A method and computing device for geometric calibration of a ground-based lunar radar

Abstract

The application relates to the technical field of radar calibration, and provides a ground-based lunar radar geometric calibration method. The method comprises the following steps: processing radar echo original data into a first radar image, each pixel in the first radar image having a range-Doppler coordinate; performing coordinate projection on each pixel in the first radar image to obtain the geographic coordinates of each pixel; performing homonym matching according to the geographic coordinates of each pixel to obtain a plurality of homonym point pairs; each homonym point pair comprises an oblique range-Doppler coordinate and a lunar surface optical image coordinate; performing geometric calibration according to the plurality of homonym point pairs to determine correction parameters; the correction parameters comprise a radar lower point oblique range correction amount, a radar lower point Doppler frequency correction amount, a range direction sampling frequency correction amount and a pulse repetition frequency correction amount; and obtaining corrected lunar surface geographic longitude and latitude coordinates according to the correction parameters. The application solves the correction amount from the radar parameter level, thereby realizing high-precision geographic positioning error correction that is universal in the full field of view.

Description

Technical Field

[0001] This application relates to the field of radar calibration technology, and in particular to a ground-based lunar radar geometric calibration method and computing device. Background Technology

[0002] Ground-based lunar radar imaging is of great significance in lunar observation and exploration research. Compared with spaceborne observation systems, ground-based radar has advantages such as flexible deployment, lower cost, and the ability to achieve long-term continuous observation. It can acquire high-resolution information on the lunar surface scattering characteristics and topography without relying on spacecraft platforms.

[0003] The observation geometry of ground-based lunar radar imaging differs significantly from that of spaceborne radar. First, the ground-based lunar radar beam covers the entire lunar hemisphere, rendering the field-of-view plane assumption invalid and requiring the consideration of the relative motion between the Earth and the Moon to achieve focused imaging. Second, the vast distance between the Earth and the Moon results in a round-trip time of approximately 2.5 seconds for the radar signal, necessitating consideration of the influence of optical travel time, differentiation of signal transmission and reception locations, and correction of reception time differences at different locations on the lunar surface. Finally, the Moon's apparent rotation speed is extremely small, requiring a longer synthetic aperture time for focusing, which introduces significant range migration errors that need to be corrected using a high-order slant range history model.

[0004] Due to the aforementioned observation geometry characteristics, radar parameter errors such as initial slant range error and azimuth time error can lead to complex geolocation errors. These errors vary in magnitude and direction at different locations within the field of view, making it difficult to effectively correct them using common image registration methods such as translation and stretching.

[0005] Therefore, there is an urgent need to develop a geometric calibration method based on the geometry of ground-based lunar radar observations to achieve high-precision geolocation error correction applicable across the entire field of view. Summary of the Invention

[0006] In view of this, this application provides a ground-based lunar radar geometric calibration method and computing device, which solves the correction amount from the radar parameter level, thereby achieving high-precision geolocation error correction applicable to the entire field of view.

[0007] In a first aspect, this application provides a geometric calibration method for a ground-based lunar radar, the method comprising: processing raw radar echo data into a first radar image, wherein each pixel in the first radar image has range Doppler coordinates. The geographic coordinates of each pixel in the first radar image are obtained by performing coordinate projection. Based on the geographic coordinates of each pixel, perform corresponding point matching to obtain multiple pairs of corresponding points; each pair of corresponding points includes slant range Doppler coordinates. The coordinates of the lunar surface optical image are The correction parameters are determined by geometric calibration based on multiple pairs of corresponding points; the correction parameters include the radar slant range correction. Radar Doppler frequency correction Range-direction sampling frequency correction amount Pulse repetition frequency correction amount The corrected lunar surface geographic latitude and longitude coordinates are obtained based on the correction parameters. .

