A method for identifying and associating marine vessels based on AIS and space-borne SAR data

Abstract

The patent belongs to the field of remote sensing data marine application processing, and particularly relates to a marine ship correlation identification method based on AIS and satellite-borne SAR data. The method mainly comprises the following steps: step one: acquiring AIS data based on a SAR satellite imaging time; step two: AIS data cleaning and processing; step three: ship track recovery in an AIS time interval; step four: ship state deduction in the AIS time interval; the deduction comprises time-distance vector or time-azimuth vector speed mapping relationship establishment; step five: ship real position revision based on a satellite-borne SAR imaging principle; step six: AIS and satellite-borne SAR data marine ship target accurate correlation processing; and step seven: satellite-borne SAR marine image system geometric error calculation. The method deduces ship real position, speed, heading and other state information at the satellite imaging time through data processing, mapping relationship establishment, distance speed decomposition and other modes, and can effectively quantify the degree of displacement of a moving target when imaged by a SAR.

Description

Technical Field

[0001] This patent belongs to the field of remote sensing data marine application processing, specifically a method for identifying and associating marine vessels based on AIS and spaceborne SAR data. Background Technology

[0002] Maritime surveillance is of great significance for safeguarding national maritime rights and interests and ensuring national maritime security. Among these, the surveillance of maritime vessels is of great value in applications such as maritime situation awareness, maritime rights protection, maritime traffic, fisheries management, and combating maritime terrorism. The fusion processing of maritime target information from spaceborne Synthetic Aperture Radar (SAR) and Automatic Identification System (AIS) solves the problem of accurate vessel target identification in spaceborne SAR images and fully realizes the complementarity of SAR and AIS information, improving the application capabilities of spaceborne SAR images in the automatic detection and identification of vessel targets. AIS is a maritime information communication platform provided by the International Lighthouse Association for vessels at sea. AIS equipment operates at frequencies of 161.975MHz and 162.025MHz and is carried by shore-based, shipborne, airborne, and spaceborne platforms. Currently, service platforms such as Shipxy.com can provide and publish AIS data from various sources globally in real time, allowing ground-based systems to collect AIS data within a region of interest and fuse it with other multi-source data.

[0003] There are two main existing fusion technologies. One is the space-based method, which involves mounting AIS receiving equipment and onboard processing equipment on a satellite platform. This allows for real-time information association and fusion in orbit, and has high application value. However, because AIS is a low-frequency signal, its receiving antenna is relatively large, which constrains the structural shape of the SAR satellite and increases the difficulty of onboard data stream design, as well as the satellite's weight and launch and development costs. Therefore, lightweight and miniaturized commercial SAR satellites typically use ground-based processing methods for information fusion to obtain information on maritime vessel targets. Currently, the commonly used method is mainly a location feature-based association method. This method matches vessel targets within the scene of the spaceborne SAR image with AIS-reported targets, and then associates them based on location features to complete the data fusion. However, due to the discontinuity of AIS data, the multi-source and complex nature of ground-acquired AIS data, the geometric system biases in SAR images, and the imaging position shift of moving vessel targets in SAR images, these errors reduce the accuracy of location feature-based association and can easily lead to erroneous association results. Summary of the Invention

[0004] The purpose of this invention is to propose a method for rapid association and identification of marine vessel targets through the fusion processing of AIS and spaceborne SAR data. The method aims to fully utilize AIS data to deduce the state information of different ships at the time of SAR imaging, compensate for the offset of moving targets on the imaging position by spaceborne SAR, calculate the system offset, and provide support data for improving the rapid matching and identification accuracy of AIS and spaceborne SAR vessel targets, as well as for evaluating and improving the positioning accuracy of spaceborne SAR image products.

