Ship passage river section width extraction method and system based on sar data

Abstract

The application discloses a ship passing river section width extraction method and system based on SAR data, which is applied to the technical field of remote sensing monitoring, and the method comprises the following steps: acquiring SAR remote sensing image data and performing pretreatment; based on a preset river section, buffer areas of an upstream and a downstream of the river are respectively divided; according to the pretreated SAR remote sensing image data, a water body area of the river is extracted, and a water body extraction result is obtained; and the river section width is calculated through the water body extraction result. The application can eliminate the influence of the complex and irregular riverbank curve and is not affected by the ship passing on the river.

Description

Technical Field

[0001] This invention relates to the field of remote sensing monitoring technology, and in particular to a method and system for extracting the width of river cross-sections for vessel passage based on SAR data. Background Technology

[0002] River cross-sectional width is an important hydrological parameter closely related to river flow. While river flow cannot be directly monitored through remote sensing, it can be obtained to some extent through correlations. If the river cross-section is rectangular, there is no one-to-one correspondence between river surface width and water level, or between water level and flow. Currently, most river cross-sections, whether naturally formed or man-made, are trapezoidal or curved, and therefore, there are one-to-one correlations between river surface width and water level, and between water level and river flow. In this case, the minimum river surface width corresponds to the minimum ecological water level, and the minimum ecological water level corresponds to the minimum ecological flow. The minimum river surface width corresponding to the minimum ecological flow is considered an indicator of ecological flow. By comparing the river surface width extracted from remote sensing with the minimum river surface width, it is possible to determine whether the ecological flow meets the standard and whether the river's ecological health is good. Therefore, river surface width is an important means of assessing the current state of the river ecosystem and is of great significance for protecting the river's ecological environment in my country.

[0003] Traditionally, river width is obtained by monitoring personnel using low-frequency ground-penetrating radar (GPR) systems at hydrological stations. However, this method relies on GPR antennas located on bridges or aerial cableways, and is limited by natural environment and economic constraints, making long-term measurements of certain areas difficult and unreliable for reference. With the development of remote sensing technology, which offers advantages such as wide coverage, low cost, and long-term data transmission, it has been applied to river width monitoring. However, optical remote sensing imagery is easily affected by cloud and rain weather, making it impossible to acquire effective images periodically, thus failing to meet the requirements for long-term, periodic monitoring. Synthetic Aperture Radar (SAR) offers the advantages of all-weather, all-time imaging, compensating for the shortcomings of optical imagery. SAR data has become an important data source for river cross-section monitoring.

[0004] Domestic and foreign scholars have applied SAR data to monitor the width of river cross sections, including ERS-1[1], Radartsat[2], SMMR[3], AMSR-E[4], JERS-1[5], etc. The method used is to extract water bodies based on the separation of land and water, and then measure the width of the river surface.

[0005] References:

[0006] [1]Brakenridge G R,Nghiem S V,Anderson E,et al.Orbital mi-crowavemeasurement of river discharge and ice status[J].WaterResources Research,2007,43,W04405.

[0007] [2]Brakenridge G R,Knox J C,Paylor E D,et al.Radar remote sens-ingaids study of the great flood of 1993[J].EOSTransactions,A-mericanGeophysical Union,1994,75(45):521-527.

[0008] [3]Smith L C.Satellite remote sensing of river inundation area,stage,and discharge:A review[J].Hydrological Processes,1997,11:1 427-1 439.

[0009] [4]Qi Huang,Di Long,Mingda Du,Chao Zeng,Gang Qiao,Xingdong Li,AizhongHou,Yang Hong,Discharge estimation in high-mountain regions with improvedmethods using multisource remote sensing:A case study of the UpperBrahmaputra River,Remote Sensing of Environment,Volume 219,2018,Pages115-134.

[0010] [5]Wenchao S, Hiroshi I, Satish B. Towards improving river discharge estimation in ungauged basins: Calibration of rainfall-runoff models based on satellite observations of river flow width at basin outlet [J]. Hydrology and Earth System Sciences, 2010, 14(10).

