A bistatic seabed geomorphology acoustic imaging method based on pixel statistical distribution weighting

Abstract

This invention relates to a bistatic seabed acoustic imaging method based on pixel statistical distribution weighting. Utilizing the characteristic that the positions of real seabed features in the imaging image are fixed while spurious features vary with the position of the receiving array, a statistical method is used to obtain the pixel distribution characteristics of the acoustic image. The statistical distribution probability of each pixel is then used as the pixel weight to perform weighted stitching of multiple acoustic images. The method includes: bistatic seabed scattering signal modeling and simulation, received signal acoustic imaging processing, pixel statistical distribution processing, and weighted stitching and fusion of multiple acoustic images. Compared with conventional bistatic acoustic imaging methods, the method proposed in this invention can effectively reduce spurious features and random noise caused by the inherent characteristics of the one-dimensional horizontal linear array in the acoustic image, highlighting real features in the acoustic image and achieving imaging and discrimination of significant seabed features.

Description

Technical Field

[0001] This invention belongs to the field of sonar imaging, and specifically relates to a bistatic seabed topographic acoustic imaging method based on pixel statistical distribution weighting, which is applicable to seabed mapping, imaging and identification of seabed targets or topographic features (seamounts, etc.). Background Technology

[0002] Bistatic seabed topographic acoustic imaging can extend the range of sonar, fully utilize forward-scattered signals from the seabed, and achieve large-scale seabed acoustic imaging. References include (Ratilal P, Lai Y, Symonds T, et al. Long range acoustic imaging of the continental shelf environment: the Acoustic Clutter Reconnaissance Experiment 2001. [J]. Journal of the Acoustical Society of America, 2005, 117(4Pt1): 1977-1998.), (Makris NC, Chia CS, Fialkowski L T. The bi-azimuthal scattering distribution of an abyssal hill [J]. The Journal of the Acoustical Society of America, 1999, 106(5): 2491-2512.), (Swee Chia C, Makris NC, Fialkowski L TA. Comparison of bistatic scattering from two geologically distinct abyssal hills [J]. The Journal of the Acoustical Society of America, 1999, 106(5): 2491-2512.), and (Swee Chia C, Makris NC, Fialkowski L TA. Comparison of bistatic scattering from two geologically distinct abyssal hills [J]. The Journal of the Acoustical Society of America, 1999, 106(5): 2491-2512.). America, 2000, 108(5):2053-2070.) proposed a bistatic seabed topographic acoustic imaging system, using two ships to carry a vertical transmitting array and a horizontal towed array, respectively. The transmitting ship is fixed in the test area, while the receiving ship travels along different survey lines and receives forward-scattered signals from the seabed to obtain an image of the seabed topographic acoustic intensity distribution. However, in conventional bistatic acoustic imaging, the one-dimensional spatial characteristics of the receiving towed array cause left-right blurring, resulting in false features that are symmetrical about the receiving array axis in the seabed topographic image. With limited prior environmental information, it is difficult to distinguish the true scattering features of the seabed, affecting the identification of seabed targets or topographic features in the acoustic image.

[0003] Currently, the main methods to solve this left-right ambiguity phenomenon are: 1) Using dual-horizontal towed arrays and multi-horizontal towed arrays at the receiving end, and using the time delay difference of the signal arriving at each receiving array to distinguish them. This method has high requirements for equipment performance and is relatively complex to implement in terms of engineering technology; 2) Using the maneuvering of the receiving ship and judging based on the changing trend of the incident angle of the forward-scattered signal from the seabed. However, this method requires waiting for the receiving ship to complete the maneuver and for the receiving towed array to be straightened before making a judgment, which has limitations and lag; 3) Estimating the array shape of the receiving towed array and then performing back-end processing. This method has high requirements for the array shape estimation method and is difficult to implement. Summary of the Invention

[0004] Technical problems to be solved

[0005] To address the issue of left-right blurring caused by the one-dimensional characteristics of the receiving towed array in existing bistatic seabed acoustic imaging methods, which leads to false features in the seabed acoustic imaging image, this invention provides a bistatic seabed acoustic imaging method based on pixel statistical distribution weighting.

