A waveform re-tracking method and device for a total focusing synthetic aperture radar altimeter

Abstract

The present invention provides a waveform re-tracking method and device for a fully focused synthetic aperture radar altimeter, which belongs to the field of remote sensing water level inversion, including: selecting daily transit waveform data of a river section in a research sub-area, obtaining data of multiple echo waveforms; extracting multi-scale adaptive sub-peaks, and using a method based on local extreme value scanning to identify multiple strong sub-peak signals of each echo waveform; counting sub-waveform information; performing sub-waveform segmentation processing to obtain segmented sub-waveforms and sub-waveform peaks; re-tracking each type of segmented sub-waveform and calculating its corresponding height; the height is the elevation value in the elevation sequence; removing the abnormal value of the single-day elevation sequence; and obtaining the elevation sequence of each segmented sub-waveform. The present invention can realize high-precision river water level monitoring.

Description

Technical Field

[0001] The invention belongs to the field of remote sensing water level inversion, and in particular relates to a waveform re-tracking method and device for a total focusing synthetic aperture radar altimeter. Background Art

[0002] Satellite altimetry technology has been widely used to estimate water surface height in coastal and inland waters, and is particularly important in areas where there is no or insufficient measured data.

[0003] The limitation of traditional radar altimeters lies in the along-track resolution, that is, the ability to distinguish between two targets on the surface. The along-track resolution of the original pulse-limited radar altimeter was the size of the pulse-limited footprint. The synthetic aperture radar (SAR) / delay Doppler altimeter (DDA) altimeter transmits pulses at a high pulse repetition frequency (PRF) to ensure their coherence, which significantly increases the number of independent observations obtained from a single scatterer on the surface, reducing the along-track resolution by an order of magnitude. In 2017, Egido and Smith developed the theory of fully focused coherent processing and developed a fully focused synthetic aperture radar (FFSAR) processing algorithm, which enables the SAR altimeter along-track resolution to reach the theoretical limit, which is about half the length of the antenna along the track.

[0004] When measuring the water level of inland waters, the altimeter is often affected by the echo signals of complex ground objects. This situation can usually be improved by waveform re-tracking, that is, correcting the initial distance calculated from the two-way time of the pulse echo. Many research results have been achieved on inland water waveform retracking, such as the OCOG retracking algorithm with center of gravity offset (Scott RF, Baker SG, Birkett CM, et al. A COMPARISON OF THE PERFORMANCE OF THE ICE AND OCEAN TRACKING MODES OF THE ERS-1 RADAR ALTIMETER OVER NONOCEAN SURFACES[J]. Geophysical Research Letters, 1994, 21(7): 553-556.); the NPPR (Narrow Primary Peak Retracker) retracking algorithm based on the main peak of the maximum reflected energy of the water surface (Jain M, Andersen OB, Dall J, et al. Sea surface height determination in the Arctic using Cryosat-2 SAR data from primary peak empirical retrackers[J]. Advances in Space Research, 2015, 55(1): 40-50.); MWaPP (The Multiple Waveform PerstistentPeak) algorithm based on the consistency of the water echo height of the front and rear footprint points (Villadsen H, Deng X, Andersen OB, et al. Improved inland waterlevels from SAR altimetry using novel empirical and physical retrackers[J].Journal of Hydrology. 2016, 537: 234-247.); combined with the 50% threshold re-rooting algorithm and the ICE-1 re-tracking algorithm (TIC), it is applied to extract water levels in high mountainous areas with complex terrain (Huang Q, Long D, Du M, et al.Animproved approach to monitoring Brahmaputra River water levels usingretracked altimetry data[J]. Remote Sensing of Environment, 2018: 112-128.);MSMWTR (Multi-subwaveform Multi-weight Threshold Retracker) algorithm based on multi-subwaveform multi-weight adaptation(Yuan C, Gong P, Zhang H, et al. Monitoring water level changesfrom retracked Jason-2 altimetry data: a case study in the Yangtze River, China[J]. Remote Sensing Letters, 2017, 8(5): 399-408.);SAMOSA (SAR Altimetry Mode Studies and Applications) physical retracking algorithm(Ray, C., Martin-Puig, C.,Clarizia, MP, Ruffini, G., Dinardo, S., Gommenginger, C., & Benveniste, J.(2015). SAR Altimeter Backscattered Waveform Model. Ieee Transactions onGeoscience and Remote Sensing, 53, 911-919). .

