WO2020186915A1 - 一种注意力检测方法及系统 - Google Patents

一种注意力检测方法及系统 Download PDF

Info

Publication number
WO2020186915A1
WO2020186915A1 PCT/CN2020/071565 CN2020071565W WO2020186915A1 WO 2020186915 A1 WO2020186915 A1 WO 2020186915A1 CN 2020071565 W CN2020071565 W CN 2020071565W WO 2020186915 A1 WO2020186915 A1 WO 2020186915A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
ear
user
bioelectric
impedance
Prior art date
Application number
PCT/CN2020/071565
Other languages
English (en)
French (fr)
Inventor
倪刚
杨晖
查钧
唐卫东
李皓
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP20773834.5A priority Critical patent/EP3932303A4/en
Publication of WO2020186915A1 publication Critical patent/WO2020186915A1/zh
Priority to US17/475,658 priority patent/US20220047198A1/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A61B5/372—Analysis of electroencephalograms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
    • A61B5/168—Evaluating attention deficit, hyperactivity
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state
    • A61B5/18—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state for vehicle drivers or machine operators
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/25—Bioelectric electrodes therefor
    • A61B5/279—Bioelectric electrodes therefor specially adapted for particular uses
    • A61B5/291—Bioelectric electrodes therefor specially adapted for particular uses for electroencephalography [EEG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6813—Specially adapted to be attached to a specific body part
    • A61B5/6814—Head
    • A61B5/6815—Ear
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6813—Specially adapted to be attached to a specific body part
    • A61B5/6814—Head
    • A61B5/6815—Ear
    • A61B5/6817—Ear canal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6843—Monitoring or controlling sensor contact pressure
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • A61B5/7278—Artificial waveform generation or derivation, e.g. synthesising signals from measured signals
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • A61B5/7285—Specific aspects of physiological measurement analysis for synchronising or triggering a physiological measurement or image acquisition with a physiological event or waveform, e.g. an ECG signal
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/70—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mental therapies, e.g. psychological therapy or autogenous training
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment

Definitions

  • This application relates to the field of data processing, and in particular to methods and systems for detecting driver attention during safe driving and assisted driving.
  • Inattentive driving includes any driving activity that distracts the driver’s attention, such as riding a car, eating and drinking, talking with passengers, adjusting entertainment or navigation systems, and making phone calls. It also includes the driver’s mental state or certain changes in consciousness, such as Tired and briefly approaching sleep, etc. Studies have shown that up to 30% of traffic accidents are caused by drivers' inattention. When the vehicle is running at a higher speed, if the distracted driver is not fully aware of the timely changes in the state including the path, traffic, obstacles and even the vehicle, an accident will inevitably occur.
  • the vehicle system can accurately and timely detect and detect the driver’s state.
  • the driver is inattentive, select the appropriate The timing of auxiliary reminders is of great importance to ensure safe driving.
  • the existing technology uses intelligent computer systems to determine the driver’s driving attention type based on the driver’s gaze point, line of sight, rest time, saccades, and the state of movement of surrounding objects along the driving path collected by the car system.
  • Such technologies usually require the installation of various sensors and driving computer systems in the automobile system, which is relatively expensive. Due to the complexity and diversity of the driving environment, there is a certain probability of deviation in the driver's driving attention state obtained through intelligent calculation, which affects safe driving.
  • the vehicle system can accurately determine the driver’s driving attention type by collecting the driver’s EEG signal to achieve the accurate and timely status of the driver. Detection and detection, to provide auxiliary reminders to the driver's driving behavior, provide another effective technical realization option for ensuring safe driving.
  • the embodiments of the present application provide an attention detection method and system, which can be applied to the driver's attention detection and obtain EEG signals through the ears, so that EEG signals during driving are more convenient and feasible, and the measurement is reduced. Cost, while ensuring the accuracy of EEG signal acquisition.
  • an embodiment of the present application provides a user attention detection method, the method includes: collecting a user's bioelectric signal from the user's ear by wearing a device on the ear; and obtaining the user's EEG from the user's bioelectric signal Signal; according to the user’s EEG signal based on a machine learning model to obtain the user’s attention type; wherein through the ear side wearing device, collecting user bioelectric signals from the user’s ear side specifically includes: the ear side wearing device includes a plurality of Ear signal measurement unit; determine whether the impedance between two ear signal measurement units is lower than a preset threshold; when the impedance between the two ear signal measurement units is lower than the preset threshold, from the two Each ear side signal measuring unit collects a bioelectric signal; the user's bioelectric signal is acquired according to a potential difference signal of the bioelectric signal electric signal collected by the two ear side signal measuring units.
  • the ear side wearing device is convenient to carry and is set on the ear side. It is not easy to fall off during the wearing process, which makes it more convenient and feasible to measure the user’s EEG signal during driving.
  • the difference processing method is to process the collected bioelectric signals of the user according to the characteristics of the bioelectric signals collected on the ear side. It can effectively remove the noise in the EEG signal, and at the same time consider the reason why the user may not wear it correctly, or one side In the case of equipment failure or poor signal reception, it is necessary to judge the collected bioelectric signals before performing potential difference processing to avoid signal collection and processing even when the device on the ear side cannot be received normally, resulting in inaccurate results.
  • the plurality of ear side signal measurement units include a left ear side signal measurement unit and a right ear side signal measurement unit; and the determination of whether the two ear side signal measurement units Whether the impedance of the two ears is lower than the preset threshold; when the impedance between the two ear signal measuring units is lower than the preset threshold, the bioelectric signals are collected from the two ear signal measuring units;
  • the potential difference signal of the bioelectric signal electric signal collected by the side signal measuring unit to obtain the user bioelectric signal is specifically:
  • the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold; when the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit The impedance is lower than a preset threshold; the user's bioelectric signal is acquired according to a potential difference signal between the bioelectric signal measured by the left ear signal measuring unit and the bioelectric signal measured by the right ear signal measuring unit.
  • the ear-side wearing device may obtain bioelectric signals from both sides of the left and right ears.
  • the user's bioelectric signals are obtained according to the bioelectric signals obtained from the left and right ears.
  • the ear side wearing device is a unilateral ear side wearing device
  • the multiple ear side signal measurement units include multiple unilateral ear side signal measurement units; and the determination of two Whether the impedance between the two ear side signal measurement units is lower than the preset threshold; when the impedance between the two ear side signal measurement units is lower than the preset threshold value, collect bioelectricity from the two ear side signal measurement units Signal; acquiring the user's bioelectric signal according to the potential difference signal of the bioelectric signal electric signal collected by the two ear side signal measuring units, specifically:
  • the ear-side wearing device may be a single-ear wearing device, and it is directly judged whether the single-ear wearing is normal based on the impedance between the two measurement units.
  • the impedance between the ear side signal measurement units is higher than a preset threshold; it is determined whether the impedance between two of the plurality of left ear side signal measurement units is lower than the preset threshold value, and the multiple right ear side Whether the impedance between two of the signal measurement units is lower than a preset threshold; the obtained signal is obtained according to the potential difference value of the bioelectric signal measured by the two bioelectric measuring devices in the ear canal whose impedance is lower than the preset threshold Describe the user's bioelectric signal.
  • the obtaining the user's bioelectric signal according to the potential difference signal of the bioelectric signal collected by the two ear signal measuring units is specifically:
  • the circuit obtains the potential difference signal of the bioelectric signal collected by the two ear side signal measuring units, and obtains the user's bioelectric signal from the potential difference signal.
  • the obtaining the user's attention type based on a machine learning model according to the user's EEG signal specifically includes: calculating a sample entropy value of the user's EEG signal, According to the value of the sample entropy, the user's attention type is analyzed based on the machine learning model.
  • the detecting the user’s attention type based on the user’s EEG signal specifically includes: intercepting the user’s EEG signal of a preset length of time, and obtaining data from the preset
  • the user’s EEG signal of the time length obtains N signal sampling points; the N signal sampling points are u(1), u(2),...,u(N); based on the N Signal sampling points, using u(1), u(2),..., u(N-m+1) as the starting point, respectively intercept m sampling points to construct N-m+1 m-dimensional vectors;
  • For each of the N-m+1 m-dimensional vectors calculate the average of the number of vectors whose distance between the m-dimensional vector and each other vector is less than r, and calculate the obtained N-m+
  • the first average value is obtained by the average value of one average value; based on the N signal sampling points, m+1 is sequentially intercepted with u(1), u(2),..., u(Nm) as the
  • the machine learning model is an SVM classifier; the SVM classifier is used to perform machine learning to obtain a segmentation value, and the user's attention is determined according to the segmentation value and the sample entropy value Types of.
  • the user's attention type can be obtained through the machine learning model according to the sample entropy value, and the sample entropy characteristics of the EEG signal under different attention types can be analyzed more accurately through machine learning, so as to be based on all
  • the sample entropy value of the collected EEG signals determines the type of user's attention on the wall.
  • an embodiment of the present invention provides a user attention detection system.
  • the system includes: an ear wearing device for collecting user bioelectric signals from the user's ear; The user’s brain electrical signal; an attention detection device for detecting the user’s attention type based on the user’s brain electrical signal; wherein the ear-worn device for collecting user bioelectric signals from the user’s ear specifically includes:
  • the ear side wearing device includes a plurality of ear side signal measurement units; the ear side wearing device determines whether the impedance between two ear side signal measurement units is lower than a preset threshold; when the two ear side signal measurement units are If the impedance between the units is lower than the preset threshold, the bioelectric signals are collected from the two ear side signal measuring units; the electric signal is obtained according to the potential difference signal of the bioelectric signal collected by the two ear side signal measuring units Describe the user's bioelectric signal.
  • the multiple ear side signal measurement units include a left ear side signal measurement unit and a right ear side signal measurement unit; the ear side wearing device determines two of the ear side signals Whether the impedance between the measurement units is lower than the preset threshold; when the impedance between the two ear signal measurement units is lower than the preset threshold, collect bioelectric signals from the two ear signal measurement units; The potential difference signal of the electrical signal of the bioelectric signal collected by the two ear signal measuring units obtains the bioelectric signal of the user, specifically: the ear wearing device determines that the left ear signal measuring unit and the right ear Whether the impedance between the ear side signal measurement units is lower than the preset threshold; when the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than the preset threshold; according to the left ear side A potential difference signal between the bioelectric signal measured by the signal measuring unit and the bioelectric signal measured by the right ear side signal measuring unit acquires the user's
  • the ear side wearing device is a unilateral ear side wearing device, the unilateral ear side wearing device includes a plurality of unilateral ear side signal measurement units; the ear side wearing device determines Wherein, whether the impedance between the two ear side signal measurement units is lower than a preset threshold; when the impedance between the two ear side signal measurement units is lower than the preset threshold value, collect data from the two ear side signal measurement units Bioelectric signal; acquiring the user bioelectric signal according to the potential difference signal of the bioelectric signal electric signal collected by the two ear side signal measuring units, specifically: the ear side wearing device judges two of the single Whether the impedance between the ear side signal measurement units is lower than the preset threshold; when the impedance between the two unilateral ear signal measurement units is lower than the preset threshold; according to the two unilateral ear signal measurement units The collected potential difference signal of the bioelectric signal obtains the user's bioelectric signal.
  • the attention detection device obtaining the user's attention type based on a machine learning model according to the user's EEG signal is specifically: the attention detection device calculates the The value of the sample entropy of the user's EEG signal, and the user's attention type is analyzed based on the machine learning model according to the value of the sample entropy.
  • an embodiment of the present invention provides an ear-side wearing device.
  • the device includes: a plurality of ear-side signal measurement units for collecting user bioelectric signals from the ear side; and a first determining unit for determining which Whether the impedance between the two ear signal measuring units is lower than the preset threshold; when the impedance between the two ear signal measuring units is lower than the preset threshold; the two ear signal measuring units measure The potential difference signal of the bioelectric signal collected by the bioelectric signal is used as the user's bioelectric signal; the feature decomposition unit is used to obtain the brain electric signal from the user's bioelectric signal; the attention detection unit is used to obtain the brain electric signal according to the brain The electrical signal is based on the machine learning model to obtain the user's attention type.
  • the multiple ear signal measurement units include a left ear signal measurement unit and a right ear signal measurement unit; the first disconnection unit is used to determine the left ear signal measurement unit; Whether the impedance between the ear signal measurement unit and the right ear signal measurement unit is lower than a preset threshold; when the impedance between the left ear signal measurement unit and the right ear signal measurement unit is lower than the preset Set a threshold; the user's bioelectric signal is acquired according to the potential difference signal between the bioelectric signal measured by the left ear signal measuring unit and the bioelectric signal measured by the right ear signal measuring unit.
  • the ear side wearing device is a unilateral ear side wearing device;
  • the multiple ear side signal measurement units include multiple unilateral ear side signal measurement units;
  • the first fault Unit for determining whether the impedance between the two ear signal measuring units is lower than a preset threshold; when the impedance between the two ear signal measuring units is lower than the preset threshold, the two ears
  • the potential difference signal of the bioelectric signal collected by the bioelectric signal measured by the side signal measuring unit is used as the user bioelectric signal, specifically: the first judgment unit is used to judge two of the unilateral ear signal measuring units Whether the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, and when the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, the bioelectric signals collected by the two unilateral ear signal measurement units are The potential difference signal as the user bioelectric signal.
  • the ear wearing device further includes a second judgment Unit; when the first judging unit judges that the impedance between one of the left ear side signal measurement units and one of the right ear side signal measurement units is higher than a preset threshold, the second judgment unit judges the multiple Whether the impedance between two of the left ear side signal measurement units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear side signal measurement units is lower than the preset threshold;
  • the potential difference signal of the bioelectric signal collected by the two bioelectric measuring devices in the ear canal on the side whose impedance is lower than the preset threshold is used as the user's bioelectric signal.
  • the attention detection unit obtaining the user's attention type based on a machine learning model according to the user's EEG signal is specifically: the attention detection unit calculates the The value of the sample entropy of the user's EEG signal, and the user's attention type is analyzed based on the machine learning model according to the value of the sample entropy.
  • an embodiment of the present invention provides an ear-side wearing device, characterized in that the device includes: a plurality of ear-side signal measurement units for collecting user bioelectric signals from the ear side; and a first judgment unit, It is used to determine whether the impedance between the two ear side signal measurement units is lower than the preset threshold; when the impedance between the two ear side signal measurement units is lower than the preset threshold; The potential difference signal of the bioelectric signal collected by the bioelectric signal measured by the measuring unit is used as the user bioelectric signal; the feature decomposition unit is used to obtain the brain electric signal from the user bioelectric signal; the sending unit is used to transfer all The EEG signal is sent to the signal analysis device.
  • the multiple ear signal measurement units include a left ear signal measurement unit and a right ear signal measurement unit; the first disconnect unit is used to determine the left ear signal measurement unit; Whether the impedance between the ear signal measurement unit and the right ear signal measurement unit is lower than a preset threshold; when the impedance between the left ear signal measurement unit and the right ear signal measurement unit is lower than the preset Set a threshold; the user's bioelectric signal is acquired according to the potential difference signal between the bioelectric signal measured by the left ear signal measuring unit and the bioelectric signal measured by the right ear signal measuring unit.
  • the ear side wearing device is a unilateral ear side wearing device;
  • the plurality of ear side signal measurement units includes a plurality of unilateral ear side signal measurement units;
  • the first fault Unit for determining whether the impedance between the two ear signal measuring units is lower than a preset threshold; when the impedance between the two ear signal measuring units is lower than the preset threshold, the two ears
  • the potential difference signal of the bioelectric signal collected by the bioelectric signal measured by the side signal measuring unit is used as the user bioelectric signal, specifically: the first judgment unit is used to judge two of the unilateral ear signal measuring units Whether the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, and when the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, the bioelectric signals collected by the two unilateral ear signal measurement units are The potential difference signal as the user bioelectric signal.
  • the ear wearing device further includes a second judgment Unit; when the first judging unit judges that the impedance between one of the left ear side signal measurement units and one of the right ear side signal measurement units is higher than a preset threshold, the second judgment unit judges the multiple Whether the impedance between two of the left ear side signal measurement units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear side signal measurement units is lower than the preset threshold;
  • the potential difference signal of the bioelectric signal collected by the two bioelectric measuring devices in the ear canal on the side whose impedance is lower than the preset threshold is used as the user's bioelectric signal.
  • an embodiment of the present invention provides an attention detection device.
  • the device includes: a receiving unit, configured to receive a user's EEG signal from the ear-worn device; The signal is based on a machine learning model to obtain the user's attention type.
  • the attention detection unit is specifically configured to calculate the value of sample entropy of the user’s brain electrical signal, and analyze the user’s behavior based on the machine learning model according to the value of sample entropy. Attention type.
  • the attention detection unit is specifically configured to intercept the user's EEG signal of a preset time length, and obtain data from the user's EEG signal of the preset time length
  • the signal obtains N signal sampling points; the N signal sampling points are u(1), u(2),..., u(N); based on the N signal sampling points, each is set as u(1) ), u(2),..., u(N-m+1) is the starting point and intercepts m sampling points sequentially to construct N-m+1 m-dimensional vectors; for the N-m+1 m-dimensional vectors For each m-dimensional vector in the vector, calculate the average value of the number of vectors whose distance between the m-dimensional vector and each other vector is less than r, and calculate the average value of the obtained N-m+1 average values To the first average value; based on the N signal sampling points, taking u(1), u(2),..., u(Nm) as the starting points, respectively intercept m+1 sampling points to construct Nm
  • the machine learning model is an SVM classifier; the SVM classifier is used to perform machine learning to obtain a segmentation value; the attention detection unit is based on the segmentation value and the sample entropy The value judges the user's attention type.
  • an embodiment of the present invention provides an ear-side wearing device.
  • the device includes: a plurality of ear-side signal measurement units for collecting user bioelectric signals from the ear-side; a processor for determining two of them Whether the impedance between the ear signal measuring units is lower than the preset threshold; when the impedance between the two ear signal measuring units is lower than the preset threshold; the bioelectricity measured by the two ear signal measuring units is The potential difference signal of the bioelectric signal collected by the signal is used as the user bioelectric signal; the feature decomposition unit is used to obtain the brain electrical signal from the user’s bioelectric signal; the attention detection unit is used to The signal is based on a machine learning model to obtain the user's attention type.
  • the multiple ear signal measurement units include a left ear signal measurement unit and a right ear signal measurement unit; the processor is configured to determine the left ear Whether the impedance between the signal measurement unit and the right ear signal measurement unit is lower than a preset threshold; when the impedance between the left ear signal measurement unit and the right ear signal measurement unit is lower than the preset threshold Acquire the user's bioelectric signal according to the potential difference signal between the bioelectric signal measured by the left ear signal measuring unit and the bioelectric signal measured by the right ear signal measuring unit.
  • the ear side wearing device is a unilateral ear side wearing device; the multiple ear side signal measurement units include multiple unilateral ear side signal measurement units; the processor, It is used to determine whether the impedance between two ear side signal measurement units is lower than a preset threshold; when the impedance between the two ear side signal measurement units is lower than the preset threshold value, the two ear side signal measurement units are The potential difference signal of the bioelectric signal collected by the bioelectric signal measured by the measuring unit is used as the user bioelectric signal, specifically: the processor is used to determine whether the impedance between the two unilateral ear signal measuring units is Lower than the preset threshold, when the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, the potential difference signal of the bioelectric signal collected by the two unilateral ear signal measurement units As the user's bioelectric signal.
  • the processor is further configured to: The first judging unit judges that the impedance between one of the left ear signal measurement units and one of the right ear signal measurement units is higher than a preset threshold, and the processor separately judges the multiple left ear signal measurements Whether the impedance between two of the units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear signal measuring units is lower than the preset threshold; and the impedance is lower than the The potential difference signal of the bioelectric signal collected by the two bioelectric measuring devices in the ear canal on one side of the preset threshold is used as the user's bioelectric signal.
  • the attention detection unit obtaining the user's attention type based on a machine learning model according to the user's EEG signal is specifically: the attention detection unit calculates the The value of the sample entropy of the user's EEG signal, and the user's attention type is analyzed based on the machine learning model according to the value of the sample entropy.
  • an embodiment of the present invention provides an ear-side wearing device, characterized in that the device includes: a plurality of ear-side signal measurement units for collecting user bioelectric signals from the ear side; and a processor for Determine whether the impedance between the two ear side signal measurement units is lower than the preset threshold; when the impedance between the two ear side signal measurement units is lower than the preset threshold; change the two ear side signal measurement units
  • the potential difference signal of the bioelectric signal collected by the measured bioelectric signal is used as the user bioelectric signal; the feature decomposition unit is used to obtain the brain electric signal from the user bioelectric signal; the sending unit is used to transfer the brain
  • the electrical signal is sent to the signal analysis device.
  • the multiple ear signal measurement units include a left ear signal measurement unit and a right ear signal measurement unit; the processor is configured to determine the left ear Whether the impedance between the signal measurement unit and the right ear signal measurement unit is lower than a preset threshold; when the impedance between the left ear signal measurement unit and the right ear signal measurement unit is lower than the preset threshold Acquire the user's bioelectric signal according to the potential difference signal between the bioelectric signal measured by the left ear signal measuring unit and the bioelectric signal measured by the right ear signal measuring unit.
  • the ear side wearing device is a unilateral ear side wearing device;
  • the multiple ear side signal measurement units include multiple unilateral ear side signal measurement units;
  • the processor Determine whether the impedance between the two ear side signal measurement units is lower than the preset threshold; when the impedance between the two ear side signal measurement units is lower than the preset threshold, set the two ear side signal measurement units
  • the measured potential difference signal of the bioelectric signal collected by the bioelectric signal is used as the user bioelectric signal, specifically: the processor determines whether the impedance between the two unilateral ear signal measuring units is lower than the predetermined one. Set a threshold, and when the impedance between the two unilateral ear signal measuring units is lower than a preset threshold, use the potential difference signal of the bioelectric signal collected by the two unilateral ear signal measuring units as the User bioelectric signal.
  • the processor is further configured to serve as the A judgment unit judges that the impedance between one of the left ear signal measurement units and one of the right ear signal measurement units is higher than a preset threshold, and the processor judges the plurality of left ear signal measurement units respectively Whether the impedance between two of the two is lower than a preset threshold, and whether the impedance between two of the plurality of right ear signal measuring units is lower than a preset threshold; and the impedance is lower than the preset threshold;
  • the potential difference signal of the bioelectric signal collected by the two bioelectric measuring devices in the ear canal on one side of the threshold is set as the user's bioelectric signal.
  • an embodiment of the present invention provides an attention detection device.
  • the device includes: a receiving unit, configured to receive a user's EEG signal from an ear-worn device; and a processor, configured based on the user’s EEG signal
  • the machine learning model obtains the user's attention type.
  • the processor is specifically configured to calculate a sample entropy value of the user’s brain electrical signal, and analyze the user’s attention based on a machine learning model according to the sample entropy value Types of.
  • the processor is specifically configured to intercept the user EEG signal of a preset time length, and obtain N from the user EEG signal of the preset time length Signal sampling points; the N signal sampling points are u(1), u(2),..., u(N); based on the N signal sampling points, u(1), u (2),...,u(N-m+1) is the starting point and intercepts m sampling points in turn to construct N-m+1 m-dimensional vectors; for the N-m+1 m-dimensional vectors For each m-dimensional vector, calculate the average value of the number of vectors whose distance between the m-dimensional vector and each other vector is less than r, and calculate the average value of the N-m+1 average values to obtain the first Average value; based on the N signal sampling points, taking u(1), u(2),..., u(Nm) as the starting points and intercepting m+1 sampling points sequentially to construct Nm m+1 dimensions Vector; for each m+1 dimensional vector in the Nm
  • the machine learning model is an SVM classifier; the SVM classifier is used to perform machine learning to obtain a segmentation value; the attention detection unit is based on the segmentation value and the sample entropy The value judges the user's attention type.
  • an embodiment of the present invention provides an EEG signal detection method.
  • the method includes a user's bioelectric signal collected from the ear by a plurality of ear signal measurement units; Whether the impedance is lower than the preset threshold; when the impedance between the two ear signal measuring units is lower than the preset threshold; the bioelectric signal collected from the bioelectric signals measured by the two ear signal measuring units The potential difference signal is used as the user's bioelectric signal; the brain electric signal is obtained from the user's bioelectric signal; the brain electric signal is sent to the signal analysis device.
  • the multiple ear signal measurement units include a left ear signal measurement unit and a right ear signal measurement unit; the determining step is specifically: determining the left ear signal measurement unit Whether the impedance between the signal measurement unit and the right ear signal measurement unit is lower than a preset threshold; when the impedance between the left ear signal measurement unit and the right ear signal measurement unit is lower than the preset threshold Acquire the user's bioelectric signal according to the potential difference signal between the bioelectric signal measured by the left ear signal measuring unit and the bioelectric signal measured by the right ear signal measuring unit.
  • the ear side wearing device is a unilateral ear side wearing device;
  • the multiple ear side signal measurement units include multiple unilateral ear side signal measurement units;
  • the determination step is specifically To determine whether the impedance between the two unilateral ear signal measurement units is lower than a preset threshold, and when the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, the The potential difference signal of the bioelectric signal collected by the two unilateral ear signal measuring units is used as the user's bioelectric signal.
  • the potential difference signal of the bioelectric signal collected by the measuring device is used as the user's bioelectric signal.
  • an embodiment of the present invention provides an attention detection method, the method includes: receiving a user's EEG signal from an ear-worn device; obtaining the user's attention based on a machine learning model according to the user's EEG signal Types of.
  • obtaining the user’s attention type based on the machine learning model according to the user’s EEG signal is specifically calculating the value of the sample entropy of the user’s EEG signal, according to the The value of the sample entropy is based on the machine learning model to analyze the user's attention type.
  • calculating the value of the sample entropy of the user’s EEG signal specifically includes intercepting the user’s EEG signal of a preset time length, and obtaining the value of The user’s EEG signal obtains N signal sampling points; the N signal sampling points are u(1), u(2),..., u(N); based on the N signal sampling points, respectively Taking u(1), u(2),..., u(N-m+1) as the starting point, m sampling points are successively intercepted to construct N-m+1 m-dimensional vectors; for the N-m+ For each m-dimensional vector in one m-dimensional vector, calculate the average of the number of vectors whose distance between the m-dimensional vector and each other vector is less than r, and calculate the obtained N-m+1 averages The average value of the values obtains the first average value; based on the N signal sampling points, take u(1), u(2),..., u(Nm) as the starting points and intercept
  • the machine learning model is an SVM classifier; the SVM classifier is used for machine learning to obtain a segmentation value; the attention detection unit is based on the segmentation value and the sample entropy The value judges the user's attention type.
  • the ear-side wearing device is an earplug or earphone.
  • the user's attention type may specifically be that the user's attention state is concentrated or distracted.
  • the determining whether the impedance between the two ear side signal measurement units is lower than a preset threshold may be selected from multiple ear side signal measurement units based on a preset setting Two, it can also be based on the priority setting order to select two ear side signal measurement units for comparison. In the case that the impedance is always lower than the preset threshold, you can compare the preset times to terminate the comparison, or terminate after traversing all conditions Compare.
  • the determining whether the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold may be the left ear side signal
  • the measurement unit and the right ear side signal measurement unit are both one, and they are directly compared; it is also possible that both the left ear side signal measurement unit and the right ear side signal measurement unit are multiple.
  • Two ear signal measurement units and two right ear signal measurement units are selected respectively, or two ear signal measurement units are selected for comparison based on the priority setting sequence, and the impedance is always lower than the preset
  • the comparison can be terminated by a preset number of comparisons, or the comparison can be terminated after traversing all situations.
