Three-dimensional reconstruction method and system based on two-dimensional foresight sonar image

Abstract

The invention discloses a three-dimensional reconstruction method and system based on a two-dimensional foresight sonar image, and the method comprises the steps: 1, introducing a Mahalanobis distance detection method for a collected two-dimensional foresight sonar sequence, determining an initial contour of a local binary fitting level set segmentation method, and further, carrying out the segmentation of a local binary fitting level set; using a local binary fitting (LBF) level set segmentation method to determine a to-be-reconstructed region of the target on the basis of the detected initial contour of the target; and 2, registering the target area determined in the step 1 by using a registration method based on phase correlation, and calculating the relative displacement of sonar motion. And 3, based on a result of the above steps, reconstructing a three-dimensional point cloud image of the target by using a blind deconvolution algorithm. Compared with a traditional method, three-dimensional reconstruction can be performed on the target without acquiring horizontal position information of the sonar in advance, and the method has certain anti-noise capability. The method is more suitable for three-dimensional reconstruction of the underwater target.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and in particular relates to a three-dimensional reconstruction method and system based on a two-dimensional forward-looking sonar image. Background Art

[0002] In recent years, with the continuous development of the ocean, the shape and position estimation of underwater objects has become a hot issue. Applications such as detecting pipeline leaks, detecting and identifying potential threats on the seabed, etc. all require accurate estimation of the position and shape information of underwater objects.

[0003] Although forward-looking sonar has a large horizontal aperture and can image a large range in the water, it is unable to accurately obtain the target's elevation information due to ambiguity in the vertical aperture direction, so it is impossible to perform three-dimensional imaging of underwater targets. The existing two-dimensional forward-looking sonar three-dimensional reconstruction method requires accurate control of the sonar motion trajectory so that the positions of each point of the target to be reconstructed remain consistent in the sonar image.

[0004] Therefore, there is an urgent need for a target reconstruction method based on two-dimensional forward-looking sonar images to achieve three-dimensional reconstruction of underwater targets. Summary of the invention

[0005] The present invention aims to solve the technical problems existing in the background technology and to provide a three-dimensional reconstruction method and system based on two-dimensional forward-looking sonar images, so as to realize the restoration of the elevation angle of the target using the two-dimensional forward-looking sonar images.

[0006] In order to solve the technical problem, the technical solution of the present invention is:

[0007] A three-dimensional reconstruction method based on a two-dimensional forward-looking sonar image, the method comprising:

[0008] S1: For the collected two-dimensional forward-looking sonar sequence, the initial contour in the image is determined by the Mahalanobis distance detection method. Based on the detected initial contour, the local binary fitting LBF level set segmentation method is used to accurately segment the target area. The target area to be reconstructed is determined based on the segmentation result;

[0009] S2: For the target area determined in step S1, a certain frame is selected as a reference frame through a phase correlation-based registration method, and other sonar image frames are regarded as motion frames. The registration is performed using a formula to calculate the relative displacement of the sonar;

[0010] S3: Based on the target area obtained in step S1 and the sonar relative displacement obtained in step S2, a blind deconvolution algorithm is used to reconstruct a three-dimensional point cloud image of the target.

[0011] Furthermore, in step S1, a Mahalanobis distance detection method is introduced to determine the initial contour of the local binary fitting level set segmentation method. This method detects each point in the image by means of an inner and outer sliding window to determine whether it is a target area. The test statistic is expressed as:

[0012]

[0013] Where D(x) is the test statistic of the algorithm, U 0 Indicates that there is no target in the area, event U 1 Indicates the presence of a target in the area, u s is the pixel mean of the inner window, u s is the pixel mean of the outer window, is the variance of the exterior window;

[0014] The algorithm estimates the initial contour of the target in the sonar image by distinguishing the calculated test statistic D(x) from the set threshold.