[0011] Previous methods are more applicable to general rivers, but less so to rivers with heavy ship traffic. Therefore, this application proposes a method and system for extracting the cross-sectional width of ships-traveling rivers based on SAR data. Summary of the Invention

[0012] The purpose of this application is to provide a method and system for extracting the width of river cross-sections for vessel traffic based on SAR data, aiming to solve the problem of difficulty in measuring the width of water surface in rivers with busy vessel traffic.

[0013] To achieve the above objectives, this application provides the following technical solution:

[0014] This application provides a method for extracting the width of river cross-sections for vessel passage based on SAR data, including:

[0015] Acquire SAR remote sensing image data and perform preprocessing;

[0016] Based on the pre-defined cross-section of the river, the buffer zones for the upstream and downstream of the river are divided separately;

[0017] Based on the preprocessed SAR remote sensing image data, the water body area of ​​the river is extracted to obtain the water body extraction results;

[0018] The width of the river cross section is calculated based on the water body extraction results.

[0019] Furthermore, the step of extracting the water body area of ​​the river based on the preprocessed SAR remote sensing image data to obtain the water body extraction result specifically includes the following steps:

[0020] Fitting the K-distribution to pixels in the buffer region;

[0021] The confidence interval is calculated based on the fit of the K-distribution and used as the threshold;

[0022] The pixels within the buffer region that are smaller than the threshold are refitted with the K-distribution to obtain the termination threshold.

[0023] The water body area of ​​the river is extracted based on the termination threshold, and the water body extraction result is obtained.

[0024] Furthermore, the step of fitting the K-distribution to the pixels in the buffer region specifically includes the following steps:

[0025] The formula for the K-distribution is:

[0026]

[0027] Where K v-1 Γ is the Bessel function of the second kind, order ν-1; Γ is the Gamma function; ν is the shape factor; α is the scale parameter;

[0028] The second and fourth moment methods are used to calculate the K-distribution parameters of the sample points in the buffer region. The formula for the total moments of the K-distribution is:

[0029]

[0030] Where k is the order of the moments; E is the expected mean.

[0031] The parameter estimates for ν and α are as follows:

[0032]

[0033] Where m2 and m4 are the second and fourth moments, respectively;

[0034] The probability density function of the K-distribution is obtained from its parameters.

[0035] Furthermore, the step of calculating the confidence interval based on the fit of the K distribution and using it as the threshold specifically includes the following steps:

[0036] The shape factor ν is labeled as υ0. The confidence interval is calculated based on the fit of the K-distribution, using the following method:

[0037] Divide the interval [0,1] into 1000 equal parts with an interval of 0.001, and calculate the cumulative value of the K-distribution starting from 0:

[0038]

[0039] Where f K It follows a K-distribution;

[0040] When F(s-1) < 0.95 and F(s) ≥ 0.95, the threshold is obtained.

[0041] Furthermore, the step of refitting the K-distribution to the pixels within the buffer region that are smaller than the threshold to obtain the termination threshold specifically includes the following steps:

[0042] For all pixels within the buffer region whose values ​​are less than the threshold T0, refit the K distribution; calculate the shape coefficient v1, and calculate the confidence interval as the termination threshold T1 based on the fit between v1 and the K distribution.

[0043] If the following conditions are met:

[0044] and

[0045] If the loop terminates, the K-distribution fitting is repeated.

[0046] Furthermore, the step of extracting the river's water body area based on a termination threshold to obtain the water body extraction results specifically includes the following steps:

[0047] Binarization extraction of water areas is performed based on a termination threshold T1. Areas with SAR backscattering coefficients less than the threshold T1 are considered water areas and are marked as 1; otherwise, they are considered non-water areas and are marked as 0. The formula is as follows:

[0048]

[0049] Where g(x,y) is the binarized water extraction result, and f(x,y) is the original SAR backscattering coefficient image.

