Super-resolution underwater acoustic imaging method, device and equipment for class optical diffraction tomography

Abstract

The present application relates to a super-resolution underwater acoustic imaging method, device and equipment for optical diffraction tomography-like. The method includes: initializing to obtain a random complex initial guess of the target; propagating the random complex initial guess to the wavenumber domain to obtain a low-resolution wavenumber spectrum; based on the optical diffraction tomography-like algorithm, the imaging unit rotates and scans the target overlapped at a set angular increment, and propagates the sampled backscattered acoustic signal to the spatial domain to obtain a reconstructed sub-image; transmitting the low-resolution wavenumber spectrum back to the spatial domain to obtain a low-resolution target sub-image, and performing intensity constraint on the low-resolution target sub-image according to the reconstructed sub-image and propagating it to the wavenumber domain to obtain a high-resolution wavenumber spectrum; repeating the above steps, and fusing and stitching the high-resolution wavenumber spectra of each angular region of the target for inverse transformation to reconstruct a super-resolution target image. Using this method can improve the resolution of long-distance underwater acoustic imaging.

Description

Technical Field

[0001] The present application relates to the field of underwater acoustic imaging technology, and particularly to a super-resolution underwater acoustic imaging method, device and equipment similar to optical diffraction tomography. Background Art

[0002] Underwater acoustic imaging refers to imaging by processing the backscattered acoustic signals of objects on the seabed or in water, and the quality of target imaging depends on the integrity of the information of the target scattering frequency angular spectrum diagram. At present, the mainstream underwater acoustic imaging methods at home and abroad are two-dimensional squint synthetic aperture imaging sonar and downward-looking imaging.

[0003] Among them, most two-dimensional squint synthetic aperture imaging sonars adopt the synthetic aperture squint system. The sonar is located above the sea surface or seabed, and a two-dimensional "image" is reconstructed by squint detection of the target. However, the traditional two-dimensional squint synthetic aperture imaging sonar has the following three problems, which bring great difficulties to target recognition. One is that the two-dimensional sonar image is not intuitive. This directly leads to difficulties in interpreting the shape characteristics of objects with complex shapes and objects laid near the seabed, and even misjudgment due to projection distortion. The second is that it is greatly affected by the target scattering characteristics. In practical applications, it is mainly manifested that for the same target, when the imaging perspective direction is different, the bright spot and shadow characteristics of the generated two-dimensional sonar image will change greatly. The third is that there is a blind area directly below the system. When the target is below or near the system, normal two-dimensional imaging cannot be performed.

[0004] Downward-looking imaging has the advantage of being able to obtain three-dimensional information of both the plane dimension and the depth distance dimension at the same time, and there is no downward-looking blind area. It is an imaging method that complements the squint system. However, the existing downward-looking three-dimensional imaging mainly relies on a large-scale planar array for real-aperture imaging, or uses a multi-beam method to obtain a three-dimensional sonar image through a main passive cross array. The main problem is that the imaging resolution is related to the distance, and a huge array aperture is required for high-resolution imaging at a long distance, which is difficult to achieve in engineering. Summary of the Invention

[0005] Based on this, in view of the technical problem of low imaging resolution at long distances existing in the above-mentioned existing underwater acoustic imaging, it is necessary to provide a super-resolution underwater acoustic imaging method, device and equipment similar to optical diffraction tomography.

[0006] A super-resolution underwater acoustic imaging method similar to optical diffraction tomography, the method includes:

[0007] Initialize the high-resolution image of the underwater target and obtain a random complex initial guess of the target;

[0008] Propagate the random complex initial guess to the wavenumber domain to obtain a low-resolution wavenumber spectrum of the random complex initial guess; wherein, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped;

[0009] Based on the optical diffraction tomography algorithm, the imaging unit rotates at a set angular increment and scans the target overlappedly in a circular motion to collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain, obtaining the reconstructed sub-images of each angular region of the target;

[0010] After the imaging unit rotates once at the set angular increment and samples the backscattered acoustic signals in the corresponding angular region of the target, the low-resolution wavenumber spectrum corresponding to this angular region is transmitted back to the spatial domain to obtain the low-resolution target sub-image of this angular region, and the intensity of the low-resolution target sub-image of this angular region is constrained according to the reconstructed sub-image of the same angular region of the target, and the image after intensity constraint is propagated to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region;

[0011] Repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target, fuse and splice the high-resolution wavenumber spectra of each angular region of the target and perform inverse transformation to reconstruct the super-resolution target image.