[0008] In some possible implementations, the geographic coordinates of each pixel in the first radar image are obtained by coordinate projection, including: obtaining the center time through ephemeris. radar bottom coordinates and the apparent rotational angular velocity vector of the moon Radar lower point coordinates With the lunar apparent rotation angular velocity vector All are defined in the MCMF coordinate system; according to and Determine the coordinate transformation matrix from the observation coordinate system to the MCMF coordinate system; the origin of the observation coordinate system is located at the lunar center of mass, with... Pointing to the x-axis, with The z-axis is used as the reference point, and the y-axis is perpendicular to both the x and z axes, forming a right-handed coordinate system. The range Doppler coordinates of a specified pixel in the first radar image are then used. Convert to rectangular coordinates in the observation coordinate system Based on the coordinate transformation matrix and the rectangular coordinates in the observation coordinate system, the MCMF coordinates of a specified pixel are obtained. Perform coordinate rotation to obtain the geographic coordinates of a specified pixel. .

[0009] In some possible implementations, multiple pairs of corresponding points are obtained by matching corresponding points based on the geographic coordinates of each pixel. This includes: extracting corresponding feature points from lunar optical images and second radar images respectively; the second radar image is a radar image with coordinate projection completed; converting the corresponding feature points from image coordinates to latitude and longitude coordinates, then performing inverse projection, and recalculating their corresponding range-Doppler coordinates; modeling the range-Doppler coordinates using linear transformation to obtain a geometric transformation model; and selecting multiple pairs of corresponding points from candidate feature points based on the geometric transformation model.

[0010] In some possible implementations, corresponding feature point pairs are extracted from the lunar optical image and the second radar image, respectively, including: using the SIFT algorithm to determine the first structural feature in the second radar image and the second structural feature in the lunar optical image; constructing a multi-layer pyramid image in Gaussian scale space based on the first and second structural features; detecting candidate keypoints for the multi-layer pyramid image using the differential Gaussian operator; performing sub-pixel-level localization on the candidate keypoints and eliminating unstable points, then calculating the gradient direction and magnitude distribution in the neighborhood of the keypoints to obtain feature vectors; measuring the similarity between the first and second structural features based on the Euclidean distance of the feature vectors, and selecting corresponding point pairs from the lunar optical image and the second radar image.

[0011] In some possible implementations, a geometric transformation model is obtained by modeling the distance Doppler coordinates using a linear transformation, including: randomly selecting a first sample from the set of matching points; calculating a geometric transformation matrix based on the first sample; determining the Doppler coordinate distance of each pair of matching points in the first sample based on the geometric transformation matrix; determining "interior points" and "outer points" based on the Doppler coordinate distance; interior points are matching point pairs whose distance to each other is less than a distance threshold after the geometric transformation, and outer points are matching point pairs whose distance to each other is greater than a distance threshold after the geometric transformation; removing outer points from the first sample to obtain a second sample set; and optimizing the geometric transformation model based on the second sample set.

[0012] In some possible implementations, geometric calibration is performed based on multiple sets of corresponding point pairs, including: using the slant range Doppler coordinates of one set of corresponding points in the second radar image. and coordinates in lunar optical images Construct a geometric calibration model:

[0013] In the formula The slant range Doppler coordinates of the radar's lower point. This is the slant range correction value for the radar lower point. For radar down-point Doppler frequency correction, For range-direction sampling frequency correction, This is the pulse repetition frequency correction value.

[0014] In some possible implementations, determining the correction parameters through geometric calibration based on multiple sets of corresponding point pairs further includes: establishing an error equation based on each set of corresponding points; and defining the error equation for each pair of corresponding points. When the minimum value is reached, the residuals are solved using the least squares adjustment method; the correction amount is obtained based on the condition that the residuals are less than a preset threshold. According to the correction amount Obtain the radar slant range correction amount. Radar Doppler frequency correction Range-direction sampling frequency correction amount Pulse repetition frequency correction amount .

[0015] In some possible implementations, the corrected lunar surface geographic latitude and longitude coordinates are obtained based on the correction parameters. This includes: obtaining the corrected radar image slant range Doppler coordinates based on the correction parameters. :

[0016]

[0017]

[0018] Slant range Doppler coordinates of the corrected radar image The corrected lunar geographic latitude and longitude coordinates are obtained by performing coordinate projection. .

