An alarm method and alarm device

Abstract

The application provides an alarm method and an alarm device, and relates to the technical field of automatic alarm. The alarm method is applied to an alarm device, and the alarm device is buried in a preset range of a target facility. The alarm device comprises a shell, a sound pressure sensor array rigidly connected with the shell, and a vibration velocity sensor array rigidly connected with the shell. The method comprises the following steps: acquiring a sound pressure signal collected by the sound pressure sensor array and a vibration velocity signal collected by the vibration velocity sensor array; detecting a sound-emitting target according to the sound pressure signal; detecting the direction of the sound-emitting target according to the vibration velocity signal; and sending an alarm information when the sound-emitting target is in a preset angle interval. The preset angle interval is used for indicating the direction of the target facility. The method is suitable for the process of risk alarm of the target facility, and is used for solving the problem of low alarm accuracy of current alarm products.

Description

An alarm method and alarm device Technical Field

[0001] This application relates to the field of automatic alarm technology, and in particular to an alarm method and alarm device. Background Technology

[0002] Unattended ATMs, especially those in remote areas, are more likely to become targets for vandalism. While ATMs themselves have robust anti-vandalism measures and devices, malicious individuals often resort to drilling, cutting, or other methods to damage them.

[0003] To detect acts of vandalism against ATMs, current alarm products can detect sounds in the vicinity of the ATM (such as the ATM lobby). Since the sounds of drilling, cutting, and other similar methods are usually quite loud, if a loud sound is detected, it can be assumed that the ATM has been vandalized, and an alarm message will be issued.

[0004] However, ATMs or ATM lobbies are usually located outdoors or on the street. The surrounding environment of ATMs contains a lot of other noise, such as the roar of mechanical equipment, the vibration noise of large mechanical equipment passing by, and the vibration noise of pedestrians' footsteps. Current alarm products are greatly affected by the ambient noise around ATMs, which can easily produce false alarms and have low accuracy. Summary of the Invention

[0005] To address the aforementioned technical issues, this application provides an alarm method and an alarm device that can detect the location of the sound-emitting target, thereby improving the accuracy of the alarm.

[0006] In a first aspect, this application provides an alarm method applied to an alarm device embedded within a preset range of a target facility. The alarm device includes a housing, a sound pressure sensor array rigidly connected to the housing, and a vibration velocity sensor array rigidly connected to the housing. The method includes: acquiring sound pressure signals collected by the sound pressure sensor array and vibration velocity signals collected by the vibration velocity sensor array; detecting a sound-emitting target based on the sound pressure signals; detecting the orientation of the sound-emitting target based on the vibration velocity signals; and issuing an alarm message when the sound-emitting target is within a preset angle range. The preset angle range is used to indicate the orientation of the target facility.

[0007] Optionally, detecting a sound-emitting target based on a sound pressure signal includes: generating a minimum mean square distortion-free response (MVDR) beam based on the sound pressure signal; determining whether the power of the MVDR beam is greater than a preset threshold; and if the power of the MVDR beam is greater than the preset threshold, then determining that a sound-emitting target exists.

[0008] Optionally, determining whether the power of the MVDR beam is greater than a preset threshold includes:

[0009] Determine whether the power of the MVDR beam is greater than a preset threshold using the following formula:

[0010]

[0011] Where Th represents the preset threshold; Indicates the power of the MVDR beam; R n Represents the covariance matrix of the noise array; y(t) represents the sound pressure signal when the sound-emitting target is present, y(t) = V*s(t) + n(t); s(t) represents the effective signal; n(t) represents the noise signal; V represents the array manifold. Indicates the estimated array manifold;

[0012] like in, u1 represents the eigenvector corresponding to the largest eigenvalue of the effective signal covariance matrix. Let represent the theoretical array manifold, ε represent the upper limit of error for the array manifold, and M represent the number of sound pressure sensors in the sound pressure sensor array. Then:

[0013]

[0014] in, Converted to eigenvalue decomposition form U represents the eigenvector; Γ represents the eigenvalue matrix. γ1 represents the largest eigenvalue; I represents the identity matrix.

[0015] like but:

[0016]

[0017] Among them, W a Indicates the weighting coefficients.

[0018] Optionally, detecting the orientation of the sound-emitting target based on the vibration velocity signal includes: performing line spectrum detection on the MVDR beam according to a preset frequency band; if a line spectrum exists, calculating the sound pressure spectrum at the line spectrum; the line spectrum is used to indicate the frequency of the sound signal emitted by the sound-emitting target; determining the vibration velocity spectrum of the line spectrum in a first preset direction based on the vibration velocity signal; determining the sound energy flow component in the first preset direction based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the first preset direction; determining the vibration velocity spectrum of the line spectrum in a second preset direction based on the vibration velocity signal; the first preset direction is perpendicular to the second preset direction; determining the sound energy flow component in the second preset direction based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the second preset direction; and determining the orientation of the sound-emitting target based on the sound energy flow component in the first preset direction and the sound energy flow component in the second preset direction.

[0019] Optionally, the acoustic energy flow component in the first preset direction is determined based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the first preset direction, including: calculating the acoustic energy flow component in the first preset direction according to the following formula:

[0020] I x =Re(P(f)×V) x (f));

[0021] Among them, I x P(f) represents the acoustic energy flow component in the first preset direction; P(f) represents the sound pressure spectrum at the line spectrum; V x (f) represents the velocity spectrum of the line spectrum in the first preset direction;

[0022] Based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the second preset direction, determine the sound energy flow component in the second preset direction, including: calculating the sound energy flow component in the second preset direction according to the following formula:

[0023] I y =Re(P(f)×V) x (f));

[0024] Among them, I y V represents the acoustic energy flow component in the second preset direction; x (f) indicates the vibration velocity spectrum of the line spectrum in the second preset direction;

[0025] The location of the sound-emitting target is determined based on the sound energy flow component in the first preset direction and the sound energy flow component in the second preset direction, including calculating the location of the sound-emitting target according to the following formula:

[0026] β=atan(I y / I x );

[0027] Where β represents the azimuth angle of the sound-emitting target.