[0012] In one embodiment, initialize the high-resolution image of the underwater target, and obtain the random complex initial guess of the target, denoted as where the subscript hr represents high resolution, i represents the imaginary unit, I hr is the target intensity, is the target phase, and I hr and take random values.

[0013] In one embodiment, propagate the random complex initial guess to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess, including:

[0014] Propagate the random complex initial guess of the target to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess, denoted as

[0015]

[0016] where I lr (k x , k y ) is the intensity of the low-resolution wavenumber spectrum, the subscript lr represents low resolution; (x, y) represents the spatial domain where the target is located, (k x , k y ) represents the wavenumber domain where the wavenumber spectrum is located, and F represents the Fourier transform.

[0017] In one embodiment, based on the optical diffraction tomography-like algorithm, the imaging unit rotates at a set angular increment and scans the target in an overlapping manner around it to collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain, obtaining the reconstructed sub-images of each angular region of the target, including:

[0018] Based on the optical diffraction tomography-like algorithm, a small hydrophone linear array or planar array is used as the imaging unit, and the imaging unit rotates at a set angular increment Δθ and scans the target in an overlapping manner around it to collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain, obtaining the reconstructed sub-images of each angular region of the target; where the overlapping scan sampling means that the angular region scanned each time by the imaging unit partially overlaps with the angular region scanned in the previous time.

[0019] In one embodiment, after the imaging unit rotates once at a set angular increment and samples the backscattered acoustic signals in the corresponding angular region of the target, the low-resolution wavenumber spectrum corresponding to this angular region is transmitted back to the spatial domain to obtain the low-resolution target sub-image of this angular region, including:

[0020] After the imaging unit rotates once at a set angular increment Δθ and samples the backscattered acoustic signals in the corresponding angular region of the target, by performing an inverse Fourier transform on the low-resolution wavenumber spectrum corresponding to this angular region the low-resolution wavenumber spectrum corresponding to this angular region is transmitted back to the spatial domain to obtain the low-resolution target sub-image of this angular region, denoted as

[0021]

[0022] where I l is the intensity of the low-resolution target sub-image, is the phase of the low-resolution target sub-image, F -1 represents the inverse Fourier transform, I lr (k x ,k y ) is the intensity of the low-resolution wavenumber spectrum, the subscript lr represents low resolution, (k x ,k y ) represents the wavenumber domain where the wavenumber spectrum is located, and i represents the imaginary unit.

[0023] In one embodiment, the intensity of the low-resolution target sub-image of an angular region of the target is constrained according to the reconstructed sub-image of the same angular region of the target, and the image after intensity constraint is propagated to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region, including:

[0024] Perform intensity constraint on the low-resolution target sub-image of the target in the same angular region using the reconstructed sub-image of the same angular region of the target. When performing the constraint, use the amplitude modulus measurement value of the reconstructed sub-image of the same angular region of the target Replace the intensity component in the low-resolution target sub-image of this angular region, and retain the phase component in the low-resolution target sub-image of this angular region to obtain the image after intensity constraint, denoted as

[0025] Propagate the image after intensity constraint to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region, denoted as

[0026]

[0027] where I hr (k x , k y ) is the intensity of the high-resolution wavenumber spectrum, is the phase of the high-resolution wavenumber spectrum. The subscript hr represents high resolution, i represents the imaginary unit, and F represents the Fourier transform