[0015] Further, in step S1: a local binary fitting LBF level set segmentation method is used to determine the target area to be reconstructed based on the detected initial contour of the target. The energy functional form of the method is:

[0016]

[0017] In the formula, F(φ,f 1 ,f 2 ) is the energy functional of the level set, ε LBF (φ,f 1 ,f 2 ) is the area term of the local binary fitting method, is the penalty item, The length term, μ, and v are the coefficients of the area term and the energy term, respectively;

[0018] ε LBF (φ,f 1 ,f 2 ) is in the form of:

[0019] ε LBF (φ,f 1 ,f 2 )=λ 1 ∫[∫K σ (xy)|I(y)-f 1 (x)| 2 H(φ(y))dy]dx+λ 2 ∫[∫K σ (xy)|I(y)-f2 (x)| 2 ·(1-H(φ(y)))dy]dx

[0020] In the formula, λ 1 ,λ 2 is a constant, K is the kernel function, I is the sonar image frame, H is the Heaviside function, φ is the level set function, and f 1 (x), f 2 (x) is the value near the fitting point x, which changes as the position of the fitting point changes;

[0021] The specific form is:

[0022]

[0023] Where Ω is the image domain, is the gradient symbol;

[0024] The specific form is

[0025]

[0026] Where δ is the derivative of the Heaviside function;

[0027] The algorithm minimizes the energy functional through the gradient descent method to iterate the target contour. Its formula is as follows:

[0028]

[0029] In the formula, e 1 , e 2 The forms are:

[0030] e 1 (x) = ∫ Ω K(yx)|I(x)-f 1 (y)| 2 dy

[0031] e 2 (x) = ∫ Ω K(yx)|I(x)-f 2 (y)| 2 dy

[0032] According to the discretization form of the above formula, the time step is set, and finally the refined outline of the target can be obtained.

[0033] Further, in step S2:

[0034] The image segmentation result determined in step S1, i.e., the target area, is registered using a phase-correlation-based registration method. A certain frame is selected as a reference frame, and other sonar image frames are regarded as motion frames. They are registered to the reference frame, and the relative displacement of the sonar motion is calculated. The registration formula is as follows:

[0035]

[0036] In the formula, F 1 (ξ,η) is the Fourier transform result of the reference frame image, F 2 (ξ,η) is the Fourier transform result of the motion frame image, x 0 ,y 0 is the relative displacement of sonar motion;

[0037] The formula form of the cross power spectrum of the reference frame and the motion frame is:

[0038]

[0039] Where H(ξ,η) is the cross power spectrum, * is the complex conjugate;

[0040] Perform inverse Fourier transform on the cross-power spectrum in the above equation, and estimate the relative displacement of the sonar by finding the peak value of the obtained pulse function.

[0041] Further, the step S3 specifically includes:

[0042]

[0043] Where: I rθ (z) is the gray value of the sonar image r,θ at the height z after registration, β r is the vertical directivity index corresponding to the sonar image at r, V rθ is the existence term coefficient of the target, that is, the point cloud image to be solved, ρ(z,u,n) is the cosine value of the angle between the incident sound wave and the surface normal vector n of the object; the three-dimensional point cloud of the target is reconstructed by solving the non-negative least squares solution using the discretized form of the above formula.

[0044] Furthermore, in step S3, the discrete form of the gray value formula of the sonar image r,θ at the height z after registration is as follows:

[0045]

[0046] In the formula, z i is the z coordinate of the i-th frame image sequence, and P is the total number of points in the vertical direction of the point cloud to be reconstructed;

[0047] The above formula can be organized into the following form:

[0048] I rθ =AN×P b rθ

[0049] In the formula b rθ =V krθ ;

[0050] Solve the equation using the non-negative least squares method and get b rθ The value of is the point cloud after the target 3D reconstruction.

[0051] A three-dimensional reconstruction system based on a two-dimensional forward-looking sonar image, the system is used to perform any of the above methods, the system comprising:

[0052] Target domain segmentation module: For the acquired sonar image sequence, a fast detection method based on Mahalanobis distance is used to determine the general outline of the target, and then the local binary level set method is used to accurately determine the target outline;

[0053] Registration module: Use the phase correlation algorithm to register the target contour extracted by the image segmentation module and estimate the inter-frame displacement;

[0054] 3D reconstruction module: uses the precise contour of the target segmentation module and the inter-frame displacement calculated by the registration module to perform 3D reconstruction of the target collected by the forward-looking sonar image sequence.