[0050] Furthermore, the step of calculating the river cross-sectional width using the water body extraction results specifically includes the following steps:

[0051] Using the preset cross-section of the river as a reference, with the midpoint of the preset cross-section as O, perpendicular lines are drawn to the left and right sides respectively; the perpendicular lines are numbered, with the lower side of the center O point marked as 1, 2, 3...i, and the upper side of the center O point marked as -1, -2, -3...-j;

[0052] The mean value of the binarized water body extraction results on the vertical line segment is extracted respectively, and the calculation formula is as follows:

[0053]

[0054] Where m is the mean; s is the number of pixels on the vertical segment; g ′ (x,y) represents the binarized water body extraction result on the vertical line segment;

[0055] Distinguish between the mean values ​​of the vertical segments on the upper and lower sides, and label the mean value of the water body in the i-th vertical segment on the upper side as m. i The average water value of the j-th vertical segment below is marked as n. jThe upper and lower perpendicular segments are sequentially tested. When the mean value of the perpendicular segment is less than 0.5, the test is terminated and the perpendicular segment number is marked.

[0056] The mean m of the upper vertical segment i The judgment is performed sequentially from i = 1, 2, 3... if m t ≥0.5 and m t+1 If the value is less than 0.5, then the upper termination judgment is marked as t;

[0057] The mean of the lower perpendicular segment n j The judgment is performed sequentially from j = 1, 2, 3... if n w ≥0.5 and n w+1 If the value is less than 0.5, the judgment on the lower side is terminated, and the lower bank of the river is marked as w.

[0058] The width of the river cross section is then the resolution of the SAR remote sensing image data multiplied by (t+w).

[0059] This application provides a system for extracting the width of river cross-sections for vessel passage based on SAR data, including:

[0060] Acquisition module: Acquires SAR remote sensing image data and performs preprocessing;

[0061] Region division module: Based on the preset cross-section of the river, the buffer zones of the upstream and downstream of the river are divided separately;

[0062] Extraction module: Based on the preprocessed SAR remote sensing image data, extract the water body area of ​​the river to obtain the water body extraction result;

[0063] Calculation module: Calculates the width of the river cross-section based on the water body extraction results.

[0064] This application provides an apparatus comprising a processor and a memory coupled to the processor, wherein the memory stores program instructions for implementing a method for extracting the width of a river cross-section for vessel passage based on SAR data; the processor is configured to execute the program instructions stored in the memory to extract the width of a river cross-section for vessel passage based on SAR data.

[0065] This application provides a storage medium storing processor-executable program instructions for executing a method for extracting the width of a river cross-section for ship passage based on SAR data.

[0066] This application provides a method and system for extracting the width of river cross-sections for vessel passage based on SAR data, which has the following advantages:

[0067] (1) This application only uses SAR data to calculate the width of the river cross section, without the need for other auxiliary data or historical statistics of the river, and is not affected by cloud and rain weather, and can monitor the width of the river cross section at fixed periods.

[0068] (2) This application is not affected by the incident angle and incident direction of SAR data in the river area; and for rivers with busy ship traffic, it is not affected by the ship traffic on the river; the riverbank is determined based on the mean of the vertical segment, which can eliminate the influence of the complex and irregular curve of the riverbank.

[0069] (3) This application can realize adaptive threshold calculation for water surface recognition without the need to preset water body threshold; apart from the initial preset cross section of the river, no other human intervention is required, which can achieve full automation and has the advantages of low cost, wide range, high efficiency and high accuracy. Attached Figure Description

[0070] Figure 1 This is a flowchart illustrating the method for extracting the width of a river cross-section for vessel passage based on SAR data according to Embodiment 1 of this application.

[0071] Figure 2 This is a schematic diagram of the cross-section Sentinel-1 of Embodiment 1 of this application;

[0072] Figure 3 This is a schematic diagram of a pre-defined cross-section of a river in Embodiment 1 of this application;

[0073] Figure 4 This is a schematic diagram of the river pre-section buffer zone in Embodiment 1 of this application;

[0074] Figure 5 This is a schematic diagram of the vertical segment setting of the river cross section in Embodiment 1 of this application;

[0075] Figure 6 This is a schematic diagram of the structure of the SAR data-based river cross-section width extraction system for vessels in Embodiment 2 of this application;

[0076] Figure 7 This is a schematic diagram of the device structure in Embodiment 3 of this application;

[0077] Figure 8 This is a schematic diagram of the storage medium structure of Embodiment 4 of this application. Detailed Implementation

[0078] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0079] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0080] Example 1

[0081] Please see Figure 1 This is a flowchart illustrating the method for extracting the width of a river cross-section for vessel passage based on SAR data according to Embodiment 1 of this application; the steps include:

[0082] S1: Acquire SAR remote sensing image data and perform preprocessing.