[0028] In one embodiment, fuse and splice the high-resolution wavenumber spectra of each angular region of the target and perform an inverse transform to reconstruct the super-resolution target image, including:

[0029] Fuse and splice the high-resolution wavenumber spectra of each angular region of the target, and perform an inverse Fourier transform on the fused high-resolution wavenumber spectrum of the target to reconstruct the super-resolution target image

[0030] A super-resolution underwater acoustic imaging device of a type of optical diffraction tomography, the device includes:

[0031] An initialization module, configured to initialize the high-resolution image of an underwater target and obtain a random complex initial guess of the target

[0032] A guess propagation module, configured to propagate the random complex initial guess to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess; wherein, the distribution of the complete low-resolution wavenumber spectrum is annular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped

[0033] An overlapping scanning and sampling module, configured to perform overlapping scanning and sampling around the target by rotating the imaging unit according to a set angular increment based on the optical diffraction tomography-like algorithm, collect the backscattered acoustic signals of each angular region of the target and propagate them to the spatial domain to obtain the reconstructed sub-images of each angular region of the target

[0034] The intensity constraint module is used to, after the imaging unit rotates once according to a set angular increment and samples to obtain the backscattering signal of the target in the corresponding angular region, transmit the low-resolution wavenumber spectrum corresponding to this angular region back to the spatial domain, obtain the low-resolution target sub-image of this angular region, perform intensity constraint on the low-resolution target sub-image of this angular region according to the reconstructed sub-image of the target in the same angular region, and transmit the intensity-constrained image to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region;

[0035] The super-resolution imaging module is used to repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target, fuse and splice the high-resolution wavenumber spectra of each angular region of the target, and perform inverse transformation to reconstruct the super-resolution target image.

[0036] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0037] Initialize the high-resolution image of the underwater target and obtain the random complex initial guess of the target;

[0038] Transmit the random complex initial guess to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess; among them, the distribution of the complete low-resolution wavenumber spectrum is circular ring-shaped, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped;

[0039] Based on the optical diffraction tomography-like algorithm, use the imaging unit to rotate according to the set angular increment and perform overlapping scanning and sampling around the target, collect and obtain the backscattering signals of each angular region of the target and transmit them to the spatial domain to obtain the reconstructed sub-images of each angular region of the target;

[0040] After the imaging unit rotates once according to the set angular increment and samples to obtain the backscattering signal of the target in the corresponding angular region, transmit the low-resolution wavenumber spectrum corresponding to this angular region back to the spatial domain, obtain the low-resolution target sub-image of this angular region, perform intensity constraint on the low-resolution target sub-image of this angular region according to the reconstructed sub-image of the target in the same angular region, and transmit the intensity-constrained image to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region;

[0041] Repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target, fuse and splice the high-resolution wavenumber spectra of each angular region of the target, and perform inverse transformation to reconstruct the super-resolution target image.

[0042] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the following steps are implemented:

[0043] Initialize a high-resolution image of an underwater target and obtain a random complex initial guess of the target;

[0044] Propagate the random complex initial guess to the wavenumber domain to obtain a low-resolution wavenumber spectrum of the random complex initial guess; wherein, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped;

[0045] Based on an optical diffraction tomography-like algorithm, the imaging unit rotates at a set angular increment and scans the target in an overlapping manner around the target, collects the backscattered acoustic signals in each angular region of the target and propagates them to the spatial domain to obtain reconstructed sub-images of each angular region of the target;

[0046] After the imaging unit rotates once at a set angular increment and samples the backscattered acoustic signals in the corresponding angular region of the target, the low-resolution wavenumber spectrum corresponding to this angular region is transmitted back to the spatial domain to obtain a low-resolution target sub-image of this angular region, and the low-resolution target sub-image of this angular region is intensity-constrained according to the reconstructed sub-image of the same angular region of the target, and the intensity-constrained image is propagated to the wavenumber domain to obtain a high-resolution wavenumber spectrum of this angular region;

[0047] Repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target, fuse and splice the high-resolution wavenumber spectra of each angular region of the target and perform an inverse transformation to reconstruct a super-resolution target image.