[0028] Optionally, the method further includes: if no line spectrum exists, calculating the acoustic energy flow corresponding to each frequency point within the preset frequency band; based on the acoustic energy flow corresponding to each frequency point, calculating the acoustic energy intensity corresponding to each direction; and determining the direction with the largest corresponding acoustic energy intensity as the direction of the sound-emitting target.

[0029] Optionally, the vibration velocity sensor array includes a first vibration velocity sensor group and a second vibration velocity sensor group; the first vibration velocity sensor group is used to collect vibration velocity signals in a first direction; the first direction is the same as a first preset direction; the second vibration velocity sensor group is used to collect vibration velocity signals in a second direction; the second direction is the same as a second preset direction.

[0030] Optionally, when the sound-emitting target is within a preset angle range, an alarm message is issued, including: extracting the beam signal from the MVDR beam to obtain the beam signal of the target's azimuth; matching the voiceprint features of the target's azimuth beam signal with a preset voiceprint feature library corresponding to the target event; the preset voiceprint library includes multiple target voiceprint features; when the azimuth of the sound-emitting target is within the preset angle range, if the voiceprint features of the target's azimuth beam signal match any one of the multiple target voiceprint features, an alarm message is issued.

[0031] Optionally, the method further includes: weighting the MVDR beam using weighting coefficients before extracting the beam signal from the MVDR beam; the weighting coefficients satisfy the following relationship:

[0032] W=(IC(C H C) -1 C H W a ;

[0033] Where W represents the weighting coefficient; C = [V(θ), D(θ)]; V(θ) represents the array manifold at θ, where θ is the orientation of the preset interfering object;

[0034] Optionally, the sound pressure sensor array is a first circular array; the vibration velocity sensor array is a second circular array; the second circular array is located within the first circular array.

[0035] It should be understood that current alarm products issue alarms when a loud sound is detected, making them susceptible to ambient noise and prone to false alarms. The alarm method provided in this application, after detecting the sound-emitting target using sound pressure signals, can further detect the target's location using vibration velocity signals. It only issues an alarm when the target's location is within a preset angular range used to indicate the location of the target facility. This avoids false alarms caused by noise from interfering objects outside the target facility's location, thereby improving alarm accuracy.

[0036] Secondly, this application provides an alarm device, which is embedded within a preset range of a target facility. The alarm device includes: a housing; a sound pressure sensor array rigidly connected to the housing for collecting sound pressure signals; a vibration velocity sensor array rigidly connected to the housing for collecting vibration velocity signals; and a processing control module disposed within the housing for detecting a sound-emitting target based on the sound pressure signal and detecting the orientation of the sound-emitting target based on the vibration velocity signal. When the orientation of the sound-emitting target is within a preset angle range, an alarm message is issued. The preset angle range is used to indicate the orientation of the target facility.

[0037] Thirdly, this application provides a computer program product that, when run in an alarm device, causes the alarm device to perform the steps of the related method described in the first aspect, so as to implement the method described in the first aspect.

[0038] Fourthly, this application provides a readable storage medium comprising: software instructions; when the software instructions are executed in an alarm device, the alarm device causes the alarm device to implement the method described in the first aspect above.

[0039] The beneficial effects of the second to fourth aspects mentioned above can be referred to the first aspect, and will not be repeated here. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 is a schematic diagram of the composition of the alarm system provided in an embodiment of this application;

[0042] Figure 2 is a schematic diagram of the composition of the alarm device 100 provided in the embodiment of this application;

[0043] Figure 3 is a schematic diagram of the directivity of a single vibration velocity sensor provided in an embodiment of this application;

[0044] Figure 4 is an assembly diagram provided in an embodiment of this application;

[0045] Figure 5 is a schematic diagram of the vibration velocity sensor circuit connection provided in the embodiment of this application;

[0046] Figure 6 is a flowchart illustrating the alarm method provided in an embodiment of this application;

[0047] Figure 7 is another flowchart illustrating the alarm method provided in this application embodiment;

[0048] Figure 8 is a schematic flowchart of another alarm method provided in the embodiments of this application;

[0049] Figure 9 is a schematic diagram of the installation of the alarm device provided in the embodiment of this application;

[0050] Figure 10 is a schematic flowchart of another alarm method provided in the embodiments of this application;

[0051] Figure 11 is a schematic flowchart of another alarm method provided in the embodiments of this application;

[0052] Figure 12 is a flowchart of signal classification and recognition provided in an embodiment of this application;

[0053] Figure 13 is a comparative schematic diagram of constrained beamforming and conventional beamforming provided in the embodiments of this application;

[0054] Figure 14 is a schematic diagram of the composition of the alarm device provided in the embodiment of this application. Detailed Implementation

[0055] Hereinafter, the terms "first," "second," and "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first," "second," or "third," etc., may explicitly or implicitly include one or more of that feature.

[0056] Unattended ATMs, especially those in remote and isolated areas, are more likely to become targets for vandalism. While ATMs themselves have robust anti-vandalism measures and devices, malicious individuals often resort to drilling, cutting, or other methods to damage them.

[0057] To detect acts of vandalism against ATMs, current alarm products can detect sounds in the vicinity of the ATM (such as the ATM lobby). Since the sounds of drilling, cutting, and other similar methods are usually quite loud, if a loud sound is detected, it can be assumed that the ATM has been vandalized, and an alarm message will be issued.

[0058] However, ATMs or ATM lobbies are usually located outdoors or on the street. The surrounding environment of ATMs contains a lot of other noise, such as the roar of mechanical equipment, the vibration noise of large mechanical equipment passing by, and the vibration noise of pedestrians' footsteps. Current alarm products are greatly affected by the ambient noise around ATMs, which can easily produce false alarms and have low accuracy.

[0059] Based on this, embodiments of this application provide an alarm method and an alarm device that can detect the location of the sound-emitting target and improve the accuracy of the alarm.

[0060] The following description is provided in conjunction with the accompanying drawings.

[0061] Figure 1 is a schematic diagram of the composition of an alarm system provided in an embodiment of this application. As shown in Figure 1, the alarm system may include an alarm device 100 and an alarm receiving device 200. The alarm device 100 and the alarm receiving device 200 can be connected via a wired network or a wireless network.

