WO2018205424A1 - Procédé et terminal d'identification biométrique faisant appel à la myoélectricité, et support de stockage lisible par ordinateur - Google Patents

Procédé et terminal d'identification biométrique faisant appel à la myoélectricité, et support de stockage lisible par ordinateur Download PDF

Info

Publication number
WO2018205424A1
WO2018205424A1 PCT/CN2017/094938 CN2017094938W WO2018205424A1 WO 2018205424 A1 WO2018205424 A1 WO 2018205424A1 CN 2017094938 W CN2017094938 W CN 2017094938W WO 2018205424 A1 WO2018205424 A1 WO 2018205424A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
collected
user
myoelectric
tested
Prior art date
Application number
PCT/CN2017/094938
Other languages
English (en)
Chinese (zh)
Inventor
袁晖
李凝华
Original Assignee
深圳市科迈爱康科技有限公司
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 深圳市科迈爱康科技有限公司 filed Critical 深圳市科迈爱康科技有限公司
Publication of WO2018205424A1 publication Critical patent/WO2018205424A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/15—Biometric patterns based on physiological signals, e.g. heartbeat, blood flow

Definitions

  • the present invention relates to the field of biometrics, and in particular to a myoelectric-based biometric method, a terminal, and a computer readable storage medium.
  • biometrics With the rapid advancement of biometrics, the computer is closely integrated with high-tech means such as optics, acoustics, biosensors and biostatistics to exploit the inherent physiological characteristics of the human body, such as iris, face, fingerprint, palm vein, voiceprint, etc. To identify individuals.
  • biometrics are mostly used in the field of identity information security or property security, and are rarely used in sports health.
  • the main object of the present invention is to provide a myoelectric-based biometric identification method, a terminal, and a computer readable storage medium, which aim to solve the technical problem that the biometric identification technology in the field of sports health is not popular in the prior art.
  • the present invention provides a myoelectric-based biometric recognition method, which is applied to a biometric terminal, and the biometric identification terminal includes a step frequency signal acquisition device and an electromyography signal acquisition device.
  • the myoelectric-based biometric method includes:
  • step frequency value corresponding to the step frequency signal is within a preset range, acquiring the myoelectric signal collected by the myoelectric signal acquisition device at the location to be collected, wherein the to-be-collected portion is the corresponding part of the user to be tested;
  • the present invention further provides a myoelectric-based biometric terminal, comprising: a step frequency signal collecting device, an electromyography signal collecting device, a memory, a processor, and a storage.
  • An electromyography-based biometric program on the memory and operable on the processor, wherein:
  • the step frequency signal collecting device is configured to collect a step frequency signal of the user to be tested
  • the electromyography signal collecting device is configured to collect an electromyogram signal at a part to be collected
  • the steps of the myoelectric-based biometric method as described above are implemented when the myoelectric-based biometric program is executed by the processor.
  • the present invention also provides a computer readable storage medium having a myoelectric-based biometric program stored thereon, the bioelectric-based biometric program being executed by a processor The steps of the myoelectric-based biometric method as described above are achieved.
  • the electromyogram signal collected by the myoelectric signal acquisition device at the location to be collected is acquired, and the active segment characteristics corresponding to the myoelectric signal are extracted. And sending the active segment feature value to the neural network model corresponding to the to-be-collected portion information, so that the neural network model identifies the user identity to be measured according to the active segment feature value, and receives the neural network model feedback recognition result. If the recognition result is identification pass, the identity of the user to be tested passes.
  • the bioelectricity-based biometrics method is applied to the field of sports health, and biometrics can be easily realized in the field of sports health.
  • FIG. 1 is a schematic flow chart of a first embodiment of a myoelectric-based biometric identification method according to the present invention
  • FIG. 2 is a schematic flow chart of a second embodiment of a myoelectric-based biometric identification method according to the present invention.
  • step S60 in FIG. 1 is a schematic diagram showing the refinement process of step S60 in FIG. 1;
  • FIG. 4 is a schematic flow chart of a third embodiment of a myoelectric-based biometric identification method according to the present invention.
  • FIG. 5 is a schematic diagram showing the refinement process of step S90 in FIG. 4;
  • FIG. 6 is a schematic structural diagram of a terminal in a hardware operating environment according to an embodiment of the present invention.
  • FIG. 1 is a schematic flow chart of a first embodiment of a bioelectricity-based biometric identification method according to the present invention.
  • the myoelectric based biometric method comprises:
  • Step S40 acquiring the information of the part to be collected, acquiring the step frequency signal of the user to be tested collected by the step frequency signal collecting device, and detecting whether the step frequency value corresponding to the step frequency signal is within a preset range;
  • the selection of the portion to be collected generally selects a muscle having a large area and a significant contraction.
  • the gastrocnemius or tibialis anterior muscle at the lower leg is selected.
  • the site to be harvested is the tibialis anterior muscle of the left and right calves.
  • the step frequency signal acquisition device can adopt an acceleration sensor, and the number of acceleration changes of the user to be tested collected by the acceleration sensor (the acceleration changes when the person walks, and changes once every step), and the acceleration sensor is used for one minute by using a preset algorithm.
  • the acceleration data collected in the analysis is analyzed to obtain the number of times the acceleration is changed within one minute, and the number of steps of the user in one minute is obtained.
  • the preset range is manually set according to actual needs, for example, set to 90 to 130 steps/minute, and it is detected whether the step frequency value corresponding to the step frequency signal is in this range.
  • this detection process is done by a smartphone (most smartphones currently have a step-measure function) or implemented by a device such as a wristband with a step function.