[0048] The above super-resolution underwater acoustic imaging method, device and equipment of pseudo-optical diffraction tomography introduce the pseudo-optical diffraction tomography algorithm into underwater acoustic imaging. By rotating the imaging unit according to a set angular increment and performing overlapping scanning sampling around the target, the overlapping scanning sampling can solve the sparsity of sampling by the imaging unit at a long distance, obtain sufficient and complete target scattering field information, which is beneficial to accumulating redundant spatial frequency information. Further, it is possible to construct a reconstructed sub-image by using the backscattered acoustic signals in each angular region of the target acquired by collection, and perform intensity constraint on the low-resolution target sub-images in each angular region of the target through the reconstructed sub-image, so that the image intensity constraint reconstruction process proceeds in the direction of higher resolution of the target image, obtaining a high-resolution wavenumber spectrum with extremely rich spatial frequency information, even close to all the complete spatial frequency information of the target object, and through inverse transformation, reconstructing a super-resolution target image close to the real object. By using the overlapping redundant information brought by the angular transformation of the imaging unit to perform overlapping scanning sampling around the target, applying the idea of the pseudo-optical diffraction tomography algorithm, and iteratively reconstructing the super-resolution target image, the present application can significantly improve the resolution of the original underwater acoustic image of a long-distance target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a schematic flowchart of a super-resolution underwater acoustic imaging method of pseudo-optical diffraction tomography in an embodiment;

[0050] Figure 2 is a schematic diagram of the imaging unit rotating and performing overlapping scanning sampling around the target in an embodiment;

[0051] Figure 3 is a schematic diagram of overlapping scanning sampling in an embodiment;

[0052] Figure 4 is an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] In one embodiment, as Figure 1 shown, a super-resolution underwater acoustic imaging method of pseudo-optical diffraction tomography is provided, including the following steps:

[0055] Step S1, initialize the high-resolution image of the underwater target and obtain a random complex initial guess of the target.

[0056] Specifically, initialize the high-resolution image of the underwater target and obtain a random complex initial guess of the target, denoted as Among them, the subscript hr represents high resolution, i represents the imaginary unit, and I hr is the target intensity, is the target phase, and I hr and take random values. Among them, the random complex initial guess of the target can be understood as the initial value of the input loop iteration program for computational imaging. The initial guess is expressed as a complex number in the program, and this value can take random values and will change continuously during the iterative calculation process. For example can take 0, and I hr taking a constant or the intensity of any under-sampled low-resolution image of the target is acceptable.

[0057] Step S2: Propagate the random complex initial guess to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess; among them, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped.

[0058] Specifically, propagate the random complex initial guess of the target to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess, denoted as

[0059]

[0060] where I lr (k x , k y ) is the intensity of the low-resolution wavenumber spectrum, and the subscript lr represents low resolution; (x, y) represents the spatial domain where the target is located, (k x , k y ) represents the wavenumber domain where the wavenumber spectrum is located, F represents the Fourier transform, and the propagation from the spatial domain to the wavenumber domain is a single Fourier transform.

[0061] Step S3: Based on the optical diffraction tomography-like algorithm, use the imaging unit to rotate at a set angular increment and scan the target overlappedly in a circular motion, collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain to obtain the reconstructed sub-images in each angular region of the target.

[0062] Specifically, as Figure 2 shown, based on the optical diffraction tomography-like algorithm, use a small hydrophone linear array or planar array as the imaging unit, and use the imaging unit to rotate at a set angular increment Δθ and scan the target (or target) overlappedly in a circular motion, collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain to obtain the reconstructed sub-images in each angular region of the target. The optical diffraction tomography-like algorithm can be understood as scanning and collecting multiple low-resolution intensity images of the target by changing the incident light angle and iteratively reconstructing the high-resolution intensity image of the target.