[0005] However, the existing waveform re-tracking methods have some limitations, which are mainly manifested in the following aspects: 1) The current waveform re-tracking algorithm performs well in lake water level monitoring, but its processing capability for complex multi-peak waveforms on the river surface is weak, making it difficult to meet the monitoring needs of river water levels; 2) The latest FFSAR data has great potential for river water level monitoring, but the large amount of data also brings more challenges to waveform re-tracking, and the current re-tracking algorithm has unsatisfactory accuracy; 3) The terrain characteristics of the river itself and its surroundings are complex and changeable (such as river width, landform, etc.), and the waveform interference signal is serious, so the existing waveform re-tracking algorithm has poor adaptability. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a waveform re-tracking method and device for a full-focusing synthetic aperture radar altimeter, which utilizes a waveform re-tracking algorithm with adaptive multi-scale peak detection to give full play to the high sampling rate advantage of FFSAR (full-focusing synthetic aperture radar) data, accurately identify water surface signals from the complex multi-peak waveforms on the river surface, and can adapt to various types of river landforms to achieve high-precision monitoring of river water levels.

[0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] A total focusing synthetic aperture radar altimeter waveform re-tracking method comprises the following steps:

[0009] Step 1: Select the daily transit waveform data of the river section in the study sub-area and obtain data of multiple echo waveforms;

[0010] Step 2: extract multi-scale adaptive sub-peaks, and use a method based on local extreme value scanning to identify multiple strong sub-peak signals of each echo waveform;

[0011] Step 3, counting sub-waveform information, wherein the sub-waveform information includes the gate length, the starting gate position, and the ending gate position of the strong sub-peak signal of each sub-waveform;

[0012] Step 4: classify and count the frequency of the strong sub-wave peak signal according to the wave gate length of step 3 to obtain the minimum wave gate length, calculate the number of segments according to the minimum wave gate length, perform sub-waveform segmentation processing, and obtain segmented sub-waveforms and sub-waveform peaks;

[0013] Step 5, re-tracking each type of segmented sub-waveform according to the starting gate position and the ending gate position in step 3 and the sub-waveform peak value in step 4, that is, re-tracking each type of segmented sub-waveform and calculating its corresponding height; the height is the elevation value in the elevation sequence;

[0014] Step 6: Remove outliers from the daily high sequence;

[0015] Step 7, obtain the elevation sequence of each segmented sub-waveform, the elevation sequences of different segmented sub-waveforms represent the elevation sequences from different landforms; select the segmented water level of the river section in the study sub-area, and obtain the water level sequence of the river surface of the river section in the study sub-area.

[0016] The present invention also provides a waveform re-tracking device for a total focusing synthetic aperture radar altimeter, comprising the following modules:

[0017] The data acquisition module selects the daily transit waveform data of the river section in the study sub-area and obtains the data of multiple echo waveforms;

[0018] The strong sub-peak signal acquisition module extracts multi-scale adaptive sub-peaks and uses a method based on local extreme value scanning to identify multiple strong sub-peak signals of each echo waveform;

[0019] A statistical module for counting sub-waveform information, wherein the sub-waveform information includes a gate length, a starting gate position, and an ending gate position of a strong sub-wave peak signal of each sub-waveform;

[0020] The segmentation processing module classifies and counts the strong sub-wave peak signals according to the wave gate length, obtains the minimum wave gate length, calculates the number of segments according to the minimum wave gate length, performs sub-waveform segmentation processing, and obtains segmented sub-waveforms and sub-waveform peaks;