  • Step S50 when the step frequency value corresponding to the step frequency signal is within a preset range, acquiring an electromyogram signal collected by the myoelectric signal acquisition device at the to-be-acquired portion;
  • the preset range is set to 90 to 130 steps/minute
  • the step frequency value is 100
  • the step frequency corresponding to the user's step frequency signal is within a preset range
  • the myoelectric signal collected by the myoelectric signal acquisition device at the location to be collected is acquired.
  • an EMG signal acquisition device corresponds to a part to be collected, and the EMG signal acquisition device collects the EMG signal of the part to be collected through the surface electrode, and the surface electrode may be a dry electrode, a fabric electrode, a microarray electrode, or the like.
  • the surface electrode can be integrated into the smart wearable device, such as a smart wristband, an ankle guard, etc., and the collected myoelectric signal is transmitted to the signal processing terminal in the form of wireless communication or wired communication.
  • the processing terminal can be Smart terminals such as mobile phones, computers, and tablets.
  • the myoelectric signal of the left tibialis anterior muscle of the user is collected by using the myoelectric signal acquisition device 1 respectively, and the myoelectric signal of the tibialis anterior muscle of the right lower leg of the user is collected by using the myoelectric signal acquisition device 2. .
  • the processing terminal acquires the myoelectric signal of the left tibialis anterior muscle of the user, and acquires the myoelectric signal acquisition device 2 to collect the right tibialis anterior muscle of the user.
  • EMG signal the frequency of the myoelectric signal acquisition device can be set to be between 600 and 1000 Hz, and is specifically set according to actual needs.
  • Step S60 extracting an active segment feature value corresponding to the myoelectric signal, and transmitting the active segment feature value to a neural network model corresponding to the to-be-collected portion information, so that the neural network model treats the active segment feature value according to the active segment Detect user identity for identification;
  • the EMG signal is first denoised.
  • a low pass or band pass (20-450 Hz) filter can be used to filter out 0-20 Hz low frequency noise, using wavelet
  • the denoising or adaptive filtering algorithm removes power frequency interference and other high frequency noise around 50 Hz.
  • the active segment feature value corresponding to the myoelectric signal after the denoising process is extracted, and firstly, the electromyogram signal corresponding to the active segment needs to be determined.
  • the electromyographic signal of the left tibialis anterior muscle of the user to be tested is normally collected, for example, the electromyogram of the left tibialis anterior muscle of the 10 groups of the user to be tested is normally collected.
  • Signal, the average of these 10 groups of EMG signals is used as the threshold (ie, the threshold for determining the starting point of muscle activity).
  • the electromyographic signal collected by the electromechanical signal acquisition device is acquired, and the myoelectric signal is compared with the threshold value, and the part whose absolute value is greater than the threshold is the active segment muscle. electric signal.
  • the EMG integral value, the signal wave length, the absolute value mean value, the average frequency, and the average power corresponding to the active segment EMG signal can be used as the active segment feature value, which is not limited herein, and is specifically reduced or expanded according to requirements.
  • Active segment feature value two or more of the above-mentioned active segment feature values are used as neural network model input parameters, and the neural network model may be pre-stored in the memory. The neural network model has been trained in identifying the user's identity. The trained neural network model is used to identify the identity of the user to be tested based on the input feature value of the active segment. If the recognition is passed, the dialog box may be used to prompt. If the identification fails, the user may be prompted by other authentication methods through a dialog box. Identity authentication, such as passwords, fingerprints, voiceprints, etc.
  • Step S70 Receive a recognition result fed back by the neural network model, and if the recognition result is the identification pass, the identity of the user to be tested passes.
  • the dialog box may be prompted. If the identification does not pass, the user may be prompted to perform identity authentication through other authentication methods, such as password, fingerprint, voiceprint, and the like.
  • the electromyogram signal collected by the myoelectric signal acquisition device at the location to be collected is acquired, and the active segment characteristics corresponding to the myoelectric signal are extracted. And sending the active segment feature value to the neural network model corresponding to the to-be-collected portion information, so that the neural network model identifies the user identity to be measured according to the active segment feature value, and receives the neural network model feedback recognition result. If the recognition result is identification pass, the identity of the user to be tested passes.
  • the bioelectricity-based biometrics method is applied to the field of sports health, and biometrics can be easily realized in the field of sports health.
  • FIG. 2 is a schematic flow chart of a second embodiment of a bioelectricity-based biometric identification method according to the present invention.
  • step S40 before step S40,
  • Step S10 receiving the identity information of the user to be tested and the information about the part to be collected, acquiring the step frequency signal of the user to be tested collected by the step frequency signal collecting device, and detecting whether the step frequency value corresponding to the step frequency signal is within a preset range;
  • Step S20 when the step frequency value corresponding to the step frequency signal is within a preset range, acquiring an electromyogram signal collected by the myoelectric signal acquisition device at the to-be-acquired portion;
  • Step S30 extracting an active segment feature value corresponding to the myoelectric signal, and transmitting the active segment feature value to the neural network model corresponding to the user identity information to be tested and the location information to be collected, for the neural network The model is trained.
  • This embodiment is a description of the steps of training a neural network model.
  • the implementation device is composed of a terminal, a step frequency signal acquisition device, and a myoelectric signal acquisition device, and the terminal, the step frequency signal acquisition device, and the myoelectric signal acquisition device perform data communication by means of wired or wireless communication.