[0006] Technical solution

[0007] A bistatic seabed acoustic imaging method based on pixel statistical distribution weighting, characterized by the following steps:

[0008] Step 1: Modeling the bistatic seabed forward scattering signal;

[0009] Assume the transmitting array is located at the center of the observation area, and the receiving array is an N-element horizontal towed array with an element spacing of d. The transmitted signal is a linear frequency modulated signal with a center frequency of f0, a signal bandwidth of B, and a pulse width of T; p scattering coefficients σ are set on the seabed. p Point targets, simulating forward scattering from the seabed;

[0010] The transmitted signal is represented as

[0011]

[0012] Where t represents time, T represents the signal pulse width, and B represents the signal pulse width;

[0013] When the receiving array is located at a certain survey point, the received echo on the nth (n=1,2,...,N) element is:

[0014]

[0015] Where n(t) represents the noise at the receiving array element, σ p τ represents the scattering coefficient at the p-th scattering point. p This represents the time delay from the transmitting array to the p-th scattering point and then to the reference element (element 1) of the receiving array. The phase delay on the nth receiving element is expressed as...

[0016]

[0017] Where, θ p This represents the incident angle of the signal at the p-th scattering point;

[0018] Step 2: After receiving the forward-scattered signal from the seabed acquired by the towed array, the received array element signal is processed using conventional beamforming and matched filtering methods to obtain the time-domain output signal of each beam angle, i.e., the beam angle-time image; the beam angle-time image is then processed using conventional time-domain bistatic acoustic imaging methods to generate a sound intensity image of the seabed topography.

[0019] Sub-step 1: For x in the previous step n (t) performs beamforming, and its beam output can be expressed as

[0020]

[0021] in, The weighting coefficients for the nth element;

[0022] Sub-step 2: Perform matched filtering on the beam output from sub-step 1 and take the absolute value to obtain the beam angle-time image.

[0023]

[0024] Among them, R s (t) is the autocorrelation function of the transmitted signal. Represents convolution;

[0025] Sub-step 3: Using a conventional time-domain bistatic acoustic imaging method, perform acoustic imaging processing on the beam angle-time image, and calculate the imaging return point transmission delay τ from the transmitter array position, receiver array position, and imaging return point coordinates (x, y). (x,y) and incident angle θ (x,y) The amplitude A(x,y) at the imaging repositioning point coordinates (x,y) can be expressed as:

[0026] A(x,y)=|y beam -MF(θ (x,y) ,τ (x,y) (6)

[0027] Step 3: Perform pixel statistical analysis on the current frame acoustic image. Use sliding window local peak detection to process the acoustic intensity image, and simultaneously generate a hotspot statistical image with the same dimension as the acoustic image matrix. The hotspot statistical image consists of values ​​0 and 1, with values ​​above the local threshold set to 1 and values ​​below the local threshold set to 0. Perform element-wise multiplication between the original acoustic image and the hotspot statistical image to obtain the new image of the current frame.

[0028] B i (x,y)=A i (x,y).*I i (x,y) (7)

[0029] Where (x,y) are the pixel coordinates of the acoustic image, B i (x,y) represents the new image obtained by dot product operation in the i-th frame, A i (x,y) represents the original sound image of the i-th frame, I i (x,y) represents the hotspot statistics image corresponding to the i-th frame of the acoustic image;

[0030] Step 4: Use a weighted method to stitch together the acoustic images of each measuring point to finally generate an acoustic image of the seabed topography of the observation area;

[0031] The process of stitching together the acoustic images at each measurement point is as follows. Let F be the total image after stitching together the first (i-1) frames of acoustic images. i-1 (x, y), the new image obtained by multiplying the i-th frame is B. i (x,y), then the total stitching result F of the i-frame acoustic image i (x,y) is represented as

[0032] F i (x,y)=w1(x,y)F i-1 (x,y)+w2(x,y)B i (x,y) (8)

[0033] Where w1(x,y) and w2(x,y) are the images F i-1 (x,y), B i The weighting coefficients for pixels (x, y) are calculated using equation (9).

[0034]

[0035] A computer system is characterized by comprising: one or more processors, and a computer-readable storage medium for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method described above.

[0036] A computer-readable storage medium is characterized by storing computer-executable instructions, which, when executed, are used to implement the above-described method.