[0021] The re-tracking module re-tracks various types of segmented sub-waveforms according to the starting wave gate position, the ending wave gate position, and the sub-waveform peak value, that is, re-tracks each type of segmented sub-waveform and calculates its corresponding height; the height is the elevation value in the elevation sequence;

[0022] The outlier removal module removes outliers in the daily high-order series;

[0023] The sequence acquisition module obtains the elevation sequence of each segmented sub-waveform, and the elevation sequences of different segmented sub-waveforms represent the elevation sequences from different landforms; the segmented water level of the river section in the study sub-area is selected to obtain the water level sequence of the river surface of the river section in the study sub-area.

[0024] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned waveform re-tracking method for a fully focused synthetic aperture radar altimeter when executing the program.

[0025] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned total focusing synthetic aperture radar altimeter waveform re-tracking method.

[0026] Beneficial effects:

[0027] The SAMOSA algorithm is designed to re-track a single echo waveform only in the range direction (one dimension), but the echo waveform of the river is complex and has many interferences, so it is easy to re-track the wrong sub-peak and obtain the wrong elevation result. The present invention adopts the ImpAMPD (Improved Adaptive Multi-scale Peak Detection) algorithm to perform waveform re-tracking calculations on the overall echo waveform in the range and azimuth directions (two dimensions), and through sub-waveform segmentation processing, the correct signal and the wrong signal are separated in the range direction, and the correct signal is identified by its continuity characteristics, thereby achieving the purpose of weakening the influence of the interference signal.

[0028] The present invention can not only obtain results comparable to or better than other algorithms under the condition of good waveform quality, but also can strip off interference signals in complex echo waveforms under waveform pollution, accurately identify river surface signals, and have good precision performance; at the same time, it can effectively process a variety of SAR altimeter data, give full play to the advantages of FFSAR data, and improve the monitoring ability of satellite altimetry for river water levels; the present invention has a high degree of automation, is easy to implement, has good adaptability to rivers of various scales of river width, and can achieve high-precision river water level monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flowchart of a waveform re-tracking method for a total focusing synthetic aperture radar altimeter of the present invention;

[0030] Figure 2 This is a comparison chart of the accuracy of the water level obtained by different re-tracking algorithms at various sampling rates and the measured water level. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0032] like Figure 1 As shown, a waveform re-tracking method for a total focusing synthetic aperture radar altimeter of the present invention comprises the following steps:

[0033] Step 1: Select the daily transit waveform data of the river section in the study sub-area and obtain multiple echo waveform data.

[0034] Step 2: Extract multi-scale adaptive sub-peaks and accurately identify multiple strong sub-peak signals of each echo waveform using a method based on local extreme value scanning (LMS). The core of this method is to construct an LMS for each echo waveform. The local extremum matrix of N , the extraction process is as follows:

[0035] Step 2.1: Calculate the local extreme value of the echo waveform by window moving average, where the window length is The range of changes meets , different window lengths The local extreme value of Determined according to the following formula:

[0036] (1)

[0037] in, is a random number that varies in the range [0,1]. represents the first Echo waveform, Represents the echo power waveform. When L=5, it is the best value for waveform peak detection. exist and The value of , based on which the local extreme value matrix can be constructed . Indicates intersection.

[0038] Step 2.2: Calculate the local extreme value matrix The sum of each row :

[0039] (2)

[0040] Local extrema matrix The sum of each row The dimensional distribution of the zero value (i.e. the distribution of local extreme values) is included in the local extreme value matrix, and then the local extreme value matrix is ​​calculated. The row corresponding to the global minimum of the sum of all rows , Represents the dimension that contains the most local maxima. Indicates finding the parameter value that minimizes a function. In addition, use Reorganize the local extreme value matrix M by eliminating all those that satisfy Elements , and obtain a new local extremum matrix .

