Method and system for detecting and eliminating induced heave error based on single strip sounding data

Abstract

The application provides a kind of based on single strip sounding data's induced heave error detection and elimination method and system, comprising: the processing of multibeam sounding data, through original data decoding obtains roll and pitch, simultaneously through sound ray tracking, homing calculation etc. method is calculated to obtain the depth of multibeam sounding point under the depth reference surface of specified depth;Through Fourier low-pass filter, two-way difference method, Multi-start algorithm extracts the abnormal stripe period along the track direction caused by induced heave error;By establishing model and using sliding RANSAC algorithm, the corrected water depth in each period is obtained;Using SVR regression algorithm obtains induced heave error, and corrects it to ship file to re-homing calculation realizes the elimination of abnormal stripe along the track direction.The application is aimed at the problem that abnormal stripe terrain in sounding result has obvious influence, cause is not clear and is difficult to eliminate, effectively eliminates the abnormal stripe along the track direction caused by induced heave error, and has robustness to complex seabed topography.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of geodesy and surveying engineering, and particularly relates to an induced heave error detection and elimination method and system based on single strip sounding data. BACKGROUND

[0002] The multi-beam echo sounding system (MBES) has the characteristics of full coverage, full water depth range, high precision, high resolution, etc., and has become a mainstream sounding device, and is widely used in underwater topographic survey, seafloor geomorphology survey, underwater target detection and other fields. As a comprehensive system composed of multiple sensors, the MBES sounding precision will be affected by sensor precision, sensor coupling effect and complex environment, resulting in sounding data always accompanied by various errors.

[0003] Due to the missing or incorrect measurement of sound velocity profile, ship parameter and other exogenous data, errors are often introduced in the process of calculating the beam footprint of multi-beam data, which will cause obvious systematic errors in the sounding point cloud data. Among these errors, few scholars have paid attention to the induced heave error caused by the installation deviation between the transducer and the swell sensor, which has a significant impact on the measurement of the seafloor topography with changing flatness, and has a serious impact on the quality of multi-beam sounding results in shallow water areas, but there is currently a lack of relatively perfect means for detection and elimination.

[0004] Some researchers use differential evolution and Gauss-Newton optimization methods to weaken the refraction error caused by incorrect sound velocity profile; some researchers propose a migration correction method to reduce the tidal component of migration, and use the instantaneous tidal correction (ITC) algorithm to reduce the tidal residual error caused by incomplete tidal measurement or inappropriate correction model; some researchers use multi-beam sonar detection method and GPS dynamic positioning to reduce the dynamic draft error caused by uneven ship motion; some researchers propose a multi-beam sonar system installation deviation overall calibration method based on terrain feature matching, which reduces the alignment error of the transducer and the corresponding calibration residual error; some researchers propose an accurate algorithm considering the non-concentric array geometry of multi-beam, to reduce the error of non-concentric array.

[0005] Unlike these errors, which usually have large amplitudes and are widely concerned, high frequency errors usually have small amplitudes, and in many cases meet the standards of International Hydrographic Organization (IHO) S44. Although the amplitudes are small, it can also significantly affect the quality of multi-beam sounding results under high resolution requirements or in shallow water areas. Researchers have identified and analyzed such high frequency errors caused by sensor fusion errors, and these errors are divided into three categories: 1) attitude time delay, scale error; 2) surface sound speed error; 3) relative position deviation of swell sensor and transducer in X and Y directions, i.e. induced heave error. There have been a lot of researches on 1) and 2) errors, but 3) error is less concerned, and there is a lack of relatively perfect means for detection and elimination. SUMMARY

[0006] The purpose of the present application is to solve the problem of abnormal stripe topography in the sounding results, which has a significant impact, unclear causes and is difficult to eliminate. A method for detecting and eliminating induced heave error based on single strip sounding data is proposed, which can effectively eliminate the abnormal stripes along the track caused by induced heave error.

