A DEM-based ortho-rectification method for SAR images

Abstract

This invention discloses a DEM-based SAR image orthorectification method, specifically including the following steps: S1, acquiring data; S2, calculating image point coordinate set A; S3, calculating correction coefficients α and β for range and azimuth polynomials; S4, calculating direct-form image point coordinate set C; S5, calculating correction coefficients α' and β' for direct-form range and azimuth polynomials; S6, calculating mean-normalized range and azimuth polynomial correction coefficients γ and ζ; S7, SAR orthorectification. This invention relates to the field of radar image geometric processing technology. This DEM-based SAR image orthorectification method can improve correction accuracy by combining indirect and direct correction methods based on DEM. It effectively achieves the goal of correcting images in special geographical locations by using orthorectified high-precision optical images as reference images to mutually verify and fuse the correction accuracy of the two methods, thus achieving high correction accuracy.

Description

Technical Field

[0001] This invention relates to the field of radar image geometric processing technology, specifically to a method for orthorectifying SAR images based on DEM. Background Technology

[0002] Orthorectification typically involves selecting ground control points on an image and using previously acquired digital elevation model (DEM) data for that area to simultaneously correct tilt and projection differences, resampling the image into an orthophoto. Multiple orthophotos are then stitched together, color balanced, and cropped within a defined range to create an orthophoto map. Orthophotos possess both topographic and imagery characteristics, are rich in information, and can serve as a data source for GIS, thus enriching the presentation of geographic information systems.

[0003] Referring to the SAR image orthorectification method based on satellite control point library and DEM published in Chinese Patent Publication No. CN107341778B, this method can automatically utilize the ZY-3 control point library to participate in SAR image orthorectification, reducing the inefficiency caused by manual intervention or the point selection error caused by manual selection. It makes full use of the high-precision ZY-3 control point library obtained through optical means, enabling it to be effectively used for SAR orthorectification, reducing the cost of acquiring control points. By simulating SAR images with an external DEM and matching them with multi-view SAR images, the coordinate set of the high-precision ZY-3 control point library on the SAR image is determined, which is better than the effect of directly selecting points on the external DEM to participate in SAR image correction.

[0004] A comprehensive analysis of the above-mentioned patents reveals the following shortcomings: Most existing SAR image orthorectification methods rely on indirect correction using DEM simulation. This method has inherent biases, particularly in its inability to accurately correct spaceborne SAR images in mountainous or other special geographical locations. It fails to improve correction accuracy by combining DEM-based indirect and direct correction methods. Furthermore, it cannot achieve the goal of using orthorectified high-precision optical images as a reference image to cross-verify and fuse the correction accuracy of the two methods for targeted correction of spaceborne SAR images in mountainous or other special geographical locations. Consequently, it cannot achieve high correction accuracy and fails to meet the needs of actual production. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a DEM-based SAR image orthorectification method. This method solves the problem that most existing SAR image orthorectification methods rely on indirect correction using DEM simulation, which has inherent biases. In particular, it cannot accurately correct spaceborne SAR images in mountainous or other special geographical locations. Furthermore, it fails to improve correction accuracy by combining DEM-based indirect correction with direct correction methods. It also cannot achieve the goal of using orthorectified high-precision optical images as a reference image to cross-verify and fuse the correction accuracy of the two methods for targeted correction of spaceborne SAR images in mountainous or other special geographical locations. Consequently, it cannot achieve high correction accuracy and fails to meet practical production needs.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution: a SAR image orthorectification method based on DEM, specifically comprising the following steps:

[0009] S1. Data Acquisition: Collect satellite image control point data V, external reference DEM data W, original SAR image Z, and original SAR image imaging parameters U respectively. Then, perform image extraction and segmentation processing on the original SAR image Z using an image segmentation and recognition algorithm to obtain multi-view SAR image Z`.

[0010] S2. Calculate the image point coordinate set A: By simulating each data point in step S1 using the DEM model, calculate the image point coordinate set A = [a(1), a(2), ..., a(n)] of the satellite image control point data in the multi-view SAR image, where a(n) = (x n ,y n );

[0011] S3. Calculate the correction coefficients α and β of the range and azimuth polynomials: Convert the coordinate set of the satellite image control point data in the multi-view SAR image from step S2 into the coordinate set B = [b(1), b(2), ..., b(n)] of the satellite image control point data in the original SAR image, where b(n) = (t n ,k n Then, SAR image orthorectification is performed to calculate the range polynomial correction coefficient α and the azimuth polynomial correction coefficient β.