  • the judging whether the impedance between the multiple unilateral ear signal measurement units is lower than a preset threshold may be two unilateral ear signal measurement units and directly compare them; It can also be based on a preset setting to select two from multiple single-sided ear signal measurement units, or it can be based on the priority setting sequence to select two ear side signal measurement units for comparison, and the impedance is always lower than the preset
  • the threshold is set, the comparison can be terminated by the preset number of comparisons, or the comparison can be terminated after traversing all situations.
  • the multiple left ears are determined separately Whether the impedance between two of the side signal measurement units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear signal measurement units is lower than the preset threshold, it may be left, There are two ear-side signal measurement units for the right ear, which are directly compared; it can also be based on preset settings to select two from multiple left and right ear-side signal measurement units; it can also be for the left ear.
  • the side signal measurement unit selects two left ear ear side signal measurement units based on the priority setting order for comparison.
  • the impedance is always lower than the preset threshold, it can compare the preset times to terminate the comparison, or after traversing all situations Termination of comparison; for the right ear side signal measurement unit, two right ear side signal measurement units are selected for comparison based on the priority setting sequence.
  • the impedance is always lower than the preset threshold value, the preset number of comparisons can be terminated. Compare, or terminate the comparison after traversing all cases.
  • the implementation of the technical solutions of the embodiments of the present application can solve the problem of inconvenience of current technical attention judgment and the easy inaccuracy of the judgment result of attention in a moving state.
  • the driver’s EEG signals are collected through the ears, so that the user The acquisition of EEG signals is more convenient and feasible.
  • this solution can judge whether the current wearing is normal, and ensure that the signal acquisition and subsequent analysis are performed under the condition of normal acquisition to ensure the accuracy of the detection results Sex.
  • the EEG signals collected by the left and right ear canals can be processed by potential difference processing to ensure the accuracy of the collected EEG signals.
  • the sample entropy of the collected EEG signals is calculated to obtain the EEG signal in the time domain. Consistency state, and the attention is judged through the SVM classification algorithm, which can relatively accurately determine the driver’s current driving attention type, which can be used to accurately give follow-up operations during driving, such as reminding the driver Or take corresponding emergency actions.
  • FIG. 1 shows a schematic diagram of an application scenario in an embodiment of the present application
  • Figure 2a shows a schematic flowchart of a method for user attention detection provided by an embodiment of the present application
  • FIG. 2b shows a schematic flow chart of detecting whether the ear wearing device is normally worn during the user's brain electrical signal acquisition process according to an embodiment of the present application
  • FIG. 2c shows a schematic flow chart of detecting whether the unilateral wearing device is normally worn during the process of acquiring the user's brain electrical signals according to an embodiment of the present application
  • Figure 3 shows a schematic diagram of the brain waveforms of ⁇ , ⁇ , ⁇ , ⁇ , and ⁇ produced by the brain
  • FIG. 4 shows a schematic flowchart of a method for detecting user attention provided by an embodiment of the present application
  • FIG. 5 shows an implementation manner of a differential circuit in a method for acquiring a user's brain electrical signal provided by an embodiment of the present application
  • FIG. 6 shows a schematic diagram of differential processing of left and right ear EEG signals in an attention detection method provided by an embodiment of the present application
  • FIG. 7 shows a schematic diagram of differential processing of left and right ear EEG signals in an attention detection method provided by an embodiment of the present application
  • Figure 8a shows the myoelectric artifacts caused by the action of the neck joint
  • Figure 8b shows electrooculogram artifacts produced by blinking
  • Figure 9a shows a schematic diagram of the principle of SVM classification
  • Figure 9b shows a schematic diagram of the principle of SVM classification
  • FIG. 10 shows a schematic structural diagram of an attention detection system provided by an embodiment of the present application.
  • FIG. 11a shows a schematic structural diagram of an ear wearing device provided by an embodiment of the present application.
  • FIG. 11b shows a schematic structural diagram of another ear wearing device provided by an embodiment of the present application.
  • FIG. 11c shows a schematic structural diagram of another ear wearing device provided by an embodiment of the present application.
  • FIG. 11d shows a schematic structural diagram of another ear wearing device provided by an embodiment of the present application.
  • FIG. 12 shows a schematic diagram of a specific implementation form of an ear wearing device according to an embodiment of the present application
  • FIG. 13 shows a schematic diagram of a wearing position of an ear wearing device according to an embodiment of the present application
  • FIG. 14 shows a schematic structural diagram of an attention detection device provided by an embodiment of the present application.
  • FIG. 15a shows a schematic structural diagram of an ear wearing device provided by an embodiment of the present application.
  • Figure 15b shows a schematic structural diagram of an attention analysis device provided by an embodiment of the present application.
  • FIG. 16 shows a flowchart of a method for measuring user-related signals provided by an embodiment of the present application
  • FIG. 17 shows a flowchart of an attention detection method provided by an embodiment of the present application.
  • the one or more structural composition of the functions, modules, features, units, etc. mentioned in the specific embodiments of the present application can be understood as consisting of any physical or tangible components (for example, software and hardware running on a computer device). (For example, logic functions implemented by a processor or a chip), etc., and/or any other combination) are implemented in any manner.
  • the illustrated division of various devices into different modules or units in the drawings may reflect the use of corresponding different physical and tangible components in actual implementation.
  • a single module in the drawings of the embodiments of the present application may also be implemented by multiple actual physical components.
  • any two or more modules depicted in the drawings may also reflect different functions performed by a single actual physical component.
  • the embodiments of this application are mainly used for the detection of user attention, and can be specifically applied to the detection of the driver’s attention during driving, to determine whether the driver’s attention is concentrated, so as to provide immediate prompts based on the judgment result. In addition, It is applied to other scenes that need to detect the user's attention.
  • FIG. 1 is a typical application scenario of an embodiment of the present invention.
  • the ear-side wearing device 101 (specifically, earphones or earplugs) is worn on the user's ear, and the driver's bioelectric signal is collected from the ear, and the driver's bioelectric signal It is sent to the user's attention detection device 102, wherein the specific operation of the ear-side wearing device 101 may also include collecting ear-side bioelectric signals through the ear-side signal measuring unit, and obtaining the potential difference of the bioelectric signals collected by the measuring unit, for
  • the enhanced signal also eliminates the interference of external noise signals; performs de-artifact processing, and filters non-EEG frequency signals through a filter circuit (such as filtering out waveforms greater than 32Hz), and uses wavelet analysis to extract waveform features and perform subsequent digital encoding.
  • a filter circuit such as filtering out waveforms greater than 32Hz
  • the attention detection device 102 (specifically, a handheld terminal, such as a mobile phone, a PDA, a pad, etc., or a vehicle-mounted terminal device) analyzes the user's brain electrical signals. When it is judged that the attention is distracted, the corresponding follow-up operations are carried out, such as promptly reminding the driver through an alarm device to ensure driving safety.
  • the attention analysis method can be to calculate the sample entropy of the EEG signal and use the SVN algorithm to analyze the sample entropy. Classify and determine the state of attention.
  • the ear side wearing device 101 will also make a pre-judgment on whether it can collect signals normally, and judge the ear based on the impedance value between the ear side signal measurement units. Whether the side signal measurement unit fits the skin, so that different signal acquisition strategies can be selected for different situations.
  • the ear side in the embodiment of the present invention refers to the area on the human ear and near the ear where bioelectric signals can be measured, such as the inner ear canal, the auricle, the ear groove, the back of the ear, and the area around the ear.
  • the bioelectric signal is collected by deploying the ear-side signal measuring unit on the human ear area and near the human ear.
  • Fig. 13 is an exemplary way of wearing the ear-side wearing device, that is, the signal acquisition manner, according to an embodiment of the present invention, and exemplarily shows a signal acquisition manner of acquiring bioelectric signals from the inner side of the ear canal.
  • 401 is the human ear canal
  • 403 is the ear side signal measurement unit
  • 402 is the main body of the ear side wearing device
  • 404 is the user's auricle.
  • Figure 2a is a schematic flow chart of a method for acquiring user brain electrical signals provided by an embodiment of the application, and the specific flow includes:
  • S101 Collect the user's bioelectric signal from the user's ear through the ear-worn device
  • the ear-side wearing device After wearing the ear-side wearing device, turn on the EEG signal collection function of the device, and collect the user's EEG signal from the ear side through the ear-side wearing device.
  • the wearing method has been introduced above and will not be repeated here.
  • There are many ways to turn on the device including pressing the physical button on the headset, or through a user attention detection device (which can be a mobile phone or a vehicle terminal Etc.) to trigger the corresponding APP (such as touching the virtual button to start driving in the APP) to make the ear-worn device enter the working state.
  • a user attention detection device which can be a mobile phone or a vehicle terminal Etc.
  • the corresponding APP such as touching the virtual button to start driving in the APP
  • the embodiment of the present application will judge the wearing condition of the ear wearing device, and decide whether to collect data or whether to collect data according to the judgment result.
  • the data is used to analyze the types of user attention.
  • the ear-side wearing device includes a plurality of ear-side signal measurement units; it is determined whether the impedance between the two ear-side signal measurement units is lower than the preset threshold; The impedance is lower than a preset threshold; and the potential difference signal of the bioelectric signal measured by the two ear side signal measuring units is used as the user's bioelectric signal.
  • the ear side wearing device for bilateral ear canal measurement, that is, the ear side wearing device includes a left ear ear side signal measurement unit and a right ear side signal measurement unit.
  • the collection method of collecting bioelectric signals from the ear side can be further shown in the figure As shown in 2b, including:
  • S201 Determine whether the ear side wearing device can perform measurement normally by determining whether the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold.
  • the left ear side signal measurement unit and the right ear side signal measurement unit may be one or more than one.
  • the shape of the left/right ear side signal measurement unit may be an electrode during the realization process.
  • a preset threshold can be set to determine the wearing condition of the ear-side wearing device.
  • the preset judgment threshold of impedance may be the impedance value of the surface of the ear.
  • the impedance value between the two measurement units is directly obtained for judgment.
  • left ear signal measurement units and right ear signal measurement units there may be multiple measurement strategies. If arbitrarily select a left ear side signal measurement unit and a right ear side signal measurement unit to obtain the impedance value, it can also be the left ear side signal measurement unit at the preset position and the right ear side at the preset position.
  • the impedance value between the side signal measuring units is acquired only once, and it is judged whether the left and right ears are worn normally according to the acquired impedance value. Or you can set the priority order to perform matching measurements one by one, and terminate the measurement and judgment after a preset number of measurements if the preset threshold is not met, or perform measurements one by one until the impedance below the preset value is measured. It means that the ear-worn device can measure normally, otherwise it means it cannot work normally.
  • the specific measurement method in the case of multiple measurement units is not limited here.
  • the bioelectric signal collected by the left ear side signal measurement unit and the right ear side signal measurement unit acquires the user's bioelectric signal.
  • the wearing device can perform measurements normally; the user's bioelectric signal is acquired according to the potential difference signal between the bioelectric signal collected by the left ear signal measuring unit and the bioelectric signal collected by the right ear signal measuring unit.
  • the specific method for obtaining the user's bioelectric signal may include: The potential difference signal between the bioelectric signal collected by the left ear side signal measuring unit and the bioelectric signal collected by the right ear side signal measuring unit obtains the user's bioelectric signal; the ear side wearing device may also be set The reference electrode separately obtains the bioelectric signal collected by the left ear side signal measurement unit and the first potential difference signal of the reference electrode, and the bioelectric signal collected by the right ear side signal measurement unit and the second potential difference signal of the reference electrode. Potential difference signal, and then obtain the difference signal between the first potential difference signal and the second potential difference signal.
  • the impedance between the left ear signal measurement unit and the right ear signal measurement unit is lower than the preset threshold It is determined that the ear canals corresponding to the two measurement units measured can be measured normally.
  • the potential difference signal between the bioelectric signal collected by the left ear side signal measuring unit and the bioelectric signal collected by the right ear side signal measuring unit that can be measured normally after the measurement is determined as the user bioelectric signal signal.
  • the measured The ear canals corresponding to the two measurement units can be measured normally.
  • the potential difference signal between the bioelectric signal collected by the left ear side signal measuring unit and the bioelectric signal collected by the right ear side signal measuring unit that can be measured normally after the measurement is determined as the user bioelectric signal Signal, or based on the measurement result that the ear-worn device can be measured normally, select any left-ear signal measurement unit and right-ear signal measurement unit, or a pre-designated left-ear signal measurement unit And the right ear side signal measuring unit to obtain the potential difference signal of the collected bioelectric signal as the user's bioelectric signal.
  • step S50103 may be included.
  • the left ear side signal measurement unit and the right ear side signal measurement unit need to be multiple.
  • the threshold determines that at least one of the ear canals corresponding to the two measurement units cannot be measured normally.
  • the impedance between the left ear signal measurement unit and the right ear signal measurement unit is higher than the preset threshold, the measured At least one of the ear canals corresponding to the two measurement units cannot be measured normally. That is, when the impedance between one of the left ear side signal measurement units and one of the right ear side signal measurement units is higher than the preset threshold, it is determined that at least one of the ear canals corresponding to the two measurement units cannot be performed normally measuring.
  • the impedance between two of the plurality of left ear signal measurement units is lower than a preset threshold, And whether the impedance between two of the plurality of right ear signal measuring units is lower than a preset threshold.
  • this application can also set the priority order to perform the measurement between the two left/right ear side signal measurement units, and terminate the measurement and judgment after a preset number of measurements if the preset threshold is not met, or perform pair by pair After measuring until the impedance lower than the preset value is measured, it means that the ear-worn device can be measured normally. Otherwise, when all the conditions are traversed, no impedance lower than the preset value is measured, indicating that the normal measurement cannot be performed.
  • this application does not limit the specific measurement method in the case of single-side multiple measurement units.
  • the potential difference signal of the bioelectric signal collected by the left ear ear signal measuring unit and the bioelectric signal collected by the right ear ear signal measuring unit that can be measured normally after the measurement is determined as the user Bioelectric signal.
  • the potential difference signal of the bioelectric signal collected by the left ear side signal measuring unit and the bioelectric signal collected by the right ear side signal measuring unit that can be measured normally after the measurement is determined as the user's biological
  • the specific method of the electrical signal may include directly acquiring the user's bioelectric signal by the potential difference signal of the bioelectric signal collected by the two ear side signal measurement units that can normally measure one side; the ear side wearing device may also be provided with a reference electrode , Respectively acquiring the bioelectric signal collected by the one ear side signal measuring unit and the third potential difference signal of the reference electrode, and the bioelectric signal collected by the other ear side signal measuring unit and the fourth potential difference signal of the reference electrode, and then Obtain a difference signal between the third potential difference signal and the fourth potential difference signal.
  • the selection of the ear signal measurement unit can have multiple transmissions, such as directly selecting two measurement units for impedance value judgment to obtain the potential difference signal, or selecting the measurement unit according to a preset setting, or selecting arbitrarily.
  • the method of obtaining the potential difference signal in the present application may be implemented through software instructions, or may be implemented through hardware circuits.
  • the ear side wearing device for unilateral measurement, that is, the ear side wearing device only includes the left ear side signal measurement unit or the right ear side signal measurement unit.
  • the left ear side signal measurement unit or the right ear side signal measurement unit There are multiple ear signal measurement units, that is, there are multiple single ear signal measurement units.
  • the collection method of collecting bioelectric signals from the ear side may also be as shown in Fig. 2c, which further includes:
  • S211 Determine whether the ear side wearing device can perform the measurement normally by determining whether the impedance between the two signal measurement units in the unilateral ear side signal measurement unit is lower than a preset threshold.
  • the application can also set the priority order to perform the measurement between two unilateral ear signal measurement units, and terminate the measurement and judgment after the preset number of measurements if the preset threshold is not met, or perform the measurement pair by pair until the measurement
  • the impedance is lower than the preset value, it means that the ear-worn device can be measured normally. Otherwise, when all the conditions are traversed, no impedance lower than the preset value is measured, which means that the normal measurement cannot be performed.
  • the application also does not limit the specific measurement method in the case of multiple measurement units on one side.
  • the ear side wearing device can work normally measuring. After the measurement, the potential difference signal of the bioelectric signal collected by the two unilateral ear signal measuring units that are determined to be able to be measured is used as the user's bioelectric signal.
  • the specific method of using the potential difference signal of the bioelectric signal collected by the two unilateral ear signal measuring units that can be measured normally after the measurement as the user's bioelectric signal may include directly connecting the two ears that can be measured normally
  • the potential difference signal of the bioelectric signal collected by the signal measuring unit obtains the user's bioelectric signal;
  • the ear-side wearing device may also be provided with a reference electrode to separately obtain the bioelectric signal and the reference electrode collected by the one ear-side signal measuring unit And the bioelectric signal collected by the other ear signal measuring unit and the sixth potential difference signal of the reference electrode, and then the difference signal of the fifth potential difference signal and the sixth potential difference signal is obtained.
  • the potential difference signal of the bioelectric signal collected by the two unilateral ear signal measuring units that can be measured normally is used as the user's bioelectric signal, or the ear wearing device can be considered normal based on the measurement result
  • any two single-sided ear canal measurement units or two pre-designated single-sided ear canal measurement units are selected to obtain the collected potential difference signal of the bioelectric signal as the user's bioelectric signal.
  • the processing of obtaining the potential difference mentioned in the above embodiment may specifically be differential processing of the collected bioelectric signals, because it is difficult to obtain the EEG signal from the ear side, especially the ear canal, and the obtained EEG signal
  • the intensity is relatively weak, and the subsequent judgment of the signal may have a greater impact when disturbed by noise. Therefore, in order to ensure the feasibility of obtaining the electrical signal from the ear side and the accuracy of the subsequent conclusions of the user’s attention analysis, it is necessary
  • the bioelectric signal taken by the ear canal is subjected to targeted denoising processing, and the differential circuit can remove the noise in the collected bioelectric signal.
  • the ear-worn device is an electronic product.
  • circuit Although the circuit has been designed for electromagnetic shielding during operation, it may be affected by electrical signals on the circuit board and electromagnetic waves in the air under certain special scenarios, resulting in waveform distortion.
  • electrodes are attached to both ears, and signals are collected from the two ear canals to ensure the correctness of the signals.
  • FIG. 6 is the signal receiving circuit model on the two ear canals.
  • the differential circuit is used to eliminate the interference and facilitate the subsequent extraction of the correct waveform.
  • 601 is the left ear canal bioelectric signal
  • 602 is the bioelectric signal of the right ear canal
  • 603 is the noise signal
  • 601a is the bioelectric signal of the left ear canal mixed with noise
  • 602a is the bioelectric signal of the right ear canal mixed with noise
  • 604 is the result of differential processing
  • the bioelectric signal is the first bioelectric signal.
  • FIG. 6 is only an example of a situation, and 601 may be a right ear canal bioelectric signal, and 602 may be a left ear canal bioelectric signal.
  • the current differential circuit design is generally implemented directly by a chip.
  • the circuit implemented in the embodiment of the present invention is shown in Figure 5.
  • 501 and 502 are the inputs of the bioelectric signals collected by the left and right ear canals
  • 503 is the output after differential processing by the differential circuit.
  • the waveform representation is shown in Figure 7, V+ is the bioelectric signal of the left ear canal, V- is the bioelectric signal of the right ear canal, and (V+)-(V-) is the bioelectric signal after differential processing.
  • FIG. 7 is only an example of a situation, and V+ may be the bioelectric signal of the right ear canal, and V- may be the bioelectric signal of the left ear canal.
  • the EEG signal can be extracted from it according to specific application requirements.
  • the bioelectric signal includes one or more of various characteristic signals of the human body, such as an electrocardiogram ECG signal, an EOG signal, an electromyographic EMG signal, and an electroencephalogram EEG signal.
  • electrocardiogram ECG signal an electrocardiogram ECG signal
  • EOG signal an electrocardiographic EMG signal
  • electroencephalogram EEG signal an electroencephalogram EEG signal.
  • biometric signals through feature decomposition, which can be extracted according to the different spectrums of different types of signals.
  • the more commonly used method is to use the blind source separation algorithm independent component analysis (ICA) to decompose to get more information.
  • ICA blind source separation algorithm independent component analysis
  • the processing may include one or more of conventional bioelectric signal processing operations such as de-artifacting, wavelet analysis, and digital coding, to obtain an EEG signal that can more accurately and truly reflect the user's EEG characteristics. It may also be the EEG signal after other de-manipulation processing and digital conversion, which is not limited here.
  • bioelectric signal processing operations such as de-artifacting, wavelet analysis, and digital coding, to obtain an EEG signal that can more accurately and truly reflect the user's EEG characteristics. It may also be the EEG signal after other de-manipulation processing and digital conversion, which is not limited here.
  • the processing methods and functions of various processing operations are:
  • Anti-artifacts In the human body, electrical phenomena occur in many places. The most common is nerve conduction. One neuron receives stimulation and transmits bioelectricity to the next neuron. This electrical phenomenon occurs all the time as human beings survive. When it happens, every tiny human expression is closely related to the conduction of nerve currents. Not only the nerve cells, but also the organs in the human body can produce bioelectrical signals of different degrees and strengths. However, when measuring brain electricity, other bioelectrical signals are also mixed in, because they cannot be completely extracted from the coagulated brain. The original electrical signal is basically mixed with different bioelectricity from the human body, and its influence is large or small.
  • EEG signals such as heart beats, muscle movements, blinking movements, deep breathing, skin perspiration and so on.
  • the difference in temperature will also noise the strength of the bioelectric signal to varying degrees. If the ambient temperature is low, it will cause chills and jitters in a few people. These movements are relatively large and can also cause interference to the brain electricity.
  • Figure 8a shows the electromyographic artifacts produced by the neck joint action
  • Figure 8b shows the electrooculogram artifacts produced by blinking.
  • Wavelet analysis It is a time-frequency analysis method. Because the EEG signal is an unsteady signal, the traditional Fourier transform cannot extract details well (only frequency information can be extracted, time information cannot be extracted), wavelet transform It is a signal analysis method that can better reflect the time characteristics of the signal in the frequency domain. Wavelet analysis can well characterize the local characteristics of the signal.
  • Digital coding It is to digitally code the EEG signal and convert the EEG signal into a digital signal.
  • the anti-artifact processing can be performed on the bioelectric signal, or it can be performed on the posterior brain electrical signal from the feature extraction.
  • the obtained EEG signal can be used to analyze and judge the user's attention, and the extracted user's EEG signal can also be executed in step S103.
  • S103 Obtain the user's attention type based on the machine learning model according to the user's brain electrical signal. Specifically:
  • the usual processing method is to extract the attention feature of the obtained EEG signal.
  • the EEG signal is the EGG (electroencephalogram) signal, which is the external manifestation of brain activity. Different brain activities are manifested as EEG signals with different characteristics. Studies have shown that the presence of people can be clearly detected through the detected EEG signals.
  • Delta wave The frequency is distributed between 1Hz and 4Hz, and the amplitude is between 20uv and 200uv. The performance is more obvious in Dingye and the pituitary gland, and it is more prominent in the infant or the period of immature intellectual development, just like theta wave , Also belongs to slow wave. Under normal circumstances, delta waves only exist in states such as extreme lack of oxygen, deep sleep, or brain diseases.
  • Theta wave The frequency is distributed between 4Hz and 7Hz, and the amplitude is between 20uf and 40uf. It is a slow wave, which mainly appears in the occipital and parieto-occipital regions, and the left and right sides are symmetrical. It can generally be detected when a person is sleepy or lightly sleeping At the same time, there is a universal connection with people's psychological state. This wave usually appears when the central nervous system is in a state of depression, frustration, or drowsiness.
  • Alpha wave The frequency is distributed between 8Hz and 12Hz, and the amplitude is between 25uf and 75uf. It mainly appears in the parieto-occipital region, and the two sides are basically synchronized. This is the basic rhythm of normal EEG. The wave change is more obvious when in the state of thinking or resting. When the individual has a targeted activity, opens his eyes or receives other stimuli, the wave disappears and the ⁇ wave replaces it.
  • Beta wave The frequency is distributed between 14Hz and 30Hz, and the amplitude is about half of the delta wave. It mainly appears in the forehead and central area. The frequency of this wave significantly represents the level of excitement in the cerebral cortex. The individual is awake and sleeping Will appear.
  • step S103 can be further divided as shown in Fig. 4:
  • the process of obtaining sample entropy based on EEG signals includes:
  • A Intercept the EEG signal of a preset time length, and obtain N signal sampling points from the EEG signal of the preset time length, u(1), u(2),...,u(N) .
  • sampling points are sampling points at equal time intervals, and the preset time length of interception is optional.
  • Bi(r) (number of X(j)such that d[X(i),X(j)] ⁇ r)/(Nm), i ⁇ j, the value range of i is [1,N-m +1], the value range of j is [1,N-m+1] other than i, and r is a preset value.
  • the value of r can be related to the standard deviation ⁇ value of the above sampling point. The value can be between 0.1 ⁇ and 0.3 ⁇ .
  • Ai(r) (number of Y(j)such that d[Y(i),Y(j)] ⁇ r)/(Nm-1), i ⁇ j, the value range of i is [1,Nm ], the value range of j is [1,Nm] other than i, r is a preset value, for example, the value of r can be related to the standard deviation ⁇ value of the above sampling point, and the value can be from 0.1 ⁇ to Between 0.3 ⁇ .
  • A-F is not fixed.
  • B, C and D, E can be carried out before B and C, or can be carried out at the same time, or partially overlapping in time.
  • the user's attention is judged according to the obtained sample entropy value.
  • users or product developers can set one or more preset values based on historical experience.
  • the segmentation value used to distinguish between concentration and distraction is used to distinguish between Being awake is the split value of sleep.
  • the segmentation value used to distinguish between concentration and distraction when it is greater than or equal to this segmentation value, it means that it is concentrated, and when it is less than or equal to this segmentation value, it means it is distracted.
  • the size and number of the segmentation values are determined according to the number of attention states to be distinguished and the type of attention states.
  • the SVM classifier can also be used for machine learning model training to obtain the segmentation value, and the user's attention type can be determined according to the segmentation value and the sample entropy value.
  • the model training method is to use a variety of EEG signal samples of a certain length of time with different attention types, and calculate and obtain the sample entropy value of the EEG signal sample, and use the sample entropy value and the corresponding attention type to construct the sample. Train the SVM model. Then the trained model is used for subsequent attention analysis, that is, input the sample entropy value of the corresponding EEG signal, and output the corresponding attention type, or the probability of the attention type.
  • SVM is a discriminative classifier defined by a classification hyperplane. Given a set of labeled training samples, the algorithm will output an optimal hyperplane to classify new samples (test samples).
  • Figures 9a and 9b are schematic diagrams of obtaining an optimal hyperplane. Dots and squares represent two different types of data. For a linearly separable set composed of two-dimensional coordinate points, if a dividing line can be found that is separated from the two types of sample points As far as possible, it is considered to be the optimal hyperplane of this two-dimensional coordinate space, that is, the solid line in Figure 9b.
  • the purpose of SVM machine learning is to find a hyperplane, and it can distinguish between the two types of data while being able to maximize the distance to the nearest training sample. That is, the optimal segmentation hyperplane maximizes the training sample boundary.