[0005] To achieve this objective, the implementation of the present invention includes the following steps:

[0006] Step 1: Acquire AIS data based on SAR satellite imaging time;

[0007] Step 2: AIS data cleaning and processing;

[0008] Step 3: Restore ship trajectory within the AIS time interval;

[0009] Step 4: Ship status simulation within the AIS time interval (establishment of time-range (azimuth) velocity mapping relationship);

[0010] Step 5: Correction of the ship's true position based on the principle of spaceborne SAR imaging;

[0011] Step Six: Precise correlation processing of AIS and spaceborne SAR data with marine vessel targets;

[0012] Step 7: Calculation of geometric errors of the spaceborne SAR marine image system.

[0013] As a further technical solution of the present invention, in step one: based on the imaging latitude and longitude range, imaging time and the azimuth angle of the satellite at the imaging time of the SAR satellite, ship AIS data within the latitude and longitude range are filtered and downloaded.

[0014] As a further technical solution of the present invention, in step two: the data is preprocessed using the Vondrak data smoothing method based on time-latitude (longitude) data, and the correct AIS sequence is obtained after deleting erroneous AIS data; the cleaned data is vectorized into points, and the existing data is classified according to the relative position information and the existing attributes of the ship type, and the discontinuous trajectories under different ships are extracted.

[0015] In the Vondrak data smoothing method, for a given data sequence... ,in It is time. It refers to latitude and longitude; the basic assumption of the Vondrak data smoothing method is:

[0016] ;

[0017] in: In the formula: It is a smoothed value; express The third derivative; It is power; These are undetermined coefficients; This is called the approximation degree; This is called smoothness; It is called the smoothing coefficient, and the value of the smoothing coefficient determines the smoothness of the data curve.

[0018] As a further technical solution of the present invention, in step three: based on the cubic spline interpolation method, the AIS data is sorted according to time, the latitude and longitude of the starting point of time is selected as the origin of coordinates, and the time difference of the AIS data before and after the calculation is used as the interpolation sequence to obtain the cubic spline interpolation of the time-longitude trajectory and the time-latitude trajectory respectively. The ship's navigation trajectory is obtained by merging the above two cubic spline curves.

[0019] As a further technical solution of the present invention, in step four: the tangent equation of any point is established based on the navigation trajectory to obtain the heading at any time; based on the time-velocity information and heading information of different ships, combined with satellite imaging parameters, the velocity component decomposition and linear interpolation are used to establish the velocity mapping relationship in the time-range direction and the velocity-azimuth direction; and by merging the velocity values ​​in the above directions, the ship velocity values ​​corresponding to different times are obtained.

[0020] As a further technical solution of the present invention, in step five: combining the satellite azimuth at the time of satellite imaging, and the calculated actual ship position and course, the ship on the nearest SAR image along the distance direction is obtained as the target associated ship, and the azimuth of the ship is offset and compensated.

[0021] The azimuth offset of the ship target is represented as:

[0022] ;

[0023] in, This indicates the azimuth offset of the ship target. Indicates the closest slant range between the radar and the target. This indicates the range velocity of the ship target. Indicates the radar's angle of incidence. Indicates radar speed.

[0024] As a further technical solution of the present invention, in step six: the ship in the SAR scene at the imaging time is subjected to azimuth offset processing, and then the corrected SAR target position and the AIS inferred position are associated, and the SAR target position with the smallest distance along the range direction and the AIS inferred position are the same ship.

[0025] Based on the SAR satellite imaging plan, SAR imaging is performed at time t; based on the established time-latitude and longitude coordinates, the theoretical position of the ship at that imaging time is deduced by substituting the imaging time t value, as well as the tangent angle formed based on that position. The ship's heading at that time is further obtained by using the relative position of the fore and aft; the ship's speed at that imaging time is obtained by using the time-velocity mapping relationship.

[0026] As a further technical solution of the present invention, in step seven: combining the satellite azimuth at the time of satellite imaging, and the calculated actual ship position and heading, the distance difference is calculated with the corresponding ship position on the compensated SAR image, and the offset distance of the ship in the range direction at the time of SAR imaging is decomposed to obtain the offset distance of the ship in the range direction at the time of SAR imaging. This offset distance is used as the quantification value of the systematic error in the range direction when the SAR images the moving target.