[0063] Among them, the imaging unit rotates at a set angular increment Δθ, which means that the angular increment is the same for each rotation. For example, it may be 10 degrees or 20 degrees for each rotation to ensure coverage of the entire 360 degrees around the target.

[0064] Among them, as Figure 3 shown, overlapping scan sampling means that the angular region scanned by the imaging unit each time partially overlaps with the angular region scanned in the previous scan. The purpose of doing this is to solve the sparsity of the hydrophone array sampling at long distances through overlapping scan sampling, improve the wavenumber spectrum information density, and the overlap is conducive to accumulating redundant spatial frequency information, enabling more spatial frequency information to be used for target image reconstruction, which can constrain the reconstruction process to a higher resolution direction of the target image. Moreover, the overlapping scan sampling method brings rich redundant information due to the scan overlap, meets the conditions of excellent prior information, and well solves the deblurring problem that may occur in image reconstruction. Figure 2 and Figure 3 In it, θ represents the azimuth angle and k represents the acoustic wave vector.

[0065] In addition, a small hydrophone linear array or planar array is used as the imaging unit to collect the spatial frequency information within the fan-shaped sub-aperture corresponding to the array aperture at one time. Only a specific aperture overlap is required between multiple array sub-aperture samplings, thus relaxing the strict phase dependence relationship of the circular synthetic aperture monostatic transceiver system for the 360-degree circle.

[0066] Step S4, after the imaging unit rotates once at the set angular increment and samples to obtain the backscattering signal of the target in the corresponding angular region, the low-resolution wavenumber spectrum corresponding to this angular region is back-projected to the spatial domain to obtain the low-resolution target sub-image of this angular region, and the low-resolution target sub-image of this angular region is intensity-constrained according to the reconstructed sub-image of the same angular region of the target, and the intensity-constrained image is propagated to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region.

[0067] Specifically, after the imaging unit rotates once at the set angular increment Δθ and samples to obtain the backscattering signal of the target in the corresponding angular region, by performing an inverse Fourier transform on the low-resolution wavenumber spectrum corresponding to this angular region, the low-resolution wavenumber spectrum corresponding to this angular region is back-projected to the spatial domain to obtain the low-resolution target sub-image of this angular region, denoted as

[0068]

[0069] where I l is the intensity of the low-resolution target sub-image, is the phase of the low-resolution target sub-image, and F -1Denotes the inverse Fourier transform. The propagation from the wavenumber domain to the spatial domain is an inverse Fourier transform in the program, I lr (k x , k y ) is the intensity of the low-resolution wavenumber spectrum. The subscript lr indicates low resolution. (k x , k y ) represents the wavenumber domain where the wavenumber spectrum is located, and i represents the imaginary unit. Among them, the inverse Fourier transform can be implemented using the fast inverse Fourier transform function in the Matlab program.

[0070] Specifically, the intensity of the low-resolution target sub-image in the same angular region of the target is constrained according to the reconstructed sub-image in the same angular region of the target. When constraining, the amplitude modulus measurement value of the reconstructed sub-image in the same angular region of the target is used to replace the intensity component in the low-resolution target sub-image in this angular region, and the phase component in the low-resolution target sub-image in this angular region is retained to obtain the image after intensity constraint, denoted as

[0071] The image after intensity constraint is propagated to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region, denoted as

[0072]

[0073] Among them, I hr (k x , k y ) is the intensity of the high-resolution wavenumber spectrum, is the phase of the high-resolution wavenumber spectrum. The subscript hr indicates high resolution, i represents the imaginary unit, and F represents the Fourier transform.

[0074] The high-resolution wavenumber spectrum obtained through intensity constraint has richer spatial frequency information in this angular region compared to the low-resolution wavenumber spectrum of the same angular region before transformation, and can improve the resolution of the final image.

[0075] Step S5, repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target. The high-resolution wavenumber spectra of each angular region of the target are fused and spliced, and an inverse transformation is performed to reconstruct the super-resolution target image.

