A rapid footprint generation method for remote sensing images based on RPC

Abstract

The present application relates to remote sensing image footprint technical field, especially to a kind of based on RPC's remote sensing image fast footprint generation method.The method includes the following steps: based on the remote sensing image to be handled and its metadata, image size reading and RPC parameter analysis are carried out, obtain image size data and RPC parameter data;RPC model is constructed based on RPC parameter data;Based on RPC model and image size data, the geographic coordinates of three reference corner points of image are calculated, and the geographic coordinate data of three reference corner points is obtained;Based on reference corner point geographic coordinate data, orbit radian offset factor is calculated, and orbit radian offset factor data is obtained;nonlinear calculation is carried out based on orbit radian offset factor data three reference corner points and image size data, the geographic coverage range vector data of remote sensing image is generated in the present application, to realize the normalized expression of image coverage range, it is convenient to be quickly loaded and applied in geographic information system.

Description

Technical Field

[0001] This invention relates to the field of remote sensing image footprint technology, and in particular to a method for rapid footprint generation of remote sensing images based on RPC. Background Technology

[0002] With the continuous growth in the scale of remote sensing image data acquisition, how to quickly and accurately generate the geographic coverage (i.e., image footprint) of images has become a fundamental problem in remote sensing data processing and management. Current technologies typically rely on the rational polynomial coefficient (RPC) model attached to the image, performing back-projection calculations on each of the four corner points of the image to obtain the corresponding geographic coordinates, thereby constructing the spatial coverage of the image. Since the RPC model is inherently a non-linear mapping relationship, the back-calculation from image coordinates to geographic coordinates usually requires an iterative solution, which is complex and time-consuming, especially in large-scale image batch processing or real-time applications, easily becoming a processing bottleneck. Furthermore, in high-latitude regions or when the image span is large, the curvature of the Earth and the convergence effect of meridians can lead to significant errors in simple geometric calculations, further affecting footprint accuracy. Therefore, how to reduce the computational load of RPC back-projection and improve the efficiency of footprint generation while ensuring geographic positioning accuracy has become a pressing technical problem to be solved in this field. Summary of the Invention

[0003] Therefore, it is necessary to provide a method for rapid footprint generation of remote sensing images based on RPC to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for rapid footprint generation from remote sensing images based on RPC is proposed, comprising the following steps: Step S1: Based on the remote sensing image to be processed and its metadata, perform image size reading and RPC parameter parsing to obtain image size data and RPC parameter data; Step S2: Construct an RPC model based on RPC parameter data; Step S3: Based on the RPC model and image size data, calculate the geographic coordinates of the three reference corner points of the image to obtain the geographic coordinate data of the three reference corner points; Step S4: Calculate the orbital radian offset factor based on the geographic coordinate data of the three reference corner points to obtain the orbital radian offset factor data; Step S5: Nonlinear calculations are performed on the base orbit radian offset factor data, the geographic coordinate data of the three reference corner points, and the image size data to obtain the geographic coordinate data of the fourth corner point; Step S6: Integrate the geographic coordinate data of the three reference corner points and the fourth corner point to generate and output the geographic coverage vector data of the remote sensing image.

[0005] This invention analyzes the RPC parameters of remote sensing images and constructs a back-projection model. It performs geographic coordinate calculations only on three reference corner points, and then introduces an orbital arc offset factor to nonlinearly extrapolate the fourth corner point. This avoids point-by-point RPC back-projection calculations for all corner points, significantly reducing overall computational complexity and processing time, and improving the efficiency of remote sensing image footprint generation. Simultaneously, by using the latitude cosine ratio to correct the longitude span, it effectively compensates for geometric distortions caused by the Earth's curvature and meridian convergence, maintaining high footprint positioning accuracy while reducing the number of calculations. Furthermore, by uniformly constructing closed polygons and outputting standard vector formats, it achieves a standardized expression of image coverage, facilitating rapid loading and application in geographic information systems. Therefore, this invention achieves high efficiency and improved engineering practicality in the remote sensing image footprint generation process while ensuring computational accuracy. Attached Figure Description

[0006] Figure 1 This is a flowchart illustrating the steps of a method for rapid footprint generation from remote sensing images based on RPC. Figure 2 This is the original remote sensing image; Figure 3 The original remote sensing image is overlaid on the result; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0007] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0008] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0009] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0010] To achieve the above objectives, please refer to Figures 1 to 3 A method for rapid footprint generation from remote sensing images based on RPC includes the following steps: All specific values ​​involved in this embodiment are exemplary parameters used to clearly illustrate the technical operation process and are not the only limitation of the present invention.