[0027] Let the position of a certain ship, based on AIS simulation, be... The location of the same ship on the SAR image generated under the SAR satellite imaging plan is ;

[0028] The two sets of coordinate values ​​are transformed using Gaussian forward arithmetic. Calculate the distance ;

[0029] ;

[0030] Combined with heading angle and the satellite azimuth angle corresponding to the imaging time Calculate the degree of displacement in the distance direction ;

[0031] ;

[0032] Ultimately, the degree of systematic shift in the range and azimuth directions when SAR images moving targets is quantified. .

[0033] The present invention has the following beneficial effects:

[0034] This algorithm employs cubic spline interpolation to form a complete ship trajectory within the AIS time interval. It interpolates the velocity along the range and azimuth dimensions based on the satellite orbit, creating two sets of mapping relationships, which are then merged to form the speed at any given time. By performing tangency calculations on the position at any time along the trajectory, heading information at any given time can be effectively obtained. Through processing the AIS data, ship information from the SAR data at the imaging time can be extracted, such as the ship's latitude and longitude, heading, speed, and SAR range and azimuth speeds at that time. Based on the spaceborne SAR imaging parameters, the actual ship positions on the SAR image are revised. By comparing the results calculated from the AIS data with the actual ship imaging results in the SAR image after compensating for motion deviations, the systematic errors in the range and azimuth directions of the spaceborne SAR marine scene are calculated, providing support for the geometrical fine correction of subsequent spaceborne SAR image products.

[0035] In summary, the method of this invention integrates the AIS multi-dimensional attribute mapping method and the compensation processing method for position offset caused by spaceborne SAR motion, so as to realize the rapid and accurate association and identification of ship target data in spaceborne SAR marine images, obtain more accurate position and attribute information of ship targets, and improve the application capabilities of spaceborne SAR in marine applications. Attached Figure Description

[0036] Figure 1 is a flowchart of the method steps of the present invention;

[0037] Figure 2 is a schematic diagram of the relationship between heading and satellite azimuth in this invention;

[0038] Figure 3 shows the imaging results of the SAR mission plan in this invention.

[0039] Figure 4 is a schematic diagram of the appearance of the two ships to be simulated in SAR images;

[0040] Figure 5 shows a schematic diagram of the ship trajectory repaired by cubic spline interpolation within the AIS time interval; Figure (a) shows the complete trajectory of INTERASIA HORIZON; Figure (b) shows the complete trajectory of HUGE.

[0041] Figure 6 is a schematic diagram of the time-longitude (speed) mapping relationship of each ship; where Figure (c) represents the time-longitude (speed) mapping relationship of INTERASIA HORIZON; Figure (d) represents the time-longitude (speed) mapping relationship of INTERASIA HORIZON.

[0042] Figure 7 is a schematic diagram of the ship status projection results on the SAR image at the imaging time based on AIS data. In Figure (e), the projection position of INTERASIA HORIZON at the SAR imaging time is shown; and in Figure (f), the projection position of HUGE at the SAR imaging time is shown. Detailed Implementation

[0043] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. The present invention provides a method for rapid association and identification of marine vessel targets through the fusion of AIS data and spaceborne SAR image data. It integrates AIS multi-dimensional attribute mapping methods and spaceborne SAR motion-induced positional offset compensation methods to achieve rapid and accurate association and identification of ship target data in spaceborne SAR marine images, obtaining relatively precise position and attribute information of the ship targets and improving the capabilities of spaceborne SAR marine applications.

[0044] Please refer to Figure 1. The method for rapid association and identification of marine vessel targets through AIS and spaceborne SAR data fusion processing described in this invention mainly includes the following steps:

[0045] Step 1: Acquire AIS data based on SAR satellite imaging time: Based on the SAR satellite imaging plan, this mainly includes the imaging latitude and longitude range, imaging time, and the corresponding azimuth angle of the satellite at the imaging time. Filter and download ship AIS data collected by the U.S. Coast Guard within the latitude and longitude range;

[0046] Step 2: AIS Data Cleaning and Processing: AIS data often contains errors. Common errors include duplicate AIS data during transit, speeds exceeding the normal range, trajectory position deviations, and accelerations exceeding the normal range. Based on time-latitude / longitude data, the Vondrak data smoothing method is used for preprocessing. After removing erroneous AIS data, the correct AIS sequence is obtained.