[0041] Among them, L refers to the size of the local extreme value matrix in the echo waveform in the azimuth direction, N is the size of the local extreme value matrix in the echo waveform in the distance direction; m is the local extreme value; is the "row number" of the local extreme value matrix where the local extreme value is located, indicating the first row of the waveform set along the track. echo waveform, that is, the number of the local extreme value matrix; is the upward index of the local extremum in the local extremum matrix;

[0042] Step 2.3: By calculating the new local extreme value matrix The standard deviation of each column To determine the peak value of the waveform, where Index That is, the gate number of the detected peak value, and the detected peak value is set as the end gate of the sub-waveform For each echo waveform, a set of end gates of a single sub-waveform can be obtained Considering that some low-power peaks caused by noise may be detected, only when the power value is greater than 0.003 The detected peak is considered valid only when 0.003 is the set empirical threshold.

[0043] (3)

[0044] (4)

[0045] in, Indicates the threshold value. Indicates the echo power waveform.

[0046] In addition, to determine the starting gate of each sub-waveform, the echo waveform needs to be normalized by dividing it by the maximum power value; then the power difference of the continuous gates of the normalized waveform is calculated to obtain the power difference waveform; finally, the ending gate of each sub-waveform is determined. , search the power difference waveform from back to front, the first gate with a power value lower than 0.001 is the starting gate of the corresponding sub-waveform .

[0047] Based on the above steps, the end gate set All sub-waveform ending gates in the same way are searched, and finally the sub-waveform starting gate set can be calculated. , and the end gate set Together they form N sub-waveforms. Considering that some data may have a sub-waveform width of only 1-2 gates, since the sub-waveform width is too narrow to perform re-tracking calculations, it is necessary to expand several gates forward and backward to form a new sub-waveform. Previous studies have shown that the sub-waveform width requires at least 5 gates to be optimal, so the present invention calculates the difference between the end gate and the start gate to check the length of the sub-waveform, and for special sub-waveforms, expands 2-3 gates backward or 2-3 gates forward (giving priority to expanding the gates backward) to make the gate width of the sub-waveform reach 5.

[0048] Step 3: Count the sub-waveform information, including:

[0049] After multi-scale adaptive sub-peak extraction, traverse each row of echo waveform in the azimuth direction, and calculate the gate length of each sub-waveform according to the starting gate and ending gate position set of each sub-waveform in each row (of each) echo waveform, and record it in the allGateLengths array. Then, count each gate length and its corresponding frequency of occurrence. Then sort by frequency, select the gate length with the smallest length from the three gate lengths with the highest frequency, and record it as the minimum gate length. .

[0050] Step 4: performing sub-waveform segmentation processing, including:

[0051] According to the echo waveform gate number and minimum gate length Calculate the number of echo waveform gate segments :

[0052] (5)

[0053] There may still be multiple sub-peaks in each segment, so the minimum gate length can be considered to be halved, the purpose is to further subdivide the echo to remove the interference signal, which is called the narrow gate scheme. The original scheme is called the wide gate scheme. The narrow gate scheme is suitable for rivers with large widths, and the wide gate scheme is suitable for rivers with small widths.

[0054] Traverse the gate position of each sub-waveform peak of each echo waveform, determine the segment to which it belongs, count the first three segments with the largest number of sub-waveform peaks and the corresponding number of peaks, store the starting gate and ending gate positions of the sub-waveforms of each segment and the corresponding echo waveform row number .

[0055] Step 5: re-tracking various types of segmented sub-waveforms, that is, re-tracking each type of segmented sub-waveform and calculating its corresponding height, including:

[0056] First, traverse each type of sub-waveform and obtain the starting gate set of each sub-waveform of this type. , end wave gate set , echo waveform line number , create a temporary row newRow with the same size as a single echo waveform, and set all its values ​​to 0. Then assign the value of the sub-waveform to the corresponding gate position of the temporary row newRow, which is determined by the starting gate set of the sub-waveform and end gate set Sure.