[0019] Currently, China possesses world-class radar observation stations such as FAST, SYISR, and QJISR, enabling high-precision lunar imaging. This provides the hardware foundation for the application of this method, and its implementation will effectively contribute to the production of high-precision lunar radar image products.

[0020] In a second aspect, this application provides a computing device, comprising: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute the method as described in any one of the first aspects.

[0021] Thirdly, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to perform the method as described in any one of the first aspects. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the various embodiments disclosed in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only a few embodiments disclosed in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] The accompanying drawings used in the description of the embodiments or prior art are briefly introduced below.

[0024] Figure 1 A schematic flowchart illustrating a ground-based lunar radar geometric calibration method provided in this application embodiment;

[0025] Figure 2 This is a schematic diagram of the matching results of corresponding points provided in the embodiments of this application;

[0026] Figure 3 This is a schematic diagram illustrating the geometric calibration effect in the slant-range Doppler domain provided in an embodiment of this application.

[0027] Figure 4 A ground-based lunar radar geometric calibration device is provided as an embodiment of this application;

[0028] Figure 5 A computing device provided in an embodiment of this application. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be described below with reference to the accompanying drawings.

[0030] In the description of the embodiments of this application, the words "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a specific manner.

[0031] In the description of the embodiments in this application, the term "and / or" is merely a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, and A and B existing simultaneously. Furthermore, unless otherwise stated, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple terminals refer to two or more terminals.

[0032] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0033] In the description of the embodiments in this application, "some embodiments" are mentioned, which describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0034] In the description of the embodiments of this application, the terms "first, second, third, etc." or module A, module B, module C, etc. are used only to distinguish similar objects and do not represent a specific ordering of objects. It is understood that, where permitted, a specific order or sequence can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0035] In the description of the embodiments of this application, the reference numerals for the steps, such as S110, S120, etc., do not necessarily indicate that the steps will be executed in this manner. Where permissible, the order of the steps can be interchanged or executed simultaneously.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0037] Existing technologies have certain shortcomings in the geometric calibration of lunar radar images. For example, the paper "GeometricCorrection of Bistatic SAR Moon Mapping Results Based on the RPC Model" proposes to use the RPC model for lunar radar image registration, but this method does not fully consider the observation geometry and only relies on the RPC model for geometric correction. Therefore, it is not suitable for correcting image geolocation errors under full field of view and large-scale conditions.

[0038] In another methodological example, to address the range alignment problem in lunar imaging by ground-based radar, the relative motion effect in the range direction is eliminated through steps such as pulse compression and point selection fitting. Although this method is related to RDM image processing by ground-based lunar radar, it does not involve the analysis of observational geometric factors and the establishment of high-order slant range history models required for geometric calibration. It lacks the ability to systematically model and quantitatively correct the sources of geometric errors, and therefore cannot meet the requirements for high-precision geometric calibration of lunar radar imaging.

[0039] This application proposes a geometric calibration method for ground-based lunar radar. A corresponding plugin has been developed that, based on a ground-based lunar radar imaging program, automatically obtains radar parameter corrections and the lunar latitude and longitude coordinates corresponding to each pixel in the radar image by inputting radar RDM images, radar imaging parameters, and a reference lunar DOM image. The radar imaging parameters include the range sampling frequency and the pulse repetition frequency; the radar parameter corrections include radar slant range correction, radar slant Doppler frequency correction, range sampling frequency correction, and pulse repetition frequency correction. The radar RDM image can be designated as the first radar image, and the lunar DOM image as the lunar optical image.

[0040] Figure 1 A flowchart illustrating the geometric calibration method for ground-based lunar radar provided in this application embodiment. Figure 1 As shown, it includes the following steps S11-S15.

[0041] S11, the raw radar echo data is processed into a first radar image, and each pixel in the first radar image has range Doppler coordinates.