[0047] In the Vondrak data smoothing method, for a given data sequence ,in It is time. It refers to latitude and longitude. The basic assumption of the Vondrak data smoothing method is:

[0048] ;

[0049] in: In the formula: It is a smoothed value; express The third derivative; It is power; These are undetermined coefficients; This is called the approximation degree; This is called smoothness. It is called the smoothing coefficient, and the value of the smoothing coefficient determines the smoothness of the data curve.

[0050] The cleaned data is vectorized into points, and the existing data is classified according to the relative position information and existing attributes such as ship type to extract discontinuous trajectories under different ships.

[0051] Step 3: Ship trajectory recovery within the AIS time interval: After data cleaning, a relatively accurate AIS data sequence is obtained. Based on cubic spline interpolation, the AIS data is sorted by time, and the latitude and longitude of the starting point of the time are selected as the coordinate origin. The time difference between the AIS data before and after the calculation is used as the interpolation sequence to obtain the cubic spline interpolation of the time-longitude trajectory and the time-latitude trajectory, respectively. The ship's navigation trajectory is obtained by merging the above two cubic spline curves.

[0052] The specific principle of cubic spline interpolation is as follows, taking the longitude of a ship as an example, let the ship's longitude be... It is a function of time, and within the time interval... The interior satisfies the cubic spline function ,but According to the boundary conditions, in and The following conditions must be met at any time:

[0053] ;

[0054] Based on the boundary values ​​of the sequence segment to be interpolated The spline function coefficients can then be obtained to determine the longitude. Regarding time The function expression is then used to calculate the longitude value of the corresponding trajectory point based on the interpolation time interval. The interpolation method for the latitude value of the trajectory sequence is similar and will not be described in detail here.

[0055] Step 4: Ship Status Deduction within the AIS Time Interval: Based on the above trajectory, establish the tangent equation for any point to obtain the heading at any given time. Based on the time-velocity and heading information of different ships, combined with satellite imaging parameters, establish the time-range velocity and velocity-azimuth velocity mapping relationships using velocity component decomposition and linear interpolation. By merging the velocity values ​​in each direction, obtain the ship velocity values ​​corresponding to different times.

[0056] As shown in Figure 2 and Table 1, let the time be... The speed is The heading angle is The satellite azimuth angle corresponding to the imaging time is The velocity calculation method in the distance direction is shown in equation (1), and the velocity calculation method in the azimuth direction is shown in equation (2). For the velocity in the east-west longitude direction... and speed in the north and south latitude directions By performing linear interpolation on t, we obtain linear functions of velocity with respect to time in each direction. For Velocity in the distance direction at any time t The calculation formula is shown in equation (3); for The velocity in the azimuth direction at any time t Similarly, by combining the velocity components in both directions, we can obtain the ship's speed at any time t. The calculation formula is shown in equation (4).

[0057] Table 1

[0058] ;

[0059] Step 5: Correction of Ship's True Position Based on Spaceborne SAR Imaging Principles: The range motion of a moving ship causes a shift in its Doppler center frequency, which manifests as an azimuth shift in the SAR image. This azimuth shift of the ship target can be expressed as:

[0060] ;

[0061] in, This indicates the azimuth offset of the ship target. Indicates the closest slant range between the radar and the target. This indicates the range velocity of the ship target. Indicates the radar's angle of incidence. Indicates radar speed.

[0062] By combining the satellite azimuth at the time of satellite imaging with the calculated actual ship position and heading, the ship on the nearest SAR image along the range direction is obtained as the target associated ship, and the azimuth of the ship is offset compensation is performed.

