Three-dimensional sound source localization imaging method, device and equipment based on two-dimensional convex array and medium

Abstract

The invention provides a three-dimensional sound source positioning imaging method, device and equipment based on a two-dimensional convex array and a medium, and the method comprises the steps: obtaining a time domain signal p (r, phi, y, t) passively received by a two-dimensional convex array ultrasonic transducer in a sound velocity non-uniform medium, the array elements of the two-dimensional convex array ultrasonic transducer are distributed on a cylindrical surface, and a cylindrical coordinate reference system is (r, phi, y, t); r is the extrapolation radius of the probe of the two-dimensional convex array ultrasonic transducer, and R is the distribution surface where the array elements of the convex array ultrasonic transducer are located; according to the time domain signal p (r, phi, y, t), determining the sound velocity distribution of a propagation medium, and generating an aligned sound velocity diagram to determine the slowness u (r, phi, y) of the medium; the time domain signal p (r, phi, y, t) is converted in the time t, the circumferential direction phi and the longitudinal direction y, a spiral wave spectrum Pn (r, ky, omega) is generated, omega represents the time angular frequency, n is the circumferential wavelength, and ky is the longitudinal wavelength; and according to the spiral wave spectrum Pn (r, ky, omega) and the medium slowness u (r, phi, y), determining a frequency domain sound field of a next layer, and according to the frequency domain sound field, generating a passive sound image I (r, phi, y). The scheme can accurately position the sound source.

Description

Technical Field

[0001] This invention relates to the field of ultrasound imaging technology, and in particular to a three-dimensional sound source localization imaging method, device, equipment and medium based on a two-dimensional convex array. Background Technology

[0002] Currently, one-dimensional convex arrays have been applied to passive acoustic source localization in ultrasound therapeutic monitoring, i.e., passive acoustic imaging. Passive acoustic imaging methods applicable to the sound velocity of one-dimensional convex arrays are mainly based on time-delay accumulation integration, and this method can be extended to two-dimensional convex arrays. However, complex environments contain diverse media types and various heterogeneous sound velocity components (i.e., non-uniform media), causing distortion in sound waves during propagation and affecting image reconstruction quality. In non-uniform sound velocity media, the time-delay accumulation integration method calculates the time delay by integrating along the sound propagation path, adjusts the waveform based on the time delay, and then coherently accumulates the adjusted signal to calculate the signal energy as image intensity information. However, this method has high algorithm complexity and is computationally time-consuming. In contrast, frequency-wavenumber domain methods (such as the sound velocity heterogeneous angle spectrum method) have lower computational complexity. In a Cartesian coordinate system, the sound velocity heterogeneous angle spectrum method can overcome the distortion problem caused by sound velocity heterogeneity. However, the sound velocity heterogeneous angle spectrum method cannot be directly applied to convex arrays. Summary of the Invention

[0003] This invention provides a three-dimensional sound source localization imaging method, apparatus, device, and medium based on a two-dimensional convex array to solve the above-mentioned technical problems.

[0004] This invention provides a three-dimensional sound source localization imaging method based on a two-dimensional convex array, the method comprising:

[0005] The time-domain signal p(r,φ,y,t) passively received by a two-dimensional convex array ultrasonic transducer in a non-uniform sound velocity medium is obtained. The array elements of the two-dimensional convex array ultrasonic transducer are distributed on a cylindrical surface, and the cylindrical coordinate reference system is (r,φ,y), where r is the extrapolation radius of the probe of the two-dimensional convex array ultrasonic transducer, and R is the distribution surface of the array elements of the convex array ultrasonic transducer.

[0006] Based on the time-domain signal p(r,φ,y,t), the sound velocity distribution of the propagation medium is determined, and an aligned sound velocity map is generated to determine the medium slowness u(r,φ,y);

[0007] The time-domain signal p(r,φ,y,t) is transformed along time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n (r,k y ,ω), where ω represents the time angular frequency, n is the circumferential wavelength, and k y Longitudinal wavelength;

[0008] According to the spiral spectrum P n (r,k y The frequency domain sound field of the next layer is determined by using the medium slowness u(r,φ,y) and ω, and a passive sound image I(r,φ,y) is generated based on the frequency domain sound field.