[0012] S4. Calculate the direct image point coordinate set C: Directly import the satellite image control point data V, external reference DEM data W, original SAR image Z, and original SAR image imaging parameters U collected in step S1 into the satellite SAR image recognition system for recognition processing to calculate the image point coordinate set C = [c(1), c(2), ..., c(n)] of the direct satellite image control point data in the multi-view SAR image, where c(n) = (u n ,w n );

[0013] S5. Calculate the correction coefficients α` and β` of the direct range and azimuth polynomials: Perform SAR image orthorectification on the image point coordinate set C = [c(1),c(2),…,c(n)] of the direct satellite image control point data in the multi-view SAR image from step S4, and calculate the correction coefficients α` of the direct range polynomial and β` of the direct azimuth polynomial.

[0014] S6. Calculate the correction coefficients γ and ζ of the mean-valued range and azimuth polynomials: The correction coefficients α` and β` of the direct-form range polynomial obtained in step S5 are mean-valued by the correction coefficients α and β of the range polynomial obtained in step S3, respectively, to obtain the mean-valued range polynomial correction coefficient γ and the mean-valued azimuth polynomial correction coefficient ζ.

[0015] S7. SAR Orthorectification: Based on the averaged range polynomial correction coefficient γ and the averaged azimuth polynomial correction coefficient ζ obtained in step S6, the original SAR image is orthorectified to obtain the SAR orthorectified DOM image.

[0016] Preferably, the image extraction and segmentation processing in step S1 using the image segmentation and recognition algorithm specifically includes the following steps:

[0017] T1. The original SAR image Z is first binarized using a binarization function to obtain foreground and background information from the original SAR image, resulting in a binarized image.

[0018] T2. Remove noise from the image after binarization in step T1, and extract the data information features from the image using a feature extraction algorithm;

[0019] T3. Based on the feature data information extracted in step T2, the image is divided into multi-view SAR images Z'.

[0020] Preferably, the SAR image orthorectification processing in step S3 is as follows:

[0021] The coordinate values ​​of the satellite image control point data in the multi-view SAR image coordinate set A = [a(1), a(2), ..., a(n)] are selected and subtracted from the corresponding coordinate values ​​of the satellite image control point data in the original SAR image coordinate set B = [b(1), b(2), ..., b(n)]. The specific process is as follows:

[0022]

[0023] Where, x n y represents the x-coordinate of the image point in the multi-view SAR image, which is the control point data of the satellite image. n t represents the ordinate value of the image point in the multi-view SAR image, which is the control point data of the satellite image. n k represents the x-coordinate value of the satellite image control point data in the original SAR image. n α represents the ordinate value of the satellite image control point data in the original SAR image, α is the polynomial correction coefficient in the range direction, and β is the polynomial correction coefficient in the azimuth direction.

[0024] Preferably, the SAR image orthorectification processing of the image point coordinate set C of the direct satellite image control point data in the multi-view SAR image in step S5 is as follows:

[0025] The coordinate values ​​of the control point data in the direct satellite imagery control point data set C = [c(1), c(2), ..., c(n)] in the multi-view SAR image are selected, and the difference is calculated with the corresponding coordinate values ​​of the control point data in the original SAR imagery coordinate set B = [b(1), b(2), ..., b(n)]. The specific details are as follows:

[0026]

[0027] Among them, u n y represents the x-coordinate of the image point in a multi-view SAR image based on the control point data of a direct satellite image. n t represents the ordinate value of the image point in the multi-view SAR image, which is the control point data of the satellite image. n w represents the x-coordinate value of the satellite image control point data in the original SAR image. n α` represents the ordinate value of the direct satellite image control point data in the original SAR image, α` represents the polynomial correction coefficient of the direct range direction, and β` represents the polynomial correction coefficient of the direct azimuth direction.

[0028] Preferably, the medianization process in step S6 specifically includes the following steps:

[0029] E1. Add the polynomial correction coefficients α' and α' of the direct-form range direction to obtain the mean-valued polynomial correction coefficients γ of the range direction. Let γ be the polynomial correction coefficients of the range direction.

[0030] E2. Add the correction coefficients β` of the direct-form azimuth polynomial and the correction coefficients β of the azimuth polynomial, and the difference is the correction coefficient ζ of the mean-valued azimuth polynomial, denoted as:

[0031] Preferably, in step S2, the DEM model is used to simulate each data in step S1 to obtain the coordinate relationship of the one-to-one mapping between the circumscribed rectangular digital elevation model RefDEM data, the DEM simulated SAR image, the DEM simulated SAR image, and the multi-view SAR image radar coordinate space. The SAR simulated image consistent with the multi-view SAR image radar coordinate space is calculated. Then, grayscale matching is performed on the SAR simulated image and the multi-view SAR image to calculate the new coordinate relationship of the one-to-one mapping between the DEM simulated SAR image and the multi-view SAR image radar coordinate space.