[0062] The alarm device 100 can be installed within a preset range of the target facility to detect the sound-emitting target and its location. The specific composition and detection process of the alarm device 100 can be referred to the following embodiments, and will not be repeated here.

[0063] In some possible embodiments, the alarm device 100 may also be used to send alarm information to the alarm receiving device 200 when a sound-emitting target is detected and the sound-emitting target is within the azimuth range of the target facility.

[0064] For example, the alarm device 100 may have a preset voiceprint library corresponding to the target event (e.g., the AMT destruction event). When the alarm device 100 detects a sound-emitting target and the sound-emitting target is within the location range of the target facility, it can use the preset voiceprint library to match the detected signal. If the signal matches the voiceprint features in the preset voiceprint library, it can send an alarm message to the alarm receiving device 200.

[0065] As described above, the alarm device 100 and the alarm receiving device 200 can be connected via a wired network or a wireless network. This wired or wireless network may include one or more media or devices capable of transmitting alarm information from the alarm device 100 to the alarm receiving device 200. The one or more communication media may include wireless, and / or wired communication media, such as wireless spectrum (including Bluetooth spectrum or radio frequency (RF) spectrum) or one or more physical transmission lines.

[0066] In other possible embodiments, the alarm device 100 may also be connected to a speaker or an alarm light. The alarm device 100 may also be used to issue an audible alarm or a light alarm via a speaker when a sound-emitting target is detected and the sound-emitting target is within the location range of the target facility.

[0067] The following describes the composition of the alarm device 100.

[0068] Figure 2 is a schematic diagram of the composition of the alarm device 100 provided in the embodiment of this application. As shown in Figure 2, the alarm device 100 may include: a vector sensor 110, a pre-processing module 120, a signal processing module 130, a main control module 140, a transmission module 150, a power supply module 160, and a housing 170.

[0069] The vector sensor 110 may include a sound pressure sensor array 111 and a vibration velocity sensor array 112.

[0070] The sound pressure sensor array 111 can be used to acquire sound pressure signals.

[0071] The sound pressure sensor array 111 may include multiple sound pressure sensors. These sound pressure sensors may be, for example, piezoelectric ceramic sound pressure sensors or other types of omnidirectional sound pressure sensors. This application embodiment does not limit this.

[0072] The vibration velocity sensor array 112 can be used to acquire vibration velocity signals.

[0073] The vibration velocity sensor array 112 may include multiple vibration velocity sensors. These sensors may be, for example, moving-coil vibration velocity sensors or other types of vibration velocity sensors. This application embodiment does not impose any limitations on this.

[0074] Moving-coil velocity sensors exhibit high sensitivity in the low-frequency range, effectively picking up low-frequency vibration signals. Furthermore, the velocity sensor itself has a figure-eight orientation, which effectively suppresses interference.

[0075] For example, Figure 3 is a schematic diagram of the directivity of a single vibration velocity sensor provided in an embodiment of this application. As shown in Figure 3, within the range of 0° to 360°, the signal strength is relatively high at 90° and 270° (shown as 1 in Figure 3), and the signal strength gradually decreases on both sides of 90° and 270°. The signal strength is 0 at 0° and 180°. The signal strength change presents an "8" shape.

[0076] The sound pressure sensor array 111 and the vibration velocity sensor array 112 can be rigidly connected to the housing 170. For example, the sound pressure sensor array 111 and the vibration velocity sensor array 112 can be fixed to the housing with screws, or the sound pressure sensors in the sound pressure sensor array 111 and the vibration velocity sensors in the vibration velocity sensor array 112 can be provided with external threads, and the housing 170 can be provided with threaded holes, so that the sound pressure sensors and vibration velocity sensors can be connected by threads, etc. The embodiments of this application do not limit the specific method of rigid connection.

[0077] For example, FIG4 is an assembly schematic diagram provided in an embodiment of the present application. As shown in FIG4(a), the housing 170 may include a top shell 171 and a bottom shell 172 (FIG4 shows an example of the height of the top shell 171 and the bottom shell 172 after assembly being 80 mm).

[0078] As shown in Figure 4(b), the sound pressure sensor array 111 can specifically be a ring array (for example, it can be called a first ring array) rigidly connected to the bottom shell 172. The first ring array can include eight evenly distributed sound pressure sensors (that is, the angle difference between two adjacent sound pressure sensors in the sound pressure sensor array 111 is 45°). Each sound pressure sensor is circular with a diameter of 15 mm, and the diameter of the first ring array is 300 mm.

[0079] The vibration velocity sensor array 112 can also be specifically a ring array (for example, it can be called a second ring array) rigidly connected to the bottom shell 172. The second ring array can include four vibration velocity sensors evenly distributed (that is, the angle difference between two adjacent vibration velocity sensors in the vibration velocity sensor array 112 is 90°). Each vibration velocity sensor is cylindrical with a length of 25 mm and a ground diameter of 15 mm. The second ring array is located inside the first ring array, and the diameter of the second ring array is 60 mm.

[0080] It should be noted that the specific sensor size and array size in Figure 4 are examples only. Other values ​​can also be used for the sensor size and array size, and this application embodiment does not limit them.

[0081] The preprocessing module 120 can be used to amplify and filter signals.

[0082] For example, Figure 5 is a schematic diagram of the vibration velocity sensor circuit connection provided in an embodiment of this application. As shown in Figure 5, taking the vibration velocity sensor array 112 shown in Figure 4(b) above, which includes 4 vibration velocity sensors, as an example, the 4 vibration velocity sensors can be divided into two vibration velocity sensor groups. The two vertical vibration velocity sensors in Figure 5 can be divided into the first vibration velocity sensor group, which can be used to collect vibration velocity signals in the first direction (that is, the vertical direction in Figure 5, shown as Vx in Figure 5); the two horizontal vibration velocity sensors in Figure 5 can be divided into the second vibration velocity sensor group, which can be used to collect vibration velocity signals in the second direction (that is, the horizontal direction in Figure 5, shown as Vy in Figure 5).