[0042] In some possible implementations, the raw radar echo data can be processed into a first radar image through steps such as range pulse compression, range migration and Doppler frequency shift compensation, and azimuth FFT coherent accumulation, thereby achieving range-Doppler imaging. Each pixel in the image has range-Doppler coordinates. .

[0043] S12, Perform coordinate projection on each pixel in the first radar image to obtain the geographic coordinates of each pixel. .

[0044] To obtain the geographic coordinates of a pixel, coordinate projection is required. In some possible implementations, step S12 includes the following steps:

[0045] S121, first obtain the center time through ephemeris. radar bottom coordinates and the apparent rotational angular velocity vector of the moon Due to the complexity of the moon's apparent motion, The following numerical approximation method is generally used for calculation:

[0046] (1)

[0047] The radar lower point coordinates in equation (1) With the lunar apparent rotation angular velocity vector All are defined in the MCMF coordinate system.

[0048] S122, then according to and Determine the coordinate transformation matrix from the observation coordinate system to the MCMF coordinate system.

[0049] The origin of the observation coordinate system is located at the Moon's center of mass, with... Pointing as Axis, with Pointing to the z-axis, making The axes point perpendicular to the x and z axes, forming a right-handed system.

[0050] The coordinate transformation matrix from the observation coordinate system to the MCMF coordinate system. It can be written as:

[0051] (2)

[0052] The coordinate transformation matrix can be obtained. .

[0053] S123, the range Doppler coordinates of the specified pixel in the first radar image. Convert to rectangular coordinates in the observation coordinate system .

[0054] Distance Doppler coordinates of a given pixel in the first radar image It can be converted to rectangular coordinates in the observation coordinate system:

[0055]

[0056] In the formula The average radius of the moon is 1738 km. The center wavelength of the radar. This is the apparent rotational angular velocity of the moon. The distance-Doppler coordinates of the radar sub-point can be directly obtained by extracting the coordinates of the minimum slant range location from the first radar image. When observing the northern hemisphere of the moon, Calculations are taken as positive signs, while observations in the Southern Hemisphere are taken as negative signs.

[0057] S124, based on coordinate transformation matrix and rectangular coordinates in the observation coordinate system MCMF coordinates of a specified pixel Perform coordinate rotation to obtain the geographic coordinates of a specified pixel. .

[0058] Specifically, the rectangular coordinates of the observation coordinate system are obtained. Then, for MCMF coordinates Perform the following coordinate rotations directly:

[0059]

[0060] The coordinate projection is then completed in the following manner:

[0061]

[0062]

[0063] In the formula Longitude The coordinates are represented by latitude. The RDM image with the coordinate projection completed is denoted as the second radar image.

[0064] S13, perform corresponding point matching based on the geographic coordinates of each pixel to obtain multiple pairs of corresponding points; the multiple pairs of corresponding points include slant range Doppler coordinates. Coordinates in lunar optical DOM images Specifically, it includes the following steps S131-S134.

[0065] S131, during the matching of corresponding points, feature points are extracted from the lunar surface optical image and the second radar image respectively.

[0066] To ensure stable feature extraction across heterogeneous images, in some possible implementations, S131 may take the following steps to extract feature points:

[0067] S1311, First, the SIFT algorithm is used to select the structural features of two images.

[0068] For example, the SIFT algorithm can be used to select the first structural features of the lunar surface optical image, and the SIFT algorithm can be used to select the second structural features of the second radar image.

[0069] S1312, construct a multi-layer pyramid image in Gaussian scale space based on the first and second structural features, detect candidate key points by differential Gaussian operator, perform sub-pixel level precise positioning of candidate key points and remove unstable points, and then calculate the gradient direction and magnitude distribution in the neighborhood of candidate key points to form a 128-dimensional feature vector.

[0070] S1313 Finally, the Euclidean distance between feature vectors is used to measure the similarity between the first and second structural features, and stable and reliable feature point pairs are selected simultaneously from lunar optical images and second radar images.