[0076] Specifically, repeat the above steps until the low-resolution wavenumber spectrum of the initial random complex conjecture is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target. Then, fuse and splice the high-resolution wavenumber spectra of each angular region of the target to obtain a wavenumber spectrum with extremely rich spatial frequency information, even close to all the complete spatial frequency information of the target object. Next, perform an inverse Fourier transform on the fused high-resolution wavenumber spectrum of the target or use other algorithms of synthetic aperture sonar to reconstruct a super-resolution target image close to the real object.

[0077] It should be understood that although Figure 1 the steps in the flowchart of Figure 1 are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless specifically stated in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover,

[0078] In one embodiment, a super-resolution underwater acoustic imaging device based on optical diffraction tomography is provided, including:

[0079] An initialization module, configured to initialize the high-resolution image of an underwater target and obtain a random complex initial conjecture of the target;

[0080] A conjecture propagation module, configured to propagate the random complex initial conjecture to the wavenumber domain to obtain a low-resolution wavenumber spectrum of the random complex initial conjecture; wherein, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped;

[0081] An overlapping scanning and sampling module, configured to perform overlapping scanning and sampling around the target by rotating the imaging unit at a set angular increment based on the optical diffraction tomography algorithm, collect the backscattered acoustic signals of each angular region of the target and propagate them to the spatial domain to obtain the reconstructed sub-images of each angular region of the target;

[0082] An intensity constraint module, configured to, after the imaging unit rotates once at the set angular increment and samples the backscattered acoustic signals of the corresponding angular region of the target, transmit the low-resolution wavenumber spectrum of this angular region back to the spatial domain to obtain the low-resolution target sub-image of this angular region, perform intensity constraint on the low-resolution target sub-image of this angular region according to the reconstructed sub-image of the same angular region of the target, and transmit the intensity-constrained image to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region;

[0083] The super-resolution imaging module is used to repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of all angular regions of the target, fuse and splice the high-resolution wavenumber spectra of all angular regions of the target, and perform an inverse transform to reconstruct the super-resolution target image.

[0084] For the specific limitations of the super-resolution underwater acoustic imaging device for optical diffraction tomography-like, reference can be made to the limitations of the super-resolution underwater acoustic imaging method for optical diffraction tomography-like in the above text, which will not be elaborated here. Each module in the above super-resolution underwater acoustic imaging device for optical diffraction tomography-like can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to each of the above modules.

[0085] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 4 shown. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a super-resolution underwater acoustic imaging method for optical diffraction tomography-like. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0086] Those skilled in the art can understand that Figure 4 the structure shown in

[0087] is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0088] Initialize the high-resolution image of the underwater target and obtain a random complex initial guess of the target;

[0089] Propagate the random complex initial guess to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess; among them, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped;

[0090] Based on the optical diffraction tomography-like algorithm, use the imaging unit to rotate at a set angular increment and surround the target for overlapping scanning sampling, collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain to obtain the reconstructed sub-images in each angular region of the target;

[0091] After the imaging unit rotates once at a set angular increment and samples the backscattered acoustic signals in the corresponding angular region of the target, transmit the low-resolution wavenumber spectrum corresponding to this angular region back to the spatial domain to obtain the low-resolution target sub-image in this angular region, and perform intensity constraint on the low-resolution target sub-image in the same angular region of the target according to the reconstructed sub-image in this angular region, and transmit the intensity-constrained image to the wavenumber domain to obtain the high-resolution wavenumber spectrum in this angular region;

[0092] Repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra in each angular region of the target, fuse and splice the high-resolution wavenumber spectra in each angular region of the target and perform inverse transformation to reconstruct the super-resolution target image.