[0083] In this embodiment, Sentinel-1A data with a resolution of 10 meters is used as an example. The input Sentinel-1A VV polarization data is preprocessed, including but not limited to thermal noise removal, radiometric calibration, and Doppler topographic correction.

[0084] Please see Figure 2 This is a schematic diagram of the cross-section Sentinel-1 of Embodiment 1 of this application. In the diagram, black represents the Xijiang River in Zhaoqing City, and white dots represent passing vessels.

[0085] The impact of passing vessels on river cross-section extraction is mainly reflected in two aspects:

[0086] Some small boats will only sail close to the shore. If the resolution of the SAR remote sensing image data is not high, such as Sentinel-110 meter resolution, it will cause the boat to be connected to the riverbank in the SAR image, resulting in an underestimation of the cross-sectional width during the extraction of cross-sectional width.

[0087] The antennas of large ships can produce sidelobes in SAR images. If the sidelobes pass through the river cross-section, they will affect the extraction of the real water body. A large number of sidelobes will make the threshold for water-land segmentation too large, which will cause some near-shore low-reflection areas to be classified as water bodies, thus expanding the extraction range of water bodies and leading to an overestimation of the cross-sectional width.

[0088] Therefore, this application only uses SAR remote sensing image data to calculate the width of the river cross section, without the need for other auxiliary data or historical statistics of the river, and is not affected by cloud and rain weather, and can monitor the width of the river cross section at fixed periods.

[0089] S2: Based on the preset cross-section of the river, the buffer zones of the upstream and downstream of the river are divided respectively.

[0090] In this embodiment, a buffer is performed upstream and downstream of the river based on a pre-defined cross-section of the river, with the buffer distance being consistent with the pre-defined cross-section of the river.

[0091] Please see Figure 3 This is a schematic diagram of a preset cross-section of a river in Embodiment 1 of this application.

[0092] Please see Figure 4 This is a schematic diagram of the river's preset cross-section buffer zone in Embodiment 1 of this application. The preset cross-section of the river is 700 meters, therefore the buffer zones upstream and downstream are both 700 meters.

[0093] S3: Based on the preprocessed SAR remote sensing image data, extract the water body area of ​​the river to obtain the water body extraction result.

[0094] In this embodiment, the step of extracting the water body area of ​​the river based on the preprocessed SAR remote sensing image data to obtain the water body extraction result specifically includes steps S31 to S34. The implementation of each step is described in detail below.

[0095] S31: Fitting the K-distribution to the pixels in the buffer region.

[0096] The formula for the K-distribution is:

[0097]

[0098] Where K v-1 Γ is the Bessel function of the second kind, order ν-1; Γ is the Gamma function; ν is the shape factor; α is the scale parameter;

[0099] The second and fourth moment methods are used to calculate the K-distribution parameters of the sample points in the buffer region. The formula for the total moments of the K-distribution is:

[0100]

[0101] Where k is the order of the moments; E is the expected mean.

[0102] The parameter estimates for ν and α are as follows:

[0103]

[0104] Where m2 and m4 are the second and fourth moments, respectively;

[0105] The probability density function of the K-distribution is obtained from its parameters.

[0106] S32: Calculate the confidence interval based on the fit of the K distribution and use it as the threshold.

[0107] Label the shape factor ν as υ0, and calculate the 95% confidence interval based on the fit of the K distribution. The 95% confidence interval is calculated as follows:

[0108] Divide the interval [0,1] into 1000 equal parts at intervals of 0.001, and calculate the cumulative value of the K-distribution starting from 0:

[0109]

[0110] Where f K It follows a K-distribution;

[0111] When F(s-1) < 0.95 and F(s) ≥ 0.95, the threshold is obtained.

[0112] S33: Refit the K-distribution to the pixels within the buffer area that are smaller than the threshold to obtain the termination threshold.

[0113] For all pixels within the buffer region whose values ​​are less than the threshold T0, the K distribution fitting in step S31 is performed again; the shape coefficient v1 is calculated, and based on the fitting of v1 and the K distribution, the 95% confidence interval is calculated as the termination threshold T1.