  • the SVM classifier will output one or more segmentation values to determine the attention corresponding to the sample entropy of the user’s EEG signal status.
  • the segmentation value can be one or more, such as a segmentation value used to distinguish between concentration and distraction, and a segment value used to distinguish whether it is awake or sleep. For example, for the segmentation value used to distinguish between concentration and distraction, when it is greater than or equal to this segmentation value, it means that it is concentrated, and when it is less than or equal to this segmentation value, it means it is distracted.
  • the sample entropy analysis method only needs shorter data to get a robust estimate. It is an attention analysis method with better anti-noise and anti-interference ability.
  • Fig. 10 is a diagram of an exemplary attention detection system according to an embodiment of the present invention. include:
  • the ear side wearing device 11000 is used to collect the user's bioelectric signal from the ear side; obtain the brain electric signal from the user's bioelectric signal;
  • the ear wearing device 11000 can be specifically divided into unilateral or bilateral measurement.
  • the ear wearing device 11000 is for unilateral measurement. Its structure is shown in Figure 11a.
  • the unilateral ear signal measurement unit 1011 is used to measure from the left or right ear canal. The ear canal obtains the user's bioelectric signal.
  • the ear-mounted device 11000 is used for bilateral measurement, and its structure is shown in Figure 11b.
  • the left-ear ear signal measuring unit 101a is used to obtain user bioelectric signals from the left ear canal
  • the right ear ear-side signal measuring unit 101b is used to measure from the right.
  • the ear canal obtains the user's bioelectric signal.
  • the ear side wearing device 11000 corresponding to bilateral measurement, that is, the ear side wearing device includes a left ear ear side signal measuring unit 101a and a right ear side signal measuring unit 101b.
  • the ear side wearing device 11000 determines the left ear side signal Whether the impedance between the measuring unit 101a and the right ear signal measuring unit 101b is lower than the preset threshold is used to determine whether the ear wearing device can perform measurement normally.
  • the potential difference signal between the bioelectric signal collected by the left ear side signal measuring unit and the bioelectric signal collected by the right ear side signal measuring unit is used to obtain the user's bioelectric signal, and when the judgment result is the ear side wearing device
  • the measurement cannot be performed normally, and it is determined whether the impedance between two of the plurality of left ear signal measuring units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear signal measuring units is Whether the impedance between the two is lower than the preset threshold.
  • the user's bioelectric signal is acquired according to the potential difference signal of the bioelectric signal collected by the two bioelectric measuring devices in the ear canal on the side whose impedance is lower than the preset threshold.
  • the specific judgment method can be parameter steps S201-S6203.
  • the ear side wearing device 11000 for unilateral ear canal measurement, that is, the ear side wearing device 11000 only includes the left ear side signal measurement unit or the right ear side signal measurement unit 1011.
  • the left ear side signal measurement The unit or the right ear side signal measurement unit needs to be multiple, that is, there are multiple single ear side signal measurement units.
  • the ear side wearing device 11000 judges whether the ear side wearing device can perform the measurement normally by determining whether the impedance between the two signal measurement units in the unilateral ear side signal measuring unit is lower than a preset threshold.
  • the device can perform measurement normally, and obtain the user's bioelectric signal according to the potential difference signal of the bioelectric signal collected by two of the plurality of unilateral ear signal measuring units. Refer to steps S211-S212 for specific judgment methods.
  • the attention detection device 1200 detects the attention type of the user according to the brain electrical signal.
  • Fig. 11c is an exemplary structure diagram of an ear wearing device 11000 with attention detection capability according to an embodiment of the present invention.
  • the attention detection device and the ear wearing device can also be integrated in At the same time, as shown in FIG. 11, an ear-side wearing device 1100 integrated with an attention detection function corresponding to an embodiment of the present application.
  • the device includes:
  • the ear side signal measuring unit 111 is used to collect user bioelectric signals from the ear side.
  • the ear signal measuring unit 111 may include a left ear signal measuring unit 111a and a right ear signal measuring unit 111b.
  • the ear side signal measurement unit 111 may only include the unilateral ear side signal measurement unit 111c.
  • the feature decomposition unit 112 is configured to extract brain electrical signals from the user's bioelectric signals.
  • the attention detection unit 113 is configured to obtain the user's attention classification result based on the EEG signal based on a machine learning model.
  • a machine learning model For the specific analysis method, please refer to the above specific embodiment, which will not be repeated here.
  • the first judging unit 114 is configured to judge whether the impedance between the two ear side signal measuring units is lower than the preset threshold; when the impedance between the two ear side signal measuring units is lower than the preset threshold; The potential difference signal of the bioelectric signal collected by the bioelectric signal measured by the two ear side signal measuring units is used as the user's bioelectric signal.
  • the ear-side wearing device may optionally include a second judgment unit.
  • the first breaking unit 114 is used to determine whether the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold (the specific judgment method has been introduced above, and it is not here. Repeat); when the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold; the bioelectric signal collected by the left ear side signal measurement unit and The potential difference signal of the bioelectric signal collected by the right ear signal measuring unit is used as the user's bioelectric signal.
  • a preset threshold the specific judgment method has been introduced above, and it is not here. Repeat
  • the second judging unit 115 is used for when the first judging unit 114 judges that the ear-mounted device cannot be measured normally (the specific judging method has been introduced above and will not be repeated here), the second judging unit 115 respectively judges Whether the impedance between two of the plurality of left ear signal measuring units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear signal measuring units is lower than a preset Threshold; and the potential difference signal of the bioelectric signal collected by two bioelectric measurement devices in the ear canal on the side whose impedance is lower than the preset threshold is used as the user's bioelectric signal.
  • the second judging unit 115 is an optional unit, and the second judging unit 115 is applied to the case where there are multiple ear-side signal measurement units for the left ear and multiple right-ear ear-side signal measurement units.
  • the ear-side wearing device includes a first judging unit 114 for judging whether the impedance between the two unilateral ear-side signal measurement units is lower than a preset threshold (specifically, the judgment method is It has been introduced in the article and will not be repeated here.)
  • a preset threshold specifically, the judgment method is It has been introduced in the article and will not be repeated here.
  • the embodiment of the present invention also discloses a method for measuring a user's EEG signal, as shown in Figure 16: where steps S1601 and S1602 are the same as in Figure 2, and S1603 is sending the EEG signal to a signal analysis device.
  • the signal analysis device In the embodiment of the present application, it may specifically be an attention detection device.
  • the corresponding embodiment of the present invention also discloses an ear-side wearing device for measuring user-related signals.
  • the ear-side signal measuring unit 121 is used for collecting user bioelectric signals from the ear side.
  • the ear signal measuring unit 121 may include a left ear signal measuring unit 121a and a right ear signal measuring unit 121b.
  • the ear side wearing device 120 is a unilateral measurement device, the ear side signal measurement unit 121 may only include the unilateral ear side signal measurement unit 121c.
  • the feature decomposition unit 122 is used to obtain brain electrical signals from the user's bioelectric signals.
  • the sending unit 123 is configured to send the biometric signal to a signal analysis device; the signal analysis device may specifically be an attention detection device in the embodiment of the present application.
  • the first judging unit 124 is configured to judge whether the impedance between the two ear side signal measuring units is lower than the preset threshold; when the impedance between the two ear side signal measuring units is lower than the preset threshold; The potential difference signal of the bioelectric signal collected by the bioelectric signal measured by the two ear side signal measuring units is used as the user's bioelectric signal.
  • the ear-side wearing device may optionally further include a second judgment unit 125.
  • the first disconnection unit 124 is used to determine whether the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold (the specific determination method has been introduced above, and it is not here. Repeat); when the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold; the bioelectric signal collected by the left ear side signal measurement unit and The potential difference signal of the bioelectric signal collected by the right ear signal measuring unit is used as the user's bioelectric signal.
  • a preset threshold the specific determination method has been introduced above, and it is not here. Repeat
  • the second judging unit 125 is used for when the first judging unit 124 judges that the ear-side wearing device cannot be measured normally (the specific judging method has been introduced above and will not be repeated here), the third judging unit judges the Whether the impedance between two of the plurality of left ear signal measuring units is lower than a preset threshold, and whether the impedance between two of the plurality of right ear signal measuring units is lower than the preset threshold And the potential difference signal of the bioelectric signal collected by the two bioelectric measuring devices of the ear canal on the side whose impedance is lower than the preset threshold is used as the user's bioelectric signal.
  • the second judging unit 125 is an optional unit, and the second judging unit 125 is applied to the case where there are multiple ear-side signal measurement units for the left ear and multiple right-ear ear-side signal measurement units.
  • the first judgment unit 124 of the ear-side wearing device is used to judge whether the impedance between the two unilateral ear-side signal measurement units is lower than a preset threshold (the specific judgment method is described above It has been introduced and will not be repeated here), when the impedance between the two unilateral ear signal measurement units is lower than a preset threshold, the two of the multiple unilateral ear signal measurement units are collected The potential difference signal of the bioelectric signal is used as the user's bioelectric signal.
  • Fig. 12 is a schematic diagram of a specific product structure of the ear-side wearing device in Fig. 11.
  • the ear-side wearing device can have multiple forms, for example, it can be in the form of earphones or earplugs.
  • the ear-side wearing device given in this example The device is in the form of an earplug, but it is not limited in this application.
  • the ear-side wearing device includes an earplug body 301, a flexible electrode carrier 302 and a plurality of surface flexible electrodes 303.
  • the flexible electrode carrier 302 provides a sufficiently elastic support to ensure that the multiple flexible electrodes 303 attached to the surface of the flexible electrode carrier 302 form a close fit with the surface of the user's ear, ensuring stable collection of the user's brain wave signals.
  • Section 310 exemplarily presents a structure of a surface flexible electrode 303, including a biosensing flexible electrode 303A, a biosensing flexible electrode 303B, and a grounded common flexible electrode 303G with an equiangular 120 degree distribution, and 304 is an earplug hole.
  • the earplug body 301 is connected to a common flexible electrode that is grounded.
  • the grounded common flexible electrode can also be realized by electric shock of the electrode on the auricle support.
  • the wearing schematic diagram of the ear-side wearing device in the form of earplugs in Figure 12 is shown in Figure 13, where 401 is the user’s ear canal, 402 is the earplug for EEG signal measurement, 403 is the flexible electrode, and 404 is the user’s auricle . It can be seen from FIG.
  • the ear-worn device may also include a communication module for receiving or sending EEG signals, and optionally an attention detection unit for analyzing the user's attention type through EEG signals.
  • the ear side signal measurement unit in Figures 11a-d can be implemented by a flexible electrode.
  • Step S1702 is the same as S603 in Fig. 7, and S1701 is receiving EEG signals from the ear-worn device.
  • the embodiment of the present invention also discloses an attention detection device 130, as shown in FIG. 14.
  • the device includes:
  • the receiving unit 131 is configured to receive EEG signals from the ear-side wearing device
  • the attention detection unit 132 is configured to obtain the attention type of the user according to the EEG signal.
  • the attention detection unit 132 analyzes the user's attention type according to the EEG signal.
  • the attention detection unit may be a user's terminal device such as a mobile phone, or other wearable or portable terminal, or Set in a server in the cloud.
  • the attention detection unit includes a sample entropy acquisition module and an attention recognition module.
  • the sample entropy acquisition module is used to acquire the sample entropy based on the EEG signal
  • the process of obtaining sample entropy based on EEG signals includes:
  • A Intercept the EEG signal of a preset time length, and obtain N signal sampling points from the EEG signal of the preset time length, u(1), u(2),...,u(N) .
  • sampling points are sampling points at equal time intervals, and the preset length of time intercepted from the brain electrical signal can be set according to the needs of analysis.
  • Bi(r) (number of X(j)such that d[X(i),X(j)] ⁇ r)/(Nm), i ⁇ j, the value range of i is [1,N-m +1], the value range of j is [1,N-m+1] other than i, and r is a preset value.
  • the value of r can be related to the standard deviation ⁇ value of the above sampling point. The value can be between 0.1 ⁇ and 0.3 ⁇ .
  • Ai(r) (number of Y(j)such that d[Y(i),Y(j)] ⁇ r)/(Nm-1), i ⁇ j, the value range of i is [1,Nm ], the value range of j is [1,Nm] other than i, r is a preset value, for example, the value of r can be related to the standard deviation ⁇ value of the above sampling point, and the value can be from 0.1 ⁇ to Between 0.3 ⁇ .
  • A-F is not fixed.
  • B, C and D, E can be carried out before B and C, or can be carried out at the same time, or partially overlapping in time.
  • the attention recognition module is used to judge the user's attention state based on the sample entropy value obtained from the collected EEG signal.
  • the attention recognition module can include:
  • the SV classifier is used for machine learning to obtain the segmentation value; specifically, it can be through the SVM machine learning method. After inputting multiple sample entropy values and their corresponding attention states, the SVM classifier will output one or more segmentation values. To determine the attention state corresponding to the sample entropy of the user’s EEG signal.
  • the SVM classifier can be set in the attention recognition module, or it can be set in other devices for training to obtain the segmentation value, and then the segmentation value is sent to the attention recognition module, or manually set by the user or developer according to the training result.
  • the judgment module is used to judge the user's attention type according to the segmentation value and the sample entropy value.
  • the segmentation value can be one or more, such as a segmentation value used to distinguish between concentration and distraction, and a segment value used to distinguish whether it is awake or sleep.
  • a segmentation value used to distinguish between concentration and distraction when it is greater than or equal to this segmentation value, it means that it is concentrated, and when it is less than or equal to this segmentation value, it means it is distracted.
  • the specific implementation form of the attention detection device 130 may be a handheld terminal, a vehicle-mounted terminal, or other devices that can be used to calculate and analyze brain electrical signals.
  • Fig. 15a is a schematic diagram of the processor structure corresponding to the ear wearing device of the embodiment of the present application.
  • the ear-side wearing device 1400 integrated with the attention detection function may include one or more processors 1406, one or more memories 1401, and a feature decomposition unit 1403.
  • the ear wearing device may further include a communication unit 1405.
  • the processor 1406 can be respectively connected to the memory 1401, the measuring electrode 1402, the feature decomposition circuit 1403 and other components through a bus. They are described as follows:
  • the processor 1406 is the control center of the ear wearing device, and various interfaces and lines are used to connect various components of the ear wearing device.
  • the processor 1406 may further include one or more processing cores.
  • the processor 1400 can determine whether the measurement electrode can be measured normally (whether the ear-worn device can be measured normally) by executing program instructions, and perform user attention analysis based on the measurement signal.
  • the processor 1406 may be a dedicated processor or a general-purpose processor, when the processor 1406 is a general-purpose processor, the processor 1406 runs or executes software programs (instructions) and/or modules stored in the memory 1401.
  • the memory 1401 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.
  • the memory 1401 may further include a memory controller to provide the processor 1400 and the input unit to access the memory 1401.
  • the memory 1401 may be specifically used to store software programs (instructions) and collected user bioelectric signals.
  • the ear side signal measuring unit 1402 is used to collect user bioelectric signals from the ear side.
  • the ear side signal measurement unit 1402 may include a left ear side signal measurement unit and a right ear side signal measurement unit.
  • the ear side signal measurement unit 1402 may only include a unilateral ear side signal measurement unit.
  • the ear signal measuring unit 1402 is usually implemented by hardware.
  • the ear signal measuring unit 1402 may be an electrode, and the ear signal measuring unit 1402 may be one or more.
  • the feature decomposition unit 1403 is used to obtain brain electrical signals from the user's bioelectric signals.
  • the feature decomposition unit 1403 is usually implemented by hardware, such as a feature decomposition circuit and an ICA component.
  • the communication unit 1405 is used to communicate with the ear-worn device and other devices through wireless or wired communication technology, such as cellular mobile communication technology, WLAN, Bluetooth, etc.
  • the ear-side wearing device in the embodiments of the present application may include more or fewer components than those shown in the figure, or a combination of certain components, or different component arrangements.
  • the ear wearing device may further include a speaker, a camera, etc., which will not be repeated here.
  • the processor 1406 can read and analyze the measurement signal stored in the memory 1401 to determine whether the measurement electrode can be measured normally (whether the ear-worn device can be measured normally), and perform user attention analysis based on the measurement signal .
  • the processor 1406 can read and analyze the measurement signal stored in the memory 1401 to determine whether the measurement electrode can be measured normally (whether the ear-worn device can be measured normally), and perform user attention analysis based on the measurement signal .
  • the processor 1406 is configured to determine whether the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold (the specific determination method has been introduced above , Not repeat them here); when the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold; according to the left ear side signal measurement unit
  • the potential difference signal between the bioelectric signal and the bioelectric signal collected by the right ear side signal measuring unit acquires the user’s bioelectric signal; when it is determined that the ear side wearing device cannot be measured normally (the specific determination method has been described above , I will not repeat them here), respectively determine whether the impedance between two of the multiple left ear signal measurement units is lower than a preset threshold, and whether the impedance between the multiple right ear signal measurement units Whether the impedance between the two is lower than a preset threshold; and obtaining the user's bioelectricity according to the potential difference signal of the bioelectric
  • the processor 1406 is configured to determine whether the impedance between the two unilateral ear signal measurement units is lower than a preset threshold (the specific determination method has been introduced above, and will not be repeated here), When the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, the obtained signal is obtained according to the potential difference signal of the two collected bioelectric signals among the multiple unilateral ear signal measurement units. Describe the user's bioelectric signal.
  • the processor 1406 is further configured to obtain the attention type of the user according to the brain electrical signal.
  • the specific analysis method please refer to the above specific embodiment, which will not be repeated here.
  • FIG. 14 is only an implementation of the ear-side wearing device of the present application, the processor 1406 and the memory 1401 in the ear-side wearing device may also be integratedly deployed in possible embodiments. .
  • FIG. 14 may also be an ear side wearing device for measuring user brain electrical signals according to an embodiment of the present invention. It may include one or more processors 1406, one or more memories 1401, an ear side signal measuring unit 1402, and feature decomposition. Unit 1403. In specific implementation, the ear-side wearing device may further include a communication unit 1405 (including a sending unit and a receiving unit). The processor 1406 can be respectively connected to the memory 1401, the measuring electrode 1402, the feature decomposition circuit 1403 and other components through a bus. They are described as follows:
  • the processor 1406 is the control center of the ear wearing device, and various interfaces and lines are used to connect various components of the ear wearing device.
  • the processor 1406 may further include one or more processing cores.
  • the processor 1400 can determine whether the measurement electrode can be measured normally (whether the ear-worn device can be measured normally) by executing program instructions.
  • the processor 1406 may be a dedicated processor or a general-purpose processor, when the processor 1406 is a general-purpose processor, the processor 1406 runs or executes software programs (instructions) and/or modules stored in the memory 1401.
  • the memory 1401 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.
  • the memory 1401 may further include a memory controller to provide the processor 1400 and the input unit to access the memory 1401.
  • the memory 1401 may be specifically used to store software programs (instructions) and collected user bioelectric signals.
  • the ear side signal measuring unit 1402 is used to collect user bioelectric signals from the ear side.
  • the ear side signal measurement unit 1402 may include a left ear side signal measurement unit and a right ear side signal measurement unit.
  • the ear side signal measurement unit 1402 may only include a unilateral ear side signal measurement unit.
  • the ear signal measuring unit 1402 is usually implemented by hardware.
  • the ear signal measuring unit 1402 may be an electrode, and the ear signal measuring unit 1402 may be one or more.
  • a feature decomposition unit 1403 may be further included for obtaining brain electrical signals from the user's bioelectric signals.
  • the feature decomposition unit 1403 is usually implemented by hardware, such as a feature decomposition circuit and an ICA component.
  • the communication unit 1405 is used to communicate with the ear-worn device and other devices through wireless or wired communication technology, such as cellular mobile communication technology, WLAN, Bluetooth, etc., for collecting and processing bioelectric signals or brain electricity
  • the signal is sent to the signal analysis device; the signal analysis device may specifically be an attention detection device in the embodiment of the present application.
  • the signal analysis device can also be a sleep measurement device, an emotion detection device and other needs A device that obtains information by analyzing EEG signals.
  • the ear-side wearing device in the embodiments of the present application may include more or fewer components than those shown in the figure, or a combination of certain components, or different component arrangements.
  • the ear wearing device may further include a speaker, a camera, etc., which will not be repeated here.
  • the processor 1406 may read and analyze the measurement signal stored in the memory 1401 to determine whether the measurement electrode can be measured normally (whether the ear-worn device can be measured normally), and the type of user attention based on the measurement signal Analysis.
  • the measurement signal stored in the memory 1401 to determine whether the measurement electrode can be measured normally (whether the ear-worn device can be measured normally), and the type of user attention based on the measurement signal Analysis.
  • the processor 1406 is configured to determine whether the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold (the specific determination method has been introduced above , Not repeat them here); when the impedance between the left ear side signal measurement unit and the right ear side signal measurement unit is lower than a preset threshold; according to the left ear side signal measurement unit
  • the potential difference signal between the bioelectric signal and the bioelectric signal collected by the right ear side signal measuring unit acquires the user’s bioelectric signal; when it is determined that the ear side wearing device cannot be measured normally (the specific determination method has been described above , I will not repeat them here), respectively determine whether the impedance between two of the multiple left ear signal measurement units is lower than a preset threshold, and whether the impedance between the multiple right ear signal measurement units Whether the impedance between the two is lower than a preset threshold; and obtaining the user's bioelectricity according to the potential difference signal of the bioelectric
  • the processor 1406 is configured to determine whether the impedance between the two unilateral ear signal measurement units is lower than a preset threshold (the specific determination method has been introduced above, and will not be repeated here), When the impedance between the two unilateral ear signal measurement units is lower than the preset threshold, the obtained signal is obtained according to the potential difference signal of the two collected bioelectric signals among the multiple unilateral ear signal measurement units. Describe the user's bioelectric signal.
  • FIG. 14 is only an implementation of the ear-side wearing device of the present application.
  • the processor 1406 and the memory 1401 in the ear-side wearing device may also be integratedly deployed in possible embodiments.
  • FIG. 15b is a schematic structural diagram of another terminal form of an attention detection device provided by an embodiment of the present application.
  • the attention detection device may include one or more processors 1500 and one or more memories. 1501.
  • the attention detection device may further include an input unit 1506, a display unit 1503, a communication unit 1502 and other components.
  • the processor 2011 may be connected to the memory 1501, the communication unit 1502, the input unit 1506, the display unit 1503 and other components through the bus. . They are described as follows:
  • the processor 1500 is the control center of the attention detection device, and various interfaces and lines are used to connect various components of the attention detection device.
  • the processor 1500 may further include one or more processing cores.
  • the processor 1500 can perform attention detection of the brain electrical signal by executing program instructions.
  • the processor 1500 may be a dedicated processor or a general-purpose processor, when the processor 1500 is a general-purpose processor, the processor 1500 runs or executes software programs (instructions) and/or modules stored in the memory 1501.
  • the memory 1501 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.
  • the memory 1501 may further include a memory controller to provide the processor 1500 and the input unit 1506 to access the memory 1501.
  • the memory 1501 may be specifically used to store software programs (instructions) and brain electrical signals.
  • the input unit 1506 can be used to receive numeric or character information input by the user, and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.
  • the input unit 1506 may include a touch-sensitive surface 1505 and other input devices 1507.
  • the touch-sensitive surface 1505 is also called a touch screen or a touch pad, which can collect user touch operations on or near it, and drive the corresponding connection device according to a preset program.
  • other input devices 1507 may include, but are not limited to, one or more of a physical keyboard, function keys, trackball, mouse, and joystick.
  • the display unit 1503 can be used to display the search request input by the user or the search result provided to the user by the search device and various graphical user interfaces of the search device. These graphical user interfaces can be composed of graphics, text, icons, videos, and any combination thereof .
  • the display unit 1503 may include a display panel 1504.
  • the display panel 1504 may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc.
  • the touch-sensitive surface 1505 and the display panel 1504 are used as two independent components, in some embodiments, the touch-sensitive surface 1505 and the display panel 1504 may be integrated to implement input and output functions.
  • the touch-sensitive surface 1505 may cover the display panel 1504. When the touch-sensitive surface 1505 detects a touch operation on or near it, it is transmitted to the processor 1500 to determine the type of the touch event, and then the processor 1500 determines the type of the touch event. The corresponding visual output is provided on the display panel 1504.
  • the communication unit 1502 is configured to communicate with the ear-worn device and other devices through wireless or wired communication technology, such as cellular mobile communication technology, WLAN, Bluetooth, etc. It is used to receive the EEG signal sent by the ear-side wearing device, and possibly return a prompt signal to the ear-side wearing device according to the judgment result, or to prompt directly through a speaker, or to display a reminder interface through the display unit 1503.
  • wireless or wired communication technology such as cellular mobile communication technology, WLAN, Bluetooth, etc.
  • the retrieval device in the embodiments of the present application may include more or less components than those shown in the figure, or a combination of certain components, or different component arrangements.
  • the retrieval device may further include a speaker, a camera, etc., which will not be repeated here.
  • the processor 1500 may read, analyze and determine the brain electrical signal stored in the memory 1501 to realize the detection of the user's attention type based on the brain electrical signal in step S103 of the embodiment of the present application. include:
  • the sample entropy is acquired based on the EEG signal, and the acquisition process of the sample entropy has been described in detail above, so it will not be repeated here.
  • the user's attention state is judged based on the value of sample entropy obtained from the collected EEG signals.
  • the user's attention type is judged.
  • FIG. 15b is only an implementation of the retrieval device of the present application, the processor 1500 and the memory 1501 in the retrieval device may also be integrated deployment in possible embodiments.
  • the above embodiments it may be implemented in whole or in part by software, hardware, firmware or any combination.
  • software it can be implemented in the form of a computer program product in whole or in part.
  • the computer program product includes one or more computer instructions, and when the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part.
  • the processor may be a general-purpose processor or a special-purpose processor.
  • the retrieval device may be one or a computer network composed of multiple retrieval devices.
  • the computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a network site, computer, server, or data center.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer, and may also be a data storage device such as a server or a data center integrated with one or more available media.
  • the usable medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape, etc.), an optical medium (such as a DVD, etc.), or a semiconductor medium (such as a solid state hard disk), and so on.
  • the execution subject may be ASIC, FPGA, CPU, GPU, etc., which may be implemented by hardware or software, and the memory may be volatile, such as DDR, SRAM, HDD, SSD, etc. Or non-volatile storage devices.
  • the data retrieval device can be applied to a variety of scenarios, such as a server used in a video surveillance system, and may be in the form of a PCIe expansion card.
  • ASIC and FPGA are hardware implementations, that is, the method of this application is implemented by means of hardware description language during hardware design;
  • CPU and GPU are software implementations, that is, the method of this application is implemented by means of software program codes during software design. .