[0063] Step 6: Precise association processing of AIS and spaceborne SAR data for ocean vessel targets: The azimuth offset of the ships in the SAR scene at the imaging time is processed, and then the corrected SAR target positions and AIS-inferred positions are associated. The SAR target position with the smallest distance along the range direction and the AIS-inferred position are the same ship.

[0064] The SAR satellite imaging plan acquires SAR images at time t. Based on the time-latitude and longitude coordinates established in the previous steps, by inputting the imaging time t, the theoretical position of the ship at that imaging time can be deduced, as well as the tangent angle formed based on that position. The ship's heading at that time can then be obtained through the relative position. Finally, the ship's speed at that imaging time is obtained using the time-velocity mapping relationship.

[0065] Step 7: Calculation of geometric error of spaceborne SAR marine image system: Combine the satellite azimuth at the time of satellite imaging with the calculated actual ship position and heading, calculate the distance difference with the corresponding ship position on the compensated SAR image, and decompose it to obtain the ship's offset distance in the range direction at the time of SAR imaging. This offset distance is used as the quantified value of the system error in the range direction when the SAR images moving targets.

[0066] Let the position of a certain ship based on AIS simulation be... The location of the same ship on the SAR image generated under the SAR satellite imaging plan is The two sets of coordinate values ​​are transformed using Gaussian forward arithmetic. Calculate the distance .

[0067] ;

[0068] Combined with heading angle and the satellite azimuth angle corresponding to the imaging time Calculate the degree of displacement in the distance direction .

[0069] ;

[0070] Ultimately, the degree of systematic shift in the range and azimuth directions when SAR images moving targets is quantified. .

[0071] This method, based on the imaging time and regional location of each target in a satellite-borne SAR image, acquires AIS data within the imaging range through a service platform. By cleaning the AIS data and removing outliers, effective AIS attribute information is obtained. Ship trajectories established from AIS data are discrete spatiotemporal sequences. By combining latitude, longitude, time, and speed, mapping relationships are constructed across different dimensions to form continuous ship trajectories corresponding to the AIS time period within the latitude and longitude range. Simultaneously, using the constructed mapping relationships across dimensions, the latitude, longitude, speed, and heading of different ships at the SAR imaging time are extrapolated. Furthermore, since the range motion of moving targets causes a shift in their Doppler center frequency, resulting in an azimuth shift in the SAR image, this method combines satellite parameters at the satellite imaging time with the extrapolated ship position and heading information to reposition the ships in the SAR image, fully utilizing multi-dimensional feature information to improve correlation and recognition accuracy.

[0072] Example

[0073] Please refer to Figure 3. In this embodiment, AS01 SAR data is selected for the experiment. The data is in TOPS mode, with a range resolution of 4.6m and an azimuth resolution of 10.0m, and the polarization mode is HH. The satellite is scheduled to begin imaging a certain area on February 27, 2024, at 02:52:37.698 (UTC). The longitude range of the area is 113.951°~115.166°, and the latitude range is 20.973°~22.459°. The satellite azimuth angle corresponding to the imaging time is 192.65°, and the radar center incidence angle is... The angle is 28.75°, the near-range distance is 581,675 m, and the radar velocity is 7,171.6 m / s. The final image is shown in Figure 3.

[0074] Please refer to Figure 4. Download the corresponding AIS data based on the latitude and longitude range, clean and process the AIS data, and finally filter out the ships within the latitude and longitude range. In this experiment, the imaging times of two ships (ship names: INTERASIA HORIZON and HUGE) were selected to verify the implementation of this algorithm.

[0075] The specific implementation and process results are as follows: After data cleaning, a relatively accurate AIS data sequence was obtained, and detailed information is shown in Table 2.

[0076]

[0077] Please refer to Figure 5. The ship's navigation trajectory is obtained by cubic spline interpolation based on the cubic spline interpolation method and combined with the cleaned AIS data.