[0037] Beneficial effects

[0038] This invention utilizes the characteristic that the positions of real seabed features in a geomorphic imaging image are fixed, while false features change with the position of the receiving towed array. It proposes a bistatic seabed geomorphic acoustic imaging method based on pixel statistical distribution weighting. The proposed method uses sliding window peak detection to process seabed geomorphic acoustic images generated by conventional bistatic acoustic imaging methods, obtaining the pixel statistical distribution features in different frames of acoustic images. The pixel statistical distribution probability is introduced as the pixel weight of the acoustic image, and multiple frames of seabed geomorphic acoustic images are weighted and stitched together. This effectively suppresses false features in the seabed geomorphic acoustic images, enhances real features, achieves imaging and discrimination of significant seabed features, and improves the overall seabed geomorphic acoustic imaging effect. Attached Figure Description

[0039] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0040] Figure 1 This is a schematic diagram of the bistatic configuration of the method of the present invention.

[0041] Figure 2 This is a flowchart illustrating the method of the present invention.

[0042] Figure 3 This is a schematic diagram showing the distribution of the transmitting array, receiving array, and preset seabed target in an implementation example of the method of the present invention.

[0043] Figure 4 a, Figure 4 b represents a single-frame acoustic image obtained from different receiving locations using conventional imaging methods.

[0044] Figure 5 a, Figure 5 b is the single-frame acoustic imaging image obtained by pixel statistical processing in step three.

[0045] Figure 6 This is the stitched result of acoustic imaging images obtained through conventional acoustic imaging methods. Figure 6 (a) is a composite image of the acoustic imaging of the observation area; Figure 6 (b) is a magnified view of a local area of ​​the true feature region; Figure 6 (c) is a magnified view of the false feature region.

[0046] Figure 7 This is a probability map of pixel hotspot statistical distribution obtained by the method of the present invention.

[0047] Figure 8 This is the stitched result of the acoustic imaging image obtained by the method proposed in this invention. Figure 8 (a) is a composite image of the acoustic imaging of the observation area; Figure 8 (b) is a magnified view of the actual feature area.

[0048] Figure 9 The image stitching result is the result of processing the experimental data using conventional acoustic imaging methods.

[0049] Figure 10 This is a probability map showing the statistical distribution of pixel hotspots obtained by processing experimental data using the method of this invention.

[0050] Figure 11 The image stitching result is obtained after processing the experimental data using the method proposed in this invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention 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 merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0052] This invention provides a bistatic seabed acoustic imaging method based on pixel statistical distribution weighting. First, a bistatic seabed scattering signal model is constructed. Two ships, one carrying a vertical transmitting array and the other a horizontal towed array, are used. The transmitting ship is fixed at the center of the observation area and transmits the signal, while the receiving ship moves and receives the forward-scattered signal containing seabed feature information. Next, beamforming and pulse compression are used to obtain the beam angle-time image of the sonar imaging. Conventional bistatic acoustic imaging methods are then used to obtain the corresponding seabed acoustic image of the observation area. Taking advantage of the fixed location of real seabed features and the variation of blurred features with the position of the receiving towed array in the terrain imaging image, a sliding window peak detection process is used to process the current acoustic image. Statistical methods are used to obtain the pixel distribution characteristics of the acoustic image, and corresponding hotspot statistical images (represented by values ​​0 and 1) are generated simultaneously. Finally, the pixel statistical distribution probability is used as the image pixel weight to perform weighted stitching of multiple frames of seabed acoustic images, effectively enhancing the real features in the imaging image while suppressing false features and random noise.

[0053] like Figure 1 , 2 As shown, a bistatic seabed acoustic imaging method based on pixel statistical distribution weighting includes the following steps:

[0054] Step 1: Modeling the bistatic seabed forward scattering signal;

[0055] like Figure 3 As shown, the launching vessel is fixed in the center of the observation area at a depth of 60m. The receiving array is an N-element horizontal towed linear array with an element spacing of d. The transmitted signal is a linear frequency modulated signal with a center frequency of f0, a signal bandwidth of B, and a pulse width of T. p scattering coefficients of σ are set on the seabed. p Point targets are used to simulate forward scattering from the seabed.

[0056] The transmitted signal is represented as

[0057]

[0058] Figure 3 The diagram shows the distribution of the transmitting array, receiving array, and seabed target in an embodiment of the method of the present invention. ☆ represents the position of the transmitting array, △ represents the position of the horizontal receiving array, → represents the heading of the receiving array, and black □ represents a preset point target.