[0007] The method for detecting and eliminating induced heave error based on single strip sounding data provided by the present application comprises the following steps,

[0008] Step 1, processing of multi-beam sounding data, obtaining roll and pitch through original data decoding, and calculating the depth of multi-beam sounding points under the specified depth reference surface;

[0009] Step 2, extraction of abnormal stripe period along the track caused by induced heave error;

[0010] Step 3, obtaining the corrected water depth in each period, the implementation includes the following sub-steps,

[0011] Step 3.1, determining the multi-beam sounding induced heave error correction model, the correction model takes into account the consistency principle of seabed topography trend, establishes the relationship between water depth and induced heave error, introduces the fitting trend term, and then randomly selects a part of points in each period extracted in step 2 to establish a regression equation to obtain the parameters in the correction model;

[0012] Step 3.2, testing all data in the current period of the sounding sequence with the obtained model, and classifying the data applicable to the model as local points;

[0013] Step 3.3, if the local points do not meet a certain number, go to step 3.1; if they meet a certain number, re-calculate the model parameters with all local points;

[0014] Step 3.4, re-evaluate the model with all local points;

[0015] Step 3.5: Repeat steps 3.1 to 3.4 a certain number of times, and select the model with the best evaluation in step 3.4 as the result for water depth correction;

[0016] Step 4, Detection and elimination of induced heave error, is achieved through the following sub-steps:

[0017] Step 4.1: Determine the correction model. The correction model takes the corrected water depth in each cycle as the model input and establishes the relationship between the water depth before and after correction and the induced surge error. Then, the SVR regression algorithm is used to solve the induced heave error.

[0018] Step 4.2: Correct the induced heave error to the ship file, recalculate the multibeam echo sounding, and eliminate abnormal stripes along the navigation track.

[0019] Furthermore, in step 1, the depth of the multibeam echo sounder point below the specified depth reference plane is calculated using ray tracking and repositioning calculation methods.

[0020] Furthermore, the specific implementation of step 2 includes the following sub-steps:

[0021] Step 2.1: Extract the depth sequence of the central beam region from the processed bathymetry data, and use Fourier low-pass filtering to remove high-frequency terrain details and noise, which facilitates the extraction of the period.

[0022] Step 2.2: Use the bidirectional difference algorithm to initially extract the maximum and minimum values ​​in the depth sounding sequence from the Fourier low-pass filtered data;

[0023] Step 2.3: Starting with the maximum and minimum values ​​extracted by the bidirectional difference algorithm, the Multi-start algorithm is used to determine the peaks and troughs of the abnormal stripes along the track direction.

[0024] Furthermore, in step 2.1, the frequency threshold selection for the Fourier low-pass filter is based on the spectrum of the sounding sequence and the roll and pitch time series.

[0025] Furthermore, in step 3.1, the multibeam bathymetry-induced heave error correction model is as follows:

[0026] Δh i =h i -(xsinP i -ysinR i cosP i )-(kDis+d h )

[0027]

[0028] Where, Δh iTo correct the water depth, h i is the initial depth of the i ping in the sounding data sequence, Dis is the along-track distance of the i sequence to the first sequence in the current area, ping represents a transmitting and receiving process of a multi-beam sonar, and represents a vertical along-track sounding point; (kDis+d h ) represents a fitting term of the seabed topography trend, P i , R i are the pitch and roll at the corresponding moment, x and y are the relative position deviations of the swell sensor and the transducer in the X and Y directions, that is, the induced heave error, k1 and k2 are the start and end sequence points of the period extracted in step 2.

[0029] Further, in step 3.2, the evaluation criteria suitable for the model are as follows:

[0030] |h i -(xsinP i -ysinR i cosP i )-(kDis+d h )|<ε

[0031] Wherein, ε is set to the water depth measurement accuracy of the current instrument device.

[0032] Further, in step 3.4, the method for evaluating the model is as follows:

[0033]

[0034] Further, the correction model determined in step 4.1 is:

[0035] h i -h i '=sinP i x-sinR i cosP i y

[0036] Wherein, h i is the original sounding or original average sounding of the i ping, and h i ' is the corrected water depth.

[0037] The present application also provides a system for detecting and eliminating induced heave error based on single strip sounding data, comprising the following modules:

[0038] A preprocessing module for processing multi-beam sounding data, obtaining roll and pitch through original data decoding, and calculating the depth of multi-beam sounding points under a specified depth reference surface;

[0039] A period extraction module for extracting the period of abnormal stripes along the track caused by induced heave error;

[0040] The corrected water depth acquisition module is configured to acquire the corrected water depth in each cycle, and the implementation manner comprises the following sub-steps,

[0041] Step 3.1, determining a multi-beam sounding induced heave error correction model, which takes into account the principle of seabed topography consistency, links the water depth with the induced heave error, and introduces a fitting trend item, then a part of points in the region are randomly selected to establish a regression equation to obtain the parameters;