[0057] Afterwards, two re-tracking algorithms, PTR (Point Target Retracker) and Threshold (threshold method), are used to perform waveform re-tracking calculations on the waveform of the temporary row newRow. Among them, Threshold is an extension of the OCOG (center of gravity offset method) waveform re-tracking algorithm, Threshold = thres · OCOG, Threshold represents the result of the threshold method, OCOG is the result of the center of gravity offset method, and the threshold thres can be optimized according to the same track standard deviation. The smaller the same track standard deviation, the better the threshold, and the same track standard deviation is the standard deviation of the height sequence.

[0058] Traverse the elevation sequence of each type of sub-waveform segment, calculate the standard deviation and median of the elevation sequence, remove the elevation values ​​that exceed 3 times the standard deviation of the sequence median, and repeat the above process 3 times.

[0059] Step 6: Remove outliers in the single-day high sequence, including:

[0060] Extract height data, the same orbit number is the same day data, use the nearest neighbor linear interpolation method to filter out the outliers in the single-day height sequence, then calculate the mean of the single-day height sequence, and use this value as the basis to remove the original height sequence Abnormal values ​​outside the 8m range, and finally obtain the water levels of several observation points on a single day .

[0061] Step 7: Select the segmented water level to obtain the water level sequence of the river surface, including:

[0062] Previously, the first three segments with the largest number of sub-peaks and the sub-waveforms to which they belong have been stored. In general, the first segment with the largest number of sub-peaks can be directly used as the correct signal segment. In extreme cases, if the result of the first segment is extremely poor (based on the same track standard deviation and the number of observations), the secondary segmentation can be considered based on the actual performance of the waveform in the study area.

[0063] Compared with the widely recognized SAMOSA+ waveform re-tracking method, the present invention (abbreviated as ImpAMPD) has the following technical advantages:

[0064] (1) Accuracy comparison:

[0065] The accuracy of the two algorithms was compared in five study sub-areas in the middle and upper reaches of the Yangtze River, namely, Zhicheng, Shashi, Hankou, Zhutuo, and Huantan. The results of the verification of the measured water level accuracy are shown in Table 1. The comparison between the algorithm water level and the measured water level sequence is as follows: Figure 2 shown.

[0066] High sampling rate is the advantage and feature of FFSAR data. The algorithm of the present invention is mainly aimed at FFSAR data; "1280Hz", "640Hz" and "160Hz" represent the sampling rates of FFSAR data; "ImpAMPD_OCOG" represents the water level sequence obtained by using the ImpAMPD algorithm and selecting the OCOG algorithm in the sub-waveform re-tracking calculation; "ImpAMPD_PTR" represents the water level sequence obtained by using the ImpAMPD algorithm and selecting the PTR algorithm in the sub-waveform re-tracking calculation; "SAMOSA+" represents the water level sequence obtained by using the SAMOSA+ re-tracking algorithm.

[0067] The ImpAMPD algorithm can adapt to rivers of various widths, and achieved the best accuracy results in the four river sections of Zhicheng, Shashi, Hankou, and Huantan, with STDD (standard deviation of difference) of 0.18m, 0.26m, 0.47m, and 0.36m respectively.

[0068] The ImpAMPD algorithm can fully utilize the high sampling rate advantage of FFSAR data, but in wide river areas, the high sampling rate has a negative impact on the performance of the SAMOSA+ algorithm.