[0011] Step S1: Based on the remote sensing image to be processed and its metadata, perform image size reading and RPC parameter parsing to obtain image size data and RPC parameter data; Step S2: Construct an RPC model based on RPC parameter data; Step S3: Based on the RPC model and image size data, calculate the geographic coordinates of the three reference corner points of the image to obtain the geographic coordinate data of the three reference corner points; Step S4: Calculate the orbital radian offset factor based on the geographic coordinate data of the three reference corner points to obtain the orbital radian offset factor data; Step S5: Nonlinear calculations are performed on the base orbit radian offset factor data, the geographic coordinate data of the three reference corner points, and the image size data to obtain the geographic coordinate data of the fourth corner point; Step S6: Integrate the geographic coordinate data of the three reference corner points and the fourth corner point to generate and output the geographic coverage vector data of the remote sensing image.

[0012] In this embodiment, remote sensing image data to be processed and its corresponding metadata file are acquired, wherein the metadata file contains an RPC parameter file. After acquiring the data, a size reading operation is performed on the remote sensing image to extract the image's width and height, thereby forming image size data for subsequent determination of corner image coordinates. Simultaneously, the RPC parameter file is parsed to extract rational polynomial coefficients, normalization parameters, and offset parameters, constructing RPC parameter data in a unified format. Through the above processing, image size data and RPC parameter data are output. The RPC parameter data is input to step S2 for RPC model construction, while the image size data is retained for determining corner image coordinates in step S3.

[0013] Based on the RPC parameter data obtained in step S1, a back-projection positioning model from image space to ground space is constructed. A nonlinear mapping relationship between image coordinates and geographic coordinates is established according to the rational polynomial coefficients in the RPC parameters, and an iterative solution mechanism for calculating geographic coordinates from image coordinates is configured. During the back-projection solution, the initial value of the geographic coordinates is set to the normalized center point, and this initial value is substituted into the RPC model to calculate the corresponding image coordinates. Subsequently, by calculating the residual between the calculated value and the actual image coordinates, a Jacobian matrix is ​​constructed using numerical differentiation methods, forming a linear correction equation. This equation is used to iteratively update the geographic coordinates until the correction amount meets a preset convergence threshold or reaches the maximum number of iterations, thereby obtaining a stable back-projection solution model. Through the above processing, the RPC model is obtained and output to step S3 for corner geographic coordinate calculation.

[0014] Based on the image size data obtained in step S1, the positions of the three reference corner points in the image space are first determined: the upper left corner (0,0), the upper right corner (Width,0), and the lower left corner (0,Height). Then, the image coordinates of these three corner points are input into the RPC model constructed in step S2, and back-projection calculation is performed to obtain the corresponding geographic coordinates, which are denoted as the geographic coordinates of the upper left corner, the upper right corner, and the lower left corner. This step outputs the geographic coordinate data of the three reference corner points, which is then input into step S4 for calculating the orbital radian offset factor, and simultaneously input into step S5 as the basis data for calculating the fourth corner point.

[0015] Based on the geographic coordinate data of the three reference corner points obtained in step S3, the latitude coordinates of the upper left and lower left corner points are extracted. Considering that the Earth is an ellipsoid, meridians exhibit a convergence effect at different latitudes, meaning that the actual ground distance corresponding to the same longitude difference differs at different latitudes. Therefore, an orbital arc offset factor is introduced to correct for scale variations in the longitude direction. Specifically, the orbital arc offset factor is obtained by calculating the ratio of the latitude cosine of the upper left corner point to that of the lower left corner point, which characterizes the longitude scale scaling relationship caused by latitude variations. After the above processing, the orbital arc offset factor data is output and input into step S5 for the correction calculation of the longitude direction of the fourth corner point.