[0093] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0094] Initialize the high-resolution image of the underwater target and obtain a random complex initial guess of the target;

[0095] Propagate the random complex initial guess to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess; among them, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped;

[0096] Based on the optical diffraction tomography-like algorithm, use the imaging unit to rotate at a set angular increment and surround the target for overlapping scanning sampling, collect the backscattered acoustic signals in each angular region of the target and propagate them to the spatial domain to obtain the reconstructed sub-images in each angular region of the target;

[0097] After the imaging unit rotates once at a set angular increment and samples to obtain the backscattering signal of the target in the corresponding angular region, the low-resolution wavenumber spectrum corresponding to this angular region is transmitted back to the spatial domain to obtain the low-resolution target sub-image of this angular region. Then, the low-resolution target sub-image of this angular region is intensity-constrained based on the reconstructed sub-image of the same angular region of the target, and the intensity-constrained image is propagated to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region;

[0098] Repeat the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of all angular regions of the target. Then, fuse and splice the high-resolution wavenumber spectra of all angular regions of the target and perform an inverse transform to reconstruct the super-resolution target image.

[0099] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in this application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0100] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0101] The above-described embodiments merely represent several implementation manners of the present application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A super-resolution underwater acoustic imaging method similar to optical diffraction tomography, characterized in that, The method includes: Initializing a high-resolution image of an underwater target and obtaining a random complex initial guess of the target; Propagating the random complex initial guess to the wavenumber domain to obtain a low-resolution wavenumber spectrum of the random complex initial guess; wherein, the distribution of the complete low-resolution wavenumber spectrum is annular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped; Based on an optical diffraction tomography-like algorithm, using an imaging unit to rotate at a set angular increment and scan the target overlappedly in a circular motion, collecting the backscattered acoustic signals in each angular region of the target and propagating them to the spatial domain to obtain reconstructed sub-images of each angular region of the target; After the imaging unit rotates once at the set angular increment and samples the backscattered acoustic signals in the corresponding angular region of the target, transmitting the low-resolution wavenumber spectrum corresponding to this angular region back to the spatial domain to obtain a low-resolution target sub-image of this angular region, and performing intensity constraint on the low-resolution target sub-image of this angular region according to the reconstructed sub-image of the same angular region of the target, and transmitting the intensity-constrained image to the wavenumber domain to obtain a high-resolution wavenumber spectrum of this angular region; Repeating the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target, fusing and stitching the high-resolution wavenumber spectra of each angular region of the target and performing inverse transformation to reconstruct a super-resolution target image.

2. The method according to claim 1, characterized in that Initialize a high-resolution image of an underwater target and obtain a random complex initial guess of the target, denoted as where the subscript hr represents high resolution, i represents the imaginary unit, I hr is the target intensity, is the target phase, and I hr and take random values.

3. The method according to claim 2, characterized in that, Propagating the random complex initial guess to the wavenumber domain to obtain a low-resolution wavenumber spectrum of the random complex initial guess, including: Random complex initial guess of the target Propagate it to the wavenumber domain to obtain the low-resolution wavenumber spectrum of the random complex initial guess, denoted as where I lr (k x , k y ) is the intensity of the low-resolution wavenumber spectrum, and the subscript lr indicates low resolution; (x, y) represents the airspace where the target is located, and (k x , k y ) represents the wavenumber domain where the wavenumber spectrum is located, and F represents the Fourier transform.

4. The method according to claim 3, characterized in that, Based on an optical diffraction tomography-like algorithm, using an imaging unit to rotate at a set angular increment and scan the target overlappedly in a circular motion, collecting the backscattered acoustic signals in each angular region of the target and propagating them to the spatial domain to obtain reconstructed sub-images of each angular region of the target, including: Based on an optical diffraction tomography-like algorithm, using a small hydrophone linear array or planar array as the imaging unit, and using the imaging unit to rotate at a set angular increment Δθ and scan the target overlappedly in a circular motion, collecting the backscattered acoustic signals in each angular region of the target and propagating them to the spatial domain to obtain reconstructed sub-images of each angular region of the target; wherein, the overlapped scanning sampling means that the angular region scanned by the imaging unit each time has partial overlap with the angular region scanned in the previous time.