[0009] In one embodiment of the present invention, the time-domain signal p(r,φ,y,t) is transformed along time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n (r,k y ,ω), including:

[0010] Perform a Fourier transform on the time-domain signal p(r,φ,y,t) with respect to time t to convert the time-domain signal p(r,φ,y,t) into a time-frequency domain signal.

[0011] Time-frequency domain signal Perform Fourier transforms in the circumferential φ and longitudinal y directions to generate the helical spectrum P. n (r,k y ,ω).

[0012] In one embodiment of the present invention, based on the spiral spectrum P n (r,k y Based on the medium slowness u(r,φ,y), the frequency domain sound field of the next layer is determined, and a passive acoustic image I(r,φ,y) is generated according to the frequency domain sound field, including:

[0013] The medium slowness u(r,φ,y) is decomposed into the average slowness u0 and the deviation Δu(r,φ,y) between the current layer slowness and the average slowness.

[0014] Set r = r m r m+1 =r m +Δr,r m r is the extrapolation radius of the current layer. m+1 Let Δr be the extrapolation radius of the next layer, and Δr be the difference in extrapolation radii, to determine the interval [r]. m ,r m+1 Medium slowness u(r) within the range m+1 ,φ,y).

[0015] In one embodiment of the present invention, based on the spiral spectrum P n (r,k y Based on the medium slowness u(r,φ,y), the frequency domain sound field of the next layer is determined, and a passive acoustic image I(r,φ,y) is generated according to the frequency domain sound field, including:

[0016] The acoustic wavelength λ, imaging effect requirements, and extrapolation radius difference Δr are determined, and the frequency domain sound field of the next layer is determined based on these parameters. The imaging effects include low standard and high standard;

[0017] If the imaging effect is low, the frequency domain sound field of the next layer is updated using the stepwise spiral spectral method.

[0018] If the imaging effect is of a high standard, and the difference in the extrapolated radius is less than or equal to half a wavelength, then the first-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), and then the spiral spectrum P n (r m+1 ,k y Perform an inverse Fourier transform on ω to determine the frequency domain sound field of the next layer.

[0019] If the imaging effect is of a high standard and the extrapolated diameter difference is greater than half a wavelength, then the second-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), and then the spiral spectrum P n (r m+1 ,k y Perform an inverse Fourier transform on ω to determine the frequency domain sound field of the next layer.

[0020] In one embodiment of the present invention, the first-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), including:

[0021]

[0022] in,

[0023]

[0024]

[0025] This is a Hankel equation of the second kind, nth order. This is a Hankel equation of the first kind, order n.

[0026] In one embodiment of the present invention, the second-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,ky ,ω), including:

[0027]

[0028] in,

[0029] The spiral spectrum of the next layer updated by the first-order direct integration method

[0030] This is the inverse Fourier transform.

[0031] In one embodiment of the present invention, the stepwise spiral spectral method is used to update the frequency domain sound field of the next layer. include:

[0032]

[0033] The present invention also provides a three-dimensional sound source localization imaging device based on a two-dimensional convex array, comprising:

[0034] The signal acquisition module is used to acquire the time-domain signal p(r,φ,y,t) passively received by the two-dimensional convex array ultrasonic transducer in a non-uniform sound velocity medium. The array elements of the two-dimensional convex array ultrasonic transducer are distributed on a cylindrical surface, and the cylindrical coordinate reference system is (r,φ,y), where r is the extrapolation radius of the probe of the two-dimensional convex array ultrasonic transducer, and R is the distribution surface of the array elements of the convex array ultrasonic transducer.

[0035] The medium slowness determination module is used to determine the sound velocity distribution of the propagation medium based on the time-domain signal p(r,φ,y,t) and generate an aligned sound velocity map to determine the medium slowness u(r,φ,y);

[0036] The signal transformation module is used to transform the time-domain signal p(r,φ,y,t) in time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n (r,k y ,ω), where ω represents the time angular frequency, n is the circumferential wavelength, and k y Longitudinal wavelength;

[0037] Image generation module, used to generate images based on the helical spectrum P n (r,k y The frequency domain sound field of the next layer is determined by using the medium slowness u(r,φ,y) and ω, and a passive sound image I(r,φ,y) is generated based on the frequency domain sound field.

[0038] The present invention also provides an electronic device, the electronic device comprising:

[0039] One or more processors and memory,

[0040] The memory stores a computer program, which, when executed by the one or more processors, causes the device to perform the method described above.