[0055] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any one of the above-mentioned three-dimensional reconstruction methods based on two-dimensional forward-looking sonar images is implemented.

[0056] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any one of the above-mentioned three-dimensional reconstruction methods based on a two-dimensional forward-looking sonar image.

[0057] Compared with the prior art, the advantages of the present invention are:

[0058] Compared with the traditional method, the present invention does not need to obtain the X and Y displacements of the sonar in advance, but only needs to obtain the relative displacement of the Z axis of the sonar to perform three-dimensional reconstruction of the target. In addition, since the target position and outline are segmented in advance, the interference of the background can be eliminated, and it has a certain anti-noise ability. The method of the present invention is more suitable for three-dimensional reconstruction of underwater targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of the three-dimensional reconstruction method based on the two-dimensional forward-looking sonar image according to the present invention;

[0060] Figure 2 This is the result of the three-dimensional reconstruction method based on the two-dimensional forward-looking sonar image described in the present invention. DETAILED DESCRIPTION

[0061] The specific implementation mode of the present invention is described below in conjunction with embodiments:

[0062] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to match the contents disclosed in the specification so that people familiar with this technology can understand and read them, and are not used to limit the conditions under which the present invention can be implemented. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical content disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0063] At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" cited in this specification are only for the convenience of description and are not used to limit the scope of implementation of the present invention. Changes or adjustments to their relative relationships should be regarded as the scope of implementation of the present invention without substantially changing the technical content.

[0064] Embodiment 1:

[0065] Combination Figure 1 The present invention proposes a three-dimensional reconstruction method based on a two-dimensional forward-looking sonar image, the method comprising:

[0066] Step 1: The present invention introduces a Mahalanobis distance detection method to determine the initial contour of the local binary fitting level set segmentation method. This method detects each point in the image by means of an inner and outer sliding window to determine whether it is a target area. The test statistic is expressed as:

[0067]

[0068] Where D(x) is the test statistic of the algorithm, U 0 Indicates that there is no target in the area, event U 1 Indicates the presence of a target in the area, u s is the pixel mean of the inner window, u b is the pixel mean of the outer window, is the variance of the exterior window.

[0069] The algorithm estimates the initial contour of the target in the sonar image by distinguishing the calculated test statistic D(x) from the set threshold.

[0070] Furthermore, in step 1, a local binary fitting (LBF) level set segmentation method is used to determine the target area to be reconstructed based on the detected initial contour of the target. The energy functional form of this method is:

[0071]

[0072] In the formula, F(φ,f 1 ,f 2 ) is the energy functional of the level set, ε LBF (φ,f 1 ,f 2 ) is the area term of the local binary fitting method, is the penalty item, The length term, μ, υ are the coefficients of the area term and the energy term respectively.

[0073] ε LBF (φ,f 1 ,f 2 ) is in the form of:

[0074] ε LBF (φ,f 1 ,f 2 )=λ 1 ∫[∫K σ (xy)|I(y)-f 1 (x)| 2 H(φ(y))dy]dx+λ 2 ∫[∫K σ (xy)|I(y)-f 2 (x)| 2 ·(1-H(φ(y)))dy]dx

[0075] In the formula, λ 1 ,λ 2 is a constant, K is the kernel function, I is the sonar image frame, H is the Heaviside function, φ is the level set function, and f 1 (x), f 2 (x) is the value near the fitting point x, which changes as the position of the fitting point changes.

[0076] The specific form is:

[0077]

[0078] Where Ω is the image domain, is the gradient symbol;

[0079] The specific form is

[0080]

[0081] Where δ is the derivative of the Heaviside function.

[0082] The algorithm minimizes the energy functional through the gradient descent method to iterate the target contour. Its formula is as follows:

[0083]

[0084] In the formula, e 1 , e 2 The forms are

[0085] e 1 (x) = ∫ Ω K(yx)|I(x)-f 1 (y)| 2 dy

[0086] e 2 (x) = ∫ Ω K(yx)|I(x)-f 2 (y)| 2 dy

[0087] According to the discretization form of the above formula, the time step is set, and finally the refined outline of the target can be obtained.