[0114] If the following conditions are met:

[0115] and

[0116] If the loop terminates, then repeat steps S31 to S33.

[0117] S34: Extract the water body area of ​​the river based on the termination threshold to obtain the water body extraction result.

[0118] Binarization extraction of water areas is performed based on a termination threshold T1. Areas with SAR backscattering coefficients less than the threshold T1 are considered water areas and are marked as 1; otherwise, they are considered non-water areas and are marked as 0. The formula is as follows:

[0119]

[0120] Where g(x,y) is the binarized water extraction result, and f(x,y) is the original SAR backscattering coefficient image.

[0121] It is understandable that this application extracts the water body area of ​​a river based on preprocessed SAR remote sensing image data to obtain the water body extraction result. Conventional methods mainly employ threshold segmentation, with thresholds often obtained empirically. However, in rivers with heavy ship traffic, factors such as ships and radar sidelobes can cause empirical thresholds to result in an underestimation of the water surface area. Furthermore, for the same river, its location in the SAR image varies, leading to different SAR signal incident angles and consequently different SAR signal values ​​on the river surface. Therefore, a single empirical threshold cannot be applied to the entire river.

[0122] Therefore, this application achieves adaptive threshold calculation for water surface identification without the need for pre-setting water body thresholds; apart from the initial preset river cross-section, no other human intervention is required, enabling full automation with advantages of low cost, wide coverage, high efficiency, and high accuracy. Furthermore, it is unaffected by the incident angle and direction of SAR data in the river area, and for rivers with heavy vessel traffic, it remains unaffected by vessel traffic.

[0123] S4: Calculate the width of the river cross-section based on the water body extraction results.

[0124] In this embodiment, the preset cross-section of the river is used as a reference, with the midpoint of the preset cross-section being O. Perpendicular segments are made every 10 meters to the left and right sides. The 10-meter interval is set according to the resolution of Sentinel-1 SAR data. The length of the perpendicular segment is 450 meters. The length of 450 meters is set according to the maximum length of the boats that can pass on the river. If the maximum length of the boat is n, then the length of the perpendicular segment is set to 3n. Since the maximum length of the boat on the Xijiang River is 150 meters, it is set to 450 meters here.

[0125] Please see Figure 5 This is a schematic diagram of the vertical line segment setting of the river cross section in Embodiment 1 of this application. Each line segment is numbered, with the lower side of the center point O sequentially marked as 1, 2, 3...i, and the upper side of the center point O sequentially marked as -1, -2, -3...-j.

[0126] The mean value of the binarized water body extraction results on the vertical line segment is extracted respectively, and the calculation formula is as follows:

[0127]

[0128] Where m is the mean; s is the number of pixels on the vertical segment; g ′ (x,y) represents the binarized water body extraction result on the vertical line segment.

[0129] To distinguish the mean values ​​of the vertical segments on the upper and lower sides, the mean value of the water body in the i-th vertical segment on the upper side is labeled as m. i The average water value of the j-th vertical segment below is marked as n. jThe upper and lower perpendicular segments are sequentially tested. When the mean value of the perpendicular segment is less than 0.5, the test is terminated and the perpendicular segment number is marked.

[0130] The mean m of the upper vertical segment i The judgment is performed sequentially from i = 1, 2, 3... if m t ≥0.5 and m t+1 If the value is less than 0.5, then the upper termination judgment is marked as t;

[0131] The mean of the lower perpendicular segment n j The judgment is performed sequentially from j = 1, 2, 3... if n w ≥0.5 and n w+1 If the value is less than 0.5, the judgment on the lower side is terminated, and the lower bank of the river is marked as w.

[0132] The width of the river cross section is the resolution of the SAR remote sensing image data multiplied by (t+w). In this example, the resolution is set to 10, so the actual water cross section of this section of the river is 10(t+w).