[0042] Step 3.2, testing all data in the current cycle of the sounding sequence by using the obtained model, and classifying the data applicable to the model as local points;

[0043] Step 3.3, if the local points do not meet a certain number, turning to step 3.1, and if the local points meet a certain number, recalculating the model parameters by using all the local points;

[0044] Step 3.4, reevaluating the model by using all the local points;

[0045] Step 3.5, repeating steps 3.1 to 3.4 for a certain number of times, selecting the best model in step 3.4 as the result, and performing water depth correction;

[0046] The error elimination module is configured to detect and eliminate the induced heave error, and the implementation manner comprises the following sub-steps,

[0047] Step 4.1, determining a correction model, which takes the corrected water depth in each cycle as the model input, establishes a link between the water depth before and after correction and the induced heave error, and solves the induced heave error by using the SVR regression algorithm;

[0048] Step 4.2, correcting the induced heave error to the ship file, and re-performing the multi-beam sounding homing calculation to eliminate the abnormal stripes along the track direction.

[0049] Further, the specific implementation manner of the cycle extraction module comprises the following sub-steps,

[0050] Step 2.1, extracting the depth sequence of the central beam region in the processed sounding data, and performing high-frequency terrain detail and noise elimination by using the Fourier low-pass filtering to facilitate the extraction of the cycle;

[0051] Step 2.2, preliminarily extracting the maximum and minimum values in the sounding sequence by using the bidirectional difference algorithm on the data after the Fourier low-pass filtering;

[0052] Step 2.3, taking the maximum and minimum values extracted by the bidirectional difference algorithm as the starting points, and determining the wave peaks and troughs of the abnormal stripes along the track direction by using the Multi-start algorithm.

[0053] This invention proposes a technique for detecting and eliminating induced heave errors based on single-strip bathymetry data. Addressing the issues of significant topographical influence, unclear causes, and difficulty in eliminating anomalous stripes in bathymetry results, this method first determines a calibration window by extracting the period of the anomalous stripes. Then, it estimates the induced heave error using all calibrated bathymetry data within these windows. During this process, the sliding RANSAC algorithm is employed to avoid the influence of seabed targets and topographic features, while the SVR algorithm is used to suppress the influence of other noise, improving the estimation accuracy of model parameters. This method effectively eliminates anomalous stripes along the navigation track caused by induced heave errors. Furthermore, this method is robust to complex seabed topography and has fewer application limitations. Attached Figure Description

[0054] Figure 1 This is a flowchart of an embodiment of the present invention;

[0055] Figure 2 The images shown are seabed topographic maps before and after the induced heave error correction obtained in the final embodiment of the present invention. Detailed Implementation

[0056] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0057] This invention provides a method for eliminating induced heave error based on single-strip depth sounding data, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0058] Step 1: Processing of multibeam echo sounding data. Roll and pitch are obtained by decoding the raw data. At the same time, the depth of the multibeam echo sounding point under the specified depth reference plane is calculated by acoustic ray tracking and repositioning calculation methods.

[0059] In practice, the example is based on data collected by the SeaBat T50-P multibeam sonar produced by Reason Corporation. The data comes from a certain sea area. Existing multibeam data processing software is used to process the raw data, and finally the pitch and roll P and R of the depth sounding point are obtained, as well as the depth h below the tidal reference level at the corresponding time.

[0060] Step 2, extraction of abnormal fringe periods along the flight path caused by induced heave error, is achieved through the following sub-steps.

[0061] Step 2.1: Extract the depth sequence of the central beam region from the processed bathymetry data, and use Fourier low-pass filtering to remove high-frequency terrain details and noise, which facilitates the extraction of the period.

[0062] In implementation, the central beam area is set as the central 5 beams, and the value of 5 is to avoid the influence of random error in multi-beam sounding, and can be selected as 2-10, which has little effect on the final result. The depth sequence is obtained by averaging; the frequency threshold of Fourier low-pass filtering can be selected by referring to the spectrum of the sounding sequence and the roll and pitch time sequence, because the induced heave error is directly related to roll and pitch.

[0063] Step 2.2, using bidirectional difference algorithm to extract the maximum and minimum values in the sounding sequence after Fourier low-pass filtering;

[0064] The bidirectional difference algorithm is an existing algorithm, so it is not described in detail.