[0071] This lays the foundation for subsequent geometric matching.

[0072] Due to the influence of complex observation geometry, corresponding point pairs in lunar optical images and second radar images differ significantly in position and orientation. To reduce the projection bias caused by geometric differences, the following steps S132 can be performed.

[0073] S132 converts the feature point pairs from image coordinates to latitude and longitude coordinates, then performs inverse projection and recalculates their corresponding distance-Doppler coordinates.

[0074] This effectively reduces nonlinear distortion caused by differences in imaging perspective and projection transformation, and improves the consistency of matching points in spatial positioning.

[0075] S133, a geometric transformation model is obtained by using linear transformation based on distance Doppler coordinates.

[0076] In the range Doppler domain, the positioning error caused by differences in radar parameters mainly manifests as translation, scaling, or slight rotation. In the method provided in this application, this positioning error can be modeled using linear transformation.

[0077] To select true pairs of identical points from candidate feature points, the RANSAC algorithm can be used for robust matching.

[0078] This algorithm can effectively eliminate outliers by random sampling and model fitting in the presence of mismatches, thereby obtaining a stable geometric transformation model.

[0079] In some possible implementations, S133 performs modeling through the following steps:

[0080] S1331, repeatedly select the first sample from the matching point set and calculate the geometric transformation matrix.

[0081] S1332, determine the Doppler coordinate distance of each pair of matching points in the first sample according to the geometric transformation matrix, and determine the "inner point" and "outer point" according to the Doppler coordinate distance of each pair of matching points.

[0082] Matching point pairs whose distance to each other after geometric transformation is less than a distance threshold are designated as interior points; matching point pairs whose distance to each other after geometric transformation is greater than a distance threshold are designated as exterior points.

[0083] S1333, remove outliers from the first sample to obtain the second sample set; recalculate the geometric transformation matrix based on the second sample set to obtain the optimized geometric transformation model.

[0084] Through this iterative screening and optimization process, the geometric transformation model obtained by the RANSAC algorithm can significantly reduce the impact of erroneous matching pairs and ensure reliable matching of corresponding points under complex observation geometry conditions.

[0085] S134, based on the optimized geometric transformation model, corresponding point matching can obtain multiple pairs of corresponding points for geometric calibration.

[0086] For example, Figure 2 This is a schematic diagram illustrating the matching results of corresponding points provided in an embodiment of this application. For example... Figure 2 As shown, (a) is a schematic diagram of the coordinates of multiple sets of corresponding points in the lunar optical DOM image, and (b) is a schematic diagram of the slant range Doppler coordinates of multiple sets of corresponding points after removing outliers.

[0087] S14. Perform geometric calibration based on multiple pairs of corresponding points to determine the correction parameters.

[0088] Let the slant range Doppler coordinates of the corresponding point in the second radar image be . In the lunar optical image, the coordinates are .

[0089] In some possible implementations, step S14 determines the correction parameters through the following steps:

[0090] S141, based on the slant range Doppler coordinates of one set of corresponding points in the second radar image from a plurality of sets of corresponding points. and coordinates in the lunar optical image Construct the following geometric calibration model:

[0091]

[0092] In the formula The slant range Doppler coordinates of the radar's lower point. This is the slant range correction value for the radar lower point. For radar down-point Doppler frequency correction, For range-direction sampling frequency correction, This is the pulse repetition frequency correction value.

[0093] S142, For each group of points with the same name, an error equation can be further established. :

[0094]

[0095] in: These are the linear transformation coefficients. The residual is the result of the linear transformation.

[0096]

[0097]

[0098] S143, Error equation for pairs of corresponding points When the minimum value is reached, the residuals are solved using the least squares adjustment method.

[0099] Specifically, Newton's iteration method can be further constructed to... As an unknown quantity, the initial value Set as The iterative formula is:

[0100]

[0101] Among residuals Solve using the least squares adjustment method to obtain the error equations for all pairs of corresponding points. It reaches the minimum value.