[0041] The present invention also provides a computer-readable storage medium having a computer program stored thereon that, when executed by one or more processors, causes a device to perform the method described thereon.

[0042] The beneficial effects of this invention are as follows: This invention proposes a three-dimensional acoustic source localization imaging method based on a two-dimensional convex array. Using a two-dimensional convex array ultrasonic transducer, compared to a two-dimensional planar array ultrasonic transducer, the two-dimensional convex array ultrasonic transducer provides a wider imaging range, is more suitable for large-field-of-view imaging, and achieves high-precision imaging. Compared to the spiral wave spectrum method, the three-dimensional acoustic source localization imaging method can accurately locate the acoustic source in a non-uniform sound velocity medium, and more effectively suppresses distortion errors and artifacts. Compared to the time-delay accumulation integration method, it has the advantages of low algorithm complexity and low computation time. Furthermore, this method can correct the phase distortion caused by sound velocity heterogeneity, providing key technical support for real-time, precise monitoring and guidance of abdominal ultrasound cavitation therapy. This method is a general solution to the weakly heterogeneous wave equation of sound velocity, serving as a fast solution method for cylindrical wave fields. It can also be used to reconstruct B-mode ultrasound images or as a forward model for solving cylindrical sound velocity fields. Attached Figure Description

[0043] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0044] In the attached diagram:

[0045] Figure 1 This is a schematic diagram of a cylindrical coordinate reference system provided in an embodiment of the present invention.

[0046] Figure 2 This is a flowchart of a three-dimensional sound source localization imaging method based on a two-dimensional convex array, provided as an embodiment of the present invention.

[0047] Figure 3 This is a schematic diagram of the coordinate system for imaging a two-dimensional convex array in a non-uniform sound velocity medium, according to an embodiment of the present invention.

[0048] Figure 4 This is a schematic diagram of the two-dimensional convex array provided in Experimental Example 1 of the present invention.

[0049] Figure 5This is a schematic diagram of the simulation settings provided in Embodiment 1 of the present invention.

[0050] Figure 6 The voxels are those reconstructed using the 3D HHWS method and the 3D HWS method as provided in Embodiment 1 of the present invention.

[0051] Figure 7 This is a schematic diagram of the positioning error of voxels reconstructed by the 3D HHWS method and the 3D HWS method in Embodiment 1 of the present invention.

[0052] Figure 8 The diagram shows the simulation setup (a) provided in Embodiment 2 of the present invention, and the voxels reconstructed by the 3D HWS method (b) and the 3D HHWS method (c). Detailed Implementation

[0053] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0054] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. The drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0055] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0056] Figure 1 This is a schematic diagram of a cylindrical coordinate reference system provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the cylindrical coordinate reference system is (r,φ,y), where φ represents the circumferential direction, y represents the longitudinal direction, r is the extrapolation radius of the two-dimensional convex array ultrasonic transducer probe, and R is the distribution surface of the array elements of the convex array ultrasonic transducer.

[0057] Figure 2This is a flowchart illustrating a three-dimensional sound source localization imaging method based on a two-dimensional convex array, as provided in an embodiment of the present invention. Figure 2 As shown, based on different sound field propagation calculation processes, three methods can be used: the first-order direct integration method, the second-order direct integration method, and the step-by-step spiral spectral method. Among these, the step-by-step spiral spectral method has the simplest algorithm, but it has a slight error compared to the direct integration method. The second-order direct integration method is more accurate than the first-order direct integration method and has lower requirements for the precision of the axial propagation step size (Δr). Therefore, it is recommended to use the step-by-step spiral spectral method for initial verification of the imaging effect, and then switch to the direct integration method when improvements are needed. Use the first-order direct integration method when the axial interval Δr ≤ λ / 2, and use the second-order direct integration method when Δr > λ / 2, where λ is the wavelength of the sound wave.

[0058] Step S100: Acquire time-domain signal p(r,φ,y,t)

[0059] To monitor passive sound sources, a two-dimensional convex array ultrasonic transducer was used to passively receive the time-domain signal p(r,φ,y,t) in a non-uniform sound velocity medium. The array elements of the two-dimensional convex array ultrasonic transducer are distributed on an arc of radius R, specifically as follows: Figure 3 As shown.