[0065] Step 2.3, using the maximum and minimum values extracted by the bidirectional difference algorithm as the starting point, using the Multi-start algorithm to determine the wave peaks and troughs of the abnormal stripes along the track direction.

[0066] In the example, the range of the Multi-start algorithm is set to 25 sequence values, which should be determined according to the period of the abnormal stripes along the track direction caused by the induced heave error, and 25 is approximately half a period.

[0067] Step 3, obtaining the corrected water depth in each period, the implementation includes the following sub-steps,

[0068] Step 3.1, determining the multi-beam sounding induced heave error correction model, randomly selecting a part of points in each period extracted in step 2 to establish a regression equation to obtain the parameters in the correction model;

[0069] In implementation, one-half of the sounding sequence points are randomly selected, and one-half can be adjusted up and down, which has little effect on the final result. Another multi-beam sounding induced heave error correction model is as follows:

[0070] Δh i = h i - (x sin P i - y sin R i cos P i ) - (kDis+d h )

[0071]

[0072] Where, Δh i is the corrected water depth, h i is the initial depth of the i ping (ping represents a multi-beam sonar transmission and reception process, which is represented as a vertical track direction sounding point) in the sounding data sequence, Dis is the distance along the track direction from the i sequence to the first sequence in the current area, (kDis+d h) represents the fitting term of the seafloor topography trend, and the linear fitting of each small area can accurately represent the overall topographic trend of the seafloor. i , R i are the pitch and roll at the corresponding time, x and y are the relative position deviations between the swell sensor and the transducer in the X and Y directions, i.e. the induced heave error, k1 and k2 are the start and end sequence points of the extracted period in step 2.

[0073] Step 3.2, test all data in the current period of the obtained model with the sounding sequence, and classify the data applicable to the model as local points;

[0074] In the example, the selected point area is selected as the current processing abnormal stripe period and the adjacent periods before and after, and the evaluation criteria applicable to the model are as follows:

[0075] |h i -(xsinP i -ysinR i cosP i )-(kDis+d h )|<ε

[0076] Wherein, ε is set as the water depth measurement accuracy of the current instrument device, and the other parameters have the same meaning as before.

[0077] Step 3.3, if the local points do not meet a certain number, go to step 3.1; if they meet a certain number, re-calculate the model parameters with all local points;

[0078] Step 3.4, re-evaluate the model with all local points;

[0079] In the example, the method for evaluating the model is as follows:

[0080]

[0081] Wherein, the parameters have the same meaning as before.

[0082] Step 3.5, repeat steps 3.1 to 3.4 for a certain number of times, select the best model evaluated in step 3.4 as the result, and perform water depth correction;

[0083] In the example, the water depth correction is as follows:

[0084] h i '=h i -(sinP i x'-sinR i cosP i y')

[0085] In the formula, h iwhere h is the original depth, h' is the corrected depth, x' and y' are the relative position offset errors of the transducer and surge sensor in the horizontal direction solved by the model, and other parameters have the same meaning as before.

[0086] Step 4, detection and elimination of induced heave error, the implementation includes the following sub-steps,

[0087] Step 4.1, determining a correction model, taking the corrected depth in each cycle as the model input, and solving the induced heave error by using the SVR regression algorithm;

[0088] In the embodiment, the SVR regression is an existing algorithm, and will not be described in detail, and the correction model is as follows:

[0089] h i -h i '=sinP i x-sinR i cosP i y

[0090] where h is the original depth, h' is the corrected depth, x' and y' are the relative position offset errors of the transducer and surge sensor in the horizontal direction solved by the model, and other parameters have the same meaning as before. i i h is the original depth, h' is the corrected depth, x' and y' are the relative position offset errors of the transducer and surge sensor in the horizontal direction solved by the model, and other parameters have the same meaning as before.

[0091] Step 4.2, correcting the induced heave error to the ship file, and re-performing the multi-beam depth homing calculation to eliminate the abnormal stripes along the track direction.

[0092] In the embodiment, the Caris multi-beam data post-processing software is used to re-perform the multi-beam depth homing calculation, eliminate the induced heave error, and obtain the seabed topographic map after the abnormal stripes along the track direction are eliminated.

[0093] In the specific implementation, the method provided by the technical scheme of the present application can be automatically run by a computer software technology, and the system device of the method, such as a computer readable storage medium storing the corresponding computer program of the technical scheme of the present application and a computer device including the running corresponding computer program, should also be within the protection scope of the present application.