  • the terminal may be a smart terminal such as a mobile phone or a computer, and the terminal acquires a signal collected by the step frequency signal collecting device and the myoelectric signal collecting device, and analyzes and processes the signal.
  • the user's identity information is manually entered by the user, as well as the site information to be collected (eg, the left calf anterior muscle).
  • the step frequency signal acquisition device can adopt an acceleration sensor, and the number of acceleration changes of the user to be tested collected by the acceleration sensor (the acceleration changes when the person walks, and changes once every step), and the acceleration sensor is used for one minute by using a preset algorithm.
  • the acceleration data collected in the analysis is analyzed to obtain the number of times the acceleration is changed within one minute, and the number of steps of the user in one minute is obtained.
  • the preset range is manually set according to actual needs, for example, set to 90 to 130 steps/minute, and it is detected whether the step frequency value corresponding to the step frequency signal is in this range.
  • this detection process is done by a smartphone (most smartphones currently have a step-measure function) or implemented by a device such as a wristband with a step function.
  • an EMG signal acquisition device corresponds to a part to be collected, and the EMG signal acquisition device collects the EMG signal of the part to be collected through the surface electrode, and the surface electrode may be a dry electrode, a fabric electrode, a microarray electrode, or the like.
  • the surface electrode can be integrated into the smart wearable device, such as a smart wristband, an ankle guard, etc., and the collected myoelectric signal is transmitted to the signal processing terminal through wireless communication or wired communication, and after receiving the myoelectric signal, the terminal receives the electromyogram signal.
  • the EMG signal is denoised.
  • a low pass or band pass (20-450 Hz) filter can be used to filter out 0-20 Hz low frequency noise, using wavelet denoising or adaptive filtering.
  • the algorithm removes power frequency interference and other high frequency noise around 50 Hz.
  • the active segment feature value corresponding to the myoelectric signal after the denoising process is extracted, and firstly, the electromyogram signal corresponding to the active segment needs to be determined.
  • the electromyographic signal of the left tibialis anterior muscle of the user to be tested is normally collected, for example, the electromyographic signal of the left tibialis anterior muscle of the user to be tested during normal standing is collected.
  • the myoelectric signal is used as a threshold (ie, a threshold for determining the starting point of muscle activity).
  • the electromyographic signal collected by the electromechanical signal acquisition device at the location to be collected is acquired, and the myoelectric signal is compared with the threshold value, and the portion larger than the threshold is the active segment myoelectric signal. .
  • the EMG integral value, the signal wave length, the absolute value mean value, the average frequency, and the average power corresponding to the active segment EMG signal are used as the active segment feature values, and are not limited herein, and the active segment feature values are specifically reduced or expanded as needed.
  • the active segment feature value is input as a parameter to the neural network model for training the neural network model. The above steps S10 to S30 are repeated several times to obtain a trained neural network model.
  • the EMG information of the user A is collected through steps S10 to S30, and the neural network model is trained based on the acquired EMG information of the user A. , get the trained neural network model.
  • the EMG information of User A is collected, and the corresponding active segment feature value is obtained.
  • the active segment feature value is input into the trained neural network model, and the neural network model determines whether the user corresponding to the current active segment feature value is A, if yes, the feedback recognition pass result, if not, the feedback recognition failure result.
  • the electromyogram signal collected by the myoelectric signal acquisition device at the location to be collected is acquired, and the activity corresponding to the myoelectric signal is extracted. Segment feature value, the active segment feature value is input as a parameter to the neural network model for training the neural network model as a basis for subsequent recognition.
  • step S50 includes:
  • step frequency value corresponding to the step frequency signal is within a preset range, acquiring the myoelectric signal collected by the myoelectric signal acquisition device at the preset sampling frequency at the to-be-acquired portion.
  • the preset range is set to 90 to 130 steps/minute
  • the step frequency value is 100
  • the step frequency corresponding to the user's step frequency signal is within a preset range
  • the myoelectric signal collected by the myoelectric signal acquisition device at the location to be collected is acquired.
  • an EMG signal acquisition device corresponds to a part to be collected, and the EMG signal acquisition device collects the EMG signal of the part to be collected through the surface electrode, and the surface electrode may be a dry electrode, a fabric electrode, a microarray electrode, or the like.
  • the surface electrode can be integrated into the smart wearable device, such as a smart wristband, an ankle guard, etc., and the collected myoelectric signal is transmitted to the signal processing terminal in the form of wireless communication or wired communication.
  • the processing terminal can be Smart terminals such as mobile phones, computers, and tablets.
  • the myoelectric signal of the left tibialis anterior muscle of the user is collected by using the myoelectric signal acquisition device 1 respectively, and the myoelectric signal of the tibialis anterior muscle of the right lower leg of the user is collected by using the myoelectric signal acquisition device 2. .
  • the processing terminal acquires the myoelectric signal of the left tibialis anterior muscle of the user, and acquires the myoelectric signal acquisition device 2 to collect the right tibialis anterior muscle of the user.
  • EMG signal the frequency of the myoelectric signal acquisition device can be set to be between 600 and 1000 Hz, and is specifically set according to actual needs.
  • the acquired myoelectric signal acquisition device collects the myoelectric signal of the to-be-acquired portion at a preset sampling frequency (600-1000 Hz), which is convenient for identification and filtering. Noise, electromagnetic interference signals. Further improve the accuracy of subsequent identification.