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Physics & Mathematics (AREA)
  • Biophysics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Veterinary Medicine (AREA)
  • Psychiatry (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Physiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychology (AREA)
  • Developmental Disabilities (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Hospice & Palliative Care (AREA)
  • Social Psychology (AREA)
  • Child & Adolescent Psychology (AREA)
  • Educational Technology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Otolaryngology (AREA)
  • Evolutionary Computation (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

一种用户注意力检测方法及系统，通过耳侧佩戴装置（1100），从耳侧采集用户的脑电信号；并当判断所述耳侧佩戴装置（1100）能够从所述用户的左耳道和右耳道均能采集到脑电信号时，对所述用户的左耳道和右耳道的脑电信号进行差分处理，得到脑电信号；基于所述脑电信号检测所述用户的注意力类型。该方法和系统，能够通过耳道更加方便快捷的获取到用户的脑电信号，并能够随时随地的对用户的注意力情况进行检测，可应用于自动驾驶领域用于对驾驶员的注意力进行智能检测，并基于检测结果对用户进行预警或者进行驾驶模式的切换。

Description

一种注意力检测方法及系统 技术领域
本申请涉及数据处理领域，尤其涉及安全驾驶和辅助驾驶过程中驾驶员注意力的检测方法及系统。
背景技术
随着社会的发展，汽车的普及，安全驾驶成为保证交通安全的重要关注问题之一，其中驾驶员的状态又是影响安全驾驶的重要因素之一。注意力不集中的驾驶包括任何转移驾驶员注意力的驾驶活动，如乘车、吃喝、与乘客交谈、调整娱乐或导航系统以及打电话，还包括驾驶员的精神状态或者意识若干变化，比如因疲劳而短暂地接近睡眠状态等。研究表明，有高达30％的交通事故是因为驾驶员注意力不集中造成的。当车辆以较高速度行驶时，分散了注意力的驾驶员如果不能充分意识到包括路径、交通、障碍物甚至车辆的状态的及时变化，事故就不可避免的发生了。
在自动驾驶的L1和L2等级，由于驾驶员一直负责驾驶过程和车辆控制，因此车辆系统对驾驶员的状态的准确和及时的检测和检测，当驾驶员注意力不集中的时候，选择适当的时机进行辅助提醒，对确保安全驾驶的具有十分重要的地位。
现有的技术根据汽车系统采集的驾驶员注视点、视线、休息时间、眼跳、沿行驶路径的周围物体的运动的状态等信号，通过智能计算机系统来判断驾驶员的行驶注意力类型，这类技术通常需要在汽车系统中安装各类传感器，行车电脑系统等，代价比较大。由于驾驶环境的复杂和多样性，通过智能计算获得驾驶员的行驶注意力状态存在一定概率的偏差，影响了安全驾驶。
近年来，随着由于脑电波信号采集和分析的技术的长足发展，车辆系统通过采集驾驶员脑电EEG信号，来准确判断驾驶员的行驶注意力类型，实现驾驶员的状态的准确和及时的检测和检测，对驾驶员的驾驶行为进行进行辅助提醒，对确保安全驾驶提供了另一种有效的技术实现选择。
然而如何更加方便且准确的获取驾驶员的脑电信号，如何通过脑电信号准确的确定驾驶员的状态成为安全驾驶中的研究重点。
发明内容
本申请实施例提供了一种注意力检测方法及系统，能够应用于驾驶员的注意力检测，通过耳侧获取脑电信号，使得在驾驶过程中脑电信号的获取更加便捷可行，降低了测量成本，同时还能保证脑电信号获取的准确性。
第一方面，本申请实施例提供一种用户注意力检测方法，所述方法包括：通过耳侧佩戴装置，从用户耳侧采集用户生物电信号；从所述用户生物电信号中获取用户脑电信号；根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型；其中通过耳侧佩戴装置，从用户耳侧采集用户生物电信号具体包括：所述耳侧佩戴装置包括多个耳侧信号测量单元；判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号。
上述方法中通过耳侧获取脑电信号，更加方便快捷，耳侧佩戴装置携带方便，设 置在耳侧，佩戴过程中不易脱落，使得在驾驶过程中测量用户脑电信号更加方便可行，同时采用电位差处理的方式对所采集用户生物电信号进行处理是针对耳侧采集生物电信号的特性的处理方式，能够有效去除脑电信号中的噪声，同时考虑用户可能没有正确佩戴的原因，或者一侧设备故障或是信号接收不良的情况，在进行电位差处理之前需要对采集的生物电信号进行判断，避免在耳侧佩戴装置不能正常接收时依然进行信号采集并处理，导致结果的不准确。
在所述第一方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
即耳侧佩戴装置可以是从左右耳双侧来获取生物电信号，判断两侧佩戴正常的情况下，根据从左右耳获取的生物电信号来获取用户生物电信号。
在所述第一方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置，所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；所述判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值；根据所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
根据上述方式，耳侧佩戴装置可以是单耳佩戴装置，直接根据两个测量单元之间的阻抗来判断单耳是否佩戴正常。
在所述第一方面的某些实现方式下，所述左耳侧信号测量单元为多个；所述右耳侧信号测量单元为多个；当其中一个左耳侧信号测量单元和其中一个右耳侧信号测量单元之间的阻抗高于预设阈值；分别判断所述多个左耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置测量的生物电信号的电位差值信号获取所述用户生物电信号。
根据上述方式，如果仅有一侧耳道能够获取到脑电信号，可以继续判断是否有一侧佩戴正常，如果有一侧佩戴正常上述实施方式下依然可以正确的获取到用户的生物电信号。
在所述第一方面的某些实现方式下，所述根据所述两个耳侧信号测量单元采集的 生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：通过差分电路获取所述两个耳侧信号测量单元采集的生物电信号的电位差值信号，并将所述电位差值信号获取所述用户生物电信号。
在所述第一方面的某些实现方式下，所述根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
在所述第一方面的某些实现方式下，所述基于所述用户脑电信号检测所述用户的注意力类型具体为：截取预设时间长度的所述用户脑电信号，从所述预设时间长度的所述用户脑电信号获得N个信号采样点；所述N个信号采样点为，u(1),u(2),...,u(N)；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；针对所述N-m+1个m维向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；针对所述N-m个m+1维向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值；基于所述第一平均值与所述第二平均值比值来计算样本熵(SampEn)的值。
在所述第一方面的某些实现方式下，所述机器学习模型为SVM分类器；采用SVM分类器进行机器学习获得分割值，根据所述分割值和所述样本熵值判断用户的注意力类型。
通过上述方式，根据样本熵的值通过机器学习模型来获取用户的注意力类型，可以通过机器学习的方式，更准确的分析出不同注意力类型下的脑电信号的样本熵特性，从而基于所采集的脑电信号的样本熵值确定用户挡墙的注意力类型。
第二方面，本发明实施例提供一种用户注意力检测系统，所述系统包括：耳侧佩戴装置，用于从用户耳侧采集用户生物电信号；用于从所述用户生物电信号中获取用户脑电信号；注意力检测装置，用于基于所述用户脑电信号检测所述用户的注意力类型；其中所述耳侧佩戴装置，用于从用户耳侧采集用户生物电信号具体包括：所述耳侧佩戴装置包括多个耳侧信号测量单元；所述耳侧佩戴装置判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号。
在所述第二方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述耳侧佩戴装置判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：所述耳 侧佩戴装置判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第二方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置，所述单侧耳侧佩戴装置包括多个单侧耳侧信号测量单元；所述耳侧佩戴装置判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：所述耳侧佩戴装置判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值；根据所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第二方面的某些实现方式下，所述左耳侧信号测量单元为多个；所述右耳侧信号测量单元为多个；当其中一个左耳侧信号测量单元和其中一个右耳侧信号测量单元之间的阻抗高于预设阈值；所述耳侧佩戴装置判断分别判断所述多个左耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置测量的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第二方面的某些实现方式下，所述注意力检测装置根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：所述注意力检测装置计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
第三方面，本发明实施例提供一种耳侧佩戴装置，所述装置包括：多个耳侧信号测量单元，用于从耳侧采集的用户生物电信号；第一判断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；特征分解单元，用于从所述用户生物电信号获得脑电信号；注意力检测单元，用于根据所述脑电信号信号基于机器学习模型获得所述用户的注意力类型。
在所述第三方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述第一断单元，用于判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第三方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置；所 述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；所述第一断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号，具体为：所述第一判断单元用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第三方面的某些实现方式下，所述左耳耳侧信号测量单元为多个；所述右耳耳侧信号测量单元为多个；所述耳侧佩戴装置还包括第二判断单元；当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，所述第二判断单元分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第三方面的某些实现方式下，所述注意力检测单元根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：所述注意力检测单元计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
第四方面，本发明实施例提供一种耳侧佩戴装置，其特征在于，所述装置包括：多个耳侧信号测量单元，用于从耳侧采集的用户生物电信号；第一判断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；特征分解单元，用于从所述用户生物电信号获得脑电信号；发送单元，用于将所述脑电信号发送给信号分析装置。
在所述第四方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述第一断单元，用于判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第四方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置；所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；所述第一断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号，具体为：所述第一判断单元用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所 述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第四方面的某些实现方式下，所述左耳耳侧信号测量单元为多个；所述右耳耳侧信号测量单元为多个；所述耳侧佩戴装置还包括第二判断单元；当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，所述第二判断单元分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
第五方面，本发明实施例提供一种注意力检测装置，所述装置包括：接收单元，用于从耳侧佩戴装置接收用户脑电信号；注意力检测单元，用于根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型。
在所述第五方面的某些实现方式下，所述注意力检测单元具体用于，计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
在所述第五方面的某些实现方式下，，所述注意力检测单元具体用于，截取预设时间长度的所述用户脑电信号，从所述预设时间长度的所述用户脑电信号获得N个信号采样点；所述N个信号采样点为，u(1),u(2),...,u(N)；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；针对所述N-m+1个m维向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；针对所述N-m个m+1维向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值；基于所述第一平均值与所述第二平均值比值来计算样本熵(SampEn)的值。
在所述第五方面的某些实现方式下，所述机器学习模型为SVM分类器；采用SVM分类器进行机器学习获得分割值；所述注意力检测单元根据所述分割值和所述样本熵值判断用户的注意力类型。
第六方面，本发明实施例提供一种耳侧佩戴装置，所述装置包括：多个耳侧信号测量单元，用于从耳侧采集的用户生物电信号；处理器，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；特征分解单元，用于从所述用户生物电信号获得脑电信号；注意力检测单元，用于根据所述脑电信号信号基于机器学习模型获得所述用户的注意力类型。
在所述第六方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述处理器，用于判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第六方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置；所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；所述处理器，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号，具体为：所述处理器用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第六方面的某些实现方式下，所述左耳耳侧信号测量单元为多个；所述右耳耳侧信号测量单元为多个；所述处理器还用于，当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，所述处理器分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第六方面的某些实现方式下，所述注意力检测单元根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：所述注意力检测单元计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
第七方面，本发明实施例提供一种耳侧佩戴装置，其特征在于，所述装置包括：多个耳侧信号测量单元，用于从耳侧采集的用户生物电信号；处理器，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；特征分解单元，用于从所述用户生物电信号获得脑电信号；发送单元，用于将所述脑电信号发送给信号分析装置。
在所述第七方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述处理器，用于判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取 所述用户生物电信号。
在所述第七方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置；所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；所述处理器，判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号，具体为：所述处理器判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第七方面的某些实现方式下，所述左耳耳侧信号测量单元为多个；所述右耳耳侧信号测量单元为多个；所述处理器还用于当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，所述处理器分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
第八方面，本发明实施例提供一种注意力检测装置，所述装置包括：接收单元，用于从耳侧佩戴装置接收用户脑电信号；处理器，用于根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型。
在所述第八方面的某些实现方式下，所述处理器具体用于，计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
在所述第八方面的某些实现方式下，所述处理器具体用于，截取预设时间长度的所述用户脑电信号，从所述预设时间长度的所述用户脑电信号获得N个信号采样点；所述N个信号采样点为，u(1),u(2),...,u(N)；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；针对所述N-m+1个m维向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；针对所述N-m个m+1维向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值；基于所述第一平均值与所述第二平均值比值来计算样本熵(SampEn)的值。
在所述第八方面的某些实现方式下，所述机器学习模型为SVM分类器；采用SVM分类器进行机器学习获得分割值；所述注意力检测单元根据所述分割值和所述样本熵值判断用户的注意力类型。
第九方面，本发明实施例提供一种脑电信号检测方法，所述方法包括通过多个耳 侧信号测量单元从耳侧采集的用户生物电信号；判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；从所述用户生物电信号获得脑电信号；将所述脑电信号发送给信号分析装置。
在所述第九方面的某些实现方式下，所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；所述判断步骤具体为，判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
在所述第九方面的某些实现方式下，所述耳侧佩戴装置为单侧耳侧佩戴装置；所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；所述判断步骤具体为，判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
在所述第九方面的某些实现方式下，所述左耳耳侧信号测量单元为多个；所述右耳耳侧信号测量单元为多个；当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
第十方面，本发明实施例提供一种注意力检测方法，所述方法包括：从耳侧佩戴装置接收用户脑电信号；根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型。
在所述第十方面的某些实现方式下，根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为，计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
在所述第十方面的某些实现方式下，计算所述用户脑电信号的样本熵的值具体为，截取预设时间长度的所述用户脑电信号，从所述预设时间长度的所述用户脑电信号获得N个信号采样点；所述N个信号采样点为，u(1),u(2),...,u(N)；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；针对所述N-m+1个m维向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量； 针对所述N-m个m+1维向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值；基于所述第一平均值与所述第二平均值比值来计算样本熵(SampEn)的值。
在所述第十方面的某些实现方式下，所述机器学习模型为SVM分类器；采用SVM分类器进行机器学习获得分割值；所述注意力检测单元根据所述分割值和所述样本熵值判断用户的注意力类型。
在以上各方面的某些实现方式下，所述耳侧佩戴装置为耳塞或者耳机。
在以上方面的某些实现方式下，所述用户的注意力类型具体可以是用户的注意力状态为集中，或者注意力分散。
在以上方面的某些实现方式下，所述多个耳侧信号测量单元中的多个为两个或者两个以上。
在以上方面的某些实现方式下，所述判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值，可以是基于预先的设定来从多个耳侧信号测量单元中选取两个，也可以是基于优先级的设定顺序选取两个耳侧信号测量单元进行比较，在阻抗一直低于预设阈值的情况下，可以比较预设次数终止比较，或者遍历所有情况之后终止比较。
在以上方面的某些实现方式下，所述判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值，可以是左耳耳侧信号测量单元和右耳耳侧信号测量单元均为一个，直接进行比较；也可以左耳耳侧信号测量单元和右耳耳侧信号测量单元均为多个，基于预先的设定来从多个左耳耳侧信号测量单元和多个右耳耳侧信号测量单元中分别选取两个，也可以是基于优先级的设定顺序选取两个耳侧信号测量单元进行比较，在阻抗一直低于预设阈值的情况下，可以比较预设次数终止比较，或者遍历所有情况之后终止比较。
在以上方面的某些实现方式下，所述判断所述多个单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，可以是单侧耳侧信号测量单元为两个，直接进行比较；也可以是基于预先的设定来从多个单侧耳侧信号测量单元中选取两个，也可以是基于优先级的设定顺序选取两个耳侧信号测量单元进行比较，在阻抗一直低于预设阈值的情况下，可以比较预设次数终止比较，或者遍历所有情况之后终止比较。
在以上方面的某些实现方式下，当判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，可以是左、右耳耳侧信号测量单元分别为两个，直接进行比较；也可以是基于预先的设定来分别从多个左、右耳耳侧信号测量单元中选取两个；也可以是对于左耳耳侧信号测量单元，基于优先级的设定顺序选取两个左耳耳侧信号测量单元进行比较，在阻抗一直低于预设阈值的情况下，可以比较预设次数终止比较，或者遍历所有情况之后终止比较；对于右耳耳侧信号测量单元，基于优先级的设定顺序选取两个右耳耳侧信号测量单元进行比较，在阻抗一直低于预设阈 值的情况下，可以比较预设次数终止比较，或者遍历所有情况之后终止比较。
可以看到，实施本申请实施例的技术方案，能够解决当前技术注意力判断不方便，以及在移动状态中判断注意力结果容易不准确的问题，通过耳部采集驾驶员脑电信号，使得用户脑电信号的采集更加方便可行，同时由于电极贴合度的要求，本方案可以判断当前佩戴是否正常，并确保在正常可采集的情况下进行信号的采集以及后续的分析，确保检测结果的准确性。同时通过电位差处理的方式处理左右耳道所采集的脑电信号能够保证采集到的脑电信号的准确性，对采集处理后的脑电信号进行样本熵计算，获取脑电信号时域上的一致性状态，并通过SVM分类算法，对注意力进行判定，可以相对准确地判断出驾驶员当下的行驶注意力类型，可以用于在驾驶过程中准确地给出后续操作，如向司机进行提醒或采取相应的应急操作。
附图说明
为了更清楚地说明本申请实施例或背景技术中的技术方案，下面将对本申请实施例或背景技术中所需要使用的附图进行说明。
图1示出了本申请实施例中的一种应用场景示意图；
图2a示出了本申请实施例提供的一种用户注意力检测的方法流程示意图；
图2b示出了本申请实施例提供的一种用户脑电信号获取过程中检测耳侧佩戴装置是否佩戴正常的流程示意图；
图2c示出了本申请实施例提供的一一种用户脑电信号获取过程中检测单侧耳侧佩戴装置是否佩戴正常的流程示意图；
图3示出了大脑产生的α、β、γ、θ、δ脑波形的示意图；
图4示出了本申请实施例提供的一种用户注意力检测的方法流程示意图；
图5示出了本申请实施例提供的一种用户脑电信号获取方法中的差分电路实现方式；
图6示出了本申请实施例提供的一种注意力检测方法中的左右耳脑电信号差分处理原理图；
图7示出了本申请实施例提供的一种注意力检测方法中的左右耳脑电信号差分处理示意图；