[0088] Step 2: Use the phase-correlation-based registration method to register the target area determined in step 1. Select a frame as the reference frame, regard other sonar image frames as motion frames, register them to the reference frame, and calculate the relative displacement of the sonar motion. In the case of only rigid translation, the relationship between the motion frame and the reference frame is as follows:

[0089] f 2 (x,y)=f 1 (xx 0 ,yy 0 )

[0090] Where f 1 (x,y),f 2 (x, y) are the reference frame and motion frame respectively, x 0 ,y 0 is the relative displacement of the x and y axes.

[0091] Furthermore, its Fourier transform form is

[0092]

[0093] Furthermore, the formula form of the cross power spectrum of the reference frame and the motion frame is

[0094]

[0095] Where H(ξ,η) is the cross power spectrum and * is the complex conjugate.

[0096] Perform inverse Fourier transform on the cross-power spectrum in the above equation, and estimate the relative displacement of the sonar by finding the peak value of the obtained pulse function.

[0097] Step 3: Based on the sonar image sequence segmentation results obtained in step 1 and the sonar relative displacement obtained in step 2, a blind deconvolution algorithm is used to reconstruct the three-dimensional point cloud image of the target. The formula is as follows:

[0098]

[0099] Where I rθ (z) is the gray value of the sonar image r,θ at the height z after registration, β r is the vertical directivity index corresponding to the sonar image at r, V rθ is the existence term coefficient of the target, that is, the point cloud image to be solved, and ρ(z,u,n) is the cosine value of the angle between the incident sound wave and the surface normal vector n of the object.

[0100] Furthermore, the discrete form of the above equation is as follows:

[0101]

[0102] In the formula, z i is the z coordinate of the i-th frame image sequence, and P is the total number of points in the vertical direction of the point cloud to be reconstructed

[0103] Furthermore, the above formula can be organized into the following form:

[0104] I rθ =A N×P b rθ

[0105] In the formula b rθ =V krθ .

[0106] Furthermore, the present invention uses the non-negative least squares method to solve the equation and obtains b rθ The value of is the point cloud after the target 3D reconstruction.

[0107] Figure 2 The results of using the algorithm to process the simulated sonar image sequence. The left side is one of the sonar image sequence frames, and the right side is the point cloud reconstruction result.

[0108] Embodiment 2:

[0109] The present invention proposes a three-dimensional reconstruction system based on forward-looking sonar images, which can be used to implement the above-mentioned three-dimensional reconstruction method based on forward-looking sonar images. Specifically, the system includes:

[0110] Target domain segmentation module: For sonar image sequences, a fast detection method based on Mahalanobis distance is used to determine the general outline of the target, and then a local binary level set method is used to accurately determine the target outline;

[0111] Registration module: using the phase correlation algorithm, register the target contour extracted by the image segmentation module and estimate the inter-frame displacement;

[0112] 3D reconstruction module: Use the target position information from the segmentation module and the inter-frame displacement calculated by the registration module to perform 3D reconstruction of the target collected by the forward-looking sonar image sequence.

[0113] Embodiment 3:

[0114] This embodiment provides a terminal device, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a three-dimensional reconstruction method based on forward-looking sonar images, including the following steps:

[0115] Step 1: For the collected two-dimensional forward-looking sonar sequence, since the traditional local binary fitting level set segmentation method is sensitive to the initial contour, the wrong initial contour will affect the final contour segmentation result of the target. The present invention introduces a Mahalanobis distance detection method to determine the initial contour of the local binary fitting level set segmentation method. This method detects each point in the image by means of internal and external sliding windows.

[0116] Step 2: Use the phase correlation-based registration method to register the target area determined in step 1. Select a frame as the reference frame, regard other sonar image frames as motion frames, register them to the reference frame, and calculate the relative displacement of the sonar motion. The registration formula is as follows:

[0117]

[0118] In the formula, F 1 (ξ,η) is the Fourier transform result of the reference frame image, F 2 (ξ,η) is the Fourier transform result of the motion frame image, x 0 ,y 0 is the relative displacement of sonar movement.