[0016] Based on the geographic coordinate data of the three reference corner points obtained in step S3 and the orbital arc offset factor data obtained in step S4, a nonlinear extrapolation is performed on the fourth corner point (lower right corner point). First, the longitude span at the top of the image is calculated based on the longitude difference between the upper left and upper right corner points. Then, this longitude span is multiplied by the orbital arc offset factor to obtain the corrected predicted longitude span at the bottom of the image. Based on this, the predicted longitude span is superimposed on the longitude coordinates of the lower left corner point to obtain the longitude coordinates of the fourth corner point. For the latitude direction, since the orbital change of the remote sensing image is small during the imaging time, the change in the latitude direction can be approximately regarded as a linear relationship. Therefore, the upper right corner point is extended based on the latitude span on the left to obtain the latitude coordinates of the fourth corner point. Through the combination of the above nonlinear correction of longitude and approximately linear extrapolation of latitude, the complete geographic coordinate data of the fourth corner point is obtained, and this data is output to step S6.

[0017] Based on the geographic coordinates of the three reference corner points obtained in step S3 and the geographic coordinates of the fourth corner point obtained in step S5, the four corner points are sorted and connected in a preset clockwise or counterclockwise order to construct a closed polygon. After forming the closed polygon, a corresponding geospatial reference system is assigned to the polygon, and it is encapsulated and output according to a preset vector data format to obtain the geographic coverage vector data of the remote sensing image. This completes the entire process from remote sensing image input to footprint vector output.

[0018] Please refer to [link / reference needed] for further information. Figure 3 The image presents a typical marine scene containing multiple islands and seabed topography, marked by a red rectangular border. The "rapid footprint generation method for remote sensing images based on RPC" can be directly applied to the image. The RPC parameters attached to the image are analyzed, and a geometric positioning model is constructed. The precise latitude and longitude of the three corner points (upper left, upper right, and lower left) within the image border are obtained through rigorous model calculation. Then, based on the Earth's curvature and the latitudinal range covered by the image, the orbital arc offset factor K is calculated. Using this factor and the coordinates of the three known corner points, the geographic coordinates of the lower right corner point are deduced.

[0019] Preferably, step S1 includes the following steps: Step S11: Obtain the remote sensing image to be processed and its associated RPC parameter file; Step S12: Obtain the image size data; Step S13: Read the RPC parameter file, extract the RPC parameters, and obtain the RPC parameter data.

[0020] In this embodiment, the remote sensing image data to be processed is determined as the input object. The remote sensing image is either a raw image acquired by a satellite sensor or standard image data after preprocessing. Simultaneously with acquiring the remote sensing image, based on the image's metadata index relationship or file naming rules, the corresponding RPC parameter file is located. This RPC parameter file describes the geometric mapping relationship between image space and ground space. Through this method, the pairing and acquisition of the remote sensing image and the RPC parameter file are completed. The remote sensing image is used as the input data source for step S12, and the RPC parameter file is used as the input data source for step S13.

[0021] Based on the remote sensing image data acquired in step S11, a size information reading operation is performed on the image. Specifically, the number of columns and rows of the image are extracted from the file header information or image data structure, and used as the image width and height, respectively, thus forming image size data. This image size data is used to characterize the image's boundary in image space and serves as the basis for subsequently determining the corner coordinates. After the size reading is completed, the image size data is output and retained for corner position calculations in subsequent steps. Simultaneously, the RPC parameter file acquired in step S11 is passed to step S13 for parameter parsing processing.

[0022] Based on the RPC parameter file obtained in step S11, a reading operation is performed. Specifically, the RPC parameter file is structured and parsed to extract key parameters for constructing the RPC model, including rational polynomial coefficients, normalized scale parameters, and offset parameters. These parameters are then uniformly formatted to form RPC parameter data. The rational polynomial coefficients describe the nonlinear mapping relationship between image coordinates and ground coordinates. The normalized and offset parameters are used to perform scale normalization and center shift processing on the input and output variables to ensure the stability of the model calculation. After parameter extraction, the RPC parameter data is output and used as input data for constructing the RPC model in subsequent step S2.

[0023] Preferably, step S13 includes the following steps: Step S131: Read the RPC parameter file; Step S132: Extract rational polynomial coefficients, normalization parameters, and offset parameters from the RPC parameter file to obtain RPC parameter data.