  • a preset sampling frequency 600-1000 Hz
  • FIG. 3 is a schematic diagram of the refinement process of step S60 in FIG.
  • step S60 includes:
  • Step S601 performing signal denoising processing on the myoelectric signal
  • Step S602 extracting, according to a preset differential threshold method, an active segment feature value corresponding to the EMG signal after the signal denoising process;
  • Step S603 Send the active segment feature value to the neural network model corresponding to the to-be-collected portion information, so that the neural network model identifies the user identity to be tested according to the active segment feature value.
  • the EMG signal is first denoised.
  • a low pass or band pass (20-450 Hz) filter can be used to filter out 0-20 Hz low frequency noise, using wavelet
  • the denoising or adaptive filtering algorithm removes power frequency interference and other high frequency noise around 50 Hz.
  • the active segment feature value corresponding to the myoelectric signal after the denoising process is extracted, and firstly, the electromyogram signal corresponding to the active segment needs to be determined.
  • the electromyographic signal of the left tibialis anterior muscle of the user to be tested is normally collected, for example, the electromyogram of the left tibialis anterior muscle of the 10 groups of the user to be tested is normally collected.
  • Signal, the average of these 10 groups of EMG signals is used as the threshold (ie, the threshold for determining the starting point of muscle activity).
  • the electromyographic signal collected by the electromechanical signal acquisition device is acquired, and the myoelectric signal is compared with the threshold value, and the part whose absolute value is greater than the threshold is the active segment muscle. electric signal.
  • the EMG integral value, the signal wave length, the absolute value mean value, the average frequency, and the average power corresponding to the active segment EMG signal can be used as the active segment feature value, which is not limited herein, and is specifically reduced or expanded according to requirements. Active segment feature value.
  • the EMG signal is denoised and extracted by a preset differential threshold method to further improve the accuracy of the data, thereby improving the accuracy of subsequent recognition results.
  • step S601 includes:
  • the EMG signal is denoised by using a wavelet denoising algorithm or an adaptive filtering algorithm.
  • the EMG signal is first denoised.
  • a low pass or band pass (20-450 Hz) filter can be used to filter out 0-20 Hz low frequency noise, using wavelet
  • the denoising or adaptive filtering algorithm removes power frequency interference and other high frequency noise around 50 Hz.
  • the denoising process is performed on the myoelectric signal by using a wavelet denoising algorithm or an adaptive filtering algorithm, which can improve the accuracy of the data, thereby improving the accuracy of the subsequent recognition results.
  • FIG. 4 is a schematic flow chart of a third embodiment of a bioelectricity-based biometric identification method according to the present invention.
  • step S60 After step S60,
  • Step S70 Receive a recognition result fed back by the neural network model. If the recognition result is that the recognition fails, the first dialog box is popped up, and the user to be tested is prompted to perform identity authentication.
  • Step S80 receiving authentication information input based on the operation of the user to be tested, and detecting whether the authentication information is legal information;
  • Step S90 If the authentication information is legal information, the identity of the user to be tested is passed, and the feature value of the active segment is sent to the neural network model corresponding to the identity information of the user to be tested and the location information to be collected, for The neural network model is trained.
  • a first dialog box is popped up for prompting the user to be tested to perform identity authentication.
  • the method of identity authentication is not limited here, and is set according to the actual situation.
  • the verification mode is that the user inputs the verification code. If the verification code is consistent with the pre-stored verification code, the authentication passes (ie, the identification is passed), or the user fingerprint is collected, and if the collected fingerprint information is the same as the pre-stored fingerprint. If the information is consistent, the authentication is passed (ie, the identification is passed).
  • the user is further authenticated by other preset verification methods. If the authentication is passed, it indicates that the neural network model has an error in determining the identity of the user to be tested.
  • the active segment feature value is sent to the user identity information to be tested and the neural network model corresponding to the location information to be collected, so as to train the neural network model to further improve the accuracy of the recognition result.
  • FIG. 5 is a schematic diagram of the refinement process of step S90 in FIG.
  • step S90 includes:
  • Step S901 If the authentication information is legal information, the identity of the user to be tested is passed, and a second dialog box is popped up, so that the user to be tested selects whether the motion amount is increased within a preset time period;
  • Step S902 receiving a feedback instruction triggered by the operation of the user to be tested, and if the feedback instruction is no, sending the active segment feature value to the user identity information to be tested and the neural network model corresponding to the location information to be collected, For training the neural network model.
  • the authentication is passed by other means. Then extracting the myoelectric integral value X corresponding to the electromyogram signal of the active segment, and the myoelectric integral value X is the same as the historical myoelectric integral value Y (which may be all the myoelectric integral values of the part to be tested corresponding to the historically collected user)
  • the mean value is compared to obtain the value Z of Z
  • a dialog box is popped up for the user to be tested to select whether to increase the amount of exercise within the preset time (for example, whether the display interface displays the amount of exercise within 4-8 hours.
  • the user feedback is not, it indicates that the neural network model has an error in the judgment of the user identity to be tested, then the activity segment feature value is sent to the user identity information to be tested and the neural network model corresponding to the part information to be collected, for the nerve