[0032] Preferably, the circumscribed rectangular digital elevation model (RefDEM) data includes the width, height, starting point geographic coordinates, latitudinal resolution, longitudinal resolution, and elevation value of the circumscribed rectangular digital elevation model (RefDEM) in the external reference DEM data;

[0033] By acquiring the width nWidthDEM and height nHeightDEM of the circumscribed rectangular digital elevation model RefDEM, two real empty matrices of size (nHeightDEM, nWidthDEM) are created to store the circumscribed rectangular digital elevation model RefDEM data and the DEM-simulated SAR image data, respectively. A complex empty matrix of size (nHeightDEM, nWidthDEM) is created to store the coordinate relationship data of the one-to-one mapping between the DEM-simulated SAR image and the multi-view SAR image radar coordinate space.

[0034] Preferably, the coordinate relationship between the circumscribed rectangular digital elevation model (RefDEM) data and the multi-view SAR image radar coordinate space is calculated based on the SAR rigorous imaging model.

[0035] (III) Beneficial Effects

[0036] This invention provides a DEM-based SAR image orthorectification method. Compared with existing technologies, it has the following advantages: This DEM-based SAR image orthorectification method specifically includes the following steps: S1, data acquisition; S2, calculating image point coordinate set A; S3, calculating the correction coefficients α and β of the range and azimuth polynomials; S4, calculating the direct-type image point coordinate set C; S5, calculating the correction coefficients α' and β' of the direct-type range and azimuth polynomials; S6, calculating the mean-normalized range and azimuth polynomial correction coefficients γ and ζ; S7, SAR orthorectification. This method combines indirect and direct correction methods based on DEM to improve correction accuracy. It effectively achieves the goal of using orthorectified high-precision optical images as reference images to mutually verify and fuse the correction accuracy of the two methods, thereby making targeted corrections for spaceborne SAR images in special geographical locations such as mountainous areas. It achieves high correction accuracy and can meet the needs of actual production. Attached Figure Description

[0037] Figure 1 This is a flowchart of the orthophoto correction method of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Please see Figure 1 The present invention provides three technical solutions: a SAR image orthorectification method based on DEM, specifically including the following embodiments:

[0040] Example 1

[0041] A method for orthorectifying SAR images based on DEM, specifically including the following steps:

[0042] S1. Data Acquisition: Collect satellite image control point data V, external reference DEM data W, original SAR image Z, and original SAR image imaging parameters U respectively. Then, perform image extraction and segmentation processing on the original SAR image Z using an image segmentation and recognition algorithm to obtain multi-view SAR image Z`.

[0043] S2. Calculate the image point coordinate set A: By simulating each data point in step S1 using the DEM model, calculate the image point coordinate set A = [a(1), a(2), ..., a(n)] of the satellite image control point data in the multi-view SAR image, where a(n) = (x n ,y n );

[0044] S3. Calculate the correction coefficients α and β of the range and azimuth polynomials: Convert the coordinate set of the satellite image control point data in the multi-view SAR image from step S2 into the coordinate set B = [b(1), b(2), ..., b(n)] of the satellite image control point data in the original SAR image, where b(n) = (t n ,k n Then, SAR image orthorectification is performed to calculate the range polynomial correction coefficient α and the azimuth polynomial correction coefficient β.

[0045] S4. Calculate the direct image point coordinate set C: Directly import the satellite image control point data V, external reference DEM data W, original SAR image Z, and original SAR image imaging parameters U collected in step S1 into the satellite SAR image recognition system for recognition processing to calculate the image point coordinate set C = [c(1), c(2), ..., c(n)] of the direct satellite image control point data in the multi-view SAR image, where c(n) = (u n ,w n );

[0046] S5. Calculate the correction coefficients α` and β` of the direct range and azimuth polynomials: Perform SAR image orthorectification on the image point coordinate set C = [c(1),c(2),…,c(n)] of the direct satellite image control point data in the multi-view SAR image from step S4, and calculate the correction coefficients α` of the direct range polynomial and β` of the direct azimuth polynomial.