[0024] In this embodiment, a file reading operation is performed on the RPC parameter file obtained in step S11. First, the storage path and file format of the RPC parameter file are determined, and the RPC parameter file is opened and loaded. Then, according to preset data reading rules, the file content is parsed item by item, and each parameter field in the file is loaded into memory in key-value pairs or structured data form, forming a raw parameter data set for subsequent processing. After reading is completed, the raw parameter data set is output and used as input data in step S132 for further parameter extraction and classification processing.

[0025] Based on the original parameter data set obtained in step S131, each parameter item is filtered and extracted. Rational polynomial coefficient parameters are extracted from the original parameter data set, including the coefficients used to construct the numerator and denominator polynomials; normalization parameters are also extracted, including the scale factors and normalization range parameters for row, column, and geographic coordinates; and offset parameters are extracted, including the center offset between image coordinates and geographic coordinates. After parameter extraction, the above parameters are organized and packaged according to a preset data structure to form RPC parameter data in a unified format, so that it can be directly called for model construction in subsequent steps. After packaging, the RPC parameter data is output and used as input data for constructing the RPC model in step S2.

[0026] Preferably, step S2 includes the following steps: Step S21: Based on the RPC parameter data, construct a back-projection positioning model from image space to ground space to obtain the RPC model; Step S22: Perform back projection calculation from image coordinates to geographic coordinates in the RPC model.

[0027] In this embodiment, firstly, in step S21, a back-projection positioning model from image space to ground space is constructed based on the RPC parameter data obtained in step S13. Specifically, the rational polynomial coefficients in the RPC parameter data are used as core mapping parameters to establish a nonlinear functional relationship between image coordinates and geographic coordinates. Simultaneously, normalization parameters are used to perform scale normalization on the image coordinates and geographic coordinates, and offset parameters are used to perform center translation on the normalized variables, thereby forming a function expression structure under a unified coordinate reference. Based on this, the row coordinates and column coordinates are respectively represented... The RPC function is represented as a rational polynomial function of geographic longitude, latitude, and elevation variables. It describes the mapping relationship between image space and ground space through the ratio of the numerator polynomial to the denominator polynomial, thus constructing a complete RPC function model structure. Since the RPC model is a forward mapping model from geographic coordinates to image coordinates, it needs to implement the back projection process of obtaining geographic coordinates from image coordinates. Therefore, a corresponding back-solution mechanism is configured while constructing the model to provide a computational basis for subsequent back projection calculations. After completing the above model construction, the RPC model is output and used as the input for step S22. Subsequently, in step S22, based on the RPC model constructed in step S21, back-projection solution is performed on the given image coordinates. The image coordinates to be solved are used as the observation input. First, an initial estimate of the geographic coordinates is set, and this initial estimate is substituted into the RPC model to calculate the corresponding predicted image coordinate values. Then, a residual term is constructed by comparing the difference between the predicted value and the actual image coordinates. The RPC function is locally linearized based on the numerical differentiation method to calculate the partial derivatives with respect to geographic longitude and latitude, thereby constructing the Jacobian matrix. Based on this, a linear correction equation is established to iteratively update the geographic coordinates. In each iteration, the process of "substitution calculation - residual construction - partial derivative solution - parameter correction" is repeated until the update amount of the geographic coordinates is less than the preset convergence threshold or the maximum number of iterations is reached, thereby obtaining a stable geographic coordinate solution corresponding to the input image coordinates, completing the back-projection solution process from image coordinates to geographic coordinates, and outputting the solution results for subsequent corner geographic coordinate calculation steps.

[0028] It should be added that, based on the parsed RPC parameters, a back-projection localization model is constructed from the image space (Col, Row) to the ground space (Lon, Lat). To achieve the back-projection from image coordinates to geographic coordinates, an iterative estimation method is needed to solve a system of nonlinear equations. Let the initial geographic coordinates be the normalized center: ; Using current geographic coordinates with average elevation Substitute them into the RPC model to calculate the corresponding image coordinates. Calculate the calculated values ​​and the coordinates of the real image. The residuals. Using numerical differentiation, the partial derivative matrix is ​​calculated, and a system of linear equations is constructed: ； in The Jacobian matrix is ​​obtained through numerical differentiation. Update the coordinates until the correction is less than a threshold (e.g., ...). (or reach the maximum number of iterations.)

