A full bone conduction brain machine vibration pickup device and method thereof

Abstract

The present invention discloses a full bone conduction brain-computer vibration pickup device and method thereof, which relates to the field of sound vibration pickup, and includes: a signal acquisition module, the signal acquisition module includes at least one bone conduction sensor, and the position of the bone conduction sensor is determined according to a position algorithm; a signal processing module, the signal processing module receives at least one bone vibration signal collected by the signal acquisition module, obtains a bone filtering signal, obtains a bone amplification signal, and a noise reduction unit performs noise reduction processing on the bone amplification signal; a sound restoration module, the sound restoration module restores the processed effective sound signal to a recognizable sound, optimizes the recognizable sound, and obtains a target output sound; a control module; and a power supply module. By setting the signal acquisition module, the signal processing module, and the sound restoration module, the sound with the best pickup effect is selected from multiple sounds, and the noise is eliminated more fully.

Description

Technical Field

[0001] The present invention relates to the field of sound vibration pickup, and in particular to a full bone conduction brain-machine vibration pickup device and a method thereof. Background Art

[0002] Traditional sound pickup devices, such as microphones, mainly receive sound signals through air conduction. However, in complex environmental noise, this method has poor noise immunity and the accuracy of sound pickup is easily disturbed. Especially in scenarios with high precision requirements such as human-machine control and medical diagnosis, the performance of traditional sound pickup devices often cannot meet the needs.

[0003] In recent years, bone conduction technology has gradually attracted attention. Bone conduction pickup devices restore sound by directly collecting human bone vibration signals, and have the advantages of strong noise resistance and good privacy protection. However, most existing bone conduction pickup devices are limited to signal collection in a single part, such as the ear or throat, resulting in limited sound pickup effects, and there is a lot of room for improvement. Summary of the invention

[0004] In order to solve the above technical problems, a full bone conduction brain-machine vibration pickup device and method are provided. This technical solution solves the problem that the existing bone conduction pickup devices proposed in the above background technology are mostly limited to signal collection in a single part, such as the ear or throat, resulting in limited sound pickup effect.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A full bone conduction brain machine vibration pickup device, comprising:

[0007] A signal acquisition module, wherein the signal acquisition module includes at least one bone conduction sensor, and the at least one bone conduction sensor is respectively arranged at the bones on the top of the head, the left and right temporal bones, the back of the head bones and the throat bones, and the positions of the bone conduction sensors are determined according to a position algorithm, and the bone conduction sensors collect bone vibration signals at the positions;

[0008] A signal processing module, wherein the signal processing module receives at least one bone vibration signal collected by the signal collection module, and the signal processing module is composed of a filtering unit, an amplifying unit and a noise reduction unit. The filtering unit performs filtering processing on the collected bone vibration signal to obtain a bone filtered signal, the amplifying unit performs amplifying processing on the bone filtered signal to obtain a bone amplified signal, and the noise reduction unit performs noise reduction processing on the bone amplified signal. The bone vibration signal is filtered, amplified and noise-reduced to obtain an effective sound signal;

[0009] A sound restoration module receives a valid sound signal, restores the processed valid sound signal to a recognizable sound, optimizes the recognizable sound, and obtains a target output sound;

[0010] A control module, wherein the control module is connected to the signal processing module and the sound restoration module, controls the operating parameters of the signal processing module and the sound restoration module, monitors the operating conditions of the signal processing module and the sound restoration module, and issues an early warning for abnormal conditions;

[0011] A power module, wherein the power module provides power support for the signal acquisition module, the signal processing module, the sound restoration module and the control module.

[0012] Preferably, the position of the bone conduction sensor is determined according to a position algorithm comprising the following steps:

[0013] Perform three-dimensional modeling on the top bones of the head, the left and right temporal bones, the back bones of the head, and the throat bones to obtain a three-dimensional model of the overall skeleton;

[0014] At least one sampling point is uniformly taken on the surface of the overall skeletal three-dimensional model;

[0015] The sampling points that appear in the bones of the top of the head, the left and right temporal bones, the bones of the back of the head, and the bones of the larynx are respectively regarded as the head sampling point, the temporal sampling point, the spoon sampling point, and the laryngeal sampling point;

[0016] Vibration capture is performed simultaneously at the head sampling point, the temporal sampling point, the spoon sampling point and the laryngeal sampling point to obtain at least one vibration intensity;

[0017] The head sampling point with the largest vibration intensity is used as the target head sampling point, the temporal sampling point with the largest vibration intensity is used as the target temporal sampling point, the spoon sampling point with the largest vibration intensity is used as the target spoon sampling point, and the laryngeal sampling point with the largest vibration intensity is used as the target laryngeal sampling point;

[0018] The target head sampling point, target temporal sampling point, target spoon sampling point and target laryngeal sampling point are used as the setting positions of the bone conduction sensors.