[0133] It is understood that this application calculates the river cross-section width based on the binarized water body extraction results. Existing technologies estimate the average river width using the ratio of water area to river length. However, on rivers with heavy ship traffic, the binarized water body extraction results are affected by ships and radar sidelobes, resulting in large areas of voids and thus inaccurate river width figures. To address this, this step proposes a method of sequentially judging from the cross-section center to the riverbank, eliminating the influence of ships and obtaining an accurate river cross-section width. Therefore, the riverbank discrimination based on the mean of the vertical segment in this application can eliminate the influence of complex and irregular riverbank curves, resulting in high accuracy and good stability.

[0134] In summary, Embodiment 1 of this application acquires and preprocesses SAR remote sensing image data without requiring other auxiliary data or historical river statistics, and is unaffected by cloud or rainy weather, enabling fixed-period monitoring of river cross-section width. It extracts the river's water body area from the preprocessed SAR remote sensing image data, obtaining water body extraction results and achieving adaptive threshold calculation for water surface identification, thus achieving full automation. The calculation of river cross-section width using the water body extraction results is unaffected by vessel traffic on the river, and the riverbank identification is based on the mean of vertical segments, eliminating the influence of complex and irregular riverbank curves. Therefore, the proposed method for extracting river cross-section width based on vessel traffic using SAR data has high accuracy and good stability.

[0135] Example 2

[0136] Please see Figure 6This is a schematic diagram of the structure of the SAR data-based river cross-section width extraction system for vessels, as described in Embodiment 2 of this application; the specific content includes:

[0137] Acquisition module: Acquires SAR remote sensing image data and performs preprocessing;

[0138] Region division module: Based on the preset cross-section of the river, the buffer zones of the upstream and downstream of the river are divided separately;

[0139] Extraction module: Based on the preprocessed SAR remote sensing image data, extract the water body area of ​​the river to obtain the water body extraction result;

[0140] Calculation module: Calculates the width of the river cross-section based on the water body extraction results.

[0141] Example 3

[0142] Please see Figure 7 This is a schematic diagram of the device structure in Embodiment 3 of this application. The device 50 includes a processor 51 and a memory 52 coupled to the processor 51.

[0143] The memory 52 stores program instructions for implementing the above-described method for extracting the width of the river cross-section for ship passage based on SAR data.

[0144] The processor 51 is used to execute program instructions stored in the memory 52 to extract the width of the river cross-section for ship passage based on SAR data.

[0145] The processor 51 can also be referred to as a CPU (Central Processing Unit).

[0146] Processor 51 may be an integrated circuit chip with signal processing capabilities. Processor 51 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor or any conventional processor.

[0147] Example 4

[0148] Please see Figure 8This is a schematic diagram of the storage medium in Embodiment 4 of this application. The storage medium in this embodiment stores a program file 61 capable of implementing all the above methods. This program file 61 can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or devices such as computers, servers, mobile phones, and tablets.

[0149] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0150] The above description is only a preferred embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0151] Although embodiments of this application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the appended claims and their equivalents.

[0152] Of course, the present invention may have many other embodiments. Based on this embodiment, other embodiments obtained by those skilled in the art without any creative effort are all within the scope of protection of the present invention.

