CN-116649988-B - Brain wave information processing method, system, equipment and medium

CN116649988BCN 116649988 BCN116649988 BCN 116649988BCN-116649988-B

Abstract

The embodiment of the invention discloses a brain wave information processing method, a system, equipment and a medium, wherein the brain wave information processing method comprises the steps of recording and collecting brain wave signals under different stimulation conditions, preprocessing the brain wave signals, analyzing the preprocessed brain wave signals in a time domain, a frequency spectrum and a time-frequency domain power spectrum by a signal analysis method, extracting and marking characteristic signals, classifying and identifying advanced information contained in the brain wave signals by a trained classifier model, carrying out sensory enhancement, motor imagery learning and emotion evolution regulation and control by taking the classified signals as input and using a reinforcement learning algorithm to form inverse brain wave regulation and control signals, carrying out feedback excitation on a brain to be tested by using the inverse brain wave regulation and control signals, storing the brain wave signals containing sensory information, motor imagery and emotion information, the inverse brain wave regulation signals and the brain wave signals after feedback excitation, and establishing a subject advanced brain wave information database.

Inventors

CHANG YU
CAI MINGCHUN
WANG WANYI
ZHANG SHENGYAO
JIANG YUGUANG
ZHANG YUNCHI

Assignees

北京电子工程总体研究所

Dates

Publication Date: 20260508
Application Date: 20230512

Claims (9)

1. The brain wave information processing method is characterized by comprising the following steps of Recording and collecting the changes of brain electrical and brain magnetic signals at different brain regions under different stimulation conditions, obtaining brain wave signals and preprocessing the brain wave signals; analyzing the time domain, the frequency spectrum and the time-frequency domain power spectrum of the preprocessed brain wave signals by using a signal analysis method, and extracting and marking characteristic signals; training a classifier model by using the marked characteristic signals, and classifying and identifying sensory information, motor imagery information and emotion information contained in brain wave signals by using the trained classifier model; taking the classified signals as input, and carrying out sensory enhancement, motor imagery learning and emotion evolution regulation and control by utilizing a reinforcement learning algorithm in the twin brain to form inverse brain wave regulation and control signals; the inverse brain wave regulation signals are utilized to perform feedback excitation on the brain to be tested, and the potential capability of the brain to be tested in the aspects of sense organ, motor imagery and emotion regulation is stimulated; Storing brain wave signals containing sensory information, motor imagery and emotion information, inverse brain wave regulation signals and brain wave signals after feedback excitation, and establishing a subject advanced brain wave information database; The classifying signals are used as input, the strengthening learning algorithm in the twin brain is utilized to carry out sensory enhancement, motor imagery learning and emotion evolution regulation and control, and the forming of the inverse brain wave regulation and control signals comprises the following steps: Using a thinking control mechanism and emotion nerve regulation and control output by a neural loop model in the twin brain as sense, motor imagery and emotion intelligent bodies by using a reinforcement learning algorithm, and interacting with a specific simulation scene and a conventional simulation scene to perform sense enhancement and motor imagery learning training; And under the supervision and unsupervised conditions, the method respectively performs self-learning and self-evolution to realize emotion evolution regulation and control, and forms inverse brain wave regulation and control signals through sensory, transportation imagination and emotion regulation and control stimulation.
2. The brain wave information processing method according to claim 1, wherein, The classifying and identifying the sensory information, the motor imagery information and the emotion information contained in the brain wave signals by using the trained classifier model comprises the following steps: classifying the sensory information, the categories including visual, auditory, olfactory, and tactile; Classifying the motor imagery information, wherein the classification comprises a motor instruction and muscle control; The emotional information is classified by a discrete emotional model or a continuous emotional model, and the categories include anger, fear, expect, sadness, aversion, surprise, acceptance, and happiness.
3. The brain wave information processing method according to claim 1, wherein, The method further comprises forming an effective external stimulus based on smell, taste, vision, hearing and touch; recording and collecting the changes of brain wave signals of different brain region positions under different stimulation conditions, obtaining brain wave signals and preprocessing the brain wave signals comprises the following steps: recording and collecting the changes of the brain electrical signals and the brain magnetic signals at different brain regions of a subject under the condition of stimulation of different effective external stimulus sources by using non-implantable brain electromagnetic signal collecting equipment, and carrying out noise reduction, filtering, artifact removal, compressing and silencing instant signals and amplifying special waveforms on the brain electrical signals and the brain magnetic signals at different brain regions to obtain preprocessed brain wave signals.
4. An electroencephalogram information processing system, characterized by comprising A brain wave stimulation module configured to generate an external stimulus; The brain wave signal acquisition module is configured to acquire brain electrical and brain magnetic signals at different brain region positions under different stimulation conditions to obtain brain wave signals; the characteristic extraction and selection module is configured to perform time domain, frequency spectrum and time-frequency domain power spectrum analysis on the brain wave signals by using a signal analysis method, and extract and mark characteristic signals; The artificial intelligent classifier model module is configured to train a classifier model based on the marked characteristic signals, and classify and identify sensory information, motor imagery information and emotion information contained in the brain wave signals by utilizing the trained classifier model; The twin brain simulation model is configured to carry out sensory enhancement, motor imagery learning and emotion evolution regulation and control by utilizing a reinforcement learning algorithm to form an inverse brain wave regulation and control signal, and comprises a thinking control mechanism and emotion nerve regulation and control output by a nerve loop model in the twin brain are used as sensory, motor imagery and emotion intelligent bodies by utilizing the reinforcement learning algorithm, and interact with a specific simulation scene and a conventional simulation scene to carry out sensory enhancement and motor imagery learning training; The brain wave signal feedback module is configured to utilize the inverse brain wave regulation and control signal to perform feedback excitation on the brain to be tested, and excite potential abilities of the brain to be tested in sense organs, motor imagination and emotion regulation and control; The brain wave storage module is configured to store brain wave signals containing sensory information, motor imagery and emotion information output by a tested brain, inverse brain wave regulation signals output by a twin brain simulation model and brain wave signals after feedback excitation output by the brain wave signal feedback module, and a brain wave information database of the tested brain is established.
5. The brain wave information processing system according to claim 4, wherein, The external stimulus is an effective external stimulus including smell, taste, vision, hearing and touch; The brain wave signal acquisition module comprises a wearable voltage and electromagnetic array acquisition sensor.
6. The brain wave information processing system according to claim 4, further comprising, And the brain wave signal preprocessing module is configured to perform noise reduction, filtering, artifact removal, silence instantaneous signal compression and special waveform amplification processing on the time waveform signals, and sends the processed brain wave signals to the feature extraction and selection module.
7. The brain wave information processing system according to claim 4, wherein, The classifier model includes at least one of decision trees, multiple logistic regression, random forests, SVMs, and deep learning.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of claims 1-3 when the program is executed by the processor.
9. A computer readable storage medium, on which a computer program is stored, which program, when being executed by a processor, implements the method according to any of claims 1-3.