  • the network model is trained to further improve the accuracy of the recognition result; if the user feedback is yes, it indicates that the neural network model is correct in determining the identity of the user to be tested.
  • This recognition fails because the user's myoelectric signal occurs due to the increased amount of exercise. A large change is made, ignoring the characteristic value of the active segment corresponding to the EMG signal of the active segment.
  • the neural network model is trained in a targeted manner, and the accuracy of the neural network model in subsequent determination is improved.
  • the active segment feature values include:
  • the EMG signal corresponds to the active segment EMG integral value, signal wave length, absolute value mean, average frequency, average power, energy characteristic value, root mean square, zero crossing point, and average amplitude difference.
  • the myoelectric integral value is a simple electromyography time domain feature quantity, which is commonly used in the pretreatment of EMG (electromyographic signal) active segment detection and in some clinical applications.
  • IEMG is defined as the absolute value of the EMG signal over a period of time
  • the absolute value mean is the absolute value of the amplitude of the EMG signal over a period of time
  • Energy eigenvalues include sum of squared magnitude (SSI), mean squared value (VAR);
  • Root mean square is a commonly used time domain feature of EMG signal analysis
  • WL is a characteristic quantity describing the complexity of the EMG signal.
  • AAC which characterizes the severity of the change in signal amplitude
  • the zero-crossing point is a representation of the feature quantity describing the frequency information of the myoelectric signal in the time domain. ZC is the number of times the signal amplitude passes through zero.
  • a threshold is generally set in the calculation. When the signal difference between the two sides of the zero point is greater than the threshold, the zero crossing is considered as Effective.
  • the feature value of the active segment can be expanded or reduced according to actual needs, and is not limited herein.
  • the active segment feature values used may be all of the above or one or both of the above. Set according to the actual situation.
  • the active segment feature value only a small amount of specific information corresponding to the myoelectric signal can be selected as the active segment feature value.
  • the active segment eMG integral value corresponding to the myoelectric signal is used as the active segment feature value, which can simplify the recognition process and can Convenient biometrics in the field of sports and health.
  • an embodiment of the present invention further provides a myoelectric-based biometric terminal.
  • FIG. 6 is a schematic structural diagram of a terminal in a hardware operating environment according to an embodiment of the present invention.
  • the terminal may be a PC, or may be a terminal device such as a smart phone, a tablet computer, or a portable computer.
  • the terminal may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a human body feature collection component 1006 (including a step frequency signal acquisition device, a myoelectric signal acquisition device), and communication.
  • Bus 1002 The step frequency signal acquisition device is configured to collect the step frequency signal of the user to be tested, the myoelectric signal acquisition device is configured to collect the myoelectric signal at the location to be collected, and the communication bus 1002 is used to implement connection communication between the components.
  • the user interface 1003 can include a display, an input unit such as a keyboard, and the optional user interface 1003 can also include a standard wired interface, a wireless interface.
  • the network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a WI-FI interface).
  • the memory 1005 may be a high speed RAM memory or a stable memory (non-volatile) Memory), such as disk storage.
  • the memory 1005 can also optionally be a storage device independent of the aforementioned processor 1001.
  • the terminal may further include a camera, RF (Radio) Frequency, RF) circuits, sensors, audio circuits, WiFi modules, and more.
  • sensors such as light sensors, motion sensors, and other sensors.
  • the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display according to the brightness of the ambient light, and the proximity sensor may turn off the display and/or when the mobile terminal moves to the ear. Backlighting.
  • the gravity acceleration sensor can detect the magnitude of acceleration in each direction (usually three axes), and can detect the magnitude and direction of gravity when stationary, and can be used to identify the posture of the mobile terminal (such as horizontal and vertical screen switching, Related games, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tapping), etc.; of course, the terminal can also be equipped with other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. Let me repeat.
  • terminal structure shown in FIG. 6 does not constitute a limitation to the terminal, and may include more or less components than those illustrated, or a combination of certain components, or different component arrangements.
  • an operating system may be included in the memory 1005 as a computer storage medium.
  • a network communication module may be included in the memory 1005 as a computer storage medium.
  • a user interface module may be included in the memory 1005 as a computer storage medium.
  • a myoelectric-based biometric program may be included in the memory 1005 as a computer storage medium.
  • the network interface 1004 is mainly used to connect to the background server for data communication with the background server;
  • the user interface 1003 is mainly used for connecting the client (user end), and performing data communication with the client;
  • 1001 can be used to invoke the myoelectric-based biometric program stored in memory 1005 and perform the steps of the myoelectric-based biometric method as described above.
  • the specific embodiment of the myoelectric-based biometric terminal of the present invention is substantially the same as the above embodiments of the myoelectric-based biometric identification method, and is not described herein.
  • an embodiment of the present invention further provides a computer readable storage medium, where the electromyography-based biometric program is stored, and the electromyography-based biometric program is implemented by a processor. The steps of the myoelectric-based biometric method described above.
  • portions of the technical solution of the present invention that contribute substantially or to the prior art may be embodied in the form of a software product stored in a storage medium (such as a ROM/RAM as described above). , a disk, an optical disk, including a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present invention.
  • a terminal device which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