[0094] The present application also provides a system for detecting and eliminating induced heave error based on single strip depth data, comprising the following modules:

[0095] A preprocessing module is used for processing multi-beam depth data, obtaining roll and pitch through original data decoding, and calculating the depth of the multi-beam depth point under a specified depth reference surface;

[0096] A cycle extraction module is used for extracting the cycle of abnormal stripes along the track direction caused by the induced heave error.​

[0097] a corrected water depth acquisition module for acquiring the corrected water depth in each cycle, the implementation mode comprising the following sub-steps,

[0098] Step 3.1, determining a multi-beam sounding induced heave error correction model, the model taking into account the principle of seabed topography trend consistency, establishing a relationship between the water depth and the induced heave error, introducing a fitting trend term, then randomly selecting a part of points in the region to establish a regression equation to obtain the parameters;

[0099] Step 3.2, testing all data in the current cycle of the sounding sequence with the obtained model, and classifying the data applicable to the model as local points;

[0100] Step 3.3, if the local points do not meet a certain number, go to step 3.1; if they meet a certain number, re-calculate the model parameters with all local points;

[0101] Step 3.4, re-evaluating the model with all local points;

[0102] Step 3.5, repeating steps 3.1 to 3.4 for a certain number of times, selecting the best evaluated model in step 3.4 as the result, and performing water depth correction;

[0103] an error elimination module for detecting and eliminating the induced heave error, the implementation mode comprising the following sub-steps,

[0104] Step 4.1, determining a correction model, the correction model taking the corrected water depth in each cycle as the model input, establishing a relationship between the corrected and uncorrected water depths and the induced heave error, and thus solving the induced heave error by using the SVR regression algorithm;

[0105] Step 4.2, correcting the induced heave error to the ship file and re-performing the multi-beam sounding homing calculation to eliminate the abnormal stripes along the track.

[0106] The specific implementation modes of each module correspond to each step, and the present application will not be described.

[0107] Although the preferred embodiments of the present application are described above in conjunction with the drawings, the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative and not limiting, and those of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims, which all belong to the protection scope of the present application.

Claims

1. A method for detecting and removing induced heave error based on single swath bathymetric data, characterized in that, The method comprises the following steps: Step 1, processing of multi-beam sounding data, roll and pitch are obtained by decoding original data, and the depth of the multi-beam sounding point under the specified depth reference surface is calculated; Step 2, extraction of abnormal stripe periods along the track direction caused by induced heave error; Step 3, obtaining the corrected water depth in each period, the implementation includes the following sub-steps, Step 3.1, determining a multi-beam sounding induced heave error correction model, the correction model considers the principle of consistency of seabed topography trend, establishes a relationship between water depth and induced heave error, and introduces a fitting trend term, then, a part of points in each period extracted in step 2 is randomly selected to establish a regression equation to obtain the parameters in the correction model; Step 3.2, testing all data in the current period of the sounding sequence by using the obtained model, and the data suitable for the model are classified as local points; Step 3.3, if the local points do not meet a certain number, go to step 3.1; if the local points meet a certain number, re-calculate the model parameters by using all the local points; Step 3.4, re-evaluate the model by using all the local points; Step 3.5, repeat steps 3.1 to 3.4 for a certain number of times, select the best model in step 3.4 as the result, and perform water depth correction; Step 4, detection and elimination of induced heave error, the implementation includes the following sub-steps, Step 4.1, determining a correction model, the correction model takes the corrected water depth in each period as the model input, establishes a relationship between the corrected and uncorrected water depths and the induced heave error, and thus solves the induced heave error by using an SVR regression algorithm; Step 4.2, correcting the induced heave error to the ship file, re-performing multi-beam sounding homing calculation, and eliminating the abnormal stripes along the track direction.

2. The method for detecting and eliminating induced heave error based on single swath bathymetric data according to claim 1, characterized in that: In step 1, the depth of the multi-beam sounding point under the specified depth reference surface is calculated by the sound ray tracking and homing calculation method.