[0069] Table 1

[0070]

[0071] The present invention also provides a waveform re-tracking device for a total focusing synthetic aperture radar altimeter, comprising the following modules:

[0072] The data acquisition module selects the daily transit waveform data of the river section in the study sub-area and obtains the data of multiple echo waveforms;

[0073] The strong sub-peak signal acquisition module extracts multi-scale adaptive sub-peaks and uses a method based on local extreme value scanning to identify multiple strong sub-peak signals of each echo waveform;

[0074] A statistical module for counting sub-waveform information, wherein the sub-waveform information includes a gate length, a starting gate position, and an ending gate position of a strong sub-wave peak signal of each sub-waveform;

[0075] The segmentation processing module classifies and counts the strong sub-wave peak signals according to the wave gate length, obtains the minimum wave gate length, calculates the number of segments according to the minimum wave gate length, performs sub-waveform segmentation processing, and obtains segmented sub-waveforms and sub-waveform peaks;

[0076] The re-tracking module re-tracks various types of segmented sub-waveforms according to the starting wave gate position, the ending wave gate position, and the sub-waveform peak value, that is, re-tracks each type of segmented sub-waveform and calculates its corresponding height; the height is the elevation value in the elevation sequence;

[0077] The outlier removal module removes outliers in a single-day high-order series;

[0078] The sequence acquisition module obtains the elevation sequence of each segmented sub-waveform, and the elevation sequences of different segmented sub-waveforms represent the elevation sequences from different landforms; the segmented water level of the river section in the study sub-area is selected to obtain the water level sequence of the river surface of the river section in the study sub-area.

[0079] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned waveform re-tracking method for a fully focused synthetic aperture radar altimeter when executing the program.

[0080] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned total focusing synthetic aperture radar altimeter waveform re-tracking method.

[0081] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The solutions in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and interpreted scripting language JavaScript, etc.

[0082] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0085] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0086] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A waveform re-tracking method for a total focusing synthetic aperture radar altimeter, characterized in that: The steps include: Step 1: Select the daily transit waveform data of the river section in the study sub-area and obtain data of multiple echo waveforms; Step 2: extracting multi-scale adaptive sub-peaks, using a method based on local extreme value scanning to identify multiple strong sub-peak signals of each echo waveform, including: Step 2.1: Calculate the local extreme value of the echo waveform by window moving average, where the window length is The range of changes meets , different window lengths The local extreme value of Determined according to the following formula: （1） in, is a random number that varies in the range [0,1]. represents the first An echo waveform also indicates the row number of the local extreme value matrix where the local extreme value is located in the azimuth direction; represents the echo power waveform, m is the local extreme value, exist and The value of , based on which the local extreme value matrix is ​​constructed ; L=5 is the optimal value for waveform peak detection, L refers to the size of the local extreme value matrix in the azimuth direction of the echo waveform, and N is the size of the local extreme value matrix in the distance direction of the echo waveform; Indicates intersection; is the upward index of the local extremum in the local extremum matrix; Step 2.2: Calculate the local extreme value matrix The sum of each row : （2） Local extrema matrix The sum of each row The dimension distribution containing 0 values ​​is then calculated to obtain the local extreme value matrix The sum of each row The row corresponding to the global minimum of , represents the dimension containing the most local maxima, It means finding the parameter value that makes a function reach the minimum value; use Reorganize the local extreme value matrix M by eliminating all those that satisfy Elements , and obtain a new local extremum matrix ; Step 2.3: By calculating the new local extreme value matrix The standard deviation of each column To determine the peak value of the waveform, where the standard deviation of each column is satisfied Index That is, the gate number of the detected peak value, and the detected peak value is set as the end gate of the sub-waveform , for each echo waveform, get the end gate set of a single sub-waveform ; When the power value is greater than 0.003 When , the detected peak value is considered valid; （3） （4） in, Indicates the threshold value; To determine the starting gate of each sub-waveform, the echo waveform is normalized by dividing it by the maximum power value; then the power difference of the continuous gates of the normalized waveform is calculated to obtain the power difference waveform; finally, the ending gate of each sub-waveform is determined. , search the power difference waveform from back to front, the first gate with a power value lower than 0.001 is the starting gate of the corresponding sub-waveform ; End gate All sub-waveform ending gates in the same way are searched, and finally the sub-waveform starting gate set is calculated. , and the end gate set Together they form N sub-waveforms; Step 3, counting sub-waveform information, wherein the sub-waveform information includes the gate length, the starting gate position, and the ending gate position of the strong sub-peak signal of each sub-waveform; Step 4: classify and count the frequency of the strong sub-wave peak signal according to the wave gate length of step 3 to obtain the minimum wave gate length, calculate the number of segments according to the minimum wave gate length, perform sub-waveform segmentation processing, and obtain segmented sub-waveforms and sub-waveform peaks; Step 5, re-tracking each type of segmented sub-waveform according to the starting gate position and the ending gate position in step 3 and the sub-waveform peak value in step 4, that is, re-tracking each type of segmented sub-waveform and calculating its corresponding height; the height is the elevation value in the elevation sequence; Step 6: Remove outliers from the daily high sequence; Step 7, obtain the elevation sequence of each segmented sub-waveform, the elevation sequences of different segmented sub-waveforms represent the elevation sequences from different landforms; select the segmented water level of the river section in the study sub-area, and obtain the water level sequence of the river surface of the river section in the study sub-area.