L'invention concerne un procédé et un terminal d'identification biométrique faisant appel à la myoélectricité, et un support de stockage lisible par ordinateur. Le procédé comprend les étapes suivantes : lorsqu'une valeur de fréquence de foulée correspondant à un signal de fréquence de foulée d'un utilisateur à détecter est comprise dans une plage prédéfinie, acquérir un signal myoélectrique collecté par un appareil de collecte de signal myoélectrique auprès d'une partie à collecter, extraire une valeur caractéristique de segment actif correspondant au signal myoélectrique, et envoyer la valeur caractéristique de segment actif à un modèle de réseau neuronal correspondant à des informations concernant la partie à collecter, de telle sorte que le modèle de réseau neuronal identifie, en fonction de la valeur caractéristique de segment actif, l'identité de l'utilisateur à détecter, l'identité de l'utilisateur à détecter réussissant l'identification si un résultat d'identification indique que l'identification est réussie. Au moyen du procédé, une identification biométrique peut être commodément réalisée dans le domaine du sport et de la santé par application du procédé d'identification biométrique faisant appel à la myoélectricité au domaine du sport et de la santé.
PCT/CN2017/094938 2017-05-09 2017-07-28 Procédé et terminal d'identification biométrique faisant appel à la myoélectricité, et support de stockage lisible par ordinateur WO2018205424A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710323706.5A CN107169432A (zh) 2017-05-09 2017-05-09 基于肌电的生物识别方法、终端及计算机可读存储介质
CN201710323706.5 2017-05-09