Description

Brain wave information processing method, system, equipment and medium Technical Field The present invention relates to the field of computers. And more particularly, to a brain wave information processing method, system, device and medium. Background An electroencephalogram signal (EEG) is mainly formed by summing up postsynaptic potentials synchronously occurring in a large number of neurons in the cortex, and records the change of electric waves when brain activities are recorded, and is an overall reflection of the electrophysiological activities of brain nerve cells on the surface of the cerebral cortex or scalp. The traditional electroencephalogram signal analysis mostly utilizes visual inspection to distinguish artifacts according to experience by an expert, and gives evaluation according to amplitude, frequency, transient distribution and the like of an electroencephalogram signal waveform. This allows the reading and analysis of the electroencephalogram information to stay on a subjective level at all times. In fact, the brain electrical signals simultaneously contain advanced brain wave information such as human physiological senses, motor imagination, emotion, thinking and the like, and the research on the identification and analysis method of the advanced brain wave information has supporting effect on development of human brain and development of brain-computer interface technology. Research into senses, motor imagery, emotions is a interdisciplinary field that involves computer science, psychology, cognition and neurology. Compared with the facial recognition and the motion for carrying out the recognition of sense, motor imagery and emotion, the brain wave signal based on the analysis can eliminate the interference of artificial camouflage or change and obtain higher recognition accuracy. Although the acquisition of the electroencephalogram signals is improved, the method is a difficult problem in current research, and is faced with the fact that various noise, low signal-to-noise ratio and asymmetric unstable electroencephalogram signals are mixed, and feature extraction and selection are effectively carried out, so that sensory, motor imagery and emotion classification recognition are carried out. Currently, by analyzing human brain electrical signals (EEG) by means of artificial intelligence technology, a conclusion is drawn based on brain activity of a population from an artificial intelligence analysis viewpoint, but no effective method has been found for classification recognition of advanced brain wave information such as sense, motor imagery, emotion, etc. Disclosure of Invention The invention aims to provide a brain wave information processing method, a brain wave information processing system, brain wave information processing equipment and a brain wave information processing medium, so as to solve at least one of the problems of the related technology. In order to achieve the above purpose, the invention adopts the following technical scheme: the first aspect of the present invention provides a brain wave information processing method, comprising Recording and collecting the changes of brain electrical and brain magnetic signals at different brain regions under different stimulation conditions, obtaining brain wave signals and preprocessing the brain wave signals; analyzing the time domain, the frequency spectrum and the time-frequency domain power spectrum of the preprocessed brain wave signals by using a signal analysis method, and extracting and marking characteristic signals; training a classifier model by using the marked characteristic signals, and classifying and identifying sensory information, motor imagery information and emotion information contained in brain wave signals by using the trained classifier model; taking the classified signals as input, and carrying out sensory enhancement, motor imagery learning and emotion evolution regulation and control by utilizing a reinforcement learning algorithm in the twin brain to form inverse brain wave regulation and control signals; the inverse brain wave regulation signals are utilized to perform feedback excitation on the brain to be tested, and the potential capability of the brain to be tested in the aspects of sense organ, motor imagery and emotion regulation is stimulated; Storing brain wave signals containing sensory information, motor imagery and emotion information, inverse brain wave regulation signals and brain wave signals after feedback excitation, and establishing a subject advanced brain wave information database. Optionally, the classifying and identifying the sensory information, the motor imagery information and the emotion information contained in the brain wave signal by using the trained classifier model includes: classifying the sensory information, the categories including visual, auditory, olfactory, and tactile; Classifying the motor imagery information, wherein the classification comprises a motor inst