[0047] S6. Calculate the correction coefficients γ and ζ of the mean-valued range and azimuth polynomials: The correction coefficients α` and β` of the direct-form range polynomial obtained in step S5 are mean-valued by the correction coefficients α and β of the range polynomial obtained in step S3, respectively, to obtain the mean-valued range polynomial correction coefficient γ and the mean-valued azimuth polynomial correction coefficient ζ.

[0048] S7. SAR Orthorectification: Based on the averaged range polynomial correction coefficient γ and the averaged azimuth polynomial correction coefficient ζ obtained in step S6, the original SAR image is orthorectified to obtain the SAR orthorectified DOM image.

[0049] Example 2

[0050] The technical solution of this embodiment of the invention differs from that of Embodiment 1 in that: in this embodiment of the invention, the image extraction and segmentation processing in step S1 using the image segmentation and recognition algorithm specifically includes the following steps:

[0051] T1. The original SAR image Z is first binarized using a binarization function to obtain foreground and background information from the original SAR image, resulting in a binarized image.

[0052] T2. Remove noise from the image after binarization in step T1, and extract the data information features from the image using a feature extraction algorithm;

[0053] T3. Based on the feature data information extracted in step T2, the image is divided into multi-view SAR images Z'.

[0054] Example 3

[0055] The technical solution of this embodiment of the invention differs from that of Embodiment 2 in that: in this embodiment of the invention, the SAR image orthorectification processing in step S3 is specifically as follows:

[0056] The coordinate values ​​of each satellite image control point data in the multi-view SAR image coordinate set A = [a(1), a(2), ..., a(n)] are selected and subtracted from the corresponding coordinate values ​​of the satellite image control point data in the original SAR image coordinate set B = [b(1), b(2), ..., b(n)]. The specific process is as follows:

[0057]

[0058] Among them, X n y represents the x-coordinate of the image point in the multi-view SAR image, which is the control point data of the satellite image. n t represents the ordinate value of the image point in the multi-view SAR image, which is the control point data of the satellite image. n k represents the x-coordinate value of the satellite image control point data in the original SAR image. n α represents the ordinate value of the satellite image control point data in the original SAR image, α is the polynomial correction coefficient in the range direction, and β is the polynomial correction coefficient in the azimuth direction.

[0059] In this embodiment of the invention, the SAR image orthorectification processing of the image point coordinate set C of the direct satellite image control point data in the multi-view SAR image in step S5 is as follows:

[0060] The coordinate values ​​of the control point data in the direct satellite imagery control point data set C = [c(1), c(2), ..., c(n)] in the multi-view SAR image are selected, and the difference is calculated with the corresponding coordinate values ​​of the control point data in the original SAR imagery coordinate set B = [b(1), b(2), ..., b(n)]. The specific details are as follows:

[0061]

[0062] Among them, u n y represents the x-coordinate of the image point in a multi-view SAR image based on the control point data of a direct satellite image. n t represents the ordinate value of the image point in the multi-view SAR image, which is the control point data of the satellite image. n w represents the x-coordinate value of the satellite image control point data in the original SAR image. n α` represents the ordinate value of the direct satellite image control point data in the original SAR image, α` represents the polynomial correction coefficient of the direct range direction, and β` represents the polynomial correction coefficient of the direct azimuth direction.

[0063] In this embodiment of the invention, the medianization process in step S6 specifically includes the following steps:

[0064] E1. Add the polynomial correction coefficients α' and α' of the direct-form range direction to obtain the mean-valued polynomial correction coefficients γ of the range direction. Let γ be the polynomial correction coefficients of the range direction.

[0065] E2. Add the correction coefficients β` of the direct-form azimuth polynomial and the correction coefficients β of the azimuth polynomial, and the difference is the correction coefficient ζ of the mean-valued azimuth polynomial, denoted as:

[0066] In this embodiment of the invention, in step S2, the DEM model is used to simulate each data in step S1 to obtain the coordinate relationship of the one-to-one mapping between the circumscribed rectangular digital elevation model RefDEM data, the DEM simulated SAR image, the DEM simulated SAR image, and the multi-view SAR image radar coordinate space. The SAR simulated image consistent with the multi-view SAR image radar coordinate space is calculated. Then, grayscale matching is performed on the SAR simulated image and the multi-view SAR image to calculate the coordinate relationship of the new DEM simulated SAR image and the multi-view SAR image radar coordinate space. The circumscribed rectangular digital elevation model RefDEM data includes the width, height, starting point geographic coordinates, latitude resolution, longitude resolution, and elevation value of the circumscribed rectangular digital elevation model RefDEM in the external reference DEM data.