[0083] The two velocity sensors in the first velocity sensor group can be connected in series and then connected in series with a differential amplifier circuit in the first direction. The two velocity sensors in the second velocity sensor group can be connected in series and then connected in series with a differential amplifier circuit in the second direction. The differential amplifier circuits in the first and second directions shown in Figure 5 can be understood as components of the pre-processing module 120. The differential amplifier circuits amplify the velocity signals and filter interference signals generated when the power supply module 160 supplies power. Specific processes can be found in related technical documents and will not be elaborated here.

[0084] The signal processing module 130 may include a detection module 131, a positioning module 132, and a classification and recognition module 133.

[0085] The detection module 131 can be used to detect the sound-emitting target based on the sound pressure signal.

[0086] The positioning module 132 can be used to detect the location of the sound-emitting target based on the vibration velocity signal.

[0087] The classification and recognition module 133 can be used to match the voiceprint features of the sound signal emitted by the sound source with a preset voiceprint feature library to determine the type of sound signal.

[0088] The specific detection and analysis process of the detection module 131, the positioning module 132, and the classification and recognition module 133 can be referred to the following embodiments, and will not be repeated here.

[0089] The main control module 140 can be used to distribute and transmit control commands and data to various modules.

[0090] For example, a command is sent to the transmission module 150 so that the transmission module 150 sends alarm information to the alarm receiving device 200.

[0091] Optionally, the signal processing module 130 and the main control module 140 can be integrated into a processing control module. The processing control can be used to detect the sound-emitting target and its location based on the sound pressure signal and vibration velocity signal, and to issue an alarm message when the target is within a preset angle range. The specific process can be referred to in the following embodiments, and will not be repeated here.

[0092] The power supply module 160 can provide the appropriate voltage required by the aforementioned modules.

[0093] It should be noted that the structural composition shown in Figure 2 above does not constitute a limitation on the alarm device 100. In addition to the components shown in Figure 2, the alarm device 100 may include more or fewer components than shown (e.g., only including the sound pressure sensor array 111, the vibration velocity sensor array 112, and the processing control module (the module obtained after integrating the signal processing module 130 and the main control module 140)), or combinations of certain components, or different arrangements of components. The embodiments of this application do not impose any limitations on this.

[0094] The alarm receiving device 200 can be an electronic device with computing and processing capabilities, such as a computer or server.

[0095] The server can be a single server or a server cluster consisting of multiple servers. In some embodiments, the server cluster can also be a distributed cluster. Optionally, the server can also be implemented on a cloud platform, such as a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, inter-cloud, and multi-cloud, or any combination thereof. This application does not impose any limitations on this.

[0096] The alarm receiving device 200 can be used to receive alarm information sent by the alarm device 100.

[0097] The entity executing the alarm method provided in this application embodiment can be the alarm device 100 described above; or, it can be the processor (e.g., central processing unit (CPU)) in the alarm device 100; or, it can be an application (APP) with alarm function installed in the alarm device 100; or, it can be a functional module or functional unit in the alarm device 100 used to execute the alarm method, etc. This application embodiment does not impose any limitations on this.

[0098] The alarm method provided in the embodiments of this application is described below.

[0099] Figure 6 is a flowchart illustrating the alarm method provided in an embodiment of this application. As shown in Figure 6, the method includes steps S101 to S104.

[0100] S101, acquire the sound pressure signal collected by the sound pressure sensor array and the vibration velocity signal collected by the vibration velocity sensor array.

[0101] S102. Detect the sound-emitting target based on the sound pressure signal.

[0102] S102 can be referred to as S1021 to S1023 in Figure 7 below, and will not be repeated here.

[0103] S103. Detect the location of the sound-emitting target based on the vibration velocity signal.

[0104] S103 can be referred to as S1031 to S1037 in Figure 8 or S201 to S203 in Figure 10 below, and will not be repeated here.

[0105] S104. When the target of the sound is within the preset angle range, an alarm message is issued.

[0106] The preset angle range is used to indicate the location of the target facility.

[0107] S104 can be referred to as S1041 to S1043 in Figure 11 below, and will not be repeated here.

[0108] It should be understood that current alarm products issue alarms when a loud sound is detected, making them susceptible to ambient noise and prone to false alarms. The alarm method provided in this application, after detecting the sound-emitting target using sound pressure signals, can further detect the target's location using vibration velocity signals. An alarm is only issued when the target's location is within a preset angle range used to indicate the location of the target facility. This avoids false alarms caused by noise from interfering objects outside the target facility's location, thereby improving alarm accuracy.

[0109] The following is a description of S102.

[0110] In some possible embodiments, the alarm device can use a minimum variance distortionless response (MVDR) beam generation algorithm to generate a beam for scanning. If the beam power is high, the presence of a sound-emitting target is determined. In this case, Figure 7 is another flowchart of the alarm method provided in this application embodiment. As shown in Figure 7, the above-mentioned S102 may specifically include S1021 to S1023.

[0111] S1021. Generate MVDR beams based on sound pressure signals.

[0112] S1022. Determine whether the power of the MVDR beam is greater than the preset threshold.

[0113] In one possible implementation, the array signal is assumed to be represented by the following formula (1) when the sound-emitting target is present:

[0114] y(t)=V*s(t)+n(t), t=0,1,…:H1 Formula (1)

[0115] In formula (1), y(t) represents the sound pressure signal when the sound-emitting target is present. V represents the array manifold. s(t) represents the valid signal. n(t) represents the noise signal.

[0116] Since both the effective signal and the noise signal conform to a zero-mean Gaussian distribution, the alarm device can specifically determine whether the power of the MVDR beam is greater than the preset threshold using the following formula (2):

[0117]

[0118] In formula (2), > indicates the presence of a sound-emitting target. < indicates the absence of a sound-emitting target. Th represents the preset threshold. R n This represents the covariance matrix of the noise array.

[0119] In another possible implementation, factors such as acoustic pressure sensor position errors and environmental changes can lead to variations in the array manifold, resulting in errors in the final beam power estimation. This manifests as a decrease in the output signal-to-noise ratio during detection, ultimately causing false alarms. Therefore, the MVDR algorithm can be improved by adding multiple constraints, increasing its tolerance, and enhancing its performance.