[0019] Preferably, the filtering unit performs filtering processing on the collected bone vibration signal comprises the following steps:

[0020] In a noise-free environment, collecting at least one bone vibration sample signal;

[0021] Performing Fourier transform decomposition on the bone vibration sample signal to obtain at least one bone vibration standard signal;

[0022] The standard bone vibration signal is imaged by sampling to obtain a standard bone vibration waveform;

[0023] Obtain the period of the standard waveform of bone vibration, and use the frequency formula to calculate the frequency of the standard signal of bone vibration;

[0024] forming a filtering interval, wherein the endpoints of the filtering interval are formed by the maximum value and the minimum value of the frequency of at least one bone vibration standard signal;

[0025] Performing Fourier transform on the bone vibration signal to obtain at least one bone vibration decomposition signal;

[0026] Deleting the bone vibration decomposition signal whose frequency does not belong to the filtering interval, and combining at least one bone vibration decomposition signal after the deletion process by using inverse Fourier transform to obtain a bone filtering signal;

[0027] The Fourier transform is as follows: ,

[0028] Where, F(x) is the signal after Fourier transform, i is a unit imaginary number, e is a natural constant, f(t) is the signal before Fourier transform, t is time, and x is frequency;

[0029] The inverse Fourier transform is as follows: ,

[0030] Among them, G(x) is the signal before inverse Fourier transform, and g(t) is the signal after inverse Fourier transform;

[0031] The frequency formula is as follows: ,

[0032] Wherein, A is the frequency of the bone vibration sample signal, and a is the period of the bone vibration sample waveform.

[0033] Preferably, the amplification unit amplifies the skeleton filter signal and comprises the following steps:

[0034] Obtaining a value range after the signal is amplified, dividing the value range after the signal is amplified at equal intervals, and obtaining at least one amplified point;

[0035] Perform transmission test on the test signal whose signal strength is equal to the value at the amplification point, and take the amplification point corresponding to the test signal with the highest fidelity as the target amplification point;

[0036] The value at the target magnification point is divided by the strength of the bone filter signal to obtain the target magnification factor;

[0037] The amplification unit amplifies the bone filter signal according to the target amplification factor to obtain a bone amplification signal.

[0038] Preferably, the noise reduction unit performs noise reduction processing on the bone amplification signal, comprising the following steps:

[0039] Use the noise formula to calculate the noise coefficient of the bone filter signal;

[0040] Select the bone amplification signal corresponding to the bone filter signal with the smallest noise coefficient as the target bone amplification signal;

[0041] Acquire at least one environmental noise signal and take the average value to obtain an approximate noise signal;

[0042] Under the condition that no noise occurs in the throat of the human body and the environment is noiseless, at least one internal vibration noise of the human body is obtained and the average is taken to obtain the internal vibration noise;

[0043] Removing approximate noise signals and internal vibration noise from the target bone amplification signal to obtain an effective sound signal;

[0044] The noise formula is as follows: ,

[0045] Among them, B is the noise coefficient, H(t) is the bone filter signal, and h(t) is the bone vibration signal corresponding to the bone filter signal.

[0046] Preferably, the process of restoring the processed effective sound signal to a recognizable sound comprises the following steps:

[0047] The sound restoration module restores the effective sound signal to a recognizable sound through the speaker.

[0048] Preferably, the step of optimizing the recognizable sound to obtain the target output sound comprises the following steps:

[0049] Based on historical data, obtain the volume range of human voice communication;

[0050] Obtaining the volume of the recognizable sound, determining whether the volume of the recognizable sound falls within the volume range of human voice communication, and if so, taking the recognizable sound as the target output sound;

[0051] If not, the volume of the recognizable sound is increased until the volume of the recognizable sound falls within the volume range of human voice communication, and the recognizable sound with the increased volume is used as the target output sound.