Publications (1)

Publication Number Publication Date
WO2018205424A1 true WO2018205424A1 (fr) 2018-11-15

Family

ID=59813026

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/094938 WO2018205424A1 (fr) 2017-05-09 2017-07-28 Procédé et terminal d'identification biométrique faisant appel à la myoélectricité, et support de stockage lisible par ordinateur

Country Status (2)

Country Link
CN (1) CN107169432A (fr)
WO (1) WO2018205424A1 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2570749A (en) * 2017-09-05 2019-08-07 B Secur Ltd Wearable authentication device to prevent transfer of authentication
CN111714122A (zh) * 2020-06-22 2020-09-29 北京海益同展信息科技有限公司 肌电模式识别方法、装置、设备和计算机可读存储介质
WO2023050522A1 (fr) * 2021-09-28 2023-04-06 歌尔股份有限公司 Procédé d'interaction, dispositif terminal et support de stockage lisible
CN117547287A (zh) * 2023-11-14 2024-02-13 首都医科大学宣武医院 一种基于多生理参数的糖尿病足风险评估系统
CN117562560A (zh) * 2023-11-20 2024-02-20 北京宜善医学科技有限公司 一种康复训练中的运动效果评估方法、装置及存储介质

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109407531A (zh) * 2018-10-30 2019-03-01 深圳市心流科技有限公司 智能家居控制方法、装置及计算机可读存储介质
CN109646024A (zh) * 2019-01-09 2019-04-19 浙江强脑科技有限公司 疲劳驾驶检测方法、装置及计算机可读存储介质
CN109820523A (zh) * 2019-03-05 2019-05-31 浙江强脑科技有限公司 心理测试仪控制方法、装置及计算机可读存储介质
CN109820482A (zh) * 2019-03-05 2019-05-31 浙江强脑科技有限公司 肌肉状态检测方法、装置及计算机可读存储介质
CN109965847B (zh) * 2019-04-08 2023-11-07 清华大学 服务器和信号分析系统
CN110321871B (zh) * 2019-07-11 2023-04-07 电子科技大学成都学院 一种基于lstm的掌静脉识别系统及方法
CN110811633A (zh) * 2019-11-06 2020-02-21 中国科学院自动化研究所 基于肌电信号的身份识别方法、系统、装置
CN112773381B (zh) * 2019-11-07 2023-09-22 深圳市理邦精密仪器股份有限公司 一种肌电信号的处理方法、处理设备以及存储介质
CN111783056B (zh) * 2020-07-06 2024-05-14 诺百爱（杭州）科技有限责任公司 一种基于肌电信号识别用户身份的方法、装置和电子设备
CN112022611B (zh) * 2020-08-20 2022-06-21 无锡商业职业技术学院 一种被动式关节训练器械的识别电路结构
CN112140109B (zh) * 2020-09-10 2024-04-30 华南理工大学 基于Web网页和肌电信号的机器人远程控制系统及方法
CN112244882B (zh) * 2020-10-30 2023-06-02 北京中科心研科技有限公司 一种基于多模态生理数据的疾病预警方法和装置
CN112690807A (zh) * 2020-12-04 2021-04-23 京东数科海益信息科技有限公司 确定执行动作速度等级的方法、装置和存储介质
CN113303816A (zh) * 2021-04-30 2021-08-27 浙江大学 一种基于面部表面肌电信号的味觉感知识别方法及系统
CN114652493B (zh) * 2022-05-25 2022-08-30 深圳市心流科技有限公司 一种肌电信号控制方法、装置、肌电设备、存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1931091A (zh) * 2006-10-12 2007-03-21 北京飞天诚信科技有限公司 一种身份鉴定方法及系统
CN104573458A (zh) * 2014-12-30 2015-04-29 深圳先进技术研究院 一种基于心电信号的身份识别方法、装置及系统
CN105243312A (zh) * 2015-11-13 2016-01-13 上海傲意信息科技有限公司 一种密码系统及加解密方法
CN105279411A (zh) * 2015-09-22 2016-01-27 电子科技大学 一个基于步态生物特征的移动设备身份认证方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102073881A (zh) * 2011-01-17 2011-05-25 武汉理工大学 人体表面肌电信号消噪、特征提取和模式识别方法
US9619024B2 (en) * 2014-06-24 2017-04-11 Beijing TransBorder Information Technology Co., Ltd. Virtual input device and virtual input method
CN105807640A (zh) * 2014-12-30 2016-07-27 深圳Tcl新技术有限公司 终端控制方法和系统
CN106293057A (zh) * 2016-07-20 2017-01-04 西安中科比奇创新科技有限责任公司 基于bp神经网络的手势识别方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1931091A (zh) * 2006-10-12 2007-03-21 北京飞天诚信科技有限公司 一种身份鉴定方法及系统
CN104573458A (zh) * 2014-12-30 2015-04-29 深圳先进技术研究院 一种基于心电信号的身份识别方法、装置及系统
CN105279411A (zh) * 2015-09-22 2016-01-27 电子科技大学 一个基于步态生物特征的移动设备身份认证方法
CN105243312A (zh) * 2015-11-13 2016-01-13 上海傲意信息科技有限公司 一种密码系统及加解密方法