[0078] Based on the AIS information corresponding to the ship, a functional expression of longitude with respect to time is established. Then, based on the interpolation time interval, the longitude value of the corresponding trajectory point is obtained. Based on the above trajectory, the tangent equation of any point is established to obtain the heading at any time. Then, based on the time-velocity information of different ships, combined with the heading information and the satellite azimuth angle corresponding to the imaging time, velocity component decomposition and linear interpolation are used to establish the time-range velocity and time-azimuth velocity mapping relationship. By merging the velocity values ​​in the above directions, the ship velocity values ​​corresponding to different times are obtained. The relationship between time and longitude and the relationship between time and velocity magnitude are shown in Figure 6 below.

[0079] For each ship in the AIS data, sorted in ascending order of longitude, and using the first signal time corresponding to the smallest longitude within the time range as the benchmark, the time difference between the ships at the planned SAR satellite imaging time is calculated, as shown in Table 2. Combined with the established mapping model between time and various attribute information, the state of each ship at the SAR imaging time can be deduced, as shown in Figure 7 below, where the red dots represent the locations of the SAR imaging times.

[0080] Based on the time-latitude and longitude coordinates established in the preceding steps, and by inputting the time difference values ​​for each ship, the theoretical position of the ship at the imaging moment can be deduced, as well as the tangent angle formed based on that position. The ship's heading at that moment can then be obtained through the relative positions of the fore and aft ships. Using the time-range (azimuth) velocity mapping relationship, the ship's velocity magnitude and actual speed in the SAR range and azimuth directions at the imaging moment can be obtained.

[0081] Based on calculations and deductions, the ship's status at the moment of SAR imaging is shown in Table 3. Positive values ​​in the table represent the imaging direction along the range (azimuth) and negative values ​​represent the opposite direction (the same applies below):

[0082] Table 3. Results of Ship Status Projection on SAR Images Based on AIS Data at Imaging Time

[0083] ;

[0084] Radial velocity causes a positional shift of the ship target in the azimuth direction of the SAR image. Considering the satellite azimuth at the time of imaging (192.65°), the closest slant range between the radar and the target (581675m), the radar's center angle of incidence (28.75°), the radar velocity (7171.6m / s), and the range velocity of each ship target, the azimuth positional shift of each ship is calculated based on the relationship between radial and range velocities, as shown in the table below:

[0085] Table 4. Estimation of Azimuth Offset for Each Ship

[0086] ;

[0087] Combining the satellite azimuth at the time of satellite imaging with the calculated actual ship position and heading, the ship on the nearest SAR image along the range direction is selected as the target associated ship, and azimuth offset compensation is performed on the ship's azimuth. The position changes before and after compensation are shown in the table below:

[0088] Table 5. Changes in latitude and longitude before and after offset compensation

[0089] ;

[0090] Comparing the azimuth offset before and after compensation, it can be seen that in addition to the positional shift of the ship target in the SAR image caused by the radial velocity of the radar itself, there is also a range offset during the imaging process. By subtracting the ship target position in the range from the ship position in the AIS after compensation, the range and azimuth offsets are obtained. These offsets are the systematic errors introduced during the SAR imaging process.

[0091] Table 6. Estimation of Range-to-System Offset for Each Ship

[0092]

Claims

1. A method for marine vessel association recognition based on AIS and space-borne SAR data, characterized in that, The method comprises the following steps: Step one: obtaining AIS data based on the imaging time of SAR satellite; Step two: cleaning and processing AIS data; Step three: recovering ship trajectory in AIS time interval; Step four: deducing ship state in AIS time interval; the deduction comprises establishing time-distance or time-azimuth speed mapping relationship; Step five: revising real ship position based on the imaging principle of satellite-borne SAR; Step five specifically comprises the following steps: combining the satellite azimuth at the imaging time, the calculated actual ship position and course, obtaining the ship on the SAR image along the distance direction as the target associated ship, and performing offset compensation on the azimuth of the ship; The azimuth position offset of the ship target is represented as: wherein, represents a position offset in the azimuth direction of the ship target, represents the closest slant range of the radar to the target, represents the range direction velocity of the ship target, represents the angle of incidence of the radar, represents the radar velocity; Step six: accurate association processing of AIS and satellite-borne SAR data marine ship targets; step six specifically comprises the following steps: Performing azimuth offset processing on the ship in the SAR scene at the imaging time, then associating the revised SAR target position and the AIS deduced position, and the SAR target position and the AIS deduced position with the minimum distance along the distance direction are the same ship; Based on the SAR satellite imaging plan, the SAR is imaged at time t; based on the established time-longitude-latitude, the time t value at the imaging time is brought in, the theoretical position of the ship at the imaging time is deduced, and the tangent angle formed based on the position is used to further obtain the ship course at the imaging time; the time-speed mapping relationship is used to obtain the ship speed at the imaging time; Step seven: calculating the geometric error of the satellite-borne SAR marine image system.