[0052] Preferably, the operating parameters of the control signal processing module and the sound restoration module include the following steps:

[0053] Based on the ratio of the workload of the signal processing module and the workload of the sound restoration module, the power module is allocated to the power supply module, and the first amount of electricity and the second amount of electricity are allocated to the signal processing module and the sound restoration module respectively;

[0054] Allocating the first amount of electricity according to the ratio of the operation amounts of the filtering unit, the amplifying unit, and the noise reduction unit;

[0055] The filtering unit, the amplifying unit and the noise reduction unit operate according to the amount of electricity obtained.

[0056] Preferably, the monitoring of the operation of the signal processing module and the sound restoration module and issuing an early warning for abnormal situations comprises the following steps:

[0057] Monitor the quality of the effective sound signal generated by the signal processing module and the target output sound generated by the sound restoration module;

[0058] When there are abnormalities in the monitoring results, an early warning will be issued, otherwise, no action will be taken.

[0059] A full bone conduction brain machine vibration sound pickup method, applicable to the full bone conduction brain machine vibration sound pickup device, comprising:

[0060] At least one bone conduction sensor is respectively arranged at the bones on the top of the head, the left and right temporal bones, the bones on the back of the head and the bones on the larynx. The positions of the bone conduction sensors are determined according to a position algorithm. The bone conduction sensors collect bone vibration signals at the positions.

[0061] The collected bone vibration signal is filtered to obtain a bone filter signal, the bone filter signal is amplified to obtain a bone amplified signal, and the bone amplified signal is subjected to noise reduction. The bone vibration signal is filtered, amplified and subjected to noise reduction to obtain an effective sound signal;

[0062] Receive valid sound signals, restore the processed valid sound signals to recognizable sounds, optimize the recognizable sounds, and obtain target output sounds;

[0063] During the sound pickup process, the operating parameters of the signal processing module and the sound restoration module are controlled, the operating conditions of the signal processing module and the sound restoration module are monitored, and warnings are issued for abnormal conditions.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] By setting up a signal acquisition module, a signal processing module and a sound restoration module, sound vibrations are picked up at multiple locations, so that the sound with the best pickup effect can be selected from multiple sounds, thereby ensuring the pickup effect. In addition, based on the characteristics of bone conduction, noise reduction and filtering methods different from conventional processing methods are used for filtering, which can more fully eliminate noise. At the same time, when acquiring vibrations, the position of the bone conduction sensor is restricted to ensure better vibration extraction effects, thereby ensuring the effect of subsequent vibration processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1It is a structural schematic diagram of the full bone conduction brain machine vibration pickup device of the present invention;

[0067] Figure 2 A schematic diagram of a flow chart of determining the position of the bone conduction sensor of the present invention according to a position algorithm;

[0068] Figure 3 A schematic diagram of a flow chart of filtering the collected bone vibration signal by the filtering unit of the present invention;

[0069] Figure 4 It is a schematic diagram of the process of amplifying the skeleton filter signal by the amplifying unit of the present invention;

[0070] Figure 5 A schematic diagram of a process of noise reduction processing performed by a noise reduction unit of the present invention on a bone amplification signal;

[0071] Figure 6 A schematic diagram of a process of optimizing the recognizable sound to obtain the target output sound according to the present invention;

[0072] Figure 7 It is a flow chart of the operating parameters of the control signal processing module and the sound restoration module of the present invention;

[0073] Figure 8 A schematic diagram of a process for monitoring the operation of a signal processing module and a sound restoration module and issuing an early warning for abnormal situations according to the present invention;

[0074] Fig. 9 It is a schematic flow chart of the full bone conduction brain-machine vibration sound pickup method of the present invention. DETAILED DESCRIPTION

[0075] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0076] Reference Figure 1 As shown, a full bone conduction brain machine vibration pickup device comprises:

[0077] A signal acquisition module, wherein the signal acquisition module includes at least one bone conduction sensor, and the at least one bone conduction sensor is respectively arranged at the bones on the top of the head, the left and right temporal bones, the back of the head bones and the throat bones, and the positions of the bone conduction sensors are determined according to a position algorithm, and the bone conduction sensors collect bone vibration signals at the positions;