[0029] Where (Col, Row) represents the image space coordinates, that is, the pixel position of the target point in the remote sensing image: Col: column coordinate (horizontal pixel position). : Row coordinates (vertical pixel position), ( , ) represents ground spatial coordinates: :longitude :latitude. , The initial geographic coordinate estimate (normalized coordinates) is represented by (0, 0). RPC models typically use normalized coordinates (centered at 0). Here, (0, 0) is chosen as the initial value, which is a standard initial point. , This represents the image coordinates calculated using the RPC forward model. : Column direction error, Line direction error , This indicates the amount of correction made to the geographic coordinates in this iteration. Longitude correction Latitude correction.

[0030] Preferably, step S21 includes the following steps: Step S211: Based on the rational polynomial coefficients in the RPC parameter data, construct a polynomial function model describing the mapping relationship between image coordinates and ground coordinates; Step S212: Based on the polynomial function model, configure an iterative solver for inverse calculation of image coordinates to geographic coordinates to obtain the RPC model.

[0031] In this embodiment, step S21 is described in detail, specifically including steps S211 and S212. First, in step S211, based on the rational polynomial coefficients in the RPC parameter data obtained in step S13, a polynomial function model describing the mapping relationship between image coordinates and ground coordinates is constructed. Longitude, latitude, and elevation in the geographic coordinates are used as input variables. According to the numerator and denominator polynomial structures given in the RPC parameters, rational function expressions for row coordinates and column coordinates with respect to the above variables are established respectively, thereby forming a nonlinear function model that maps geographic coordinates to image coordinates. This polynomial function model is then used as the basic input for subsequent inverse calculation processing. Proceed to step S212; subsequently, in step S212, based on the polynomial function model constructed in step S211, a back-projection iterative solver for image coordinates to geographic coordinates is configured. Specifically, the polynomial function model is used as the forward computation unit, and an iterative solution mechanism is introduced on this basis. By setting the initial value of the geographic coordinates and continuously substituting it into the polynomial function model to calculate the corresponding image coordinates, the geographic coordinates are gradually corrected according to the difference between the calculation result and the target image coordinates, thereby forming a complete back-projection solution process. Finally, an RPC model that can be used to perform the back-projection from image coordinates to geographic coordinates is obtained, and the RPC model is output for the corner geographic coordinate calculation in subsequent steps.

[0032] Preferably, step S3 includes the following steps: Step S31: Based on the image size data, determine the image coordinates of the top left corner, top right corner, and bottom left corner; Step S32: Based on the RPC model, back-projection calculations are performed on the image coordinates of the upper left corner, upper right corner, and lower left corner respectively to obtain the geographic coordinates, which are used as the geographic coordinate data of the three reference corner points.

[0033] In this embodiment, step S3 is described in detail, specifically including steps S31 and S32. First, in step S31, the image boundary range is determined based on the image size data obtained in step S1. Taking the upper left corner of the image as the origin, the coordinate positions of three reference corner points in the image space are determined according to the image width and height. The image coordinates of the upper left corner point are (0,0), the image coordinates of the upper right corner point are (Width,0), and the image coordinates of the lower left corner point are (0,Height). Thus, the image coordinate data of the three reference corner points are obtained, and the image is then processed. The coordinate data is passed as input to step S32; then in step S32, based on the RPC model constructed in step S2, the back projection calculation is performed on the image coordinates of the three reference corner points determined in step S31. Specifically, the image coordinates of each corner point are input into the RPC model one by one, and the corresponding geographic coordinates are calculated by iterative solution to obtain the geographic coordinate results of the upper left corner point, the upper right corner point, and the lower left corner point. The above geographic coordinate results are then output as the geographic coordinate data of the three reference corner points for subsequent orbital radii offset factor calculation and fourth corner point estimation steps.

[0034] It should be added that the top left (TL), top right (TR), and bottom left (BL) corner points are selected, and the corresponding geographic coordinates are calculated using the RPC model.