[0059] In this implementation example, the transmitted signal uses a linear frequency modulated (LFM) signal with a center frequency of 1800 Hz, a bandwidth of 200 Hz, and a pulse width of 1 s. The observation area is set to 20 km × 20 km, with a sea depth of 1100 m. The transmitting array coordinates are (0 km, 0 km, 60 m), and the receiving array's initial motion coordinates are (-6 km, -3 km, 60 m), with a heading angle of 45°. Nine point targets are set on the seabed surface, with coordinates of (-4 km, 4 km, 1000 m), (-4 km, 5 km, 1000 m), (-4 km, 6 km, 1000 m), (-5 km, 4 km, 1000 m), (-5 km, 5 km, 1000 m), (-5 km, 6 km, 1000 m), (-6 km, 4 km, 1000 m), (-6 km, 5 km, 1000 m), and (-6 km, 6 km, 1000 m). The point target scattering coefficient σ is... p All values ​​are set to 1. Assuming there is no channel distortion, the received echo on the nth (n = 1, 2, ..., N) element of the receiving array can be expressed as:

[0060]

[0061] Where n(t) represents the noise at the receiving array element, σ p τ represents the scattering coefficient at the p-th scattering point. p This represents the time delay from the transmitting array to the p-th scattering point and then to the reference element (element 1) of the receiving array. The phase delay on the nth receiving element is expressed as...

[0062]

[0063] Where, θ pLet represent the incident angle of the signal at the p-th scattering point. The receiving array element signal in this example can be obtained from equation (3).

[0064] Step 2: Process the received array element signals using conventional beamforming and matched filtering methods to obtain the time-domain output signals of each beam angle, i.e., the beam angle-time image; process the beam angle-time image using conventional time-domain bistatic acoustic imaging methods to generate a sound intensity image of the seabed topography.

[0065] Sub-step 1: For x in the previous step n (t) performs beamforming, and its beam output can be expressed as

[0066]

[0067] in, is the weighting coefficient of the nth element.

[0068] Sub-step 2: Perform matched filtering on the beam output from sub-step 1 and take the absolute value to obtain the beam angle-time image.

[0069]

[0070] Among them, R s (t) is the autocorrelation function of the transmitted signal. This represents convolution.

[0071] Sub-step 3: Using a conventional time-domain bistatic acoustic imaging method, perform acoustic imaging processing on the beam angle-time image, and calculate the imaging return point transmission delay τ from the transmitter array position, receiver array position, and imaging return point coordinates (x, y). (x,y) and incident angle θ (x,y) The amplitude A(x,y) at the imaging repositioning point coordinates (x,y) can be expressed as:

[0072] A(x,y)=|y beam-MF (θ (x,y) ,τ (x,y) | (6)

[0073] According to step two, the conventional imaging results of the receiving array at different locations are as follows: Figure 4 a and Figure 4 As shown in b. Figure 4 The imaging data shows that within the x-axis range of -4km to -6km and the y-axis range of -4km to -6km, the number of bright spots corresponds to the number of preset target points, and their locations are also corresponding. Meanwhile, Figure 4 a and Figure 4 b. In the acoustic imaging, there are false features that are symmetrical about the receiving array axis, and the location of the false features changes as the position of the receiving array moves.

[0074] Step 3: Perform pixel statistical analysis on the current frame's acoustic image. Use a sliding window local peak detection process to process the acoustic intensity image, simultaneously generating a hotspot statistical image (composed of values ​​0 and 1, with values ​​above a local threshold set to 1 and values ​​below a local threshold set to 0) that matches the dimensions of the acoustic image matrix. Perform element-wise multiplication between the original acoustic image and the hotspot statistical image to obtain the new image for the current frame.

[0075] B i (x,y)=A i (x,y).*I i (x,y) (7)

[0076] Where (x,y) are the pixel coordinates of the acoustic image, B i (x,y) represents the new image obtained by dot product operation in the i-th frame, A i (x,y) represents the original sound image of the i-th frame, I i (x,y) represents the hotspot statistics image corresponding to the i-th frame of the acoustic image.