[0102] S144, when the residual If the result is less than a preset threshold, it is considered that the iteration has converged, and a correction value is output. According to the correction amount Obtain the radar slant range correction amount. Radar Doppler frequency correction Range-direction sampling frequency correction amount Pulse repetition frequency correction amount .

[0103] For example, Figure 3 This is a schematic diagram illustrating the geometric calibration effect within the slant-range Doppler domain provided in an embodiment of this application. Figure 3 As shown, (a) shows the slant range Doppler coordinates of the radar image before correction, and (b) shows the slant range Doppler coordinates of the radar image after correction.

[0104] S15, Obtain the corrected lunar surface geographic latitude and longitude coordinates based on the correction parameters. .

[0105] The second radar image can be corrected first based on the correction parameters. The slant range Doppler coordinates of the corrected radar image are then obtained. The following constraints must be met:

[0106]

[0107]

[0108] Slant range Doppler coordinates of the corrected radar image The corrected lunar geographic latitude and longitude coordinates are obtained by performing coordinate projection again. .

[0109] This application addresses the core problem of the difficulty in effectively correcting geographic positioning errors caused by the large field of view and long observation distance of ground-based lunar radar. By establishing a high-order slant range history model and coordinate projection method based on observation geometry, the correction amounts of slant range, Doppler frequency, sampling frequency and pulse repetition frequency are solved from the source of radar parameters, so as to achieve high-precision geometric calibration applicable to the entire field of view.

[0110] Nonlinear distortion caused by differences in imaging perspective is reduced by transforming latitude and longitude coordinates and inverse projection; an automated solution process for the geometric calibration model is constructed, which can directly input the first radar image, imaging parameters and reference DOM image to output the correction results, thereby improving the practicality and efficiency of the calibration method.

[0111] This application proposes a geometric calibration method based on the geometry of ground-based lunar radar observations, which specifically addresses the unique geometry and difficulty in error correction inherent in ground-based lunar radar imaging observations. This method solves for correction quantities from the source of radar parameters, eliminating geometric distortions caused by wide beam coverage, significant optical travel time effects, and prominent range migration errors. It also lays the foundation for subsequent data processing, deepens the understanding of ultra-long-range hemispherical scale imaging mechanisms, facilitates the production of high-precision lunar radar image products, and provides key technical support for lunar exploration projects.

[0112] The above is an introduction to the voice quality assessment method provided by the embodiments of this application. It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. In addition, in some possible implementations, each step in the above embodiments may be selectively executed according to the actual situation, and may be partially or fully executed, without limitation here. Furthermore, all or part of any feature of any of the above embodiments may be freely and arbitrarily combined without contradiction; the combined technical solution is also within the scope of this application.

[0113] Next, based on the above, the porous piezoelectric dielectric acoustic wave propagation device provided in the embodiments of this application will be described. For details regarding the concepts, formulas, etc., involved in the following content, please refer to the above text.

[0114] Figure 4 An apparatus is provided as an embodiment of this application. For example... Figure 4 As shown, the device 40 includes an image processing module 41, a coordinate projection module 42, a corresponding point pairing module 43, a geometric calibration module 44, and a correction module 45.

[0115] In the device, the image processing module 41 processes the raw radar echo data into a first radar image, where each pixel in the first radar image has a range Doppler coordinate. ;

[0116] The coordinate projection module 42 performs coordinate projection on each pixel in the first radar image to obtain the geographic coordinates of each pixel. ;

[0117] The corresponding point matching module 43 performs corresponding point matching based on the geographic coordinates of each pixel to obtain multiple pairs of corresponding points; each pair of corresponding points includes slant range Doppler coordinates. The coordinates of the lunar surface optical image are ;

[0118] The geometric calibration module 44 performs geometric calibration based on the multiple sets of corresponding point pairs to determine the correction parameters; the correction parameters include the radar down-point slant range correction amount. Radar Doppler frequency correction Range-direction sampling frequency correction amount Pulse repetition frequency correction amount

[0119] The correction module 45 obtains the corrected lunar surface geographic latitude and longitude coordinates based on the correction parameters. .