[0035] Result: Top left point TL: Image coordinates: ; Geographic coordinates: ; Top right point TR: Image coordinates: Geographic coordinates: Bottom left BL: Image coordinates: ; Geographic coordinates: ; in, ; indicates the position of the top left corner of the image in the image coordinate system, where: Col = 0: indicates the starting point in the column direction (the leftmost pixel), Row = 0: indicates the starting point in the row direction (the topmost pixel). This represents the geographic coordinates of the top-left corner point after RPC back projection calculation; This indicates the position of the upper right corner of the image in the image coordinate system. This point is located at the top right of the image, which is the other end of the image's horizontal boundary. This represents the geographic coordinates corresponding to the top right corner point. Through RPC back projection, the top right corner point of the image is mapped to geographic space to determine the location of the other end of the image's upper boundary. This indicates the position of the lower left corner of the image in the image coordinate system. This point is located at the bottom left of the image, which is the other end of the image's vertical boundary.

[0036] ; represents the geographic coordinates corresponding to the lower left corner point, calculated through the RPC model, and is used to determine the extension of the left boundary of the image in geographic space.

[0037] Preferably, step S4 includes the following steps: Step S41: Based on the geographic coordinate data of the three reference corner points, obtain the latitude coordinates of the upper left corner point and the lower left corner point; Step S42: Calculate the ratio of the cosine of the latitude value of the upper left corner point to the cosine of the latitude value of the lower left corner point, and use this ratio as the orbital radian offset factor to obtain the orbital radian offset factor data.

[0038] In this embodiment, step S4 is described in detail, specifically including steps S41 and S42. First, in step S41, based on the geographic coordinate data of the three reference corner points obtained in step S3, the latitude coordinate values ​​corresponding to the upper left corner point and the lower left corner point are extracted to form a latitude data pair for subsequent calculations, and the latitude data pair is passed as input to step S42. Then, in step S42, based on the latitude values ​​of the upper left corner point and the lower left corner point extracted in step S41, the cosine values ​​are calculated respectively, and the ratio between the two is calculated. The obtained ratio is used as the orbital arc offset factor to characterize the relationship of longitude scale change at different latitude positions, thereby obtaining the orbital arc offset factor data.

[0039] It should be added that the Earth is an ellipsoid, and meridians converge towards the poles. The ground distance corresponding to the same longitude difference varies with latitude. In order to accurately calculate the uncalculated lower right point, this invention introduces an orbital arc offset factor K. The K value reflects the scaling factor required for the longitude span due to the decrease (or increase) of latitude, while keeping the actual ground width approximately unchanged.

[0040] ; Preferably, step S5 includes the following steps: Step S51: Based on the geographic coordinate data of the three reference corner points, calculate the difference between the longitude of the upper right corner point and the longitude of the upper left corner point, and use it as the longitude span of the top of the image; calculate the difference between the latitude of the lower left corner point and the latitude of the upper left corner point, and use it as the latitude span of the left side of the image. Step S52: Multiply the longitude span at the top of the image with the orbital radii offset factor data to obtain the predicted longitude span at the bottom of the image after correction by the orbital radii offset factor; Step S53: Add the longitude coordinates of the lower left corner point to the predicted longitude span to obtain the longitude coordinates of the fourth corner point; Step S54: Add the latitude coordinates of the upper right corner point to the latitude span on the left to obtain the latitude coordinates of the fourth corner point, thus obtaining the complete geographic coordinate data of the fourth corner point.

[0041] In this embodiment, step S5 is described in detail, specifically including steps S51 to S54. First, in step S51, based on the geographic coordinate data of the three reference corner points obtained in step S3, the difference between the longitude of the upper right corner point and the longitude of the upper left corner point is calculated to obtain the longitude span of the top of the image. At the same time, the difference between the latitude of the lower left corner point and the latitude of the upper left corner point is calculated to obtain the latitude span of the left side of the image. The longitude span and latitude span are output as basic parameters and passed to step S52. Then, in step S52, based on the longitude span of the top of the image obtained in step S51, and combined with the orbital arc offset factor data obtained in step S4, the longitude span is corrected, that is, the longitude span of the top of the image is corrected to the orbital arc offset factor data. Multiplying by the radian offset factor yields the predicted longitude span at the bottom of the image, which is then used as input for subsequent longitude extrapolation and passed to step S53. Next, in step S53, based on the predicted longitude span obtained in step S52, it is superimposed with the longitude coordinates of the lower left corner point to obtain the longitude coordinates of the fourth corner point, which is then output and passed to step S54. Finally, in step S54, based on the latitude span on the left side of the image obtained in step S51, this latitude span is superimposed on the latitude coordinates of the upper right corner point to obtain the latitude coordinates of the fourth corner point, which is then combined with the longitude coordinates obtained in step S53 to form the complete geographic coordinate data of the fourth corner point, which is output for subsequent geographic coverage construction steps.