[0120] Because MVDR can be equivalent to the following formula (3):

[0121] max σ 2 subject to R-σ 2 VV H ≥0 Formula (3)

[0122] In formula (3), σ 2 Indicates signal power.

[0123] Furthermore, to increase tolerance, the constraints of the array manifold can be added using the following formulas (4) and (5):

[0124]

[0125] In formula (4), ε represents the theoretical array manifold. ε represents the upper limit of error for the array manifold.

[0126] ||V|| 2 =M Formula (5)

[0127] In formula (5), M represents the number of sound pressure sensors in the sound pressure sensor array.

[0128] Then, based on formulas (3) to (5), the predicted array flow pattern can be estimated:

[0129] Case 1, if in, u1 represents the eigenvector corresponding to the largest eigenvalue of the effective signal covariance matrix. Then, the predicted array manifold can be expressed as the following formula (6):

[0130]

[0131] In formula (6), Converted to eigenvalue decomposition form U represents the eigenvector. Γ represents the eigenvalue matrix. γ1 represents the largest eigenvalue. I represents the identity matrix.

[0132] Scenario 2, if The predicted array manifold can then be expressed as the following formula (7):

[0133]

[0134] In formula (7), W a Indicates the weighting coefficients.

[0135] The alarm device can then determine whether the power of the MVDR beam is greater than the preset threshold using the following formula (8):

[0136]

[0137] In formula (8), This indicates the power of the MVDR beam.

[0138] S1023. If the power of the MVDR beam is greater than the preset threshold, then it is determined that there is a sound-emitting target.

[0139] The following is a description of S103.

[0140] In some possible embodiments, Figure 8 is another schematic flowchart of the alarm method provided in the embodiments of this application. As shown in Figure 8, the above-mentioned S103 may specifically include S1031 to S1037.

[0141] S1031. Perform line spectrum detection on the MVDR beam according to the preset frequency band.

[0142] The preset frequency band can be pre-set by the administrator in the alarm device. The preset frequency band is used to characterize the frequency band of the sound signal emitted by the target event (such as the aforementioned drilling or cutting of an ATM). For example, the preset frequency band can be 100 Hz to 250 Hz, or other preset frequency bands. This application embodiment does not limit the specific value of the preset frequency band. Line spectrum detection can be referred to in related technologies, and will not be repeated here.

[0143] S1032. If there is a line spectrum in the preset frequency band, calculate the sound pressure spectrum at the line spectrum.

[0144] The line spectrum indicates the frequency of the sound signal emitted by the target. The sound pressure spectrum (SPS) is the result of the short-time Fourier transform (STFT) of the sound signal, providing the frequency content of the sound signal at different points in time. In the SPS, the horizontal axis represents time, the vertical axis represents frequency, and the color or grayscale represents the sound pressure value at that time point and frequency. Generally, the brighter the color in the SPS, the higher the sound pressure value at that location. By observing the SPS, one can understand the frequency distribution of the sound signal at different times and the intensity changes of each frequency.

[0145] S1033. Determine the vibration velocity spectrum of the line spectrum in the first preset direction based on the vibration velocity signal.

[0146] The velocity spectrum is another spectral representation used to describe sound signals, indicating the vibration velocity of each frequency component within the sound signal. The velocity spectrum is also calculated using the STFT.

[0147] S1034. Determine the acoustic energy flow component in the first preset direction based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the first preset direction.

[0148] In one possible implementation, the alarm device can specifically calculate the acoustic energy flow component in the first preset direction according to the following formula (9):

[0149] I x =Re(P(f)×V) x (f) Formula (9)

[0150] In formula (9), I x This represents the acoustic energy flow component in the first preset direction. P(f) represents the sound pressure spectrum at the line spectrum. V x (f) represents the vibration velocity spectrum of the line spectrum in the first preset direction.

[0151] S1035. Based on the vibration velocity signal, determine the vibration velocity spectrum of the line spectrum in the second preset direction.

[0152] The first preset direction is perpendicular to the second preset direction.

[0153] Optionally, as described above, the vibration velocity sensor array may include a first vibration velocity sensor group and a second vibration velocity sensor group. The first vibration velocity sensor group is used to acquire vibration velocity signals in a first direction, and the second vibration velocity sensor group is used to acquire vibration velocity signals in a second direction. In this case, the first direction may be the same as a first preset direction, and the second direction may be the same as a second preset direction.

[0154] For example, Figure 9 is a schematic diagram of the installation of the alarm device provided in an embodiment of this application. As shown in Figure 9, taking an ATM as an example, the alarm device 100 can be installed inside the ATM lobby. Figure 9 shows examples of the ATM being located at alarm device 100a° and the ATM lobby door being located at alarm device 100θ°. Assuming the first preset direction is north-south, the first direction in which the first vibration velocity sensor group collects vibration velocity signals is also north-south. Assuming the second preset direction is east-west, the second direction in which the second vibration velocity sensor group collects vibration velocity signals is also east-west.

[0155] S1036. Determine the acoustic energy flow component in the second preset direction based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the second preset direction.

[0156] In one possible implementation, the alarm device can specifically calculate the acoustic energy flow component in the second preset direction according to the following formula (10):

[0157] I y =Re(P(f)×V) y (f) Formula (10)

[0158] In formula (10), I y V represents the acoustic energy flow component in the second preset direction; y (f) represents the vibration velocity spectrum of the line spectrum in the second preset direction.

[0159] S1037. Determine the orientation of the sound-emitting target based on the sound energy flow component in the first preset direction and the sound energy flow component in the second preset direction.

[0160] In one possible implementation, the alarm device can specifically calculate the location of the sound-emitting target according to the following formula (11):

[0161] β=atan(I y / I x ) Formula (11)

[0162] Where β represents the azimuth angle of the sound-emitting target.

[0163] In some other possible embodiments, for the case where no line spectrum exists, FIG10 is another schematic flowchart of the alarm method provided in the embodiment of this application. As shown in FIG10, the method may further include S201 to S203.

[0164] S201. If there is no line spectrum within the preset frequency band, calculate the acoustic energy flow corresponding to each frequency point within the preset frequency band.