图8a示出了颈关节动作产生肌电伪迹；
图8b示出了眨眼产生的眼电伪迹；
图9a示出了SVM分类的原理示意图；
图9b示出了SVM分类的原理示意图；
图10示出了本申请实施例提供的一种注意力检测的系统结构示意图；
图11a示出了本申请实施例提供的一种耳侧佩戴装置的结构示意图；
图11b示出了本申请实施例提供的另一种耳侧佩戴装置的结构示意图；
图11c示出了本申请实施例提供的另一种耳侧佩戴装置的结构示意图；
图11d示出了本申请实施例提供的另一种耳侧佩戴装置的结构示意图；
图12示出了本申请实施例的一种耳侧佩戴装置的具体实现形式的示意图；
图13示出了本申请实施例的一种耳侧佩戴装置的佩戴位置示意图；
图14示出了本申请实施例提供的一种注意力检测装置的结构示意图；
图15a示出了本申请实施例提供的一种耳侧佩戴装置的结构示意图；
图15b示出了本申请实施例提供的一种注意力分析装置的结构示意图；
图16示出了本申请实施例提供的一种测量用户相关信号的方法的流程图；
图17示出了本申请实施例提供的一种注意力检测方法的流程图。
具体实施方式
下面结合本申请实施例中的附图对本申请的具体实现方式进行举例描述。然而本申请的实现方式还可以包括在不脱离本申请的精神或范围的前提下将这些实施例组合，比如采用其它实施例和做出结构性改变。因此以下实施例的详细描述不应从限制性的意义上去理解。本申请的实施例部分使用的术语仅用于对本申请的具体实施例进行解释，而非旨在限定本申请。
本申请的具体实施例中所提到的功能、模块、特征、单元等的一个或多个结构组成可以理解为由任何物理的或有形的组件(例如，由在计算机设备上运行的软件、硬件(例如，处理器或芯片实现的逻辑功能)等、和/或其它任何组合)以任何方式来实现。在某些实施例中，所示出的将附图中的将各种设备分成不同的模块或单元可以反映在实际实现中使用对应的不同的物理和有形的组件。可选的，本申请实施例附图中的单个模块也可以由多个实际物理组件来实现。同样，在附图中描绘的任何两个或更多个模块也可以反映由单个实际物理组件所执行的不同的功能。
关于本申请实施例的方法流程图，将某些操作描述为以一定顺序执行的不同的步骤。这样的流程图属于说明性的而非限制性的。可以将在本文中所描述的某些步骤分组在一起并且在单个操作中执行、可以将某些步骤分割成多个子步骤、并且可以以不同于在本文中所示出的顺序来执行某些步骤。可以由任何电路结构和/或有形机制(例如，由在计算机设备上运行的软件、硬件(例如，处理器或芯片实现的逻辑功能)等、和/或其任何组合)以任何方式来实现在流程图中所示出的各个步骤。
以下的说明可以将一个或多个特征标识为“可选的”。该类型的声明不应当被解释为对可以被认为是可选的特征的详尽的指示；即，尽管没有在文本中明确地标识，但其他特征可以被认为是可选的。此外，对单个实体的任何描述不旨在排除对多个这样的实体的使用；类似地，对多个实体的描述不旨在排除对单个实体的使用。最后，术语“示例性的”是指在潜在的许多实现中的一个实现。
本申请实施例主要用于用户注意力的检测，具体可以应用于驾驶过程中对驾驶员注意力的检测，判断驾驶员的注意力是否集中，从而能够根据判断结果进行即时的提示，此外也可以应用于其它需要对用户的注意力进行检测的场景。
图1为本发明实施例的一种典型的应用场景，耳侧佩戴装置101(具体可以为耳机或者耳塞)佩戴于用户耳部，从耳侧采集驾驶员生物电信号，将驾驶员生物电信号发送给用户注意力检测装置102，其中耳侧佩戴装置101具体操作可选的还可以包括 通过耳侧信号测量单元采集耳侧生物电信号，获取测量单元所采集生物电信号的电位差，用于增强信号同时排除外界杂扰信号干扰；进行去伪迹处理，并通过滤波电路滤除非脑电频率信号(比如滤除大于32Hz的波形)，并运用小波分析，提取波形特征，进行后续数字编码。注意力检测设备102(具体可以为手持终端，如手机，PDA，pad等，或者车载终端设备)对用户脑电信号进行分析。当判断发现注意力涣散时，进行相应的后续操作，如及时通过告警装置提醒驾驶员，确保行车安全，其中注意力分析方式可以是计算脑电信号的样本熵值，并通过SVN算法对样本熵进行分类，确定注意力状态。
同时为了确保耳侧佩戴装置101能够采集到准确的生物电信号，耳侧佩戴装置101还会对自身是否能够正常采集信号进行一个预判断，根据耳侧信号测量单元之间的阻抗值来判断耳侧信号测量单元是否贴合皮肤，从而分不同的情况选择不同的信号采集策略。
本发明实施例中的耳侧指的是人体耳朵上以及耳朵附近可以测量得到生物电信号的区域，如耳道内侧，耳廓，耳沟，耳背，以及耳周等位置。通过将耳侧信号测量单元部署在人耳区域上，以及人耳附近来采集生物电信号。
图13是本发明实施例的一种示例性的耳侧佩戴装置的佩戴即信号采集方式，示例性的给出了一种从耳道内侧获取生物电信号的信号采集方式。其中401为人体耳道，403为耳侧信号测量单元，402为耳侧佩戴装置的主体，404为用户的耳廓。
图2a为本申请实施例提供的一种用户脑电信号获取方法的流程示意图，具体流程包括：
S101：通过耳侧佩戴装置从用户耳侧采集用户生物电信号；
具体包括，佩戴上耳侧佩戴装置之后，开启设备的脑电信号采集功能，通过耳侧佩戴装置从耳侧采集用户的脑电信号。其中佩戴方式在上文中已经介绍过，在此不再赘述，开启设备的方式可以有多种，其中可以是摁下耳机上实体按钮，或者通过用户注意力检测装置(可以是手机，或者车载终端等)上的相应APP进行触发(如触摸APP中的开始驾驶的虚拟按钮)，使耳侧佩戴装置进入工作状态。
由于耳侧佩戴装置在配戴过程中可能会出现，脱落，或者没有正确配戴的问题，因此直接获取耳侧佩戴装置采集的信号进行处理，可能会存在因为设备脱落，或者没有正确配戴而导致测量结果不准确，从而不能正确分析出用户的注意力类型的问题，因此本申请的实施例会对耳侧佩戴装置的配戴情况进行判断，根据判断结果决定是否采集数据，或者是否将采集的数据用于进行用户注意力类型的分析。
具体包括所述耳侧佩戴装置包括多个耳侧信号测量单元；判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号的电位差值信号作为所述用户生物电信号。
对应于耳侧佩戴装置为双侧耳道测量，即耳侧佩戴装置包括左耳耳侧信号测量单元和右耳耳侧信号测量单元，从耳侧采集生物电信号的采集方式还可以进一步如图2b所示，包括：
S201：通过判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值，来判断耳侧佩戴装置是否可以正常进行测量。
判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值，以此来判断耳侧佩戴装置是否可以正常进行测量(即正常配戴)。
具体的，左耳耳侧信号测量单元和右耳耳侧信号测量单元可以为一个也可以为多个，左/右耳耳侧信号测量单元的在实现的过程中其形态可以是电极，通过电极来测量耳侧的用户生物电信号。通过判断左右耳耳侧信号测量单元之间的阻抗值来判断耳侧佩戴装置的左耳耳侧信号测量单元和所述右耳耳侧信号测量单元是否贴合耳道，即耳侧佩戴装置是否配戴正确。当左右耳耳侧信号测量单元贴合耳道时，左右耳耳侧信号测量单元之间的阻抗值较低，通常是低于耳侧表面阻抗值，当左右耳耳侧信号测量单元有一侧或者两侧均不贴合耳道时，左右耳耳侧信号测量单元之间的阻抗值较高，通常是高于耳侧表面阻抗值。因此可以设置一个预设阈值用于判断耳侧佩戴装置的配戴情况，可选的，阻抗预设的判断阈值可以是耳侧表面的阻抗值。
当左耳耳侧信号测量单元和右耳耳侧信号测量单元为一个时，直接获取两个测量单元之间的阻抗值进行判断。
当左耳耳侧信号测量单元和右耳耳侧信号测量单元为多个时，可以有多种测量策略。如任意选择一个左耳耳侧信号测量单元和一个右耳耳侧信号测量单元进行阻抗值的获取，也可以是获取预设位置的左耳耳侧信号测量单元和获取预设位置的右耳耳侧信号测量单元之间的阻抗值，只获取一次阻抗值，根据获取的阻抗值判断左右耳配戴是否正常。或者也可以是设置优先级顺序来进行逐个匹配的测量，在不满足预设阈值的情况下测量预设数量次后终止测量和判断，或者逐个进行测量直到测量到低于预设值的阻抗，则说明耳侧佩戴装置可以正常测量，否则说明无法正常工作。在此对多测量单元情况下具体的测量方法不做限定。
S202：当判断耳侧佩戴装置可以正常进行测量，根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
即当所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值，根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
具体为，当耳侧信号测量单元为左右各一个的情况下，当测量到左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值则判断耳侧佩戴装置可以正常进行测量；根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
其中根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号具体方式可以包括，直接将所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号；耳侧佩戴装置上还可以设置参考电极，分别获取所述左耳耳侧信号测量单元采集的生物电信号和参考电极的第 一电位差信号，以及所述右耳耳侧信号测量单元采集的生物电信号和参考电极的第二电位差信号，然后获取第一电位差信号和第二电位差信号的差值信号。
在耳侧信号测量单元为多个且测量次数可以为多次的情况下，当测量到左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值则判断所测量的两个测量单元对应的耳道均可以正常进行测量。将测量后判定为可以正常进行测量的所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
在耳侧信号测量单元为多个且测量次数为一次的情况下，当测量左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值则判断所测量的两个测量单元对应的耳道均可以正常进行测量。将测量后判定为可以正常进行测量的所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号，或者也可以基于测量结果认为耳侧佩戴装置为可以正常进行测量，则选择任意的左侧耳侧信号测量单元和右侧耳侧信号测量单元，或者预先指定的左侧耳侧信号测量单元和右侧耳侧信号测量单元来获取采集的生物电信号的电位差值信号作为所述用户生物电信号。
进一步当判断耳侧佩戴装置无法正常进行测量时，可以选择不再进行采集信号和注意力检测步骤。
可选的，也可以当判断耳侧佩戴装置没有正常配戴时，进一步分别判断左侧或者右侧有没有正常配戴，由此可选的，可以包括步骤S50103。
S203：当判断结果为耳侧佩戴装置无法正常进行测量，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值。将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
在此步骤下，左耳耳侧信号测量单元和右耳耳侧信号测量单元需要为多个。
在耳侧信号测量单元为多个且测量次数可以为多次的情况下，当多次测量左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗均高于预设阈值则判断所测量的两个测量单元对应的耳道至少有一个无法进行正常测量。
在耳侧信号测量单元为多个且测量次数为一次的情况下，当测量左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗高于预设阈值则判断所测量的两个测量单元对应的耳道至少有一个无法正常进行测量。即当其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，则判断所测量的两个测量单元对应的耳道至少有一个无法正常进行测量。
判断为所测量的两个测量单元对应的耳道至少有一个无法正常进行测量后，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值。同样可以有多种测量策略。如任意选择两个左/右耳耳侧信号测量单元进行阻抗值的获取，也可以是获取预设位置的两个左/右耳耳侧信号测量单元之间的阻抗值，只获取一次阻抗 值，根据获取的阻抗值判断左右耳配戴是否正常。或者也可以是设置优先级顺序来进行两个左/右耳耳侧信号测量单元之间的测量，在不满足预设阈值的情况下测量预设数量次后终止测量和判断，或者逐对进行测量直到测量到低于预设值的阻抗，则说明耳侧佩戴装置可以正常测量，否则当遍历所有情况均无测量到低于预设值的阻抗说明无法正常测量。在此同样本申请对单侧多测量单元情况下具体的测量方法不做限定。
将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号：
可以是将测量后判定为可以正常进行测量的所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
其中将测量后判定为可以正常进行测量的所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号具体方式可以包括，直接将两个可以正常测量一侧的耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号；耳侧佩戴装置上还可以设置参考电极，分别获取所述一个耳侧信号测量单元采集的生物电信号和参考电极的第三电位差信号，以及另一个耳侧信号测量单元采集的生物电信号和参考电极的第四电位差信号，然后获取第三电位差信号和第四电位差信号的差值信号。
其中耳侧信号测量单元的选取可以有多种发送，如直接选取进行阻抗值判断的两个测量单元来获取电位差值信号，也可以是按照预先设置来选择测量单元，或者是任意选取。
本申请中获取电位差值信号的方式具体可以是通过软件指令来实现，也可以是通过硬件电路来实现。
对应于耳侧佩戴装置为单侧测量，即耳侧佩戴装置仅包括左耳耳侧信号测量单元或右耳耳侧信号测量单元，在此实施方式下，左耳耳侧信号测量单元或右耳耳侧信号测量单元需要为多个，即单侧耳侧信号测量单元为多个。从耳侧采集生物电信号的采集方式还可以如图2c所示，进一步包括：
S211：通过判断所述单侧耳侧信号测量单元中的两个信号测量单元之间的阻抗是否低于预设阈值，来判断耳侧佩戴装置是否可以正常进行测量。
判断所述单侧耳侧信号测量单元中的两个信号测量单元之间的阻抗是否低于预设阈值，以此来判断耳侧佩戴装置是否可以正常进行测量(即正常配戴)。
具体的，可以有多种测量策略，如任意选择两个单侧耳侧信号测量单元进行阻抗值的获取，也可以是获取预设位置的两个单侧耳侧信号测量单元之间的阻抗值，只获取一次阻抗值，根据获取的阻抗值判断单侧耳道配戴是否正常。即判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值；低于阈值则判断为配戴正常。
或者也可以是设置优先级顺序来进行两个单侧耳侧信号测量单元之间的测量，在不满足预设阈值的情况下测量预设数量次后终止测量和判断，或者逐对进行测量直到测量到低于预设值的阻抗，则说明耳侧佩戴装置可以正常测量，否则当遍历所有情况 均无测量到低于预设值的阻抗说明无法正常测量。在此本申请同样对单侧多测量单元情况下具体的测量方法不做限定。
S212：当判断耳侧佩戴装置可以正常进行测量，将所述多个单侧耳侧信号测量单元中的两个采集的生物电信号的电位差值信号作为所述用户生物电信号。
在单个耳侧信号测量单元为多个且测量次数可以为多次的情况下，当测量到两个单侧耳侧信号测量单元之间的阻抗低于预设阈值则判断耳侧佩戴装置可以正常进行测量。将测量后判定为可以正常进行测量的两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
其中将测量后判定为可以正常进行测量的两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号具体方式可以包括，直接将两个可以正常测量耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号；耳侧佩戴装置上还可以设置参考电极，分别获取所述一个耳侧信号测量单元采集的生物电信号和参考电极的第五电位差信号，以及另一个耳侧信号测量单元采集的生物电信号和参考电极的第六电位差信号，然后获取第五电位差信号和第六电位差信号的差值信号。
在单侧耳侧信号测量单元为多个且测量次数为一次的情况下，当测量到单侧耳侧信号测量单元之间的阻抗低于预设阈值则判断所测量的两个测量单元对应的耳道均可以正常进行测量。将测量后判定为可以正常进行测量的两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号，或者也可以基于测量结果认为耳侧佩戴装置为可以正常进行测量，则选择任意的两个单侧耳道测量单元，或者预先指定的两个单侧耳道测量单元来获取采集的生物电信号的电位差值信号作为所述用户生物电信号。
上述实施方式中所提到的求电位差的处理具体可以是对所采集的生物电信号进行差分处理，因为从耳侧，尤其是耳道获取脑电信号难度较高，所获取的脑电信号强度相对较弱，受噪声干扰时对信号后续的判断可能会产生较大的影响，因此为了保证耳侧获取那电信号的可实施性，以及后续对于用户注意力分析的结论的准确性，需要对耳道采取的生物电信号进行有针对性的去噪处理，而差分电路则可以去除所采集生物电信号中的噪声。耳侧佩戴装置属于电子产品，在运行过程中，虽然电路已进行电磁屏蔽设计，但可能会在某种特殊场景下，受到电路板上电信号、空气中电磁波的影响，从而导致波形的失真，采用差分技术，在两个耳朵上都贴上电极，从两个耳道中采集信号，确保信号的正确性。
具体原理图如图6所示，是两个耳道上的信号接收电路模型，当外界噪声寄居在线路上时，通过差分电路，消除干扰，便于后续提取正确的波形，601为左耳道生物电信号，602为右耳道生物电信号，603为噪声信号，601a为参杂噪声后的左耳道生物电信号，602a为参杂噪声后的右耳道生物电信号，604为差分处理后得到的生物电信号即第一生物电信号。图6仅为一种情况的示例，也可以601为右耳道生物电信号，602为左耳道生物电信号。
当前差分电路设计上，一般采用芯片直接实现，本发明实施例实现的电路如图5所示，501、502为左右耳道采集的生物电信号的输入，503为经过差分电路差分处理后输出的第一生物电信号。波形示意见如图7所示，V+为左耳道生物电信号，V-为右耳道生物电信号，(V+)-(V-)为差分处理后的生物电信号。同样，图7仅为一种情况的示例，也可以V+为右耳道生物电信号，V-为左耳道生物电信号。
对于图2a中S101中获取的到的用户生物电信号，可以根据具体的应用需要，从中提取脑电信号。
S102：从所述用户生物电信号中获取用户的脑电信号；
所述生物电信号包括，了人体的多种特征信号如心电ECG信号，眼电EOG信号，肌电EMG信号以及脑电EEG信号中的一种或多种。可以通过进行特征分解来提取不同类的生物特征信号的方法有多种，可以根据不同类信号存在的频谱不同而进行提取，较为常用的是利用盲信源分离算法独立成分分析(ICA)分解得到多个生物特征信号的成分，通过这种方式提取脑电信号。
除了对脑电信号进行提去外。还可以对脑电信号选择性的进行一些常规的处理。所述处理可以包括去伪迹、小波分析、数字化编码等常规的生物电信号处理操作中的一个或者多个，用于得到更加准确真实能反映用户脑电特征的脑电信号。也可以是经过其他的去躁处理和数字换转换后的脑电信号，在此不做限定。其中各种处理操作的处理方式和作用为：
去伪迹：在人体内部，有很多地方都产生电现象，最常见的就是神经传导，一个神经元接收刺激后传递生物电给下一个神经元，这种电现象是随着人类的存活无时无刻都在发生的，人类的每一个微小的表情都与神经电流的传导息息相关。不仅神经细胞如此，人体内的器官也能产生不同程度、不同强弱的生物电信号，然而，在测量脑电时，其他生物电信号也同样的混杂其中，因为不能完全提取到初凝的脑电原始信号，基本都是混杂着来自人体不同的生物电，其影响或大或小。除此之外，人类的表情、肢体动作也能很大的影响脑电信号，如心脏跳动、肌肉动作、眨眼动作、深呼吸、皮肤排汗等等。同时，温度的差异也会噪声生物电信号不同程度的强弱变化，若环境温度较低，会引起少数人的寒颤及抖动，这些动作幅度都比较大，对脑电也可以造成干扰。图8a为颈关节动作产生肌电伪迹，图8b为眨眼产生的眼电伪迹。这些伪迹与有用的脑电信号混杂在一起，加大了数据处理的难度，所以在采集到脑电信号之后需要对脑电信号中的伪迹进行去除。小波分析：是一种时频分析方法，由于脑电信号是一种非稳态信号，传统的傅里叶变换无法很好的提取细节(只能提取频率信息，时间信息无法提取)，小波变换是一种信号的分析方法，能够在频域上较好的体现信号的时间特性，通过小波分析可以很好的表征信号局部特征。
数字化编码：是对脑电信号进行数字化编码，将脑电信号转化为数字信号。
其中去伪迹处理可以对生物电信号进行，也可以是针对特征提取的后脑电信号来进行。
所获取的脑电信号可以用于对用户的注意力进行分析判断，对提取的用户脑电信 号还可以执行S103步骤。
S103：根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型。具体包括：
根据所获取的脑电信号来分析用户的注意力类型，通常的处理方式是对所获取的脑电信号进行注意力特征提取。脑电信号即脑电EGG(electroencephalogram)信号，是大脑活动的外在体现，不同的大脑活动表现为具有不同特征的脑电信号。研究表明，通过检测到的脑电信号，可以清晰的检测到人处在在状态。
在平时的人类活动中，会产生α、β、γ、θ、δ波段，波形情况如图3所示。
δ波：频率分布在1Hz到4Hz之间，波幅在20uv到200uv之间，在鼎业和脑垂体的表现比较明显，在婴儿时期或者智力发育还不成熟的时期表现较为显著，和θ波一样，也属于慢波。正常情况下，δ波只存在于极度缺乏氧气，深睡状态或者存在脑科病变等状态。
θ波：频率分布在4Hz到7Hz之间，幅度在20uf到40uf之间，属于慢波，主要出现在枕部和顶枕部，且左右两边对称，在人困倦或者浅睡时一般可以测到，同时与人的心理状态存在着普遍的联系。通常在情绪低落、遇到挫折、或者困倦时，中枢神经处于抑制状态下该波会出现。
α波：频率分布在8Hz到12Hz之间，波幅在25uf到75uf之间，主要出现在顶枕部，并且双侧基本保持同步，是正常人EEG该有的基本节律。处于思考、静息状态时改波表现较为明显，当个体发生具有目标性的活动、睁眼或收到其他刺激时，该波会消失，β波取而代之。
β波：频率分布于14Hz到30Hz之间，波幅约为δ波的一半，主要出现在额头部位和中央区，该波的频率显著代表了大脑皮层的亢奋层度，个体在清醒状态和睡觉期会出现。
因此通过分析所获取脑电信号的波形特征可以判断出用户当前的注意力状态，是清醒还是睡眠，是注意力集中还是不集中。
在本申请的具体实施例S103步骤又可以进一步划分为如图4所示：
S1031：基于脑电信号来获取样本熵。
基于脑电信号来获取样本熵的获取过程包括：
A：截取预设时间长度的所述脑电信号，从所述预设时间长度的脑电信号获得N个信号采样点，u(1),u(2),...,u(N)。
通常所述采样点为等时间间隔采样点，中截取的预设时间长度可选的。
B：基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；
所构建的N-m+1个m维向量为X(1),X(2),...,X(N-m+1)，其中X(i)＝[u(i),u(i+1),...,u(i+m-1)]，1≤i≤N-m+1；m<N；
C：针对N-m+1个向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值 得到第一平均值。
针对N-m+1个向量中的每一个m维向量，统计满足以下条件的向量个数：
Bi(r)＝(number of X(j)such that d[X(i),X(j)]≤r)/(N-m),i≠j，i的取值范围为[1,N-m+1]，j的取值范围为除i以外的[1,N-m+1]，r为一个预设值，例如，r的取值可以和上述采样点的标准差δ值相关，取值可以在0.1δ至0.3δ之间。其中，d[X(i),X(j)]定义为d[X(i),X(j)]＝max|u(a)-u*(a)|,i≠j；u(a)为向量X(i)的元素，u*(a)为向量X(j)的对应维度的元素，d表示向量X(i)与X(j)的距离，向量X(i)与X(j)的距离由对应元素的差值中的最大差值决定，例如X(1)＝[2，3，4，6]，X(2)＝[4，5，7，10]，则对应元素的最大差值为|6-10|＝4，因此d[X(1),X(2)]＝4。求Bi(r)对所有i值的平均值，记为Bm(r)，即
Figure PCTCN2020071565-appb-000001
D：基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；
所构建的N-m个m+1维向量为Y(1),Y(2),...,Y(N-m)，其中X(i)＝[u(i),u(i+1),...,u(i+m)]，1≤i≤N-m；m<N；
E：针对N-m个向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值。
针对N-m个向量中的每一个m+1维向量，统计满足以下条件的向量个数：
Ai(r)＝(number of Y(j)such that d[Y(i),Y(j)]≤r)/(N-m-1),i≠j，i的取值范围为[1,N-m]，j的取值范围为除i以外的[1,N-m]，r为一个预设值，例如，r的取值可以和上述采样点的标准差δ值相关，取值可以在0.1δ至0.3δ之间。其中，d[Y(i),Y(j)]定义为d[Y(i),Y(j)]＝max|u(a)-u*(a)|,i≠j；u(a)为向量Y的元素，d表示向量Y(i)与Y(j)的距离，由对应元素的最大差值决定。求Ai(r)对所有i值的平均值，记为Am(r)，即
Figure PCTCN2020071565-appb-000002
F：基于第一平均值与第二平均值比值来计算样本熵(SampEn)的值。
SampEn＝lim(N→∞){-ln[Am(r)/Bm(r)]}。
其中A-F的顺序并不固定，比如B、C和D、E的实施并没有固定的先后顺序，D、E可以在B、C之前进行也可以同时进行，或者时间上部分重叠的实施。
S1032：基于所采集的脑电信号获得的样本熵的值判断用户的注意力状态。
根据所获得的样本熵的值来判断用户的注意力的情况。具体实现过程中可以有多种实现方式，如，可以是用户或者产品研发人员根据历史经验设定一个或多个预设数值，如用于区别注意力集中还是涣散的分割值，用于区分是清醒还是睡眠的分割值。如对于用于区别注意力集中还是涣散的分割值，大于或大于等于这个分割值时表示注意力集中，小于等于或小于这个分割值时表示注意力分散。所述分割值的大小以及个数根据所要区分的注意力状态的数量，以及注意力状态的类型来确定。
此外，也可以通过SVM分类器，用于机器学习模型训练的方式获得分割值，并根据所述分割值和所述样本熵值判断用户的注意力类型。
所述模型训练方式，为采用多种一直不同注意力类型的一定时长的脑电信号样本，并计算获取脑电信号样本的样本熵值，通过样本熵值和对应的注意力类型构建的样本来进行SVM模型的训练。然后将训练完成的模型用于后续的注意力分析，即输入相应脑电信号的样本熵值，输出对应的注意力类型，或注意力类型的概率。
SVM是一个由分类超平面定义的判别分类器，通过给定一组带标签的训练样本，算法将会输出一个最优超平面对新样本(测试样本)进行分类。图9a，9b是一个最优超平面获取的示意图，圆点和方块代表两类不同的数据，对于一个由二维坐标点构成的线性可分集合，如果能找到一条分割线离两类样本点都尽可能的远，则认为是这个二维坐标空间的最优超平面，即图9b中的实线。SVM机器学习的目的就是找一个超平面，并且它在区分两类数据的同时能够做到离他最近的训练样本的距离要最大。即最优分割超平面最大化训练样本边界。