Claims

1. A method for extracting the width of river cross-sections for vessel passage based on SAR data, characterized in that, include: Acquire SAR remote sensing image data and perform preprocessing; Based on the pre-defined cross-section of the river, the buffer zones for the upstream and downstream of the river are divided separately; Based on the preprocessed SAR remote sensing image data, the water body area of ​​the river is extracted to obtain the water body extraction results; The width of the river cross-section is calculated based on the water body extraction results. The process of extracting the water body area of ​​a river based on preprocessed SAR remote sensing image data to obtain the water body extraction result specifically includes the following steps: Fitting the K-distribution to pixels in the buffer region; The confidence interval is calculated based on the fit of the K-distribution and used as the threshold; The pixels within the buffer region that are smaller than the threshold are refitted with the K-distribution to obtain the termination threshold. The water body area of ​​the river is extracted based on the termination threshold to obtain the water body extraction result; The step of fitting the K-distribution to pixels based on the buffer region specifically includes the following steps: The formula for the K-distribution is: in for The second-order modified Bessel function; It is the Gamma function; For shape factor; For scale parameters; The second and fourth moment methods are used to calculate the K-distribution parameters of the sample points in the buffer region. The formula for the total moments of the K-distribution is: Where k is the order of the moment; It is the expected mean; and The parameter estimates are as follows: in , These are the second-order moments and the fourth-order moments, respectively. The probability density function of the K-distribution is obtained from its parameters; The step of calculating the river cross-sectional width using the water body extraction results specifically includes the following steps: Using the predetermined cross-section of the river as a reference, the midpoint of the predetermined cross-section is... Draw perpendicular lines to the left and right sides respectively; number the perpendicular lines, and mark the lower side of the center point O as 1, 2, 3...j, and the upper side of the center point O as -1, -2, -3...-i; The mean value of the binarized water body extraction results on the vertical line segment is extracted respectively, and the calculation formula is as follows: Where m is the mean; The number of pixels on the vertical line segment; The results are the binarized water body extraction results on the vertical line segment; Distinguish between the mean values ​​of the vertical segments on the upper and lower sides, and label the mean value of the water body in the i-th vertical segment on the upper side as... The average water value of the j-th vertical segment below is marked as The upper and lower perpendicular segments are sequentially tested. When the mean value of the perpendicular segment is less than 0.5, the test is terminated and the perpendicular segment number is marked. Mean of the upper vertical segment ,from The judgments are made sequentially, if and If so, the upper termination judgment mark is t; Mean of the lower perpendicular segment ,from The judgments are made sequentially, if and If the condition is met, the judgment on the lower side is terminated, and the lower bank of the river is marked as w. The width of the river cross-section is then the resolution of the SAR remote sensing image data multiplied by [the value of the data]. .

2. The method for extracting the width of a river cross-section for vessel passage based on SAR data according to claim 1, characterized in that, The step of calculating the confidence interval based on the fit of the K distribution and using it as the threshold specifically includes the following steps: Mark shape factor for The confidence interval is calculated based on the fit of the K-distribution, and the confidence interval is calculated in the following way: Divide the interval [0,1] into 1000 equal parts with an interval of 0.001, and calculate the cumulative value of the K-distribution starting from 0: in It follows a K-distribution; when and When, the threshold is obtained. .

3. The method for extracting the width of a river cross-section for vessel passage based on SAR data according to claim 2, characterized in that, The step of refitting the K-distribution to the pixels within the buffer region that are smaller than the threshold to obtain the termination threshold specifically includes the following steps: For all values ​​within the buffer region that are less than the threshold For each pixel, refit it with the K-distribution; calculate the shape coefficient. ,according to Fitting the K-distribution, the confidence interval is calculated as the termination threshold. ; If the following conditions are met: If the loop terminates, the K-distribution fitting is repeated.

4. The method for extracting the width of a river cross-section for vessel passage based on SAR data according to claim 1, characterized in that, The steps for extracting the water body area of ​​a river based on a termination threshold to obtain the water body extraction results specifically include the following steps: Termination-based threshold Binarization extraction of water body areas is performed, i.e., the SAR backscattering coefficient is less than a threshold. The water area is marked as 1; Otherwise, it is a non-water area and marked as 0; the formula is as follows: in The results are the binarized water body extraction results. This is the original SAR backscattering coefficient image.

5. A system for extracting the width of a river cross-section for vessel passage based on SAR data, used to execute the method for extracting the width of a river cross-section for vessel passage based on SAR data according to any one of claims 1-4, characterized in that, The system includes: Acquisition module: Acquires SAR remote sensing image data and performs preprocessing; Region division module: Based on the preset cross-section of the river, the buffer zones of the upstream and downstream of the river are divided separately; Extraction module: Based on the preprocessed SAR remote sensing image data, extract the water body area of ​​the river to obtain the water body extraction result; Calculation module: Calculates the width of the river cross-section based on the water body extraction results.

6. A device, characterized in that, The device includes a processor and a memory coupled to the processor, wherein the memory stores program instructions for implementing the method for extracting the width of a river cross-section for ship passage based on SAR data as described in any one of claims 1-4; the processor is used to execute the program instructions stored in the memory to implement the extraction of the width of a river cross-section for ship passage based on SAR data.

7. A storage medium, characterized in that, The device stores processor-executable program instructions for performing the method for extracting the width of a river cross-section for vessel passage based on SAR data as described in any one of claims 1-4.

Images