[0060] Step S200: Generate aligned sound velocity map

[0061] Based on the time-domain signal p(r,φ,y,t), an ultrasound image is generated. The sound velocity distribution of the propagation medium is obtained from the ultrasound image or sound velocity reconstruction, and aligned with the probe position to obtain an aligned sound velocity image. The sound velocity at the sampling point is calculated based on the aligned sound velocity image, and the medium slowness u(r,φ,y) is determined.

[0062] It is worth noting that, since this embodiment uses a non-uniform medium, the medium slowness u(r,φ,y) varies in different layers, and the slowness switches between layers as the extrapolation radius r changes.

[0063] Step S300: Determine the frequency domain sound field

[0064] Specifically, step S300 includes steps S310 to S330.

[0065] Step S310: Perform a Fourier transform on the time-domain signal p(r,φ,y,t) with respect to time t to convert the time-domain signal p(r,φ,y,t) into a time-frequency domain signal.

[0066] Specifically, a Fourier transform of the time-domain signal p(r,φ,y,t) received by the probe is performed with respect to time t. The time-domain signal p(r,φ,y,t) is transformed into the time-frequency domain to obtain the time-frequency domain signal. Recorded as:

[0067]

[0068] Step S320, convert the time-frequency domain signal Perform Fourier transforms in the circumferential φ and longitudinal y directions to generate the helical spectrum P. n (r,k y ,ω).

[0069] Specifically, time-frequency domain signal Perform Fourier transforms in the circumferential φ and longitudinal y directions to transform the time-frequency domain signal. Transforming to the spatial frequency domain, we obtain the spiral spectrum P. n (r,k y ,ω).

[0070] Step S330: Determine the sound wave wavelength λ, imaging effect requirements, and extrapolation radius difference Δr; and determine the frequency domain sound field of the next layer based on the sound wave wavelength λ, imaging effect requirements, and extrapolation radius difference Δr. The imaging effects include low standards (such as preliminary verification of imaging effects in the experimental stage) and high standards (such as imaging in practical applications such as seismic activity and medicine).

[0071] For example, suppose we know r = r m The corresponding spiral spectral distribution P n (r m ,k y Since a two-dimensional convex array can only receive sound waves emitted inward from cavitation sources located on the outer side, we only consider inward incoming waves. From the aligned sound velocity diagram, we know that the medium's slowness u(l) = 1 / c(l), where c(l) is the sound velocity in the medium at spatial coordinates l = (r, φ, y). Let r... m+1 =r m +Δr, for [r] m ,r m+1 The interval is used to decompose the slowness of the medium within the interval into two parts u(r). m+1 ,φ,y)=u0+Δu(r m+1 ,φ,y), where u0 is the average slowness (constant) of the interval, which is independent of the position coordinates φ,r,y, and Δu is the deviation between the slowness and the average slowness in the interval.

[0072] For example, if the imaging effect is of a high standard and the difference in extrapolated radius is less than or equal to half a wavelength, then the helical spectrum P of the next layer is updated using the first-order direct integration method. n (r m+1 ,k y ,ω), and then the spiral spectrum P n (r m+1,k y Perform an inverse Fourier transform on ω to determine the frequency domain sound field of the next layer.

[0073] Specifically, the first-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), is represented as:

[0074]

[0075] in,

[0076]

[0077]

[0078] This is a Hankel equation of the second kind, nth order. This is a Hankel equation of the first kind, order n.

[0079] For example, if the imaging effect is of a high standard and the extrapolated diameter difference is greater than half a wavelength, then the second-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), and then the spiral spectrum P n (r m+1 ,k y Perform an inverse Fourier transform on ω to determine the frequency domain sound field of the next layer.

[0080] Specifically, the second-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), including:

[0081]

[0082] in,

[0083]

[0084] This is the inverse Fourier transform.

[0085] It should be noted that, for ease of distinction, the spiral spectrum P of the next layer is calculated in the second-order direct integration method. n (r m+1 ,k y When ω), the spiral spectrum P calculated by the first-order direct integration method is... n (r m+1 ,ky ,ω) is denoted as

[0086] The next layer of spiral spectrum P is determined by first-order direct integration or second-order integration. n (r m+1 ,k y After ω), the spiral spectrum P n (r m+1 ,k y Performing an inverse Fourier transform on ω) yields the frequency domain sound field. Recorded as:

[0087]

[0088] For example, if the imaging effect is low-standard, the frequency domain sound field of the next layer is updated using the step-by-step spiral spectral method. Specifically:

[0089]

[0090] It should be noted that the above Hankel equation can be simplified using an asymptotic analytical solution, as shown below:

[0091]

[0092] in, for The conjugate of . The simplified calculation involving the Hankel equation is shown below:

[0093]

[0094] in,

[0095]

[0096] It is worth noting that for the non-evanescent wave portion, the asymptotic formula can accurately approximate the wave, reducing computational costs.