[0120] As an example of a software functional unit, the image processing module 41 may include code running on a computing instance. The computing instance may include at least one of a physical host (computing device), a virtual machine, or a container. Furthermore, the aforementioned computing instance may be one or more. For example, the image processing module 41 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed within the same region or in different regions. Further, the multiple hosts / virtual machines / containers used to run the code may be distributed within the same availability zone (AZ) or in different AZs, each AZ including one or more geographically proximate data centers. Typically, a region may include multiple AZs.

[0121] Similarly, multiple hosts / virtual machines / containers used to run this code can be distributed within the same Virtual Private Cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Communication between two VPCs within the same region, as well as between VPCs in different regions, requires a communication gateway to be set up within each VPC to enable interconnection between VPCs.

[0122] This application also provides a computing device 50. For example... Figure 5 As shown, the computing device 50 includes a bus 52, a processor 54, a memory 56, and a communication interface 58. The processor 54, the memory 56, and the communication interface 58 communicate with each other via the bus 52. The computing device 50 can be a computing device or a terminal device. It should be understood that this application does not limit the number of processors and memories in the computing device 50.

[0123] The bus 52 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 5 The bus 54 is represented by a single line, but this does not mean that there is only one bus or one type of bus. The bus 54 may include a path for transmitting information between various components of the computing device 50 (e.g., memory 56, processor 54, communication interface 58).

[0124] The processor 54 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0125] The memory 56 may include volatile memory, such as random access memory (RAM). The processor 54 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0126] The memory 56 stores executable program code, and the processor 54 executes the executable program code to implement the aforementioned functions respectively. Figure 4 The image processing module 41, coordinate projection module 42, corresponding point pairing module 43, geometric calibration module 44, and correction module 45 shown in the diagram perform their functions to implement all or part of the steps of the method in the above embodiments. That is, the memory 56 stores instructions for executing all or part of the steps in the method of the above embodiments.

[0127] The communication interface 58 uses transceiver modules such as, but not limited to, network interface cards and transceivers to enable communication between the computing device 50 and other devices or communication networks.

[0128] This application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the ground-based lunar radar geometric calibration method as described in any of the embodiments provided in this application.

[0129] It is understood that the processor in the embodiments of this application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0130] The method steps in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0131] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0132] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application.

Claims

1. A geometric calibration method for ground-based lunar radar, characterized in that, The method includes: The raw radar echo data is processed into a first radar image, where each pixel has a range Doppler coordinate. ; The geographic coordinates of each pixel in the first radar image are obtained by performing coordinate projection. ; Multiple pairs of corresponding points are obtained by performing corresponding point matching based on the geographic coordinates of each pixel; each pair of corresponding points includes slant range Doppler coordinates. The coordinates of the lunar surface optical image are ; Correction parameters are determined by geometric calibration based on the multiple sets of corresponding point pairs; the correction parameters include the radar slant range correction amount. Radar Doppler frequency correction Range-direction sampling frequency correction amount Pulse repetition frequency correction amount The corrected lunar surface geographic latitude and longitude coordinates are obtained based on the correction parameters. .

2. The method according to claim 1, characterized in that, The geographic coordinates of each pixel in the first radar image are obtained by projecting the coordinates of each pixel. ,include: Obtain the central time through ephemeris radar bottom coordinates and the apparent rotational angular velocity vector of the moon The radar lower point coordinates With the lunar apparent rotation angular velocity vector All are defined in the MCMF coordinate system; According to the above and Determine the coordinate transformation matrix from the observation coordinate system to the MCMF coordinate system; the origin of the observation coordinate system is located at the lunar center of mass, with the... Pointing as the x-axis, with the aforementioned The z-axis is pointed out, and the y-axis is pointed out perpendicular to the x and z axes, forming a right-handed system; The distance Doppler coordinates of a specified pixel in the first radar image Convert to rectangular coordinates in the observation coordinate system ; Based on the coordinate transformation matrix and the rectangular coordinates in the observation coordinate system, the MCMF coordinates of the specified pixel are... Perform coordinate rotation to obtain the geographic coordinates of the specified pixel. .