[0078] A signal processing module, wherein the signal processing module receives at least one bone vibration signal collected by the signal collection module, and the signal processing module is composed of a filtering unit, an amplifying unit and a noise reduction unit. The filtering unit performs filtering processing on the collected bone vibration signal to obtain a bone filtered signal, the amplifying unit performs amplifying processing on the bone filtered signal to obtain a bone amplified signal, and the noise reduction unit performs noise reduction processing on the bone amplified signal. The bone vibration signal is filtered, amplified and noise-reduced to obtain an effective sound signal;

[0079] A sound restoration module receives a valid sound signal, restores the processed valid sound signal to a recognizable sound, optimizes the recognizable sound, and obtains a target output sound;

[0080] A control module, wherein the control module is connected to the signal processing module and the sound restoration module, controls the operating parameters of the signal processing module and the sound restoration module, monitors the operating conditions of the signal processing module and the sound restoration module, and issues an early warning for abnormal conditions;

[0081] A power module, wherein the power module provides power support for the signal acquisition module, the signal processing module, the sound restoration module and the control module.

[0082] In this scheme, vibrations are acquired at the bones on the top of the head, the left and right temporal bones, the back of the head bones, and the throat bones. Since each position is affected differently by the external environment, the sound that is least affected by the environmental noise can be selected and further denoised, thereby obtaining a more fidelity sound.

[0083] In this scheme, bone conduction is used as a way to obtain sound. Due to the characteristics of bone conduction, there is a large difference between the frequency of sound conducted through bone and the frequency of sound not conducted through bone. Therefore, this property can be used to distinguish the noise in the environment and eliminate it in a targeted manner. However, due to the characteristics of bone conduction, the bone conduction sensor not only captures the sound conducted through the bones, but also the noise vibration inside the human body is also conducted through the bones, so it needs to be eliminated. In this scheme, these situations are taken into consideration and handled accordingly.

[0084] Reference Figure 2 As shown, the position of the bone conduction sensor is determined according to the position algorithm, including the following steps:

[0085] Perform three-dimensional modeling on the top bones of the head, the left and right temporal bones, the back bones of the head, and the throat bones to obtain a three-dimensional model of the overall skeleton;

[0086] At least one sampling point is uniformly taken on the surface of the overall skeletal three-dimensional model;

[0087] The sampling points that appear in the bones of the top of the head, the left and right temporal bones, the bones of the back of the head, and the bones of the larynx are respectively regarded as the head sampling point, the temporal sampling point, the spoon sampling point, and the laryngeal sampling point;

[0088] Vibration capture is performed simultaneously at the head sampling point, the temporal sampling point, the spoon sampling point and the laryngeal sampling point to obtain at least one vibration intensity;

[0089] The head sampling point with the largest vibration intensity is used as the target head sampling point, the temporal sampling point with the largest vibration intensity is used as the target temporal sampling point, the spoon sampling point with the largest vibration intensity is used as the target spoon sampling point, and the laryngeal sampling point with the largest vibration intensity is used as the target laryngeal sampling point;

[0090] The target head sampling point, target temporal sampling point, target spoon sampling point and target laryngeal sampling point are used as the setting positions of the bone conduction sensors.

[0091] When acquiring vibration, different position settings will result in different vibration acquisition. Therefore, it is necessary to ensure that the vibration is acquired as completely as possible, so it is necessary to determine the optimal position, thereby capturing the best vibration, which is more conducive to subsequent noise reduction processing, etc.

[0092] Reference Figure 3 As shown, the filtering unit performs filtering processing on the collected bone vibration signal, including the following steps:

[0093] In a noise-free environment, collecting at least one bone vibration sample signal;

[0094] Performing Fourier transform decomposition on the bone vibration sample signal to obtain at least one bone vibration standard signal;

[0095] The standard bone vibration signal is imaged by sampling to obtain a standard bone vibration waveform;

[0096] Obtain the period of the standard waveform of bone vibration, and use the frequency formula to calculate the frequency of the standard signal of bone vibration;

[0097] forming a filtering interval, wherein the endpoints of the filtering interval are formed by the maximum value and the minimum value of the frequency of at least one bone vibration standard signal;

[0098] Performing Fourier transform on the bone vibration signal to obtain at least one bone vibration decomposition signal;

[0099] Deleting the bone vibration decomposition signal whose frequency does not belong to the filtering interval, and combining at least one bone vibration decomposition signal after the deletion process by using inverse Fourier transform to obtain a bone filtering signal;

[0100] The Fourier transform is as follows: ,

[0101] Where, F(x) is the signal after Fourier transform, i is a unit imaginary number, e is a natural constant, f(t) is the signal before Fourier transform, t is time, and x is frequency;

[0102] The inverse Fourier transform is as follows: ,

[0103] Among them, G(x) is the signal before inverse Fourier transform, and g(t) is the signal after inverse Fourier transform;

[0104] The frequency formula is as follows: ,

[0105] Wherein, A is the frequency of the bone vibration sample signal, and a is the period of the bone vibration sample waveform.