[0097] Step S400: Generate passive acoustic image I(r,φ,y)

[0098] Starting from the probe cylinder r = R, the frequency domain sound field distribution in three-dimensional space is obtained by iterative calculation using the above method. The sound field intensity within a specific frequency band Ω is selected and superimposed to obtain the passive acoustic image I(r,φ,y) in polar coordinates.

[0099]

[0100] Step S500: Coordinate system transformation

[0101] The polar coordinate sampling points are transformed into an image in a Cartesian coordinate system using an interpolation algorithm, which facilitates observation and evaluation. The origin is taken as the intersection of the arc where the probe of the two-dimensional convex array ultrasonic transducer is located and the axis of symmetry, with the horizontal direction as the x-axis, the vertical direction as the y-axis, and the axial direction as the z-axis.

[0102] Experiment Example 1: Single-point sound source simulation experiment

[0103] Step S100: Acquire time-domain signal p(r,φ,y,t)

[0104] Figure 4 This is a schematic diagram of the two-dimensional convex array provided in Experimental Example 1 of the present invention. (See attached diagram.) Figure 4 As shown, a 32×32 channel two-dimensional convex array is simulated using k-Wave, where the circumferential direction φ contains 32 channels and the longitudinal direction y contains 32 channels. In the circumferential direction φ, the radius of curvature R of the array is 20 mm, and the scanning angle is 60°; in the longitudinal direction y, the spacing between adjacent channels is 0.625 mm.

[0105] Figure 5 This is a schematic diagram of the simulation settings provided in Embodiment 1 of the present invention. (See attached diagram.) Figure 5 As shown, the simulation settings include the sound velocity distribution and the positions of the array and the sound source. A two-dimensional convex array receives point source sound signals in a non-uniform sound velocity medium, with a signal sampling rate of 13.89 MHz. The background sound velocity of the medium is 1540 m / s, in which a rectangular block simulating a cyst with a sound velocity of 1400 m / s is embedded, located in the region z = 2.5–15.5 mm and y > 0.

[0106] In addition, 3×3×3 single-point sources are placed in a cubic grid, with their positions ranging from -5 to 5 mm in the x-direction, -5 to 5 mm in the y-direction, and 20 to 30 mm in the z-direction, spaced 5 mm apart. Each point source is sequentially excited by 40-cycle sine waves of 1 MHz, 2 MHz, 3 MHz, and 4 MHz, respectively.

[0107] Step S200: Generate aligned sound velocity map

[0108] The aligned sound velocity map shows the sound velocity distribution set in the simulation.

[0109] Step S300: Determine the frequency domain sound field

[0110] Taking the second-order direct integration method as an example, the received time-domain signal p(r,φ,y,t) is first subjected to a Fourier transform in the time t direction. Then, Fourier transforms are performed in the circumferential φ and longitudinal y directions to obtain the helical spectrum P. n (R,k y(ω). At depths from r = R to R + 40 mm, the spiral spectrum P of the next layer is derived iteratively at intervals of Δr = 0.2 mm. n (r m+1 ,k y In the derivation process, the average slowness u0 and the deviation Δu0 between the slowness and the average slowness are calculated by aligning the sound velocity diagram.

[0111] When using the 3D HWS method for voxel reconstruction, u0 = 1500 m / s and Δu0 = 0 m / s were set. Then, the obtained spiral spectrum P... n (r m+1 ,k y The frequency domain sound field is obtained by performing an inverse Fourier transform on (ω).

[0112] Step S400: Generate passive acoustic image I(r,φ,y)

[0113] The passive sound source image in polar coordinates is obtained by superimposing the sound field intensity of the single frequency component corresponding to the selected sound source, with a sampling grid of 201×64×64(r×φ×y).