[0119] Step 3: Based on the segmentation result obtained in step 1 and the sonar relative displacement obtained in step 2, a blind deconvolution algorithm is used to reconstruct the three-dimensional point cloud image of the target. The formula is as follows.

[0120]

[0121] Where I rθ (z) is the gray value of the sonar image r,θ at the height z after registration, β r is the vertical directivity index corresponding to the sonar image at r, V rθ is the existence term coefficient of the target, that is, the point cloud image to be solved, and ρ(z,u,n) is the cosine value of the angle between the incident sound wave and the surface normal vector n of the object.

[0122] Embodiment 4:

[0123] This embodiment provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.

[0124] The processor may load and execute one or more instructions stored in a computer-readable storage medium to implement the corresponding steps of the three-dimensional reconstruction method based on a forward-looking sonar image in the above embodiment; the processor may load and execute the following steps of one or more instructions in the computer-readable storage medium:

[0125] Step 1: For the collected two-dimensional forward-looking sonar sequence, since the traditional local binary fitting level set segmentation method is sensitive to the initial contour, the wrong initial contour will affect the final contour segmentation result of the target. The present invention introduces a Mahalanobis distance detection method to determine the initial contour of the local binary fitting level set segmentation method. This method detects each point in the image by means of internal and external sliding windows.

[0126] Step 2: Use the phase correlation-based registration method to register the target area determined in step 1. Select a frame as the reference frame, regard other sonar image frames as motion frames, register them to the reference frame, and calculate the relative displacement of the sonar motion. The registration formula is as follows:

[0127]

[0128] In the formula, F 1 (ξ,η) is the Fourier transform result of the reference frame image, F 2 (ξ,η) is the Fourier transform result of the motion frame image, x 0 ,y 0 is the relative displacement of sonar movement.

[0129] Step 3: Based on the segmentation result obtained in step 1 and the sonar relative displacement obtained in step 2, a blind deconvolution algorithm is used to reconstruct the three-dimensional point cloud image of the target. The formula is as follows.

[0130]

[0131] Where I rθ (z) is the gray value of the sonar image r,θ at the height z after registration, β r is the vertical directivity index corresponding to the sonar image at r, V rθ is the existence term coefficient of the target, that is, the point cloud image to be solved, and ρ(z,u,n) is the cosine value of the angle between the incident sound wave and the surface normal vector m of the object.

[0132] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0133] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0134] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0135] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0136] The preferred embodiments of the present invention are described in detail above, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

[0137] Many other changes and modifications may be made without departing from the concept and scope of the present invention.It should be understood that the present invention is not limited to the specific embodiments, and the scope of the present invention is defined by the appended claims.

Claims

1. A three-dimensional reconstruction method based on two-dimensional forward-looking sonar images, characterized in that: The method comprises: S1: For the collected two-dimensional forward-looking sonar sequence, the initial contour in the image is determined by the Mahalanobis distance detection method. Based on the detected initial contour, the local binary fitting LBF level set segmentation method is used to accurately segment the target area. The target area to be reconstructed is determined based on the segmentation results. S2: For the target area determined in step S1, a certain frame is selected as a reference frame through a phase correlation-based registration method, and other sonar image frames are regarded as motion frames. The registration is performed using a formula to calculate the relative displacement of the sonar; S3: Based on the target area obtained in step S1 and the sonar relative displacement obtained in step S2, a blind deconvolution algorithm is used to reconstruct a three-dimensional point cloud image of the target.

2. The three-dimensional reconstruction method based on two-dimensional forward-looking sonar images according to claim 1, characterized in that: In step S1, a Mahalanobis distance detection method is introduced to determine the initial contour of the local binary fitting level set segmentation method. This method detects each point in the image by means of an inner and outer sliding window to determine whether it is a target area. The test statistic is expressed as: In the formula, D(x) is the test statistic of the algorithm, U0 indicates that there is no target in the area, event U1 indicates that there is a target in the area, and u s is the pixel mean of the inner window, u b is the pixel mean of the outer window, is the variance of the exterior window; The algorithm estimates the initial contour of the target in the sonar image by distinguishing the calculated test statistic D(x) from the set threshold.