通过SVM机器学习方式，输入多个样本熵取值及其对应的注意力状态后，SVM分类器会输出一个或多个分割值，用于判断用户的脑电信号的样本熵所对应的注意力状态。分割值可以是一个或者多个，如用于区别注意力集中还是涣散的分割值，用于区分是清醒还是睡眠的分割值。如对于用于区别注意力集中还是涣散的分割值，大于或大于等于这个分割值时表示注意力集中，小于等于或小于这个分割值时表示注意力分散。样本熵分析方法只需要较短数据就可得出稳健的估计值,是一种具有较好的抗噪和抗干扰能力的注意力分析方法。
图10是本发明实施例示例性的一种注意力检测系统图。包括：
其中耳侧佩戴装置11000，用于从耳侧采集的用户生物电信号；从所述用户生物电信号获得脑电信号；
耳侧佩戴装置11000具体可以分为单侧或双侧测量，其中耳侧佩戴装置11000为单侧测量其结构如图11a所示，其中单侧耳侧信号测量单元1011用于从左耳道或者右耳道获取用户生物电信号。
耳侧佩戴装置11000为双侧测量其结构如图11b所示，其中左耳耳侧信号测量单元101a用于从左耳道获取用户生物电信号，右耳耳侧信号测量单元101b用于从右耳道获取用户生物电信号。
对应于双侧测量的耳侧佩戴装置11000，即耳侧佩戴装置包括左耳耳侧信号测量单元101a和右耳耳侧信号测量单元101b，耳侧佩戴装置11000通过判断所述左耳耳侧信号测量单元101a和所述右耳耳侧信号测量单元101b之间的阻抗是否低于预设阈值， 来判断耳侧佩戴装置是否可以正常进行测量，当判断耳侧佩戴装置可以正常进行测量，根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号来获取所述用户生物电信号，当判断结果为耳侧佩戴装置无法正常进行测量，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值。根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号来获取所述用户生物电信号。具体的判断方式可以参数步骤S201-S6203。
对应于耳侧佩戴装置11000为单侧耳道测量，即耳侧佩戴装置11000仅包括左耳耳侧信号测量单元或右耳耳侧信号测量单元1011，在此实施方式下，左耳耳侧信号测量单元或右耳耳侧信号测量单元需要为多个，即单侧耳侧信号测量单元为多个。耳侧佩戴装置11000通过判断所述单侧耳侧信号测量单元中的两个信号测量单元之间的阻抗是否低于预设阈值，来判断耳侧佩戴装置是否可以正常进行测量，当判断耳侧佩戴装置可以正常进行测量，根据所述多个单侧耳侧信号测量单元中的两个采集的生物电信号的电位差值信号来获取所述用户生物电信号。具体的判断方式参考步骤S211-S212。注意力检测装置1200，根据所述脑电信号检测所述用户的注意力类型。
其中耳侧佩戴装置11000具体采集脑电信号的细节在图2的S101，S102中已经详细描述过，同样注意力检测装置1200用于对用户注意力检测的技术细节也已经在图2的S102，以及图4中详细描述过，在此不再赘述。
图11c是本发明实施例示例性的一种具备注意力检测能力的耳侧佩戴装置11000的结构图，在本发明的有些实施例中，还可以将注意力检测装置和耳侧佩戴装置集成在一起，如图11所示，对应于本申请实施例的一种集成了注意力检测功能的耳侧佩戴装置1100。装置包括：
耳侧信号测量单元111，用于从耳侧采集的用户生物电信号。可选的，所述耳侧信号测量单元111可以包括左耳耳侧信号测量单元111a和右耳耳侧信号测量单元111b。当耳侧佩戴装置1100为单侧测量装置时，耳侧信号测量单元111可以仅包括单侧耳侧信号测量单元111c。
特征分解单元112，用于从所述用户生物电信号中提取脑电信号。
注意力检测单元113，用于根据所述脑电信号基于机器学习模型来获得所述用户的注意力分类结果。具体分析方式可以参考上文具体实施例，在此不再赘述。
第一判断单元114，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号。
对应于双侧测量情况，所述耳侧佩戴装置还可以可选的包括第二判断单元。
第一断单元114，用于判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)； 当所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值；将所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
第二判断单元115，用于当第一判断单元114判断所述耳侧佩戴装置无法正常测量(具体判断方式上文中已经介绍，在此不再赘述)，所述第二判断单元115分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。所述第二判断单元115为可选单元，第二判断单元115应用于所述左耳耳侧信号测量单元，右耳耳侧信号测量单元均为多个的情况。
对应于单侧测量情况，所述耳侧佩戴装置的包括第一判断单元114，用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述多个单侧耳侧信号测量单元中的两个采集的生物电信号的电位差值信号作为所述用户生物电信号。
本发明实施例还公开了一种测量用户脑电信号的方法，如图16所示：其中步骤S1601，S1602同图2相同，S1603为将所述脑电信号发送给信号分析装置，信号分析装置在本申请实施例中具体可以为注意力检测装置。
对应的本发明实施例还公开一种用于测量用户相关信号的耳侧佩戴装置，如图11d所示，耳侧信号测量单元121，用于从耳侧采集的用户生物电信号。可选的，所述耳侧信号测量单元121可以包括左耳耳侧信号测量单元121a和右耳耳侧信号测量单元121b。当耳侧佩戴装置120为单侧测量装置时，耳侧信号测量单元121可以仅包括单侧耳侧信号测量单元121c。
特征分解单元122，用于从用户生物电信号获取脑电信号。
发送单元123，用于将所述生物特征信号发送给信号分析装置；信号分析装置在本申请实施例中具体可以为注意力检测装置。
第一判断单元124，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号。
对应于双侧测量情况，所述耳侧佩戴装置还可以可选的包括第二判断单元125。
第一断单元124，用于判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)；当所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值；将所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
第二判断单元125，用于当第一判断单元124判断所述耳侧佩戴装置无法正常测 量(具体判断方式上文中已经介绍，在此不再赘述)，所述第三判断单元分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。所述第二判断单元125为可选单元，第二判断单元125应用于所述左耳耳侧信号测量单元，右耳耳侧信号测量单元均为多个的情况。
对应于单侧测量情况，所述耳侧佩戴装置的第一判断单元124，用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述多个单侧耳侧信号测量单元中的两个采集的生物电信号的电位差值信号作为所述用户生物电信号。
图12为图11中的耳侧佩戴装置的一种具体产品结构示意图，耳侧佩戴装置可以有多种形态，例如可以是耳机形态，也可以是耳塞的形态，本示例给出的耳侧佩戴装置是耳塞形态，但在本申请中并不作为限定，耳侧佩戴装置包括耳塞本体301，柔性电极载体302和多个表面柔性电极303。柔性电极载体302提供一个足够弹性的支撑，确保附着在柔性电极载体302表面的多个柔性电极303和用户的耳侧表面形成紧密贴合，确保稳定地采集用户的脑电波信号。310部分示例性地呈现了一种表面柔性电极303的构成，包括呈现等角120度分布的生物感测柔性电极303A，生物感测柔性电极303B，接地的公共柔性电极303G，304是耳塞孔。针对另一些可行的实施例，附着在柔性电极载体302表面的生物感测柔性电极303可以只有1个或2个，而将耳塞本体301连接接地的公共柔性电极。或者在其他一些可行的实施例中，接地的公共柔性电极也可以采取耳廓支架上的电极触电来实现等多种方式。图12中耳塞形态的耳侧佩戴装置的佩戴示意图如图13所示，其中401为用户的耳道，402为脑电信号测量的入耳式的耳塞，403为柔性电极，404为用户的耳廓。从图3可以看出，佩戴时柔性电极载体表面的多个柔性电极403和用户的耳道401内表面形成紧密贴合，和用户的头部形成一个测量系统。虽然图中没有示出，耳侧佩戴装置还可以，包括通信模块用于接收或者发送脑电信号，还可以可选的包括注意力检测单元用于通过脑电信号分析用户的注意力类型。
其中图11a-d中的耳侧信号测量单元可以选择通过柔性电极的方式来实现。
本发明实施例还公开了一种分析用户相关信号的方法，如图17所示：其中步骤S1702同图7中的S603相同，S1701为从耳侧佩戴装置接收脑电信号。
对应的，本发明实施例还公开一种注意力检测装置130，如图14所示。装置包括：
接收单元131，用于从耳侧佩戴装置接收脑电信号；
注意力检测单元132，用于根据所述脑电信号获得所述用户的注意力类型。
相应的，注意力检测单元132，根据所述脑电信号分析所述用户的注意力类型，所述注意力检测单元可以是用户的终端设备如手机，或其他可佩戴或便携终端，也可以是设置于云端的服务器中。
所述注意力检测单元包括样本熵获取模块，注意力识别模块。
样本熵获取模块，用于基于脑电信号来获取样本熵；
基于脑电信号来获取样本熵的获取过程包括：
A：截取预设时间长度的所述脑电信号，从所述预设时间长度的脑电信号获得N个信号采样点，u(1),u(2),...,u(N)。
通常所述采样点为等时间间隔采样点，从所述脑电信号中截取的预设时间长度可以根据分析的需要来进行设定。
B：基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；
所构建的N-m+1个m维向量为X(1),X(2),...,X(N-m+1)，其中X(i)＝[u(i),u(i+1),...,u(i+m-1)]，1≤i≤N-m+1；m<N；
C：针对N-m+1个向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值。
针对N-m+1个向量中的每一个m维向量，统计满足以下条件的向量个数：
Bi(r)＝(number of X(j)such that d[X(i),X(j)]≤r)/(N-m),i≠j，i的取值范围为[1,N-m+1]，j的取值范围为除i以外的[1,N-m+1]，r为一个预设值，例如，r的取值可以和上述采样点的标准差δ值相关，取值可以在0.1δ至0.3δ之间。其中，d[X(i),X(j)]定义为d[X(i),X(j)]＝max|u(a)-u*(a)|,i≠j；u(a)为向量X(i)的元素，u*(a)为向量X(j)的对应维度的元素，d表示向量X(i)与X(j)的距离，向量X(i)与X(j)的距离由对应元素的差值中的最大差值决定，例如X(1)＝[2，3，4，6]，X(2)＝[4，5，7，10]，则对应元素的最大差值为|6-10|＝4，因此d[X(1),X(2)]＝4。求Bi(r)对所有i值的平均值，记为Bm(r)，即
Figure PCTCN2020071565-appb-000003
D：基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；
所构建的N-m个m+1维向量为Y(1),Y(2),...,Y(N-m)，其中X(i)＝[u(i),u(i+1),...,u(i+m)]，1≤i≤N-m；m<N；
E：针对N-m个向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值。
针对N-m个向量中的每一个m+1维向量，统计满足以下条件的向量个数：
Ai(r)＝(number of Y(j)such that d[Y(i),Y(j)]≤r)/(N-m-1),i≠j，i的取值范围为[1,N-m]，j的取值范围为除i以外的[1,N-m]，r为一个预设值，例如，r 的取值可以和上述采样点的标准差δ值相关，取值可以在0.1δ至0.3δ之间。其中，d[Y(i),Y(j)]定义为d[Y(i),Y(j)]＝max|u(a)-u*(a)|,i≠j；u(a)为向量Y的元素，d表示向量Y(i)与Y(j)的距离，由对应元素的最大差值决定。求Ai(r)对所有i值的平均值，记为Am(r)，即
Figure PCTCN2020071565-appb-000004
F：基于第一平均值与第二平均值比值来计算样本熵(SampEn)的值。
SampEn＝lim(N→∞){-ln[Am(r)/Bm(r)]}。
其中A-F的顺序并不固定，比如B、C和D、E的实施并没有固定的先后顺序，D、E可以在B、C之前进行也可以同时进行，或者时间上部分重叠的实施。
注意力识别模块，用于基于所采集的脑电信号获得的样本熵的值判断用户的注意力状态。
其中注意力识别模块可以包括：
SV分类器，用于机器学习获得分割值；具体可以是通过SVM机器学习方式，输入多个样本熵取值及其对应的注意力状态后，SVM分类器会输出一个或多个分割值，用于判断用户的脑电信号的样本熵所对应的注意力状态。
其中SVM分类器可以设置与注意力识别模块中，也可以设置与其他装置中进行训练获得分割值，然后将分割值发送给注意力识别模块，或者由用户或者开发人员根据训练结果手动进行设置。
判断模块，用于根据所述分割值和所述样本熵值判断用户的注意力类型。
分割值可以是一个或者多个，如用于区别注意力集中还是涣散的分割值，用于区分是清醒还是睡眠的分割值。如对于用于区别注意力集中还是涣散的分割值，大于或大于等于这个分割值时表示注意力集中，小于等于或小于这个分割值时表示注意力分散。
其中具体技术实现细节在实现上可以采用图2中的相关说明。
注意力检测装置130的具体实现形式可以是手持终端，或是车载终端，或是其他可用于计算和分析脑电信号的装置。
图15a，是对应于是本申请实施例耳侧佩戴装置的处理器结构示意图。
如图15a所示集成了注意力检测功能的耳侧佩戴装置1400可以包括一个或者多个处理器1406、一个或多个存储器1401，特征分解单元1403。具体实现中，耳侧佩戴装置还可以进一步包括通信单元1405。处理器1406可通过总线分别连接存储器1401、测量电极1402，特征分解电路1403等部件。分别描述如下：
处理器1406是耳侧佩戴装置的控制中心，利用各种接口和线路连接耳侧佩戴装置的各个部件，在可能实施例中，处理器1406还可包括一个或多个处理核心。处理器1400可通过执行程序指令来判断测量电极是否能正常测量(耳侧佩戴装置是否能 正常测量)，以及根据测量信号进行用户注意力分析。当处理器1406可以为专用处理器也可以为通用处理器，当处理器1406为通用处理器时，处理器1406通过运行或执行存储在存储器1401内的软件程序(指令)和/或模块。
存储器1401可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地，存储器1401还可以包括存储器控制器，以提供处理器1400和输入单元对存储器1401的访问。存储器1401可具体用于存储软件程序(指令)、以及采集的用户生物电信号。
耳侧信号测量单元1402，用于从耳侧采集的用户生物电信号。可选的，所述耳侧信号测量单元1402可以包括左耳耳侧信号测量单元和右耳耳侧信号测量单元。当耳侧佩戴装置1400为单侧测量装置时，耳侧信号测量单元1402可以仅包括单侧耳侧信号测量单元。耳侧信号测量单元1402通常通过硬件方式来实现，例如耳侧信号测量单元1402可以为电极，耳侧信号测量单元1402可以为一个或者多个。
特征分解单元1403，用于从用户生物电信号获取脑电信号。特征分解单元1403通常通过硬件方式来实现，例如特征分解电路，ICA组件。
通信单元1405用于通过无线或有线通信技术和耳侧佩戴装置以及其他设备之间进行通信连接，如蜂窝移动通信技术，WLAN，蓝牙等。
本领域技术人员可以理解，本申请实施例中耳侧佩戴装置可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。例如，耳侧佩戴装置还可以进一步包括扬声器、摄像头等，在此不再赘述。
具体的，处理器1406可通过读取，并分析判断存储在存储器1401中的测量信号来判断测量电极是否能正常测量(耳侧佩戴装置是否能正常测量)，以及根据测量信号进行用户注意力分析。包括：
对应于双侧测量情况，处理器1406用于判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)；当所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号；当判断所述耳侧佩戴装置无法正常测量(具体判断方式上文中已经介绍，在此不再赘述)，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号获取所述用户生物电信号，电位差值信号可以通过处理器1406执行指令来获取，也通过电位差值获取单元即硬件电路来实现。
对应于单侧测量情况，处理器1406用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，根据所述多个单侧耳侧信号测量单元中的两个采集的生物电信号的电位差值信号获取所述用户生物电信号。
处理器1406还用于根据所述脑电信号获得所述用户的注意力类型。具体分析方式可以参考上文具体实施例，在此不再赘述。
还需要说明的是，虽然图14仅仅是本申请耳侧佩戴装置的一种实现方式，所述耳侧佩戴装置中处理器1406和存储器1401，在可能的实施例中，还可以是集成部署的。
图14还可以为本发明实施例的一种用于测量用户脑电信号的耳侧佩戴装置可以包括一个或者多个处理器1406、一个或多个存储器1401，耳侧信号测量单元1402，特征分解单元1403。具体实现中，耳侧佩戴装置还可以进一步包括通信单元1405(包括发送单元和接收单元)。处理器1406可通过总线分别连接存储器1401、测量电极1402，特征分解电路1403等部件。分别描述如下：
处理器1406是耳侧佩戴装置的控制中心，利用各种接口和线路连接耳侧佩戴装置的各个部件，在可能实施例中，处理器1406还可包括一个或多个处理核心。处理器1400可通过执行程序指令来判断测量电极是否能正常测量(耳侧佩戴装置是否能正常测量)。当处理器1406可以为专用处理器也可以为通用处理器，当处理器1406为通用处理器时，处理器1406通过运行或执行存储在存储器1401内的软件程序(指令)和/或模块。
存储器1401可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地，存储器1401还可以包括存储器控制器，以提供处理器1400和输入单元对存储器1401的访问。存储器1401可具体用于存储软件程序(指令)、以及采集的用户生物电信号。
耳侧信号测量单元1402，用于从耳侧采集的用户生物电信号。可选的，所述耳侧信号测量单元1402可以包括左耳耳侧信号测量单元和右耳耳侧信号测量单元。当耳侧佩戴装置1400为单侧测量装置时，耳侧信号测量单元1402可以仅包括单侧耳侧信号测量单元。耳侧信号测量单元1402通常通过硬件方式来实现，例如耳侧信号测量单元1402可以为电极，耳侧信号测量单元1402可以为一个或者多个。
可选的，在某些实施例中还可以包括特征分解单元1403，用于从用户生物电信号获取脑电信号。特征分解单元1403通常通过硬件方式来实现，例如特征分解电路，ICA组件。
通信单元1405用于通过无线或有线通信技术和耳侧佩戴装置以及其他设备之间进行通信连接，如蜂窝移动通信技术，WLAN，蓝牙等，用于将生物电信号或采集并处理后的脑电信号发送给信号分析装置；信号分析装置在本申请实施例中具体可以为注意力检测装置。除了注意力检测装置外，由于获取的脑电信号还可以应用于用户其他特征的分析，例如睡眠状态，情绪状态的识别，因此信号分析装置还可以是睡眠测装置，情绪检测装置等其他的需要通过脑电信号分析来获得信息的装置。
本领域技术人员可以理解，本申请实施例中耳侧佩戴装置可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。例如，耳侧佩戴装置还可以进一步包括扬声器、摄像头等，在此不再赘述。
具体的，处理器1406可通过读取，并分析判断存储在存储器1401中的测量信号来判断测量电极是否能正常测量(耳侧佩戴装置是否能正常测量)，以及根据测量信号进行用户注意力类型的分析。包括：
对应于双侧测量情况，处理器1406用于判断所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)；当所述左耳耳侧信号测量单元和所述右耳耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳耳侧信号测量单元采集的生物电信号和所述右耳耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号；当判断所述耳侧佩戴装置无法正常测量(具体判断方式上文中已经介绍，在此不再赘述)，分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号获取所述用户生物电信号，电位差值信号可以通过处理器1406执行指令来获取，也通过电位差值获取单元即硬件电路来实现。
对应于单侧测量情况，处理器1406用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值(具体判断方式上文中已经介绍，在此不再赘述)，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，根据所述多个单侧耳侧信号测量单元中的两个采集的生物电信号的电位差值信号获取所述用户生物电信号。
同样，图14仅仅是本申请耳侧佩戴装置的一种实现方式，所述耳侧佩戴装置中处理器1406和存储器1401，在可能的实施例中，还可以是集成部署的。
图15b是本申请实施例提供的一种注意力检测装置的另一种终端形式的结构示意图，如图15所示，注意力检测装置可以包括一个或者多个处理器1500、一个或多个存储器1501。具体实现中，注意力检测装置还可以进一步包括输入单元1506、显示单元1503，通信单元1502等部件，处理器2011可通过总线分别连接存储器1501、通信单元1502、输入单元1506、显示单元1503等部件。分别描述如下：
处理器1500是注意力检测装置的控制中心，利用各种接口和线路连接注意力检测装置的各个部件，在可能实施例中，处理器1500还可包括一个或多个处理核心。处理器1500可通过执行程序指令来进行脑电信号的注意力检测。当处理器1500可以为专用处理器也可以为通用处理器，当处理器1500为通用处理器时，处理器1500通过运行或执行存储在存储器1501内的软件程序(指令)和/或模块。
存储器1501可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。相应地，存储器1501还可以包括存储器控制器，以提供处理器1500和输入单元1506对存储器1501的访问。存储器1501可具体用于存储软件程序(指令)、以及脑电信号。
输入单元1506可用于接收用户输入的数字或字符信息，以及产生与用户设置以及功能控制有关的键盘、鼠标、操作杆、光学或者轨迹球信号输入。具体地，输入单 元1506可包括触敏表面1505以及其他输入设备1507。触敏表面1505也称为触摸显示屏或者触控板，可收集用户在其上或附近的触摸操作，并根据预先设定的程式驱动相应的连接装置。具体地，其他输入设备1507可以包括但不限于物理键盘、功能键、轨迹球、鼠标、操作杆等中的一种或多种。
显示单元1503可用于显示由用户输入的检索请求或检索装置提供给用户的检索结果以及检索装置的各种图形用户接口，这些图形用户接口可以由图形、文本、图标、视频和其任意组合来构成。具体的，显示单元1503可包括显示面板1504，可选的，可以采用液晶显示器(Liquid Crystal Display，LCD)、有机发光二极管(Organic Light-Emitting Diode,OLED)等形式来配置显示面板1504。虽然在图15中，触敏表面1505与显示面板1504是作为两个独立的部件，但是在某些实施例中，可以将触敏表面1505与显示面板1504集成而实现输入和输出功能。例如，触敏表面1505可覆盖显示面板1504，当触敏表面1505检测到在其上或附近的触摸操作后，传送给处理器1500以确定触摸事件的类型，随后处理器1500根据触摸事件的类型在显示面板1504上提供相应的视觉输出。
通信单元1502用于通过无线或有线通信技术和耳侧佩戴装置以及其他设备之间进行通信连接，如蜂窝移动通信技术，WLAN，蓝牙等。用于接收耳侧佩戴装置发送的脑电信号，以及可能根据判断结果向耳侧佩戴装置返回提示信号，或者是直接通过扬声器进行提示，或者通过显示单元1503显示提醒界面。
本领域技术人员可以理解，本申请实施例中检索装置可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。例如，检索装置还可以进一步包括扬声器、摄像头等，在此不再赘述。
具体的，处理器1500可通过读取，并分析判断存储在存储器1501中的脑电信号来实现本申请实施例的步骤S103中的基于所述脑电信号检测所述用户的注意力类型。包括：
基于脑电信号来获取样本熵，所述样本熵的获取过程在上文中已经详细介绍，因此在此不再赘述。
基于所采集的脑电信号获得的样本熵的值判断用户的注意力状态。
根据SVM分类器通过机器学习获得分割值和所述样本熵值判断用户的注意力类型。
处理器1500执行用户注意力分析的方法的具体实施过程可参考前文的方法实施例，这里不再赘述。
还需要说明的是，虽然图15b仅仅是本申请检索装置的一种实现方式，所述检索装置中处理器1500和存储器1501，在可能的实施例中，还可以是集成部署的。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者任意组合来实现。当使用软件实现时，可以全部或者部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令，在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述处理器可以是通用处理器或者专用处理器。所述检索装置可以是一个，也可以是多个检索装置组成的计算机网络。所述计算机指令可存储在计算机可读存储介质中，或者从一个计算机可 读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网络站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线)或无线(例如红外、微波等)方式向另一个网络站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质，也可以是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如软盘、硬盘、磁带等)、光介质(例如DVD等)、或者半导体介质(例如固态硬盘)等等。
示例性的，本申请实施例的方案，执行主体可选的可以为ASIC、FPGA、CPU、GPU等，通过硬件或软件方式实现，存储器可选的可以为DDR、SRAM、HDD、SSD等易失或非易失性的存储设备。所述数据检索装置可以应用于多种场景，例如用于视频监控系统的服务器上，示例性的可以是以PCIe扩展卡的形式。
其中ASIC、FPGA属于硬件实现，即在硬件设计时通过硬件描述语言的方式将本申请的方法落地；CPU、GPU属于软件实现，即在软件设计时通过软件程序代码的方式将本申请的方法落地。
在上述实施例中，对各个实施例的描述各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。