2. The method for AIS and space-borne SAR data based marine vessel association recognition according to claim 1, characterized in that, Step one specifically comprises the following steps: based on the imaging longitude and latitude range of the SAR satellite, the imaging time, and the satellite azimuth angle corresponding to the imaging time, the AIS data within the longitude and latitude range is screened and downloaded.

3. The method for AIS and space-borne SAR data based marine vessel association recognition according to claim 1, characterized in that, Step two specifically comprises the following steps: Based on time-longitude or latitude data, the data is preprocessed by Vondrak data smoothing method, and after deleting the wrong AIS data, the correct AIS sequence is obtained; The point position vectorization is performed on the cleaned data, the existing data is classified and processed according to the relative position information and the existing attributes of the ship type, and the discontinuous trajectory of different ships is extracted.

4. The marine ship association and identification method based on AIS and satellite-borne SAR data according to claim 3, wherein In the Vondrak data smoothing method processing, for a data sequence wherein is time, is longitude or latitude; the basic assumption of the Vondrak data smoothing method is that: wherein: , wherein: is a smoothing value; denotes a third derivative of is a weight; is a pending coefficient; is called the approximation degree; is called the smoothing degree; is called the smoothing coefficient, the value of which determines the degree of smoothing of the data curve.

5. The method for AIS and space-borne SAR data based marine vessel association recognition according to claim 1, characterized in that, Step three specifically comprises the following steps: based on the cubic spline interpolation method, the AIS data is sorted according to time, the longitude and latitude at the time starting point are selected as the coordinate origin, and the time difference of the AIS data before and after calculation is selected as the interpolation sequence, the cubic spline interpolation of the time-longitude trajectory and the time-latitude trajectory is obtained, and the ship navigation trajectory is obtained by merging the two cubic spline curves.

6. The method for AIS and space-borne SAR data based marine vessel association recognition according to claim 1, characterized in that, The step four specifically comprises the following steps: establishing tangent line equation of any point based on the sailing track to obtain the heading corresponding to any time; based on the time-speed information of different ships, the heading information, combining with the satellite imaging parameters, adopting speed component decomposition and linear interpolation to establish time-distance direction speed and speed-azimuth direction speed mapping relationship; and through merging the speed values of each direction, the ship speed values corresponding to different times are obtained.

7. The method for AIS and space-borne SAR data based marine vessel association recognition according to claim 1, characterized in that, The step seven specifically comprises the following steps: combining the satellite azimuth direction at the satellite imaging time, the actual ship position and the heading calculated, and the corresponding ship position on the SAR image after compensation, and performing distance difference to obtain the offset distance of the ship in the distance direction at the SAR imaging time, which is taken as the quantitative value of the system error in the distance direction when the SAR images the moving target.

8. The method for AIS and space-borne SAR data based marine vessel association recognition according to claim 7, characterized in that, Specifically, The position of the certain ship deduced based on AIS is The corresponding position of the same ship on the SAR image formed under the SAR satellite imaging plan is ; The two sets of coordinate values are converted into , and the distances are calculated. Combining the heading angle and the satellite azimuth angle corresponding to the imaging moment , the distance displacement degree is calculated ; The final quantification of the degree of system shift in range and azimuth direction when SAR imaging moving targets .

Images