Brain-controlled spelling method and system based on visual and auditory induction and stable decoding

Abstract

The invention relates to a brain-controlled spelling method and system based on visual and auditory induction and stable decoding, and the method comprises the steps: carrying out the coding of a target character through employing a two-stage code division multiple access coding mode based on a visual and auditory collaborative induction experiment normal form, and generating visual and auditory stimulation; sending visual and auditory stimulation to the user, and inducing the user to generate a target signal for the to-be-spelled target character; collecting an electroencephalogram signal generated by the testee; and preprocessing the electroencephalogram signals, selecting electroencephalogram signal fragments with set lengths, inputting the electroencephalogram signal fragments into the domain adaptive fusion discriminant analysis model, decoding the target signals to obtain target characters, and outputting the target characters. A to-be-spelled target character is coded by adopting a coding mode of visual and auditory collaborative induction and two-stage code division multiple access, so that the intensity of an induced specific electroencephalogram signal is enhanced; according to the method, the domain-adaptive fusion discriminant analysis recognition model is constructed, and the target characters are decoded and output, so that the character decoding output rate is improved, and brain-controlled character spelling is more stable and efficient.

Description

technical field [0001] The invention relates to the technical fields of human-computer interaction and biomedical engineering information services, in particular to a brain-controlled spelling method and system based on audio-visual induction and stable decoding. Background technique [0002] Relying on non-invasive EEG signals to establish direct information interaction and control channels will greatly improve the interactive experience of people with speech or motor dysfunction and some specific limited applications. It can be widely used in medical rehabilitation, game entertainment, smart home and military fields. . [0003] As the main sensory channel for human beings to obtain information, audio-visual senses account for the largest proportion of information acquired by various senses. Currently, the audio-visual-evoked specific EEG components used for brain-controlled spelling mainly include event-related potentials (ERPs), steady-state Visual evoked potential (SSVE...

AI Technical Summary

Problems solved by technology

[0003] As the main sensory channel for human beings to obtain information, audio-visual senses account for the largest proportion of information acquired by various senses. At present, the audio-visual-evoked specific EEG components used for brain-controlled spelling mainly include event-related potentials (ERPs), steady-state Visual evoked potential (SSVEP) and motor visual evoked potential (mVEP), etc. Audiovisual evoked ERP cannot encode characters synchronously. Encoding a character needs to traverse the character matrix completely, which makes it take a long time to output a target; SSVEP encoding can Synchronous encoding of characters, the efficiency of outputting characters is high, but long-term use is easy to make users feel visual fatigue; mVEP can improve the above shortcomings to a certain extent, but compared with SSVEP, the target encoding output rate still needs to be improved

Images