[0114] Step S500: Coordinate system transformation

[0115] After converting to a Cartesian coordinate system, the location of the simulated sound source is used as the positioning reference (l r Comparison of 3D HHWS results using sound velocity information to correct distortion with 3D HWS results using a uniform sound velocity of 1500 m / s (see attached table). Figure 6 As shown). Strong artifacts appeared in the reconstruction results of the 3DHWS method, causing the sound source to appear as two point sources (as shown). Figure 6 (a)). In contrast, the reconstruction results of the 3D HHWS method only show one sound source, consistent with the simulation settings (e.g., Figure 6 (b)). Additionally, for the 3DHWS method, the localization error of the 1MHz, 2MHz, 3MHz, and 4MHz signal sources (localization error at the peak sound intensity position l: |ll) r The thicknesses are 3.7±1.8mm, 1.4±0.7mm, 1.8±1.2mm, and 1.4±0.8mm respectively (reference). Figure 7 (a) shows the error. After using the 3DHHWS method, the errors were reduced to 1.6±1.1 mm, 0.7±0.4 mm, 0.7±0.4 mm and 0.7±0.3 mm respectively (reference). Figure 7 (b) is shown.

[0116] Experiment Example 2: Simulation Experiment of Two-Point Sound Source

[0117] Step S100: Acquire time-domain signal p(r,φ,y,t)

[0118] A 32×32 channel two-dimensional convex transducer array was simulated using k-Wave, with 32 channels circumferentially (φ) and 32 channels longitudinally (y). Circumferentially, the array's radius of curvature R = 20 mm, and the scan angle is 60°; longitudinally (y), the spacing between adjacent channels is 0.625 mm. The two-dimensional convex array received point source acoustic signals in a non-uniform acoustic medium at a sampling rate of 13.89 MHz. The background sound velocity of the medium was 1540 m / s, within which a rectangular block simulating a cyst with a sound velocity of 1400 m / s was embedded, located in the region z = 2.5–15.5 mm, where y > 0 (reference). Figure 8 (a)). Furthermore, the two single-point sources were placed separately. Location ( Figure 8 (a)). The two single-point sources were simultaneously excited by 40-cycle sine waves of 1MHz, 2MHz, 3MHz and 4MHz respectively.

[0119] Step S200: Generate aligned sound velocity map

[0120] The aligned sound velocity map shows the sound velocity distribution set in the simulation.

[0121] Step S300: Determine the frequency domain sound field

[0122] Taking the second-order direct integration method as an example, the received time-domain signal p(r,φ,y,t) is first subjected to a Fourier transform in the time t direction. Then, Fourier transforms are performed in the circumferential φ and longitudinal y directions to obtain the helical spectrum P. n (R,k y (ω). At depths from r = R to R + 40 mm, the spectrum P is derived iteratively layer by layer at intervals of Δr = 0.2 mm. n (r m+1 ,k y ,ω). Among them, the average slowness u0 and the deviation Δu0 between the slowness and the average slowness are calculated using the aligned sound velocity diagram.

[0123] When using the 3D HWS method for voxel reconstruction, u0 = 1500 m / s and Δu0 = 0 m / s were set. Then, the obtained spectrum was subjected to inverse Fourier transform to obtain...

[0124] Step S400: Generate passive acoustic image I(r,φ,y)

[0125] The passive sound source image in polar coordinates is obtained by superimposing the sound field intensity of the single frequency component corresponding to the selected sound source, with a sampling grid of 201×64×64(r×φ×y).

[0126] Step S500: Coordinate system transformation

[0127] After converting to a Cartesian coordinate system, the location of the simulated sound source is used as the positioning reference. and The results of 3D HHWS, which uses sound velocity information to correct distortion, are compared with those of 3DHWS, which uses a uniform sound velocity of 1500 m / s. For a 1 MHz sound source, the reconstruction results of the 3DHWS method show significant distortion at sound source 1, where a single point source is split into two, while sound source 2 is accompanied by strong artifacts (see reference). Figure 8 (b)). In contrast, after distortion correction using sound velocity information, the reconstruction results of the 3D HHWS method clearly show two independent sound sources, consistent with the simulation settings (see reference). Figure 8 (c)). To evaluate the distortion correction effect, the peak sound intensity positions of voxels in the regions x<0 and x≥0 were calculated, and denoted as the localization positions l1 and l2 of sound source 1 and sound source 2, respectively. The localization errors of sound source 1 and sound source 2 are respectively... and For sound source 1, compared to the 3D HWS method, the localization error of the 3D HWS method was reduced by 12.0%, 69.6%, 60.0%, and 44.4% under excitation at 1MHz, 2MHz, 3MHz, and 4MHz, respectively (Table 1). For sound source 2, compared to the 3D HWS method, the localization error of the 3D HWS method was reduced by 88.9%, 63.6%, 9.1%, and 62.5%, respectively (as shown in Table 1).