Claims (26)

  1. 一种用户注意力检测方法，其特征在于，所述方法包括：
    通过耳侧佩戴装置，从用户耳侧采集用户生物电信号；
    从所述用户生物电信号中获取用户脑电信号；
    根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型；
    其中通过耳侧佩戴装置，从用户耳侧采集用户生物电信号具体包括：
    所述耳侧佩戴装置包括多个耳侧信号测量单元；判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号。
  2. 根据权利要求1所述的方法，其特征在于：
    所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；
    所述判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
    判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
  3. 根据权利要求1所述的方法，其特征在于：
    所述耳侧佩戴装置为单侧耳侧佩戴装置，所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；
    所述判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
    判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值；
    当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值；
    根据所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
  4. 根据权利要求2所述的方法，其特征在于，所述方法包括：
    所述左耳侧信号测量单元为多个；
    所述右耳侧信号测量单元为多个；
    当其中一个左耳侧信号测量单元和其中一个右耳侧信号测量单元之间的阻抗高于预设阈值；
    分别判断所述多个左耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值， 以及所述多个右耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；
    根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置测量的生物电信号的电位差值信号获取所述用户生物电信号。
  5. 根据权利要求1-4任意一项所述的方法，其特征在于，所述方法包括：
    所述根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
    通过差分电路获取所述两个耳侧信号测量单元采集的生物电信号的电位差值信号，并将所述电位差值信号获取所述用户生物电信号。
  6. 根据权利要求1-4任意一项所述的方法，其特征在于，所述根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：
    计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
  7. 根据权利要求1-6任意一项所述的方法，其特征在于，所述基于所述用户脑电信号检测所述用户的注意力类型具体为：
    截取预设时间长度的所述用户脑电信号，从所述预设时间长度的所述用户脑电信号获得N个信号采样点；
    所述N个信号采样点为，u(1),u(2),...,u(N)；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；针对所述N-m+1个m维向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；针对所述N-m个m+1维向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值；基于所述第一平均值与所述第二平均值比值来计算样本熵(SampEn)的值。
  8. 根据权利要求6或7所述的方法，其特征在于，所述机器学习模型为SVM分类器；
    采用SVM分类器进行机器学习获得分割值，根据所述分割值和所述样本熵值判断用户的注意力类型。
  9. 一种用户注意力检测系统，其特征在于，所述系统包括：
    耳侧佩戴装置，用于从用户耳侧采集用户生物电信号；用于从所述用户生物电信号中获取用户脑电信号；
    注意力检测装置，用于基于所述用户脑电信号检测所述用户的注意力类型；
    其中所述耳侧佩戴装置，用于从用户耳侧采集用户生物电信号具体包括：
    所述耳侧佩戴装置包括多个耳侧信号测量单元；所述耳侧佩戴装置判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号。
  10. 根据权利要求9所述的系统，其特征在于：
    所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；
    所述耳侧佩戴装置判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
    所述耳侧佩戴装置判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
  11. 根据权利要求9所述的系统，其特征在于：
    所述耳侧佩戴装置为单侧耳侧佩戴装置，所述单侧耳侧佩戴装置包括多个单侧耳侧信号测量单元；
    所述耳侧佩戴装置判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，从所述两个耳侧信号测量单元采集生物电信号；根据所述两个耳侧信号测量单元采集的生物电信号电信号的电位差值信号获取所述用户生物电信号，具体为：
    所述耳侧佩戴装置判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值；根据所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号获取所述用户生物电信号。
  12. 根据权利要求10所述的系统，其特征在于，
    所述左耳侧信号测量单元为多个；
    所述右耳侧信号测量单元为多个；
    当其中一个左耳侧信号测量单元和其中一个右耳侧信号测量单元之间的阻抗高于预设阈值；
    所述耳侧佩戴装置判断分别判断所述多个左耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；根据阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置测量的生物电信号的电位差值信号获取所述用户生物电信号。
  13. 根据权利要求9-12任意一项所述的系统，其特征在于，所述注意力检测装置根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：
    所述注意力检测装置计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
  14. 一种耳侧佩戴装置，其特征在于，所述装置包括：
    多个耳侧信号测量单元，用于从耳侧采集的用户生物电信号；
    第一判断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测 量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；
    特征分解单元，用于从所述用户生物电信号获得脑电信号；
    注意力检测单元，用于根据所述脑电信号信号基于机器学习模型获得所述用户的注意力类型。
  15. 根据权利要求14所述的装置，其特征在于，
    所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；
    所述第一断单元，用于判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
  16. 根据权利要求14所述的装置，其特征在于，
    所述耳侧佩戴装置为单侧耳侧佩戴装置；
    所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；
    所述第一断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号，具体为：
    所述第一判断单元用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
  17. 根据权利要求15所述的装置，其特征在于，
    所述左耳耳侧信号测量单元为多个；
    所述右耳耳侧信号测量单元为多个；
    所述耳侧佩戴装置还包括第二判断单元；
    当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，所述第二判断单元分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
  18. 根据权利要求14-17任意一项所述的装置，其特征在于，所述注意力检测单元根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型具体为：
    所述注意力检测单元计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
  19. 一种耳侧佩戴装置，其特征在于，所述装置包括：
    多个耳侧信号测量单元，用于从耳侧采集的用户生物电信号；
    第一判断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值；将所述两个耳侧信号测 量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号；
    特征分解单元，用于从所述用户生物电信号获得脑电信号；
    发送单元，用于将所述脑电信号发送给信号分析装置。
  20. 根据权利要求19所述的装置，其特征在于，
    所述多个耳侧信号测量单元包括左耳侧信号测量单元和右耳侧信号测量单元；
    所述第一断单元，用于判断所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述左耳侧信号测量单元和所述右耳侧信号测量单元之间的阻抗低于预设阈值；根据所述左耳侧信号测量单元测量的生物电信号和所述右耳侧信号测量单元测量的生物电信号的电位差值信号获取所述用户生物电信号。
  21. 根据权利要求19所述的装置，其特征在于，
    所述耳侧佩戴装置为单侧耳侧佩戴装置；
    所述多个耳侧信号测量单元包括多个单侧耳侧信号测量单元；
    所述第一断单元，用于判断其中两个耳侧信号测量单元之间的阻抗是否低于预设阈值；当所述两个耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个耳侧信号测量单元测量的生物电信号采集的生物电信号的电位差值信号作为所述用户生物电信号，具体为：
    所述第一判断单元用于判断其中两个所述单侧耳侧信号测量单元之间的阻抗是否低于预设阈值，当所述两个所述单侧耳侧信号测量单元之间的阻抗低于预设阈值，将所述两个单侧耳侧信号测量单元采集的生物电信号的电位差值信号作为所述用户生物电信号。
  22. 根据权利要求20所述的装置，其特征在于，
    所述左耳耳侧信号测量单元为多个；
    所述右耳耳侧信号测量单元为多个；
    所述耳侧佩戴装置还包括第二判断单元；
    当所述第一判断单元判断其中一个左耳耳侧信号测量单元和其中一个右耳耳侧信号测量单元之间的阻抗高于预设阈值，所述第二判断单元分别判断所述多个左耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值，以及所述多个右耳耳侧信号测量单元中的两个之间的阻抗是否低于预设阈值；并将阻抗低于所述预设阈值的一侧耳道的两个生物电测量装置采集的生物电信号的电位差值信号作为所述用户生物电信号。
  23. 一种注意力检测装置，其特征在于，所述装置包括：
    接收单元，用于从耳侧佩戴装置接收用户脑电信号；
    注意力检测单元，用于根据所述用户脑电信号基于机器学习模型获得所述用户的注意力类型。
  24. 根据权利要求23所述的装置，其特征在于，所述注意力检测单元具体用于，计算所述用户脑电信号的样本熵的值，根据所述样本熵的值基于机器学习模型分析用户的注意力类型。
  25. 根据权利要求24所述的装置，其特征在于，所述注意力检测单元具体用于，截取预设时间长度的所述用户脑电信号，从所述预设时间长度的所述用户脑电信号获 得N个信号采样点；
    所述N个信号采样点为，u(1),u(2),...,u(N)；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m+1)为起始点依次截取m个采样点来构造出N-m+1个m维向量；针对所述N-m+1个m维向量中的每一个m维向量，计算所述m维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m+1个所述平均值的平均值得到第一平均值；基于所述N个信号采样点，分别以u(1),u(2)，…,u(N-m)为起始点依次截取m+1个采样点来构造出N-m个m+1维向量；针对所述N-m个m+1维向量中的每一个m+1维向量，计算所述m+1维向量与每个其他向量之间的距离小于r的向量的数目的平均值，计算获得的N-m个所述平均值的平均值得到第二平均值；基于所述第一平均值与所述第二平均值比值来计算样本熵(SampEn)的值。
  26. 根据权利要求23-25任意一项所述的装置，其特征在于，所述机器学习模型为SVM分类器；采用SVM分类器进行机器学习获得分割值；
    所述注意力检测单元根据所述分割值和所述样本熵值判断用户的注意力类型。
PCT/CN2020/071565 2019-03-15 2020-01-11 一种注意力检测方法及系统 WO2020186915A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP20773834.5A EP3932303A4 (en) 2019-03-15 2020-01-11 ATTENTION DETECTION METHOD AND SYSTEM
US17/475,658 US20220047198A1 (en) 2019-03-15 2021-09-15 Attention detection method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910199425.2 2019-03-15
CN201910199425.2A CN110584657B (zh) 2019-03-15 2019-03-15 一种注意力检测方法及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/475,658 Continuation US20220047198A1 (en) 2019-03-15 2021-09-15 Attention detection method and system

Publications (1)

Publication Number Publication Date
WO2020186915A1 true WO2020186915A1 (zh) 2020-09-24

Family

ID=68852452

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/071565 WO2020186915A1 (zh) 2019-03-15 2020-01-11 一种注意力检测方法及系统

Country Status (4)

Country Link
US (1) US20220047198A1 (zh)
EP (1) EP3932303A4 (zh)
CN (1) CN110584657B (zh)
WO (1) WO2020186915A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110584657B (zh) * 2019-03-15 2022-09-23 华为技术有限公司 一种注意力检测方法及系统
TWI802908B (zh) * 2021-06-15 2023-05-21 南開科技大學 提供作答建議的輔助分析系統及其方法
CN113827243B (zh) * 2021-11-29 2022-04-01 江苏瑞脑启智医疗科技有限公司 注意力评估方法及系统
CN116392127B (zh) * 2023-06-09 2023-10-20 荣耀终端有限公司 注意力检测方法及相关电子设备

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101836859A (zh) * 2009-03-17 2010-09-22 索尼公司 耳道配装单元和生物信号测量装置
CN101987017A (zh) * 2010-11-18 2011-03-23 上海交通大学 用于驾车司机警觉度测定的脑电信号识别检测方法
CN103610447A (zh) * 2013-12-04 2014-03-05 天津大学 一种基于前额脑电信号的脑力负荷在线检测方法
CN103942568A (zh) * 2014-04-22 2014-07-23 浙江大学 一种基于无监督特征选择的分类方法
CN104814735A (zh) * 2015-05-22 2015-08-05 京东方科技集团股份有限公司 判断大脑是否疲劳的方法和装置
CN105982665A (zh) * 2015-02-12 2016-10-05 中国科学院上海高等研究院 一种从人耳耳道采集脑电波信号的方法
CN106371610A (zh) * 2016-09-23 2017-02-01 重庆金瓯科技发展有限责任公司 一种基于脑电信号的驾驶疲劳的检测方法
CN106667484A (zh) * 2017-01-13 2017-05-17 南京航空航天大学 脑电采集中电极松动检测与自动报警装置及其控制方法
CN106919948A (zh) * 2015-12-28 2017-07-04 西南交通大学 一种驾驶持续性注意水平的识别方法
WO2017146956A1 (en) * 2016-02-22 2017-08-31 Persyst Development Corporation Impedance monitoring for quantitative eeg
CN107280663A (zh) * 2017-07-07 2017-10-24 南京邮电大学 一种基于不同实验难度的疲劳脑电特征研究的方法
CN109471528A (zh) * 2018-10-19 2019-03-15 天津大学 一种用于脑-机接口系统的脑-机互适应系统
CN110584657A (zh) * 2019-03-15 2019-12-20 华为技术有限公司 一种注意力检测方法及系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100482155C (zh) * 2007-05-09 2009-04-29 西安电子科技大学 基于脑机交互的注意力状态即时检测系统及检测方法
WO2010016244A1 (ja) * 2008-08-05 2010-02-11 パナソニック株式会社 運転注意量判定装置、方法およびプログラム
JP5802334B2 (ja) * 2011-08-24 2015-10-28 ヴェーデクス・アクティーセルスカプ 容量性電極を備えるｅｅｇモニタおよび脳波モニタリング方法
CA3236086A1 (en) * 2015-01-06 2016-07-14 David Burton Mobile wearable monitoring systems
CN108451505A (zh) * 2018-04-19 2018-08-28 广西欣歌拉科技有限公司 轻量入耳式睡眠分期系统

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101836859A (zh) * 2009-03-17 2010-09-22 索尼公司 耳道配装单元和生物信号测量装置
CN101987017A (zh) * 2010-11-18 2011-03-23 上海交通大学 用于驾车司机警觉度测定的脑电信号识别检测方法
CN103610447A (zh) * 2013-12-04 2014-03-05 天津大学 一种基于前额脑电信号的脑力负荷在线检测方法
CN103942568A (zh) * 2014-04-22 2014-07-23 浙江大学 一种基于无监督特征选择的分类方法
CN105982665A (zh) * 2015-02-12 2016-10-05 中国科学院上海高等研究院 一种从人耳耳道采集脑电波信号的方法
CN104814735A (zh) * 2015-05-22 2015-08-05 京东方科技集团股份有限公司 判断大脑是否疲劳的方法和装置
CN106919948A (zh) * 2015-12-28 2017-07-04 西南交通大学 一种驾驶持续性注意水平的识别方法
WO2017146956A1 (en) * 2016-02-22 2017-08-31 Persyst Development Corporation Impedance monitoring for quantitative eeg
CN106371610A (zh) * 2016-09-23 2017-02-01 重庆金瓯科技发展有限责任公司 一种基于脑电信号的驾驶疲劳的检测方法
CN106667484A (zh) * 2017-01-13 2017-05-17 南京航空航天大学 脑电采集中电极松动检测与自动报警装置及其控制方法
CN107280663A (zh) * 2017-07-07 2017-10-24 南京邮电大学 一种基于不同实验难度的疲劳脑电特征研究的方法
CN109471528A (zh) * 2018-10-19 2019-03-15 天津大学 一种用于脑-机接口系统的脑-机互适应系统
CN110584657A (zh) * 2019-03-15 2019-12-20 华为技术有限公司 一种注意力检测方法及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3932303A4 *

Also Published As

Publication number Publication date
EP3932303A4 (en) 2022-04-20
EP3932303A1 (en) 2022-01-05
CN110584657A (zh) 2019-12-20
US20220047198A1 (en) 2022-02-17
CN110584657B (zh) 2022-09-23

Similar Documents

Publication Publication Date Title
WO2020186915A1 (zh) 一种注意力检测方法及系统
Pratama et al. A review on driver drowsiness based on image, bio-signal, and driver behavior
Nguyen et al. A lightweight and inexpensive in-ear sensing system for automatic whole-night sleep stage monitoring
EP3527127B1 (en) Electronic device and control method thereof
WO2020187109A1 (zh) 一种用户睡眠检测方法及系统
Casson et al. Electroencephalogram
US20160210407A1 (en) Method and device for processing content based on bio-signals
US20240023892A1 (en) Method and system for collecting and processing bioelectrical signals
Pham et al. WAKE: a behind-the-ear wearable system for microsleep detection
Hwang et al. Driver drowsiness detection using the in-ear EEG
KR20180086547A (ko) 동작 인식 및 제어를 위한 웨어러블 장치 및 이를 이용한 동작 인식 제어 방법
CA3137818A1 (en) Multimodal hearing assistance devices and systems
Gorur et al. Glossokinetic potential based tongue–machine interface for 1-D extraction
Guettas et al. Driver state monitoring system: A review
Gangadharan et al. Drowsiness detection using portable wireless EEG
Lv et al. Design and implementation of an eye gesture perception system based on electrooculography
Han et al. Deep convolutional neural network based eye states classification using ear-EEG
Monori et al. Processing EEG signals acquired from a consumer grade BCI device
Saleem et al. A systematic review of physiological signals based driver drowsiness detection systems
Pham et al. Detection of microsleep events with a behind-the-ear wearable system
Li et al. Driver distraction from the EEG perspective: A review
Nguyen et al. LIBS: a lightweight and inexpensive in-ear sensing system for automatic whole-night sleep stage monitoring
KR20140117201A (ko) 전두엽 부착형 뇌파신호 검출 및 뇌파기반 집중력 분석 방법 및 장치
US11033214B1 (en) Wearable eye tracking system
Nguyen et al. Behind-the-ear EEG-based wearable driver drowsiness detection system using embedded tiny neural networks

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20773834

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020773834

Country of ref document: EP

Effective date: 20210927