5. The method according to claim 4, characterized in that, After the imaging unit rotates once at the set angular increment and samples the backscattered acoustic signals in the corresponding angular region of the target, transmitting the low-resolution wavenumber spectrum corresponding to this angular region back to the spatial domain to obtain a low-resolution target sub-image of this angular region, including: After the imaging unit rotates once by a set angular increment Δθ and samples to obtain the backscattering signal of the target in the corresponding angular region, by performing an inverse Fourier transform on the low-resolution wavenumber spectrum corresponding to this angular region the low-resolution wavenumber spectrum corresponding to this angular region is back-projected to the spatial domain to obtain a low-resolution target sub-image of this angular region, denoted as where I l is the intensity of the low-resolution target sub-image, is the phase of the low-resolution target sub-image, F -1 represents the inverse Fourier transform, I lr (k x , k y ) is the intensity of the low-resolution wavenumber spectrum, the subscript lr represents low resolution, (k x , k y ) represents the wavenumber domain where the wavenumber spectrum is located, and i represents the imaginary unit.

6. The method according to claim 5, characterized in that, Performing intensity constraint on the low-resolution target sub-image of this angular region according to the reconstructed sub-image of the same angular region of the target, and transmitting the intensity-constrained image to the wavenumber domain to obtain a high-resolution wavenumber spectrum of this angular region, including: Perform intensity constraint on the low-resolution target sub-image of the angle region according to the reconstructed sub-image of the same angle region of the target. When performing the constraint, use the amplitude modulus measurement value of the reconstructed sub-image of the same angle region of the target Replace the intensity component in the low-resolution target sub-image of the angle region, and retain the phase component in the low-resolution target sub-image of the angle region to obtain the image after intensity constraint, denoted as Transmitting the intensity-constrained image to the wavenumber domain to obtain a high-resolution wavenumber spectrum of this angular region, expressed as where I hr (k x , k y ) is the intensity of the high-resolution wavenumber spectrum, is the phase of the high-resolution wavenumber spectrum, the subscript hr represents high-resolution, i represents the imaginary unit, and F represents the Fourier transform.

7. The method according to claim 6, wherein Fusing and stitching the high-resolution wavenumber spectra of each angular region of the target and performing inverse transformation to reconstruct a super-resolution target image, including: Fuse and splice the high-resolution wavenumber spectra of each angular region of the target, and perform an inverse Fourier transform on the fused high-resolution wavenumber spectrum of the target to reconstruct a super-resolution target image.

8. A super-resolution underwater acoustic imaging device for optical diffraction tomography-like, characterized in that, The device includes: An initialization module for initializing the high-resolution image of the underwater target and obtaining a random complex initial guess of the target; A guess propagation module for propagating the random complex initial guess to the wavenumber domain to obtain a low-resolution wavenumber spectrum of the random complex initial guess; wherein, the distribution of the complete low-resolution wavenumber spectrum is circular, and the distribution of the low-resolution wavenumber spectrum in a certain angular region is fan-shaped; An overlapping scanning and sampling module for performing overlapping scanning and sampling around the target by rotating the imaging unit at a set angular increment based on an optical diffraction tomography-like algorithm, collecting the backscattered acoustic signals of each angular region of the target and propagating them to the spatial domain to obtain the reconstructed sub-images of each angular region of the target; An intensity constraint module for, after the imaging unit rotates once at a set angular increment and samples the backscattered acoustic signals of the corresponding angular region of the target, backpropagating the low-resolution wavenumber spectrum of this angular region to the spatial domain to obtain the low-resolution target sub-image of this angular region, performing intensity constraint on the low-resolution target sub-image of this angular region according to the reconstructed sub-image of the same angular region of the target, and propagating the intensity-constrained image to the wavenumber domain to obtain the high-resolution wavenumber spectrum of this angular region; A super-resolution imaging module for repeating the above steps until the low-resolution wavenumber spectrum of the random complex initial guess is updated and replaced by the high-resolution wavenumber spectra of each angular region of the target, fusing and splicing the high-resolution wavenumber spectra of each angular region of the target and performing an inversion transform to reconstruct a super-resolution target image.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

Images