information embedded in multi-channel EEG, we propose and implement a simple and user-friendly brain-computer system for classifying four emotions: happiness, sorrow, sadness, and tranquility. This system utilizes the fusion of convolutional attention mechanisms and fully pre-activated residual blocks, termed Attention-Convolution-based Pre-Activated Residual Network (ACPA-ResNet).In the hardware acquisition and preprocessing phase, we employ the ADS1299 integrated chip as the analog front-end and utilize the ESP32 microcontroller for initial EEG signal processing. Data is wirelessly transmitted to a PC through UDP protocol for further preprocessing. In the emotion analysis phase, ACPA-ResNet is designed to automatically extract and learn features from EEG signals, thereby enabling accurate classification of emotional states by learning time-frequency domain characteristics. ACPA-ResNet introduces an attention mechanism on the foundation of residual networks, adaptively assigning different weights to each channel. This allows it to focus on more meaningful EEG signals in both spatial and channel dimensions while avoiding the problems of gradient dispersion and explosion associated with deep network <a class="link-external link-http" href="http://architectures.Through" rel="external noopener nofollow">this http URL</a> testing on 16 subjects, our system demonstrates stable EEG signal acquisition and transmission. The novel network significantly enhances emotion recognition accuracy, achieving an average emotion classification accuracy of 95.1%.

Emotion Classification Using EEG Brain Signals and the Broad Learning System

Using electroencephalogram for emotion recognition based on filter-bank long short-term memory networks

Emotion Classification from Short-term EEG Signals in Deep Learning

Deep Learning-Based Approach for Emotion Recognition Using Electroencephalography (EEG) Signals Using Bi-Directional Long Short-Term Memory (Bi-LSTM)

EEG-Based Emotion Classification Using a Deep Neural Network and Sparse Autoencoder.

Enhancing BCI-Based Emotion Recognition Using an Improved Particle Swarm Optimization for Feature Selection

EEG Emotion Classification and Correlation Research Based on Deep Learning

An innovative EEG-based emotion recognition using a single channel-specific feature from the brain rhythm code method

Emotion recognition using effective connectivity and pre-trained convolutional neural networks in EEG signals

A Model for EEG-Based Emotion Recognition: CNN-Bi-LSTM with Attention Mechanism

Investigating of Deaf Emotion Cognition Pattern by EEG and Facial Expression Combination

Design and Implementation of an Emotion Analysis System Based on EEG Signals

Exploring Deep Learning Features for Automatic Classification of Human Emotion Using EEG Rhythms

Analysis of brain areas in emotion recognition from EEG signals with deep learning methods

An attention-based hybrid deep learning model for EEG emotion recognition

An adaptive search optimizer-based deep Bi-LSTM for emotion recognition using electroencephalogram signal

4D attention-based neural network for EEG emotion recognition

Simultaneously exploring multi-scale and asymmetric EEG features for emotion recognition

Role of machine learning and deep learning techniques in EEG-based BCI emotion recognition system: a review

Emotion recognition with convolutional neural network and EEG-based EFDMs

Research on Emotion Recognition of EEG Signal Based on Convolutional Neural Networks and High-Order Cross-Analysis