[0106] Since environmental noise is transmitted through true vibration of air, while human voice is transmitted through bone conduction vibration, there is a big difference in the frequencies of the two. In this scheme, the frequency range of bone conduction is obtained, and thus the frequency of each bone vibration decomposition signal can be judged based on Fourier decomposition. When the frequency of the bone vibration decomposition signal is not within the filtering interval, it means that it is environmental noise and needs to be removed.

[0107] Here, the bone vibration standard signal is imaged through sampling, which is achieved by setting at least one sampling point in the bone vibration standard signal, obtaining the signal value of each sampling point, and then plotting points in the coordinate system according to the time point and signal value of the sampling point to obtain the bone vibration standard waveform. This image is an approximate image.

[0108] Reference Figure 4 As shown, the amplification unit performs amplification processing on the skeleton filter signal, including the following steps:

[0109] Obtaining a value range after the signal is amplified, dividing the value range after the signal is amplified at equal intervals, and obtaining at least one amplified point;

[0110] Perform transmission test on the test signal whose signal strength is equal to the value at the amplification point, and take the amplification point corresponding to the test signal with the highest fidelity as the target amplification point;

[0111] The value at the target magnification point is divided by the strength of the bone filter signal to obtain the target magnification factor;

[0112] The amplification unit amplifies the bone filter signal according to the target amplification factor to obtain a bone amplification signal.

[0113] Signals of different strengths have different fidelity during transmission. Therefore, it is necessary to determine the optimal signal propagation strength. The target amplification point is the required strength. Therefore, when the target amplification point is determined, the bone filter signal is amplified to the value at the target amplification point.

[0114] Reference Figure 5 As shown, the noise reduction unit performs noise reduction processing on the bone amplification signal, including the following steps:

[0115] Use the noise formula to calculate the noise coefficient of the bone filter signal;

[0116] Select the bone amplification signal corresponding to the bone filter signal with the smallest noise coefficient as the target bone amplification signal;

[0117] Acquire at least one environmental noise signal and take the average value to obtain an approximate noise signal;

[0118] Under the condition that no noise occurs in the throat of the human body and the environment is noiseless, at least one internal vibration noise of the human body is obtained and the average is taken to obtain the internal vibration noise;

[0119] Removing approximate noise signals and internal vibration noise from the target bone amplification signal to obtain an effective sound signal;

[0120] The noise formula is as follows: ,

[0121] Among them, B is the noise coefficient, H(t) is the bone filter signal, and h(t) is the bone vibration signal corresponding to the bone filter signal.

[0122] Since the bone amplification signal is an amplified signal, although the noise is eliminated by filtering, the elimination is not complete. After the signal is amplified, the influence of the noise will appear again, so it needs to be eliminated again. However, due to the characteristics of bone conduction, it will capture all vibrations, environmental noise through the vibration of the air and the vibration of the internal organs of the human body. Here, the influence of environmental noise through the vibration of the air is very small. Therefore, the approximate noise signal obtained by sampling the environment at least once is sufficient to meet the needs. Since the vibration inside the human body is a periodic vibration, multiple samplings can be used to obtain a more approximate estimate.

[0123] Restoring the processed valid sound signal to a recognizable sound includes the following steps:

[0124] The sound restoration module restores the effective sound signal to a recognizable sound through the speaker.

[0125] Reference Figure 6 As shown, optimizing the recognizable sound to obtain the target output sound includes the following steps:

[0126] Based on historical data, obtain the volume range of human voice communication;

[0127] Obtaining the volume of the recognizable sound, determining whether the volume of the recognizable sound falls within the volume range of human voice communication, and if so, taking the recognizable sound as the target output sound;

[0128] If not, the volume of the recognizable sound is increased until the volume of the recognizable sound falls within the volume range of human voice communication, and the recognizable sound with the increased volume is used as the target output sound.