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2570749A (en) * 2017-09-05 2019-08-07 B Secur Ltd Wearable authentication device to prevent transfer of authentication
CN111714122A (zh) * 2020-06-22 2020-09-29 北京海益同展信息科技有限公司 肌电模式识别方法、装置、设备和计算机可读存储介质
WO2023050522A1 (fr) * 2021-09-28 2023-04-06 歌尔股份有限公司 Procédé d'interaction, dispositif terminal et support de stockage lisible
CN117547287A (zh) * 2023-11-14 2024-02-13 首都医科大学宣武医院 一种基于多生理参数的糖尿病足风险评估系统
CN117547287B (zh) * 2023-11-14 2024-04-09 首都医科大学宣武医院 一种基于多生理参数的糖尿病足风险评估系统
CN117562560A (zh) * 2023-11-20 2024-02-20 北京宜善医学科技有限公司 一种康复训练中的运动效果评估方法、装置及存储介质

Also Published As

Publication number Publication date
CN107169432A (zh) 2017-09-15

Similar Documents

Publication Publication Date Title
WO2018205424A1 (fr) Procédé et terminal d'identification biométrique faisant appel à la myoélectricité, et support de stockage lisible par ordinateur
WO2018135692A1 (fr) Dispositif pouvant être porté destiné à la reconnaissance et à la commande de mouvement, et procédé de commande de reconnaissance de mouvement l'utilisant
WO2019088769A1 (fr) Procédé et système de fourniture d'informations médicales basé sur une api ouverte
WO2018205423A1 (fr) Procédé et terminal de lecture de données audio, et support de stockage lisible par ordinateur
EP2698112B1 (fr) Détermination de contraintes en temps réel d'un individu
WO2015008936A1 (fr) Appareil de diagnostic à l'aide d'habitudes, appareil de gestion du diagnostic et procédé de diagnostic l'utilisant.
US20080294058A1 (en) Wearable Device, System and Method for Measuring a Pulse While a User is in Motion
WO2015053418A1 (fr) Casque d'écoute multifonctionnel pour le sport
WO2019039716A1 (fr) Dispositif électronique et son procédé de mesure de contrainte
WO2017067271A1 (fr) Procédé de reconnaissance d'empreintes digitales, appareil de reconnaissance d'empreintes digitales, et terminal mobile
CN112399823A (zh) 生物信息处理装置和信息处理方法
TW201225920A (en) Mechanomyographic signal input device, human-machine operating system and identification method thereof
WO2015026047A1 (fr) Interface de biosignal vestimentaire et méthode de fonctionnement d'interface de biosignal vestimentaire
US20140049627A1 (en) Systems And Methods For Sleep Monitoring
Wang et al. Biometrics on mobile phone
WO2018223505A1 (fr) Dispositif portable d'identification de démarche
WO2018190562A1 (fr) Appareil d'authentification biométrique et procédé d'authentification biométrique
WO2020042463A1 (fr) Procédé, appareil, dispositif et support de déverrouillage de contrôle d'accès basé sur la reconnaissance biométrique
JP2018504960A (ja) 人体特徴データの処理方法及び装置
WO2018223504A1 (fr) Procédé d'authentification d'identité, dispositif vestimentaire, et support d'informations lisible par ordinateur
WO2020106104A1 (fr) Dispositif électronique destiné à fournir des contenus de méditation et procédé de fonctionnement associé
CN105243312A (zh) 一种密码系统及加解密方法
WO2013141667A1 (fr) Système produisant des informations quotidiennes sur la santé et procédé produisant des informations quotidiennes sur la santé
WO2022014798A1 (fr) Système d'exécution d'enquête de satisfaction client
WO2015174732A1 (fr) Procédé pour fournir un contenu personnalisé basé sur un biosignal

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: 17908876

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17908876

Country of ref document: EP

Kind code of ref document: A1