[0128] Table 1. Positioning error in dual-point source experiments

[0129]

[0130] In another exemplary embodiment, a three-dimensional sound source localization imaging device based on a two-dimensional convex array is also provided, comprising:

[0131] The signal acquisition module is used to acquire the time-domain signal p(r,φ,y,t) passively received by the two-dimensional convex array ultrasonic transducer in a non-uniform sound velocity medium. The array elements of the two-dimensional convex array ultrasonic transducer are distributed on a cylindrical surface, and the cylindrical coordinate reference system is (r,φ,y), where r is the extrapolation radius of the probe of the two-dimensional convex array ultrasonic transducer, and R is the distribution surface of the array elements of the convex array ultrasonic transducer.

[0132] The medium slowness determination module is used to determine the sound velocity distribution of the propagation medium based on the time-domain signal p(r,φ,y,t) and generate an aligned sound velocity map to determine the medium slowness u(r,φ,y);

[0133] The signal transformation module is used to transform the time-domain signal p(r,φ,y,t) in time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n(r,k y ,ω), where ω represents the time angular frequency, n is the circumferential wavelength, and k y Longitudinal wavelength;

[0134] Image generation module, used to generate images based on the helical spectrum P n (r,k y The frequency domain sound field of the next layer is determined by using the medium slowness u(r,φ,y) and ω, and a passive sound image I(r,φ,y) is generated based on the frequency domain sound field.

[0135] It should be noted that the three-dimensional sound source localization imaging device based on a two-dimensional convex array provided in the above embodiments and the three-dimensional sound source localization imaging method based on a two-dimensional convex array provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the three-dimensional sound source localization imaging device based on a two-dimensional convex array provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0136] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, enable the electronic device to implement the three-dimensional sound source localization imaging method based on a two-dimensional convex array provided in the above embodiments.

[0137] Another aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a computer's processor, causes the computer to perform the three-dimensional sound source localization imaging method based on a two-dimensional convex array as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not assembled into the electronic device.

[0138] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the three-dimensional sound source localization imaging method based on a two-dimensional convex array provided in the various embodiments above.

[0139] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A three-dimensional sound source localization imaging method based on a two-dimensional convex array, characterized in that, include: The time-domain signal p(r,φ,y,t) passively received by a two-dimensional convex array ultrasonic transducer in a non-uniform sound velocity medium is obtained. The array elements of the two-dimensional convex array ultrasonic transducer are distributed on a cylindrical surface, and the cylindrical coordinate reference system is (r,φ,y), where r is the extrapolation radius of the probe of the two-dimensional convex array ultrasonic transducer, and R is the distribution surface of the array elements of the convex array ultrasonic transducer. Based on the time-domain signal p(r,φ,y,t), the sound velocity distribution of the propagation medium is determined, and an aligned sound velocity map is generated to determine the medium slowness u(r,φ,y); The time-domain signal p(r,φ,y,t) is transformed along time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n (r,k y ,ω), where ω represents the time angular frequency, n is the circumferential wavelength, and k y Longitudinal wavelength; According to the spiral spectrum P n (r,k y The frequency domain sound field of the next layer is determined by using the medium slowness u(r,φ,y) and ω, and a passive sound image I(r,φ,y) is generated based on the frequency domain sound field.

2. The three-dimensional sound source localization imaging method based on a two-dimensional convex array according to claim 1, characterized in that, The time-domain signal p(r,φ,y,t) is transformed along time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n (r,k y ,ω), including: Perform a Fourier transform on the time-domain signal p(r,φ,y,t) with respect to time t to convert the time-domain signal p(r,φ,y,t) into a time-frequency domain signal. Time-frequency domain signal Perform Fourier transforms in the circumferential φ and longitudinal y directions to generate the helical spectrum P. n (r,k y ,ω).