3. The method for detecting and eliminating induced heave error based on single swath bathymetric data according to claim 1, characterized in that: The specific implementation of step 2 includes the following sub-steps, Step 2.1, extracting the depth sequence of the central beam region in the processed sounding data, performing high-frequency terrain detail and noise elimination by using Fourier low-pass filtering, and facilitating the extraction of the period; Step 2.2, using a bidirectional difference algorithm to preliminarily extract the maximum and minimum values in the sounding sequence after the Fourier low-pass filtering; Step 2.3, using a Multi-start algorithm to determine the wave peaks and troughs of the abnormal stripes along the track direction, taking the maximum and minimum values extracted by the bidirectional difference algorithm as the starting points.

4. The method of claim 3, wherein the method further comprises: In step 2.1, the frequency threshold of the Fourier low-pass filtering is selected by referring to the frequency spectrum of the reference sounding sequence and the roll and pitch time sequences.

5. The method of claim 1, wherein the method further comprises: In step 3.1, the multi-beam sounding induced heave error correction model is as follows: Δh i = h i - (x sin P i - y sin R i ) cos P i - (k Dis + d h ) where Δh i is the corrected water depth, h i is the initial depth of the i th ping in the sounding data sequence, Dis is the along-track distance from the i th sequence to the first sequence in the current region, ping represents a transmitting and receiving process of the multi-beam sonar, and shows a sounding point in the vertical track direction; (kDis+d h ) represents the fitting term of the seabed topography trend, P i , R i are the pitch and roll at the corresponding moment, x and y are the relative position deviations between the swell sensor and the transducer in the X and Y directions, i.e. the induced heave error, and k1 and k2 are the start and end sequence points of the extracted period in step 2.

6. The method of claim 5, wherein the method further comprises: In step 3.2, the evaluation criteria suitable for the model are as follows: h i - (x sin P i - y sin R i cos P i - (kDis + d h )| < ε Wherein, ε is set as the water depth measurement accuracy of the current instrument device.

7. The method of claim 5, wherein the method further comprises: In step 3.4, the method for evaluating the model is as follows:

8. The method of claim 5, wherein the method further comprises: The correction model determined in step 4.1 is as follows: h i -h i ' = sinP i x - sinR i cosP i y where h i is the i-th ping raw depth or raw average depth, h i is the corrected water depth.

9. An induced heave error detection and removal system based on single swath bathymetric data, characterized by, The method comprises the following modules: A preprocessing module for processing multi-beam sounding data, roll and pitch are obtained by decoding original data, and the depth of the multi-beam sounding point under the specified depth reference surface is calculated; Period extraction module, used for inducing heave error caused by abnormal stripe along track direction period extraction; Correction depth acquisition module, used for acquiring correction depth in each period, the implementation includes the following sub-steps, Step 3.1, determine the multi-beam sounding induced heave error correction model, the model takes into account the principle of seabed topography trend consistency, establishes the relationship between water depth and induced heave error, introduces the fitting trend item, then randomly selects a part of points in the region to establish a regression equation to obtain the parameters; Step 3.2, test all data in the current period of the sounding sequence with the obtained model, and classify the data suitable for the model as local points; Step 3.3, if the local points do not meet a certain number, go to step 3.1; if it meets a certain number, recalculate the model parameters with all local points; Step 3.4, reevaluate the model with all local points; Step 3.5, repeat steps 3.1 to 3.4 for a certain number of times, select the best model in step 3.4 as the result, and correct the water depth; Error elimination module, used for detecting and eliminating induced heave error, the implementation includes the following sub-steps, Step 4.1, determine the correction model, the correction model takes the corrected water depth in each period as the model input, establishes the relationship between the corrected and uncorrected water depth and the induced heave error, and solves the induced heave error by using SVR regression algorithm; Step 4.2, correct the induced heave error to the ship file, and re-calculate the multi-beam sounding homing, to eliminate the abnormal stripes along the track direction.

10. The system for detecting and removing induced heave error based on single swath bathymetry data of claim 9, wherein: The specific implementation of the period extraction module includes the following sub-steps, Step 2.1, extract the depth sequence of the central beam region in the processed sounding data, and use Fourier low-pass filtering to remove high-frequency topographic details and noise, to facilitate period extraction; Step 2.2, use bidirectional difference algorithm to preliminarily extract the maximum and minimum values in the sounding sequence after Fourier low-pass filtering; Step 2.3, use Multi-start algorithm to determine the wave peak and wave trough of the abnormal stripe along the track direction, taking the maximum and minimum values extracted by the bidirectional difference algorithm as the starting point.

Images