[0129] When a person speaks, the volume of his or her voice may not be fixed but may fluctuate. However, there is a range of volumes that are more suitable for communication. Therefore, the recognizable sound can be optimized according to this range to ensure that the output volume is more appropriate.

[0130] Reference Figure 7 As shown, the operating parameters of the control signal processing module and the sound restoration module include the following steps:

[0131] Based on the ratio of the workload of the signal processing module and the workload of the sound restoration module, the power module is allocated to the power supply module, and the first amount of electricity and the second amount of electricity are allocated to the signal processing module and the sound restoration module respectively;

[0132] Allocating the first amount of electricity according to the ratio of the operation amounts of the filtering unit, the amplifying unit, and the noise reduction unit;

[0133] The filtering unit, the amplifying unit and the noise reduction unit operate according to the amount of electricity obtained.

[0134] Reference Figure 8 As shown, monitoring the operation of the signal processing module and the sound restoration module and issuing an early warning for abnormal situations includes the following steps:

[0135] Monitor the quality of the effective sound signal generated by the signal processing module and the target output sound generated by the sound restoration module;

[0136] When there are abnormalities in the monitoring results, an early warning will be issued, otherwise, no action will be taken.

[0137] Reference Fig. 9 As shown, a full bone conduction brain machine vibration pickup method is applicable to the above-mentioned full bone conduction brain machine vibration pickup device, comprising:

[0138] At least one bone conduction sensor is respectively arranged at the bones on the top of the head, the left and right temporal bones, the bones on the back of the head and the bones on the larynx. The positions of the bone conduction sensors are determined according to a position algorithm. The bone conduction sensors collect bone vibration signals at the positions.

[0139] The collected bone vibration signal is filtered to obtain a bone filter signal, the bone filter signal is amplified to obtain a bone amplified signal, and the bone amplified signal is subjected to noise reduction. The bone vibration signal is filtered, amplified and subjected to noise reduction to obtain an effective sound signal;

[0140] Receive valid sound signals, restore the processed valid sound signals to recognizable sounds, optimize the recognizable sounds, and obtain target output sounds;

[0141] During the sound pickup process, the operating parameters of the signal processing module and the sound restoration module are controlled, the operating conditions of the signal processing module and the sound restoration module are monitored, and warnings are issued for abnormal conditions.

[0142] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0143] In summary, the advantages of the present invention are: by setting a signal acquisition module, a signal processing module and a sound restoration module, sound vibrations are picked up at multiple locations, so that the sound with the best pickup effect can be selected from multiple sounds, thereby ensuring the pickup effect. In addition, based on the characteristics of bone conduction, a noise reduction and filtering method different from the conventional processing method is used for filtering, which can more fully eliminate the noise. At the same time, when acquiring vibration, the position of the bone conduction sensor is restricted to ensure a better vibration extraction effect, thereby ensuring the effect of subsequent vibration processing.