3. The three-dimensional sound source localization imaging method based on a two-dimensional convex array according to claim 2, characterized in that, According to the spiral spectrum P n (r,k y Based on the medium slowness u(r,φ,y), the frequency domain sound field of the next layer is determined, and a passive acoustic image I(r,φ,y) is generated according to the frequency domain sound field, including: The medium slowness u(r,φ,y) is decomposed into the average slowness u0 and the deviation Δu(r,φ,y) between the current layer slowness and the average slowness. Set r = r m r m+1 =r m +Δr,r m r is the extrapolation radius of the current layer. m+1 Let Δr be the extrapolation radius of the next layer, and Δr be the difference in extrapolation radii, to determine the interval [r]. m ,r m+1 Medium slowness u(r) within the range m+1 ,φ,y).

4. The three-dimensional sound source localization imaging method based on a two-dimensional convex array according to claim 3, characterized in that, According to the spiral spectrum P n (r,k y Based on the medium slowness u(r,φ,y), the frequency domain sound field of the next layer is determined, and a passive acoustic image I(r,φ,y) is generated according to the frequency domain sound field, including: The acoustic wavelength λ, imaging effect requirements, and extrapolation radius difference Δr are determined, and the frequency domain sound field of the next layer is determined based on these parameters. The imaging effects include low standard and high standard; If the imaging effect is low, the frequency domain sound field of the next layer is updated using the stepwise spiral spectral method. If the imaging effect is of a high standard, and the difference in the extrapolated radius is less than or equal to half a wavelength, then the first-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), and then the spiral spectrum P n (r m+1 ,k y Perform an inverse Fourier transform on ω to determine the frequency domain sound field of the next layer. If the imaging effect is of a high standard and the extrapolated diameter difference is greater than half a wavelength, then the second-order direct integration method is used to update the spiral spectrum P of the next layer. n (r m+1 ,k y ,ω), and then the spiral spectrum P n (r m+1 ,k y Perform an inverse Fourier transform on ω to determine the frequency domain sound field of the next layer.

5. The three-dimensional sound source localization imaging method based on a two-dimensional convex array according to claim 4, characterized in that, The spiral spectrum P of the next layer is updated using the first-order direct integration method. n (r m+1 ,k y ,ω), include: in, k0=ωu0; This is a Hankel equation of the second kind, nth order. This is a Hankel equation of the first kind, order n.

6. The three-dimensional sound source localization imaging method based on a two-dimensional convex array according to claim 4, characterized in that, The spiral spectrum P of the next layer is updated using the second-order direct integration method. n (r m+1 ,k y ,ω), include: in, The spiral spectrum of the next layer updated by the first-order direct integration method P n (r m+1 ,k y ,oh); This is the inverse Fourier transform.

7. The three-dimensional sound source localization imaging method based on a two-dimensional convex array according to claim 4, characterized in that, The frequency domain sound field of the next layer is updated using the step-by-step spiral spectral method. include:

8. A three-dimensional sound source localization imaging device based on a two-dimensional convex array, characterized in that, include: The signal acquisition module is used to acquire the time-domain signal p(r,φ,y,t) passively received by the two-dimensional convex array ultrasonic transducer in a non-uniform sound velocity medium. The array elements of the two-dimensional convex array ultrasonic transducer are distributed on a cylindrical surface, and the cylindrical coordinate reference system is (r,φ,y), where r is the extrapolation radius of the probe of the two-dimensional convex array ultrasonic transducer, and R is the distribution surface of the array elements of the convex array ultrasonic transducer. The medium slowness determination module is used to determine the sound velocity distribution of the propagation medium based on the time-domain signal p(r,φ,y,t) and generate an aligned sound velocity map to determine the medium slowness u(r,φ,y); The signal transformation module is used to transform the time-domain signal p(r,φ,y,t) in time t, circumferential φ, and longitudinal y to generate a spiral spectrum P. n (r,k y ,ω), where ω represents the time angular frequency, n is the circumferential wavelength, and k y Longitudinal wavelength; Image generation module, used to generate images based on the helical spectrum P n (r,k y The frequency domain sound field of the next layer is determined by using the medium slowness u(r,φ,y) and ω, and a passive sound image I(r,φ,y) is generated based on the frequency domain sound field.

9. A device, characterized in that, include: One or more processors and memory, The memory stores a computer program that, when executed by the one or more processors, causes the device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by one or more processors, causes the device to perform the method as described in any one of claims 1-7.

Images