[0042] It should be added that, using the calculated K-factor, a nonlinear extrapolation is performed on the fourth corner point (bottom right point) to correct geometric distortion. The longitude span at the top of the image is also calculated. : ; Predicting the longitude span at the bottom of an image using the K-factor : ; Calculate the longitude of the lower right point : ; For the latitude direction, since the satellite orbit's motion is relatively stable within a short imaging time, and meridian convergence mainly affects the longitude direction, the latitude direction can be approximated by an affine transformation. Calculate the latitude span on the left. : ; Calculate the latitude of the lower right point : ; Preferably, step S6 includes the following steps: Step S61: Connect the geographical coordinates of the top left corner, top right corner, fourth corner, and bottom left corner in a clockwise or counterclockwise order to form a closed polygon; Step S62: Assign the specified geospatial reference system to the polygon and output it as a standard vector format file.

[0043] In this embodiment, step S6 is described in detail, specifically including steps S61 and S62. First, in step S61, based on the geographic coordinate data of the three reference corner points obtained in step S3 and the geographic coordinate data of the fourth corner point obtained in step S5, the upper left corner point, upper right corner point, fourth corner point, and lower left corner point are sorted in a preset clockwise or counterclockwise direction, and the coordinates of each point are connected in sequence to form a closed coordinate sequence, thereby constructing a closed polygon representing the coverage area of ​​the image. This polygon data is then passed as input to step S62. Subsequently, in step S62, based on the closed polygon constructed in step S61, a preset geospatial reference system is assigned to it to clarify its spatial positioning information, and the polygon is encapsulated and output according to the standard vector data format to generate the corresponding remote sensing image geographic coverage vector file, thereby completing the construction and output of the footprint data.

[0044] It should be added that the four calculated geographic coordinate points are connected in a clockwise or counterclockwise order to construct a closed polygon: ; Finally, the polygon is assigned to a spatial reference system and output as a common vector file format to complete the footprint generation.

[0045] Preferably, step S61 includes the following steps: Step S611: Sort the geographic coordinate data of the top left corner, top right corner, fourth corner, and bottom left corner in a preset clockwise or counterclockwise direction to obtain an ordered coordinate sequence; Step S612: Based on the ordered coordinate sequence, generate a closed polygon geometry defined by the sequence.

[0046] In this embodiment, step S61 is further detailed, specifically including steps S611 and S612. First, in step S611, based on the geographic coordinate data of the three reference corner points obtained in step S3 and the geographic coordinate data of the fourth corner point obtained in step S5, the geographic coordinates of the upper left corner point, upper right corner point, fourth corner point, and lower left corner point are sorted according to a preset clockwise or counterclockwise direction to form an ordered coordinate sequence with a clear sequential relationship. This ordered coordinate sequence is then passed as input to step S612. Subsequently, in step S612, based on the ordered coordinate sequence obtained in step S611, the coordinate points are connected sequentially according to the coordinate order, and the first and last coordinates are closed to generate a closed polygon geometry defined by the coordinate sequence. This polygon is used to characterize the spatial coverage of the remote sensing image, and the output of this closed polygon is used for subsequent spatial reference assignment and vector data generation steps.