[0165] S202. Based on the acoustic energy flow corresponding to each frequency point, calculate the acoustic energy intensity corresponding to each direction.

[0166] S201 to S202 can be referred to as described in S1033 to S1036 above, and will not be repeated here.

[0167] S203. Determine the direction of the sound source as the direction of the sound energy intensity.

[0168] The following is a description of S104.

[0169] In some possible embodiments, the alarm device can directly issue an alarm message when the sound source is within a preset range.

[0170] In other possible embodiments, when the sound-emitting target is within a preset range, the alarm device can also match the target voiceprint features of the sound signal of the sound-emitting target to the sound of drilling or cutting an ATM. If a match is found, an alarm message is issued. In this case, Figure 11 is another schematic flowchart of the alarm method provided in the embodiment of this application. As shown in Figure 11, the above-mentioned S104 may specifically include S1041 to S1043.

[0171] S1041. Extract the beam signal from the MVDR beam to obtain the beam signal of the target azimuth.

[0172] The beam signal at the target azimuth can be a continuous time-domain signal or a continuous frequency-domain signal. This application does not impose any limitations on this.

[0173] S1042. Match the acoustic signature features of the target azimuth beam signal with the preset acoustic signature feature library corresponding to the target event.

[0174] The preset voiceprint library includes multiple target voiceprint features.

[0175] The establishment of a pre-defined voiceprint database mainly involves feature extraction, which can be achieved using multiscale sample entropy (MSE) or other feature extraction algorithms. Taking the MSE algorithm as an example, the MSE algorithm can obtain the time series of the signal at different scales in a coarse-grained manner. By calculating the corresponding sample entropy values ​​at different time scales, the final multiscale entropy value is obtained as the target voiceprint feature.

[0176] S1043. If the location of the sound-emitting target is within a preset angle range, and the acoustic signature of the target's azimuth beam signal matches any one of the multiple target acoustic signatures, then an alarm message is issued.

[0177] For example, alarm devices can use the broadband dynamic time warping (DTW) algorithm to match the acoustic signature features of the target azimuth beam signal with the acoustic signature features of multiple targets. This algorithm is based on dynamic programming and overcomes the problem of time series asynchrony by flexibly finding the correspondence between two time series to find their shortest distance.

[0178] Optionally, before S1042 above, the alarm device may also match the acoustic signature of the target azimuth beam signal with a preset noise feature library. The noise library may include multiple noise features. If the acoustic signature of the target azimuth beam signal does not match any of the multiple noise features, the alarm device may execute S1042. If the acoustic signature of the target azimuth beam signal matches any one of the multiple noise features, the alarm device may detect the sound pressure signal of the next frame.

[0179] Based on the understanding of the above embodiments, Figure 12 is a flowchart of signal classification and recognition provided in an embodiment of this application. As shown in Figure 12, the signal classification and recognition process may include S301 to S305.

[0180] S301, Beamforming.

[0181] S301 can be referred to in S1021 above, and will not be repeated here.

[0182] S302. Extract the acoustic signature features t(n) of the target azimuth beam signal.

[0183] S302 can be referred to in S1041 above, and will not be repeated here.

[0184] S303. Match and compare t(n) based on N_T(n).

[0185] Where N_T(n) represents the preset noise library.

[0186] S303 can be referred to the above, and will not be repeated here.

[0187] If a match is found, the sound pressure signal of the next frame is detected; if a match is not found, S304 is executed.

[0188] S304. Match and compare t(n) based on S_T(n).

[0189] Where S_T(n) represents the preset voiceprint library corresponding to the target event.

[0190] S304 can be referred to in S1042 above, and will not be repeated here.

[0191] If a match is found, execute S305; otherwise, detect the sound pressure signal of the next frame.

[0192] S305, Issue an alarm message.

[0193] In some embodiments, as shown in the installation diagram of Figure 9 above, the ATM lobby door is located at θ° of the alarm device. The sound of the door closing, footsteps, and other noises can cause fixed-directional interference to the detection and waveform output. Although MVDR beamforming can also create a dip in non-target directions, the degree of dip is related to the power of the signal and interference, and the degree of dip is not fixed. In some cases, it may not be able to effectively suppress the interference. Therefore, a fixed-directional output constraint can be added to create a fixed dip in a specific direction to suppress interference in the lobby door direction. In this case, before S1041 above, the method may further include: weighting the MVDR beam using weighting coefficients to suppress the signal strength in the preset interference direction.

[0194] The weighting coefficients can satisfy the following relationship as shown in formula (12):

[0195] W=(IC(C H C) -1 C H W a Formula (12)

[0196] In formula (12), W represents the weighting coefficient. C = [V(θ), D(θ)]. V(θ) represents the array manifold at θ, where θ is the preset interference direction (e.g., the direction of the ATM hall door mentioned above).

[0197] For example, Figure 13 is a comparative schematic diagram of constrained beamforming and conventional beamforming provided in the embodiments of this application. As shown in Figure 13, taking the azimuth of the preset interference object as 50° as an example, the constrained (or suppressed) beamforming has a suppression capability several tens of decibels higher than that of conventional beamforming.

[0198] The foregoing primarily describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the aforementioned functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0199] In an exemplary embodiment, this application also provides an alarm device that can be applied to the alarm device 100 described above. Figure 14 is a schematic diagram of the composition of the alarm device provided in this application embodiment. As shown in Figure 14, the device may include: an acquisition unit 1401 and a processing unit 1402.

[0200] The acquisition unit 1401 is used to acquire the sound pressure signal collected by the sound pressure sensor array and the vibration velocity signal collected by the vibration velocity sensor array.

[0201] The processing unit 1402 is used to detect the sound-emitting target based on the sound pressure signal; detect the location of the sound-emitting target based on the vibration velocity signal; and issue an alarm message when the sound-emitting target is within a preset angle range; the preset angle range is used to indicate the location of the target facility.

[0202] In some possible embodiments, the processing unit 1402 is specifically used to generate a minimum mean square distortion-free response (MVDR) beam based on the sound pressure signal; determine whether the power of the MVDR beam is greater than a preset threshold; and if the power of the MVDR beam is greater than the preset threshold, determine that there is a sound-emitting target.