3. The method according to claim 1, characterized in that, The process of obtaining multiple pairs of corresponding points by matching corresponding points based on the geographic coordinates of each pixel includes: Feature points with the same name are extracted from the lunar surface optical image and the second radar image respectively; the second radar image is a radar image with coordinate projection completed. The feature points with the same name are converted from image coordinates to latitude and longitude coordinates, and then inverse projection is performed to calculate their corresponding distance Doppler coordinates. A geometric transformation model is obtained by modeling the distance Doppler coordinates using a linear transformation. Based on the geometric transformation model, the multiple pairs of points with the same name are selected from the candidate feature points.

4. The method according to claim 3, characterized in that, The extraction of corresponding feature points from the lunar surface optical image and the second radar image includes: The SIFT algorithm was used to determine the first structural feature in the second radar image and the second structural feature in the lunar optical image. Construct a multi-layer pyramid image in Gaussian scale space based on the first and second structural features; For the multi-layer pyramid image, candidate key points are detected using the differential Gaussian operator; Subpixel-level localization is performed on the candidate key points and unstable points are eliminated. Then, the gradient direction and magnitude distribution are calculated in the neighborhood of the key points to obtain the feature vector. The similarity between the first and second structural features is measured by the Euclidean distance of the feature vectors, and corresponding points are selected from lunar optical images and second radar images.

5. The method according to claim 3, characterized in that, The process of modeling the distance Doppler coordinates using a linear transformation to obtain a geometric transformation model includes: Randomly select a first sample from the set of matching points, and calculate the geometric transformation matrix based on the first sample; The Doppler coordinate distance of each pair of matching points in the first sample is determined based on the geometric transformation matrix; "Interior points" and "outer points" are determined based on the Doppler coordinate distance; the interior points are matching point pairs whose distance to each other is less than a distance threshold after geometric transformation, and the outer points are matching point pairs whose distance to each other is greater than a distance threshold after geometric transformation. The second sample set is obtained by removing the outliers from the first sample; The geometric transformation model is optimized based on the second sample set.

6. The method according to claim 1, characterized in that, Based on the multiple sets of corresponding point pairs, geometric calibration is performed to determine the correction parameters, including: The slant range Doppler coordinates of one of the multiple pairs of corresponding points in the second radar image. and coordinates in the lunar optical image Input the geometric calibration model and output the correction parameters; the geometric calibration model is as follows: In the formula The slant range Doppler coordinates of the radar's lower point. This is the slant range correction value for the radar lower point. For radar down-point Doppler frequency correction, For range-direction sampling frequency correction, This is the pulse repetition frequency correction value.

7. The method according to claim 6, characterized in that, The step of determining the correction parameters by geometric calibration based on the multiple sets of corresponding point pairs further includes: Establish an error equation based on each pair of points with the same name; The error equation of the corresponding point pair When the minimum value is reached, the residuals are solved using the least squares adjustment method. The correction amount is obtained based on the condition that the residual is less than a preset threshold. ; According to the correction amount The radar slant range correction amount is obtained. Radar Doppler frequency correction Range-direction sampling frequency correction amount Pulse repetition frequency correction amount .

8. The method according to claim 7, characterized in that, The corrected lunar surface geographic latitude and longitude coordinates are obtained based on the correction parameters. ,include: The corrected radar image slant range Doppler coordinates are obtained based on the correction parameters. : The corrected radar image slant range Doppler coordinates The corrected lunar geographic latitude and longitude coordinates are obtained by performing coordinate projection. .

9. A computing device, characterized in that, The device includes: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to perform the method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a processor, causes the processor to perform the method as described in any one of claims 1-8.

Images