3. The three-dimensional reconstruction method based on two-dimensional forward-looking sonar images according to claim 1, characterized in that: In step S1: a local binary fitting LBF level set segmentation method is used to determine the target area to be reconstructed based on the detected initial contour of the target. The energy functional form of this method is: Where F(φ,f1,f2) is the energy functional of the level set, ε LBF (φ,f1,f2) is the area term of the local binary fitting method, is the penalty item, The length term, μ, υ are the coefficients of the area term and energy term respectively; ε LBF The specific form of (φ,f1,f2) is: Where λ1, λ2 are constants, K is the kernel function, I is the sonar image frame, H is the Heaviside function, φ is the level set function, f1(x), f2(x) are the values ​​near the fitting point x, which change as the position of the fitting point changes; The specific form is: Where Ω is the image domain, is the gradient symbol; The specific form is Where δ is the derivative of the Heaviside function; The algorithm minimizes the energy functional through the gradient descent method to iterate the target contour. Its formula is as follows: In the formula, the forms of e1 and e2 are: e1(x)=∫ Ω K(y-x)|I(x)-f1(y)| 2 dy e2(x)=∫ Ω K(yx)|I(x)-f2(y)| 2 of According to the discretization form of the above formula, the time step is set, and finally the refined outline of the target can be obtained.

4. The three-dimensional reconstruction method based on two-dimensional forward-looking sonar images according to claim 1, characterized in that: In step S2: The image segmentation result determined in step S1, i.e., the target area, is registered using a phase-correlation-based registration method. A certain frame is selected as a reference frame, and other sonar image frames are regarded as motion frames. They are registered to the reference frame, and the relative displacement of the sonar motion is calculated. The registration formula is as follows: Where, F1(ξ,η) is the Fourier transform result of the reference frame image, F2(ξ,η) is the Fourier transform result of the motion frame image, x0, y0 are the relative displacements of the sonar motion; The formula form of the cross power spectrum of the reference frame and the motion frame is: Where H(ξ,η) is the cross power spectrum, * is the complex conjugate; Perform inverse Fourier transform on the cross-power spectrum in the above equation, and estimate the relative displacement of the sonar by finding the peak value of the obtained pulse function.

5. The three-dimensional reconstruction method based on two-dimensional forward-looking sonar images according to claim 1, characterized in that: The step S3 specifically includes: Where: I rθ (z) is the gray value of the sonar image r,θ at the height z after registration, β r is the vertical directivity index corresponding to the sonar image at r, V rθ is the existence term coefficient of the target, that is, the point cloud image to be solved, ρ(z,u,n) is the cosine value of the angle between the incident sound wave and the surface normal vector n of the object; the three-dimensional point cloud of the target is reconstructed by solving the non-negative least squares solution using the discretized form of the above formula.

6. The three-dimensional reconstruction method based on two-dimensional forward-looking sonar images according to claim 5, characterized in that: In step S3, the discrete form of the gray value formula of the sonar image r,θ at the height z after registration is as follows: In the formula, z i is the z coordinate of the i-th frame image sequence, and P is the total number of points in the vertical direction of the point cloud to be reconstructed; The above formula can be organized into the following form: I rθ =A N×P ·b rθ In the formula b rθ =V krθ ; Solve the equation using the non-negative least squares method and get b rθ The value of is the point cloud after the target 3D reconstruction.

7. A three-dimensional reconstruction system based on two-dimensional forward-looking sonar images, characterized in that: The system is used to execute the method described in any one of claims 1 to 6, and the system comprises: Target domain segmentation module: For the acquired sonar image sequence, a fast detection method based on Mahalanobis distance is used to determine the general outline of the target, and then the local binary level set method is used to accurately determine the target outline; Registration module: Use the phase correlation algorithm to register the target contour extracted by the image segmentation module and estimate the inter-frame displacement; 3D reconstruction module: uses the precise contour of the target segmentation module and the inter-frame displacement calculated by the registration module to perform 3D reconstruction of the target collected by the forward-looking sonar image sequence.

8. A computer device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a three-dimensional reconstruction method based on a two-dimensional forward-looking sonar image according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a three-dimensional reconstruction method based on a two-dimensional forward-looking sonar image according to any one of claims 1 to 6.

Images