[0077] According to step three, the acoustic image obtained after pixel statistical processing proposed in this invention is as follows: Figure 5 a, Figure 5 As shown in b. (And) Figure 4 similar, Figure 5 The number of bright spots appearing in the image is consistent with the number of preset target points, their positions correspond, the real features are more obvious, and false features are effectively suppressed, resulting in a clean background.

[0078] Step 4: Repeat steps 1-3 to obtain the acoustic imaging results and hotspot statistics of the receiving array at different locations. Use a weighted method to stitch the acoustic images of each measuring point to finally generate an acoustic image of the seabed topography in the observation area.

[0079] The process of stitching together the acoustic images at each measurement point is as follows. Let F be the total image after stitching together the first (i-1) frames of acoustic images. i-1 (x, y), the new image obtained by multiplying the i-th frame is B. i (x,y), then the total stitching result F of the i-frame acoustic image i (x,y) is represented as

[0080] F i (x,y)=w1(x,y)F i-1 (x,y)+w2(x,y)B i (x,y) (8)

[0081] Where w1(x,y) and w2(x,y) are the images F i-1 (x,y), B i The weighting coefficients for pixels (x, y) are calculated using equation (9).

[0082]

[0083] According to step four, Figure 6 The image shown is the stitched result of acoustic imaging obtained by conventional acoustic imaging methods. Figure 7 This is a probability map of pixel hotspot statistical distribution obtained by the method of the present invention. Figure 8 This is the stitched result of the acoustic imaging image obtained by the method proposed in this invention. From... Figure 6 (a)~ Figure 6 As shown in (c), in the image stitching results processed by conventional acoustic imaging methods, a high peak value appears at the preset target point, the side lobes of the true feature imaging results are high, and obvious false features appear. Figure 7 As shown in the pixel hotspot statistical distribution probability map, at the preset target location, the pixel hotspot statistical distribution probability reaches 100%, while the probability decreases to approximately 40% due to the movement of the false feature receiver array. Statistical analysis is performed on the pixels of each frame of the acoustic image, and this hotspot statistical distribution probability is used as the acoustic image weighting coefficient. The final acoustic image stitching result is as follows: Figure 8 As shown, Figure 8 In (a) and (b), the imaging results at the preset target point are clear, and false features are effectively suppressed, while the background of the imaging image is relatively clear. Figure 6 (a) Pure.

[0084] To further verify the effectiveness of the method proposed in this invention, data processing results from a sea trial conducted in June 2021 are used for illustration. This experiment employed... Figure 1 The transmitting and receiving arrays shown are as follows: the transmitting array is a 10-element vertical array with an element spacing of 0.4m, and the transmitted signal is a linear frequency modulated signal with a center frequency of 1800Hz, a bandwidth of 200Hz, and a pulse width of 1s; the receiving array is a 96-element horizontal linear array with an element spacing of 0.416m and a sampling frequency of 16kHz.

[0085] This experiment used 40 consecutive frames of data from a certain test line segment for processing. Figure 2 The process and steps two through four are shown below. The final image result obtained after processing is as follows: Figures 9-11 As shown. This is a comparison of features in the acoustic imaging of the terrain. Figures 9-11 Contour map of the seabed topography of the test area is superimposed. Figure 9 The image stitching result is the result of processing the experimental data using conventional acoustic imaging methods. Figure 10 This is a probability map showing the statistical distribution of pixel hotspots obtained by processing experimental data using the method of this invention. Figure 11 The image stitching result is obtained after processing the experimental data using the method proposed in this invention.

[0086] contrast Figure 9 and Figure 10It can be seen that, Figure 9 Within the x-axis range of -5.5km to 7km and the y-axis range of 4.6km to 8.6km, a significant area of ​​sound intensity corresponding to the seabed topography appears, and its sound intensity distribution range corresponds to the topographic trend of the test area. Simultaneously, strong false features appear within the x-axis range of -4km to 5.5km and the y-axis range of -6.5km to -3.6km. Figure 11 The high sound intensity regions corresponding to the terrain of the test area are clear and distinct, and the sound intensity distribution range corresponds to the terrain trend of the test area. Furthermore, false features are effectively suppressed, and the overall imaging background is clear and clean. Figure 10 The pixel hotspot statistical distribution probability shows that the pixel hotspot statistical distribution probability corresponding to the terrain of the test area reaches 90%, and the distribution range is similar to... Figure 11 Correspondingly, the probability of pixel hotspot distribution within the false feature area is about 55%, and the distribution range is affected by the transmission and reception positions.