[0203] In other possible embodiments, the processing unit 1402 is specifically configured to determine whether the power of the MVDR beam exceeds the threshold according to the following formula:

[0204]

[0205] Where Th represents the preset threshold; Indicates the power of the MVDR beam; R n Represents the covariance matrix of the noise array; y(t) represents the sound pressure signal when the sound-emitting target is present, y(t) = V*s(t) + n(t); s(t) represents the effective signal; n(t) represents the noise signal; V represents the array manifold. Indicates the estimated array manifold;

[0206] like in, u1 represents the eigenvector corresponding to the largest eigenvalue of the effective signal covariance matrix. Let represent the theoretical array manifold, ε represent the upper limit of error for the array manifold, and M represent the number of sound pressure sensors in the sound pressure sensor array. Then:

[0207]

[0208] in, Converted to eigenvalue decomposition form U represents the eigenvector; Γ represents the eigenvalue matrix. γ1 represents the largest eigenvalue; I represents the identity matrix.

[0209] like but:

[0210]

[0211] Among them, W a Indicates the weighting coefficients.

[0212] In some other possible embodiments, the processing unit 1402 is specifically used to perform line spectrum detection on the MVDR beam according to a preset frequency band; if a line spectrum exists within the preset frequency band, then calculate the sound pressure spectrum at the line spectrum; the line spectrum is used to indicate the frequency of the sound signal emitted by the sound-emitting target; based on the vibration velocity signal, determine the vibration velocity spectrum of the line spectrum in a first preset direction; based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the first preset direction, determine the sound energy flow component in the first preset direction; based on the vibration velocity signal, determine the vibration velocity spectrum of the line spectrum in a second preset direction; the first preset direction is perpendicular to the second preset direction; based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the second preset direction, determine the sound energy flow component in the second preset direction; based on the sound energy flow component in the first preset direction and the sound energy flow component in the second preset direction, determine the orientation of the sound-emitting target.

[0213] In some other possible embodiments, the processing unit 1402 is specifically configured to calculate the acoustic energy flow component in a first preset direction according to the following formula:

[0214] I x =Re(P(f)×V) x (f));

[0215] Among them, I x P(f) represents the acoustic energy flow component in the first preset direction; P(f) represents the sound pressure spectrum at the line spectrum; V x (f) represents the velocity spectrum of the line spectrum in the first preset direction;

[0216] The acoustic energy flow component in the second preset direction is calculated according to the following formula:

[0217] I y =Re(P(f)×V) y (f));

[0218] Among them, I y V represents the acoustic energy flow component in the second preset direction; y (f) indicates the vibration velocity spectrum of the line spectrum in the second preset direction;

[0219] Calculate the location of the sound source using the following formula:

[0220] β=atan(I y / I x );

[0221] Where β represents the azimuth angle of the sound-emitting target.

[0222] In some other possible embodiments, the processing unit 1402 is further configured to calculate the acoustic energy flow corresponding to each frequency point within a preset frequency band if no line spectrum exists; to calculate the acoustic energy intensity corresponding to each direction based on the acoustic energy flow corresponding to each frequency point; and to determine the direction with the largest corresponding acoustic energy intensity as the direction of the sound-emitting target.

[0223] In some other possible embodiments, the vibration velocity sensor array includes a first vibration velocity sensor group and a second vibration velocity sensor group; the first vibration velocity sensor group is used to collect vibration velocity signals in a first direction; the first direction is the same as a first preset direction; the second vibration velocity sensor group is used to collect vibration velocity signals in a second direction; the second direction is the same as a second preset direction.

[0224] In some other possible embodiments, the processing unit 1402 is specifically used to extract beam signals from the MVDR beam to obtain beam signals of the target azimuth; match the voiceprint features of the target azimuth beam signals with a preset voiceprint feature library corresponding to the target event; the preset voiceprint library includes multiple target voiceprint features; if the azimuth of the sound-emitting target is within a preset angle range, and the voiceprint features of the target azimuth beam signals match any one of the multiple target voiceprint features, then an alarm message is issued.

[0225] In some other possible embodiments, the processing unit 1402 is further configured to weight the MVDR beam using weighting coefficients before extracting the beam signal from the MVDR beam, in order to suppress the signal strength at a preset interference direction; the weighting coefficients satisfy the following relationship:

[0226] W=(IC(C H C) -1 C H W a ;

[0227] Where W represents the weighting coefficient; C = [V(θ), D(θ)]; V(θ) represents the array manifold at θ, where θ is the preset interference direction;

[0228] In some other possible embodiments, the sound pressure sensor array is a first circular array; the vibration velocity sensor array is a second circular array; the second circular array is located within the first circular array.

[0229] It should be noted that the unit division in Figure 14 is illustrative and represents only one logical functional division. In actual implementation, other division methods are possible. For example, two or more functions can be integrated into one processing unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0230] In an exemplary embodiment, this application also provides a readable storage medium including software instructions that, when run on an alarm device, cause the alarm device to execute any of the methods provided in the above embodiments.

[0231] In an exemplary embodiment, this application also provides a computer program product containing computer execution instructions, which, when run on an alarm device, causes the alarm device to execute any of the methods provided in the above embodiments.

[0232] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer-executable instructions. When these computer-executable instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer-executable instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer-executable instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device including one or more servers, data centers, etc., that can be integrated with the medium. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), solid-state drives, etc.