2. The waveform re-tracking method for a total focusing synthetic aperture radar altimeter according to claim 1, characterized in that: The step 3 comprises: After multi-scale adaptive sub-peak extraction, each row of echo waveform in the azimuth is traversed, and the gate length of each sub-waveform is calculated and recorded in the allGateLengths array according to the starting gate set and the ending gate set of each sub-waveform in each row or echo waveform; then, each gate length and its corresponding frequency of occurrence are counted; then, they are sorted by frequency, and the gate length with the smallest length is selected from the three gate lengths with the highest frequency, which is recorded as the minimum gate length. .

3. The waveform re-tracking method for a total focusing synthetic aperture radar altimeter according to claim 2, characterized in that: The step 4 comprises: According to the echo waveform gate number and minimum gate length Calculate the number of echo waveform gate segments : （5） Among them, the scheme where the minimum gate length is halved is called the narrow gate scheme, and the scheme where the minimum gate length is not halved is called the wide gate scheme; the narrow gate scheme is suitable for rivers with large widths, and the wide gate scheme is suitable for rivers with small widths; Traverse the gate position of each sub-waveform peak of each echo waveform, determine the segment to which it belongs, count the first three segments with the largest number of sub-waveform peaks and the corresponding number of peaks, and store the starting gate and ending gate of the sub-waveform of each segment and the corresponding echo waveform row number .

4. The waveform re-tracking method for a total focusing synthetic aperture radar altimeter according to claim 3, characterized in that: The step 5 comprises: First, traverse each type of sub-waveform and obtain the starting gate set of each sub-waveform of this type of sub-waveform , End wave gate collection , echo waveform line number , create a temporary row newRow with the same size as a single echo waveform, and set all its values ​​to 0; then assign the value of the sub-waveform to the corresponding gate position of the temporary row newRow, and the corresponding gate position is determined by the starting gate set of the sub-waveform and end gate set Sure; Afterwards, the waveform of the temporary row newRow is re-tracked and calculated using the point target waveform re-tracking method and the threshold method to obtain a high-order sequence of each type of sub-waveform segmentation; The elevation sequence of each type of sub-waveform segment is traversed, the standard deviation and median of the elevation sequence are calculated, and the elevation values ​​exceeding 3 times the standard deviation of the sequence median are removed. The cycle is repeated 3 times to remove outliers.

5. The waveform re-tracking method for a total focusing synthetic aperture radar altimeter according to claim 4, characterized in that: The step 6 comprises: Extract height data, the same orbit number is the same day data, use the nearest neighbor linear interpolation method to filter out the outliers in the single-day height sequence, then calculate the single-day height sequence mean, and use the single-day height sequence mean as the benchmark to remove the original height sequence Abnormal values ​​outside the 8m range, and finally obtain the water levels of several observation points on a single day .

