Research On Attention Evaluation System Based On Hybrid Brain-Computer Interface And Board Games

The current development of smart phones and Internet technologies have greatly facilitated people's daily life,but they have also brought many problems.The typical one is that people's attention to doing things decreases so as the efficiency.The prerequisite for improving attention is to accurately evaluate the level of attention.Many methods for evaluating attention,such as scale analysis and image recognition,have the disadvantages of being subjective and easy to disguise.Fields such as physiology and neuroscience have confirmed that attention is related to the characteristics of human brain electrical signals.In recent years,with the development of wearable devices and brain-computer interface technology,the use of portable brain-computer interface devices to assess human attention levels has become a hot topic in current research.Starting from chess games,this article selects a typical chess game—Chinese Chess as the research carrier.First,a related game and an integrated system of EEG and EOG detection and processing are built,and then around the EOG control and attention detection some methods are carried.The work content of this article is as follows:1)Designed and implemented an integrated system that integrates EEG acquisition and visualization,multi-mode chess games,EOG detection and control,attention detection and eval-uation.Among them,the EEG acquisition and visualization module can facilitate data storage,equipment error correction,etc.The chess game has three modes: man-machine battle,network battle,and endgame deduction.2)On the basis of EOG detection and recognition,this paper proposes and designs a novel method of chess operation based on EOG control.The operation method of EOG-control chess includes three steps: turn on the EOG switch,choose the chess piece,and choose the position to move.The setting of the EOG switch is to avoid the interference of long-term flashing of the buttons on people's thinking.In addition,the goal selection is achieved by blinking syn-chronously with the target to be selected.The stimulus paradigm of the EOG button flashes is adaptive to the situation of chess.The selected number of targets adaptively adjusts the flashing interval and mode(fixed or random).3)Design EEG experiments for attention tasks and conduct research on related EEG signal processing methods.First,a novel experimental paradigm that is closely integrated with chess games is designed,which is watching the endgame deduction in order to induce the EEG char-acteristics of the participants in the state of attention.In the processing of the training sample set,this paper first performs wavelet threshold denoising on the data,and then on the basis of empirical mode decomposition,first extracts the area of the ellipse,the average distance,and the central tendency for the second-order difference plot(SODP)of the main IMFs.Then the area of the ellipse area of the complex plane trajectory of the analytical signal and the marginal spectrum energy are extracted from the Hilbert spectrum which formed another set of features.The experimental results show that the feature extraction method based on Hilbert spectrum analysis has better results.The classification accuracy rate of SVM reaches 85.82%,which is better than traditional power spectrum analysis(81.69%)and SODP-based feature extraction method(83.21%).The effect of online control experiments with 6 subjects further confirmed the effectiveness and practicability of this attention evaluation system.