[0144] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A full bone conduction brain-computer vibration pickup device, characterized in that: include: A signal acquisition module, wherein the signal acquisition module includes at least one bone conduction sensor, and the at least one bone conduction sensor is respectively arranged at the bones on the top of the head, the left and right temporal bones, the back of the head bones and the throat bones, and the positions of the bone conduction sensors are determined according to a position algorithm, and the bone conduction sensors collect bone vibration signals at the positions; A signal processing module, wherein the signal processing module receives at least one bone vibration signal collected by the signal collection module, and the signal processing module is composed of a filtering unit, an amplifying unit and a noise reduction unit. The filtering unit performs filtering processing on the collected bone vibration signal to obtain a bone filtered signal, the amplifying unit performs amplifying processing on the bone filtered signal to obtain a bone amplified signal, and the noise reduction unit performs noise reduction processing on the bone amplified signal. The bone vibration signal is filtered, amplified and noise-reduced to obtain an effective sound signal; A sound restoration module receives a valid sound signal, restores the processed valid sound signal to a recognizable sound, optimizes the recognizable sound, and obtains a target output sound; A control module, wherein the control module is connected to the signal processing module and the sound restoration module, controls the operating parameters of the signal processing module and the sound restoration module, monitors the operating conditions of the signal processing module and the sound restoration module, and issues an early warning for abnormal conditions; A power module, which provides power support for the signal acquisition module, the signal processing module, the sound restoration module and the control module; The position of the bone conduction sensor is determined according to a position algorithm and comprises the following steps: Perform three-dimensional modeling on the top bones of the head, the left and right temporal bones, the back bones of the head, and the throat bones to obtain a three-dimensional model of the overall skeleton; At least one sampling point is uniformly taken on the surface of the overall skeletal three-dimensional model; The sampling points that appear in the bones of the top of the head, the left and right temporal bones, the bones of the back of the head, and the bones of the larynx are respectively referred to as the head sampling point, the temporal sampling point, the spoon sampling point, and the laryngeal sampling point; Vibration capture is performed simultaneously at the head sampling point, the temporal sampling point, the spoon sampling point and the laryngeal sampling point to obtain at least one vibration intensity; The head sampling point with the largest vibration intensity is used as the target head sampling point, the temporal sampling point with the largest vibration intensity is used as the target temporal sampling point, the spoon sampling point with the largest vibration intensity is used as the target spoon sampling point, and the laryngeal sampling point with the largest vibration intensity is used as the target laryngeal sampling point; The target head sampling point, the target temporal sampling point, the target spoon sampling point, and the target laryngeal sampling point are used as the setting positions of the bone conduction sensors; The noise reduction unit performs noise reduction processing on the bone amplification signal, comprising the following steps: Use the noise formula to calculate the noise coefficient of the bone filter signal; Select the bone amplification signal corresponding to the bone filter signal with the smallest noise coefficient as the target bone amplification signal; Acquire at least one environmental noise signal and take the average value to obtain an approximate noise signal; Under the condition that no noise occurs in the throat of the human body and the environment is noiseless, at least one internal vibration noise of the human body is obtained and the average is taken to obtain the internal vibration noise; Removing approximate noise signals and internal vibration noise from the target bone amplification signal to obtain an effective sound signal; The noise formula is as follows: , Among them, B is the noise coefficient, H(t) is the bone filter signal, and h(t) is the bone vibration signal corresponding to the bone filter signal.

2. A full bone conduction brain-computer vibration pickup device according to claim 1, characterized in that: The filtering unit performs filtering processing on the collected bone vibration signal, comprising the following steps: In a noise-free environment, collecting at least one bone vibration sample signal; Performing Fourier transform decomposition on the bone vibration sample signal to obtain at least one bone vibration standard signal; The standard bone vibration signal is imaged by sampling to obtain a standard bone vibration waveform; Obtain the period of the standard waveform of bone vibration, and use the frequency formula to calculate the frequency of the standard signal of bone vibration; forming a filtering interval, wherein the endpoints of the filtering interval are formed by the maximum value and the minimum value of the frequency of at least one bone vibration standard signal; Performing Fourier transform on the bone vibration signal to obtain at least one bone vibration decomposition signal; Deleting the bone vibration decomposition signal whose frequency does not belong to the filtering interval, and combining at least one bone vibration decomposition signal after the deletion process by using inverse Fourier transform to obtain a bone filtering signal; The Fourier transform is as follows: , Where, F(x) is the signal after Fourier transform, i is a unit imaginary number, e is a natural constant, f(t) is the signal before Fourier transform, t is time, and x is frequency; The inverse Fourier transform is as follows: , Among them, G(x) is the signal before inverse Fourier transform, and g(t) is the signal after inverse Fourier transform; The frequency formula is as follows: , Wherein, A is the frequency of the bone vibration sample signal, and a is the period of the bone vibration sample waveform.

3. A full bone conduction brain-computer vibration pickup device according to claim 2, characterized in that: The amplification unit amplifies the skeleton filter signal and comprises the following steps: Obtaining a value range after the signal is amplified, dividing the value range after the signal is amplified at equal intervals, and obtaining at least one amplified point; Perform transmission test on the test signal whose signal strength is equal to the value at the amplification point, and take the amplification point corresponding to the test signal with the highest fidelity as the target amplification point; The value at the target magnification point is divided by the strength of the bone filter signal to obtain the target magnification factor; The amplification unit amplifies the bone filter signal according to the target amplification factor to obtain a bone amplification signal.

4. A full bone conduction brain-computer vibration pickup device according to claim 3, characterized in that: The process of restoring the processed effective sound signal to a recognizable sound comprises the following steps: The sound restoration module restores the effective sound signal to a recognizable sound through the speaker.