6. The waveform re-tracking method for a total focusing synthetic aperture radar altimeter according to claim 5, characterized in that: The step 7 comprises: For the first three stored segments with the largest number of sub-peaks and the sub-waveforms to which the segments belong, the first segment with the largest number of sub-peaks is adopted as the correct signal segment.

7. A waveform re-tracking device for a total focusing synthetic aperture radar altimeter, characterized in that: Includes the following modules: The data acquisition module selects the daily transit waveform data of the river section in the study sub-area and obtains the data of multiple echo waveforms; The strong sub-peak signal acquisition module extracts multi-scale adaptive sub-peaks and uses a method based on local extreme value scanning to identify multiple strong sub-peak signals of each echo waveform; The local extreme value of the echo waveform is calculated by window moving average, where the window length is The range of changes meets , different window lengths The local extreme value of Determined according to the following formula: （1） in, is a random number that varies in the range [0,1]. represents the first An echo waveform also indicates the row number of the local extreme value matrix where the local extreme value is located in the azimuth direction; represents the echo power waveform, m is the local extreme value, exist and The value of , based on which the local extreme value matrix is ​​constructed ; L=5 is the optimal value for waveform peak detection, L refers to the size of the local extreme value matrix in the azimuth direction of the echo waveform, and N is the size of the local extreme value matrix in the distance direction of the echo waveform; Indicates intersection; is the upward index of the local extremum in the local extremum matrix; Compute local extremum matrix The sum of each row : （2） Local extrema matrix The sum of each row The dimension distribution containing 0 values ​​is then calculated to obtain the local extreme value matrix The sum of each row The row corresponding to the global minimum of , represents the dimension containing the most local maxima, It means finding the parameter value that makes a function reach the minimum value; use Reorganize the local extreme value matrix M by eliminating all those that satisfy Elements , and obtain a new local extremum matrix ; By calculating the new local extremum matrix The standard deviation of each column To determine the peak value of the waveform, where the standard deviation of each column is satisfied Index That is, the gate number of the detected peak value, and the detected peak value is set as the end gate of the sub-waveform , for each echo waveform, get the end gate set of a single sub-waveform ; When the power value is greater than 0.003 When , the detected peak value is considered valid; （3） （4） in, Indicates the threshold value; To determine the starting gate of each sub-waveform, the echo waveform is normalized by dividing it by the maximum power value; then the power difference of the continuous gates of the normalized waveform is calculated to obtain the power difference waveform; finally, the ending gate of each sub-waveform is determined. , search the power difference waveform from back to front, the first gate with a power value lower than 0.001 is the starting gate of the corresponding sub-waveform ; End gate All sub-waveform ending gates in the same way are searched, and finally the sub-waveform starting gate set is calculated. , and the end gate set Together they form N sub-waveforms; A statistical module for counting sub-waveform information, wherein the sub-waveform information includes a gate length, a starting gate position, and an ending gate position of a strong sub-wave peak signal of each sub-waveform; The segmentation processing module classifies and counts the strong sub-wave peak signals according to the wave gate length, obtains the minimum wave gate length, calculates the number of segments according to the minimum wave gate length, performs sub-waveform segmentation processing, and obtains segmented sub-waveforms and sub-waveform peaks; The re-tracking module re-tracks various types of segmented sub-waveforms according to the starting wave gate position, the ending wave gate position, and the sub-waveform peak value, that is, re-tracks each type of segmented sub-waveform and calculates its corresponding height; the height is the elevation value in the elevation sequence; The outlier removal module removes outliers in the daily high-order series; The sequence acquisition module obtains the elevation sequence of each segmented sub-waveform, and the elevation sequences of different segmented sub-waveforms represent the elevation sequences from different landforms; the segmented water level of the river section in the study sub-area is selected to obtain the water level sequence of the river surface of the river section in the study sub-area.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of a waveform re-tracking method for a total focusing synthetic aperture radar altimeter described in any one of claims 1-6 are implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a waveform re-tracking method for a total focusing synthetic aperture radar altimeter described in any one of claims 1-6 are implemented.

Images