[0233] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple components. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0234] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

[0235] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An alarm method, characterized in that, The alarm method is applied to an alarm device, which is embedded within a preset range of the target facility. The alarm device includes a housing, a sound pressure sensor array rigidly connected to the housing, and a vibration velocity sensor array rigidly connected to the housing. The method includes: acquiring sound pressure signals collected by the sound pressure sensor array and vibration velocity signals collected by the vibration velocity sensor array; detecting the sound-emitting target based on the sound pressure signals, including: generating a minimum mean square distortion-free response (MVDR) beam based on the sound pressure signals; detecting the orientation of the sound-emitting target based on the vibration velocity signals; and issuing an alarm message when the sound-emitting target is within a preset angle range. The preset angle range is used to indicate... The location of the target facility; wherein, the step of detecting the location of the sound-emitting target based on the vibration velocity signal includes: performing line spectrum detection on the MVDR beam according to a preset frequency band; the preset frequency band is preset in the alarm device and is used to characterize the frequency band of the sound signal emitted by the target event; if there is no line spectrum within the preset frequency band, then calculate the acoustic energy flow corresponding to each frequency point within the preset frequency band; based on the acoustic energy flow corresponding to each frequency point, calculate the acoustic energy intensity corresponding to each location; determine the location of the sound-emitting target as the location of the location with the largest corresponding acoustic energy intensity; if there is a line spectrum within the preset frequency band, then determine the location of the sound-emitting target based on the sound pressure spectrum and vibration velocity spectrum at the line spectrum.

2. The method according to claim 1, characterized in that, The step of detecting the sound-emitting target based on the sound pressure signal further includes: determining whether the power of the MVDR beam is greater than a preset threshold; if the power of the MVDR beam is greater than the preset threshold, then it is determined that the sound-emitting target exists.

3. The method according to claim 2, characterized in that, The step of determining whether the power of the MVDR beam is greater than a preset threshold includes: determining whether the power of the MVDR beam is greater than the preset threshold according to the following formula: ;in, This represents the preset threshold; This indicates the power of the MVDR beam; Represents the covariance matrix of the noise array; ； This represents the sound pressure signal when the sound-emitting target is present. ； Indicates a valid signal; Indicates a noise signal; Indicates the array manifold. ； Indicates the predicted array manifold; if ,in, , ， This represents the eigenvector corresponding to the largest eigenvalue of the effective signal covariance matrix. Represents the theoretical array manifold. This indicates the upper limit of error for the array manifold. Let represent the number of sound pressure sensors in the sound pressure sensor array, then: ;in, Transformed into the eigenvalue decomposition form as ， Represents the eigenvector; Represents the eigenvalue matrix. ， Represents the largest eigenvalue; Represents the identity matrix. ;like ,but: ;in, Indicates the weighting coefficients. 。 4. The method according to claim 2, characterized in that, The step of determining the orientation of the sound-emitting target based on the sound pressure spectrum and vibration velocity spectrum at the line spectrum includes: calculating the sound pressure spectrum at the line spectrum; the line spectrum is used to indicate the frequency of the sound signal emitted by the sound-emitting target; determining the vibration velocity spectrum at the line spectrum in a first preset direction based on the vibration velocity signal; determining the sound energy flow component in the first preset direction based on the sound pressure spectrum and the vibration velocity spectrum at the line spectrum in the first preset direction; determining the vibration velocity spectrum at the line spectrum in a second preset direction based on the vibration velocity signal; the first preset direction is perpendicular to the second preset direction; determining the sound energy flow component in the second preset direction based on the sound pressure spectrum and the vibration velocity spectrum at the line spectrum in the second preset direction; and determining the orientation of the sound-emitting target based on the sound energy flow component in the first preset direction and the sound energy flow component in the second preset direction.

5. The method according to claim 4, characterized in that, The step of determining the acoustic energy flow component in the first preset direction based on the sound pressure spectrum and the vibration velocity spectrum of the line spectrum in the first preset direction includes: calculating the acoustic energy flow component in the first preset direction according to the following formula: ;in, This represents the acoustic energy flow component in the first preset direction; This represents the sound pressure spectrum at the line spectrum; The velocity spectrum of the line spectrum in the first preset direction is represented; determining the acoustic energy flow component in the second preset direction based on the sound pressure spectrum and the velocity spectrum of the line spectrum in the second preset direction includes: calculating the acoustic energy flow component in the second preset direction according to the following formula: ;in, This represents the acoustic energy flow component in the second preset direction; This indicates the velocity spectrum of the line spectrum in the second preset direction; determining the orientation of the sound-emitting target based on the sound energy flow component in the first preset direction and the sound energy flow component in the second preset direction includes: calculating the orientation of the sound-emitting target according to the following formula: ;in, This indicates the azimuth angle of the sound-emitting target.

6. The method according to claim 4, characterized in that, The vibration velocity sensor array includes a first vibration velocity sensor group and a second vibration velocity sensor group; the first vibration velocity sensor group is used to collect vibration velocity signals in a first direction; the first direction is the same as the first preset direction; the second vibration velocity sensor group is used to collect vibration velocity signals in a second direction; the second direction is the same as the second preset direction.

7. The method according to any one of claims 2-6, characterized in that, The step of issuing an alarm message when the sound-emitting target is within a preset angle range includes: extracting beam signals from the MVDR beam to obtain the beam signal of the target's azimuth; matching the voiceprint features of the target's azimuth beam signal with a preset voiceprint feature library corresponding to the target event; the preset voiceprint feature library includes multiple target voiceprint features; and issuing an alarm message if the voiceprint features of the target's azimuth beam signal match any one of the multiple target voiceprint features when the sound-emitting target's azimuth is within the preset angle range.

8. The method according to claim 7, characterized in that, The method further includes: before extracting the beam signal from the MVDR beam to obtain the beam signal at the target azimuth, weighting the MVDR beam using weighting coefficients to suppress the signal strength at a preset interference azimuth; the weighting coefficients satisfy the following relationship: ;in, Indicates the weighting coefficients; ； express directional array manifold, The orientation is the preset interference orientation; 。 9. The method according to claim 1, characterized in that, The sound pressure sensor array is a first circular array; the vibration velocity sensor array is a second circular array; the second circular array is within the first circular array.

10. An alarm device, characterized in that, The alarm device is embedded within a preset range of the target facility; the alarm device includes: a housing; a sound pressure sensor array rigidly connected to the housing for collecting sound pressure signals; a vibration velocity sensor array rigidly connected to the housing for collecting vibration velocity signals; and a processing control module disposed within the housing for detecting the sound-emitting target and the location of the sound-emitting target according to the sound pressure signal and the vibration velocity signal as described in any one of claims 1-9, and issuing an alarm message when the sound-emitting target is within a preset angle range; the preset angle range is used to indicate the location of the target facility.

Images