5. The full bone conduction brain-computer vibration pickup device according to claim 4, characterized in that: The optimizing process of the recognizable sound to obtain the target output sound comprises the following steps: Based on historical data, obtain the volume range of human voice communication; Obtaining the volume of the recognizable sound, determining whether the volume of the recognizable sound falls within the volume range of human voice communication, and if so, taking the recognizable sound as the target output sound; If not, the volume of the recognizable sound is increased until the volume of the recognizable sound falls within the volume range of human voice communication, and the recognizable sound with the increased volume is used as the target output sound.

6. A full bone conduction brain-computer vibration pickup device according to claim 5, characterized in that: The operating parameters of the control signal processing module and the sound restoration module include the following steps: Based on the ratio of the workload of the signal processing module and the workload of the sound restoration module, the power module is allocated to the power supply module, and the first amount of electricity and the second amount of electricity are allocated to the signal processing module and the sound restoration module respectively; Allocating the first amount of electricity according to the ratio of the operation amounts of the filtering unit, the amplifying unit, and the noise reduction unit; The filtering unit, the amplifying unit and the noise reduction unit operate according to the amount of electricity obtained.

7. The full bone conduction brain-computer vibration pickup device according to claim 6, characterized in that: The monitoring of the operation of the signal processing module and the sound restoration module and the early warning of abnormal situations include the following steps: Monitor the quality of the effective sound signal generated by the signal processing module and the target output sound generated by the sound restoration module; When there are abnormalities in the monitoring results, an early warning will be issued, otherwise, no action will be taken.

8. A full bone conduction brain-machine vibration pickup method, applicable to the full bone conduction brain-machine vibration pickup device as claimed in any one of claims 1 to 7, characterized in that: include: At least one bone conduction sensor is respectively arranged at the bones on the top of the head, the left and right temporal bones, the bones on the back of the head and the bones on the larynx. The positions of the bone conduction sensors are determined according to a position algorithm. The bone conduction sensors collect bone vibration signals at the positions. The collected bone vibration signal is filtered to obtain a bone filter signal, the bone filter signal is amplified to obtain a bone amplified signal, and the bone amplified signal is subjected to noise reduction. The bone vibration signal is filtered, amplified and subjected to noise reduction to obtain an effective sound signal; Receive valid sound signals, restore the processed valid sound signals to recognizable sounds, optimize the recognizable sounds, and obtain target output sounds; During the sound pickup process, control the operating parameters of the signal processing module and the sound restoration module, monitor the operating conditions of the signal processing module and the sound restoration module, and issue early warnings for abnormal conditions; The position of the bone conduction sensor is determined according to a position algorithm and comprises the following steps: Perform three-dimensional modeling on the top bones of the head, the left and right temporal bones, the back bones of the head, and the throat bones to obtain a three-dimensional model of the overall skeleton; At least one sampling point is uniformly taken on the surface of the overall skeletal three-dimensional model; The sampling points that appear in the bones of the top of the head, the left and right temporal bones, the bones of the back of the head, and the bones of the larynx are respectively referred to as the head sampling point, the temporal sampling point, the spoon sampling point, and the laryngeal sampling point; Vibration capture is performed simultaneously at the head sampling point, the temporal sampling point, the spoon sampling point and the laryngeal sampling point to obtain at least one vibration intensity; The head sampling point with the largest vibration intensity is used as the target head sampling point, the temporal sampling point with the largest vibration intensity is used as the target temporal sampling point, the spoon sampling point with the largest vibration intensity is used as the target spoon sampling point, and the laryngeal sampling point with the largest vibration intensity is used as the target laryngeal sampling point; The target head sampling point, the target temporal sampling point, the target spoon sampling point, and the target laryngeal sampling point are used as the setting positions of the bone conduction sensors; The noise reduction unit performs noise reduction processing on the bone amplification signal, comprising the following steps: Use the noise formula to calculate the noise coefficient of the bone filter signal; Select the bone amplification signal corresponding to the bone filter signal with the smallest noise coefficient as the target bone amplification signal; Acquire at least one environmental noise signal and take the average value to obtain an approximate noise signal; Under the condition that no noise occurs in the throat of the human body and the environment is noiseless, at least one internal vibration noise of the human body is obtained and the average is taken to obtain the internal vibration noise; Removing approximate noise signals and internal vibration noise from the target bone amplification signal to obtain an effective sound signal; The noise formula is as follows: , Among them, B is the noise coefficient, H(t) is the bone filter signal, and h(t) is the bone vibration signal corresponding to the bone filter signal.

Images