[0087] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the scope of the technology disclosed in the present invention, and such modifications or substitutions should all be covered within the scope of protection of the present invention.

Claims

1. A bistatic seabed acoustic imaging method based on pixel statistical distribution weighting, characterized in that... The steps are as follows: Step 1: Modeling the bistatic seabed forward scattering signal; Assume that the transmitting array is located at the center of the observation area, the receiving array is a N-element horizontal linear array with element spacing d, the transmitting signal is a linear frequency modulation signal with center frequency f0, signal bandwidth B and signal pulse width T, and p point targets with scattering coefficient σ p are set on the seabed to simulate forward scattering. The transmitted signal is represented as Where t represents time, T represents the signal pulse width, and B represents the signal pulse width; When the receiving array is located at a certain survey point, the received echo on the nth element is: Where n(t) represents the noise at the receiving array element, σ p τ represents the scattering coefficient at the p-th scattering point. p This represents the time delay from the transmitting array to the p-th scattering point and then to the reference element of the receiving array. The phase delay on the nth receiving element is expressed as... Where, θ p This represents the incident angle of the signal at the p-th scattering point; Step 2: After receiving the forward-scattered signal from the seabed acquired by the towed array, the received array element signal is processed using conventional beamforming and matched filtering methods to obtain the time-domain output signal of each beam angle, i.e., the beam angle-time image; the beam angle-time image is then processed using conventional time-domain bistatic acoustic imaging methods to generate a sound intensity image of the seabed topography. Sub-step 1: For x in the previous step n (t) performs beamforming, and its beam output can be expressed as in, The weighting coefficients for the nth element; Sub-step 2: Perform matched filtering on the beam output from sub-step 1 and take the absolute value to obtain the beam angle-time image. Among them, R s (t) is the autocorrelation function of the transmitted signal. Represents convolution; Sub-step 3: Using a conventional time-domain bistatic acoustic imaging method, perform acoustic imaging processing on the beam angle-time image, and calculate the imaging return point transmission delay τ from the transmitter array position, receiver array position, and imaging return point coordinates (x, y). (x,y) and incident angle θ (x,y) The amplitude A(x,y) at the imaging repositioning point coordinates (x,y) can be expressed as: A(x,y)=|y beam-MF (i (x,y) ,t (x,y) )| (6) Step 3: Perform pixel statistical analysis on the current frame acoustic image. Use sliding window local peak detection to process the acoustic intensity image, and simultaneously generate a hotspot statistical image with the same dimension as the acoustic image matrix. The hotspot statistical image consists of values ​​0 and 1, with values ​​above the local threshold set to 1 and values ​​below the local threshold set to 0. Perform element-wise multiplication between the original acoustic image and the hotspot statistical image to obtain the new image of the current frame. B i (x,y)=A i (x,y).*I i (x,y) (7) Where (x,y) are the pixel coordinates of the acoustic image, B i (x,y) represents the new image obtained by dot product operation in the i-th frame, A i (x,y) represents the original sound image of the i-th frame, I i (x,y) represents the hotspot statistics image corresponding to the i-th frame of the acoustic image; Step 4: Use a weighted method to stitch together the acoustic images of each measuring point to finally generate an acoustic image of the seabed topography of the observation area; The process of stitching together the acoustic images at each measurement point is as follows. Let F be the total image after stitching together the first (i-1) frames of acoustic images. i-1 (x, y), the new image obtained by multiplying the i-th frame is B. i (x,y), then the total stitching result F of the i-frame acoustic image i (x,y) is represented as F i (x,y)=w1(x,y)F i-1 (x,y)+w2(x,y)B i (x,y) (8) Where w1(x,y) and w2(x,y) are the images F i-1 (x,y), B i The weighting coefficients for pixels (x, y) are calculated using equation (9).

2. A computer system, characterized in that... include: One or more processors, a computer-readable storage medium for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method of claim 1.

3. A computer-readable storage medium, characterized in that... The device stores computer-executable instructions, which, when executed, are used to implement the method of claim 1.

Images