[0047] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for rapid footprint generation from remote sensing images based on RPC, characterized in that, Includes the following steps: Step S1: Based on the remote sensing image to be processed and its metadata, perform image size reading and RPC parameter parsing to obtain image size data and RPC parameter data; Step S2: Construct an RPC model based on RPC parameter data; Step S3: Based on the RPC model and image size data, calculate the geographic coordinates of the three reference corner points of the image to obtain the geographic coordinate data of the three reference corner points; Step S4: Calculate the orbital radian offset factor based on the geographic coordinate data of the three reference corner points to obtain the orbital radian offset factor data; Step S5: Nonlinear calculations are performed on the base orbit radian offset factor data, the geographic coordinate data of the three reference corner points, and the image size data to obtain the geographic coordinate data of the fourth corner point; Step S6: Integrate the geographic coordinate data of the three reference corner points and the fourth corner point to generate and output the geographic coverage vector data of the remote sensing image.

2. The method for rapid footprint generation of remote sensing images based on RPC according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain the remote sensing image to be processed and its associated RPC parameter file; Step S12: Obtain the image size data; Step S13: Read the RPC parameter file, extract the RPC parameters, and obtain the RPC parameter data.

3. The method for rapid footprint generation of remote sensing images based on RPC according to claim 2, characterized in that, Step S13 includes the following steps: Step S131: Read the RPC parameter file; Step S132: Extract rational polynomial coefficients, normalization parameters, and offset parameters from the RPC parameter file to obtain RPC parameter data.

4. The method for rapid footprint generation of remote sensing images based on RPC according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Based on the RPC parameter data, construct a back-projection positioning model from image space to ground space to obtain the RPC model; Step S22: Perform back projection calculation from image coordinates to geographic coordinates in the RPC model.

5. The method for rapid footprint generation of remote sensing images based on RPC according to claim 4, characterized in that, Step S21 includes the following steps: Step S211: Based on the rational polynomial coefficients in the RPC parameter data, construct a polynomial function model describing the mapping relationship between image coordinates and ground coordinates; Step S212: Based on the polynomial function model, configure an iterative solver for inverse calculation of image coordinates to geographic coordinates to obtain the RPC model.

6. The method for rapid footprint generation of remote sensing images based on RPC according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Based on the image size data, determine the image coordinates of the top left corner, top right corner, and bottom left corner; Step S32: Based on the RPC model, back-projection calculations are performed on the image coordinates of the upper left corner, upper right corner, and lower left corner respectively to obtain the geographic coordinates, which are used as the geographic coordinate data of the three reference corner points.

7. The method for rapid footprint generation of remote sensing images based on RPC according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Based on the geographic coordinate data of the three reference corner points, obtain the latitude coordinates of the upper left corner point and the lower left corner point; Step S42: Calculate the ratio of the cosine of the latitude value of the upper left corner point to the cosine of the latitude value of the lower left corner point, and use this ratio as the orbital radian offset factor to obtain the orbital radian offset factor data.

8. The method for rapid footprint generation of remote sensing images based on RPC according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: Based on the geographic coordinate data of the three reference corner points, calculate the difference between the longitude of the upper right corner point and the longitude of the upper left corner point, and use it as the longitude span of the top of the image; calculate the difference between the latitude of the lower left corner point and the latitude of the upper left corner point, and use it as the latitude span of the left side of the image. Step S52: Multiply the longitude span at the top of the image with the orbital radii offset factor data to obtain the predicted longitude span at the bottom of the image after correction by the orbital radii offset factor; Step S53: Add the longitude coordinates of the lower left corner point to the predicted longitude span to obtain the longitude coordinates of the fourth corner point; Step S54: Add the latitude coordinates of the upper right corner point to the latitude span on the left to obtain the latitude coordinates of the fourth corner point, thus obtaining the complete geographic coordinate data of the fourth corner point.

9. The method for rapid footprint generation of remote sensing images based on RPC according to claim 1, characterized in that, Step S6 includes the following steps: Step S61: Connect the geographical coordinates of the top left corner, top right corner, fourth corner, and bottom left corner in a clockwise or counterclockwise order to form a closed polygon; Step S62: Assign the specified geospatial reference system to the polygon and output it as a standard vector format file.

10. The method for rapid footprint generation of remote sensing images based on RPC according to claim 9, characterized in that, Step S61 includes the following steps: Step S611: Sort the geographic coordinate data of the top left corner, top right corner, fourth corner, and bottom left corner in a preset clockwise or counterclockwise direction to obtain an ordered coordinate sequence; Step S612: Based on the ordered coordinate sequence, generate a closed polygon geometry defined by the sequence.

Images