[0067] By acquiring the width (nWidthDEM) and height (nHeightDEM) of the circumscribed rectangular digital elevation model (RefDEM), two real empty matrices of size (nHeightDEM, nWidthDEM) are created to store the circumscribed rectangular digital elevation model (RefDEM) data and the DEM-simulated SAR image data, respectively. A complex empty matrix of size (nHeightDEM, nWidthDEM) is created to store the coordinate relationship data of the one-to-one mapping between the DEM-simulated SAR image and the multi-view SAR image radar coordinate space. The coordinate relationship of the one-to-one mapping between the circumscribed rectangular digital elevation model (RefDEM) data and the multi-view SAR image radar coordinate space is calculated according to the SAR rigorous imaging model.

[0068] In summary, this invention can improve the correction accuracy by combining indirect correction of DEM-based SAR images with direct correction methods. It effectively achieves the goal of using orthorectified high-precision optical images as reference images to cross-verify and fuse the correction accuracy of the two methods, thereby making targeted corrections for spaceborne SAR images in special geographical locations such as mountainous areas. It can achieve high correction accuracy and meet the needs of actual production.

[0069] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0070] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0071] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A DEM-based ortho-rectification method for SAR images, characterized in that: Specifically, the following steps are included: S1. Data Acquisition: Collect satellite image control point data V, external reference DEM data W, original SAR image Z, and original SAR image imaging parameters U respectively. Then, perform image extraction and segmentation processing on the original SAR image Z using an image segmentation and recognition algorithm to obtain multi-view SAR image Z`. S2, calculating the image point coordinate set : by using the DEM model to simulate each data in step S1, calculating the image point coordinate set of the satellite image control point data in the multi-view SAR image , wherein = (x i , y i ), is the horizontal coordinate value of the image point of the satellite image control point data in the multi-view SAR image, is the vertical coordinate value of the image point of the satellite image control point data in the multi-view SAR image; S3. Calculate the correction coefficients α and β of the range and azimuth polynomials: Convert the coordinate set of the satellite image control point data in the multi-view SAR image from step S2 into the coordinate set of the satellite image control point data in the original SAR image. ,in =(t i ,k i ), This represents the x-coordinate value of the satellite image control point data in the original SAR image. The vertical coordinates of the satellite image control point data in the original SAR image are given. Then, SAR image orthorectification processing is performed to calculate the polynomial correction coefficients α in the range direction and β in the azimuth direction. S4. Calculate the coordinate set of direct image points. The satellite image control point data V, external reference DEM data W, original SAR image Z, and original SAR image imaging parameters U collected in step S1 are directly imported into the satellite SAR image recognition system for recognition processing and calculation to obtain the image point coordinate set of the direct satellite image control point data in the multi-view SAR image. ,in =(u i ,w i ), This represents the x-coordinate of the image point in the multi-view SAR image based on the control point data from direct satellite imagery. The ordinate value of the image point in the multi-view SAR image representing the control point data of the direct satellite image; S5. Calculate the correction coefficients α` and β` of the direct-type range and azimuth polynomials: Find the image point coordinates of the direct-type satellite image control point data from step S4 in the multi-look SAR image. Orthorectification of SAR images is performed, and the polynomial correction coefficients α` and β` of the direct range direction and the direct azimuth direction are calculated. S6. Calculate the correction coefficients γ and ζ of the mean-valued range and azimuth polynomials: The correction coefficients α` and β` of the direct-form range polynomial obtained in step S5 are mean-valued by the correction coefficients α and β of the range polynomial obtained in step S3, respectively, to obtain the mean-valued range polynomial correction coefficient γ and the mean-valued azimuth polynomial correction coefficient ζ. S7. SAR Orthorectification: Based on the averaged range polynomial correction coefficient γ and the averaged azimuth polynomial correction coefficient ζ obtained in step S6, the original SAR image is orthorectified to obtain the SAR orthorectified DEM image.

2. The SAR image orthorectification method based on DEM according to claim 1, characterized in that: The image extraction and segmentation process performed in step S1 using the image segmentation and recognition algorithm specifically includes the following steps: T1. The original SAR image Z is first binarized using a binarization function to obtain foreground and background information from the original SAR image, resulting in a binarized image. T2. Remove noise from the image after binarization in step T1, and extract the data information features from the image using a feature extraction algorithm; T3. Based on the feature data information extracted in step T2, the image is divided into multi-view SAR images Z'.

3. The orthorectification method for SAR images based on DEM according to claim 1, characterized in that: The SAR image orthorectification processing in step S3 is as follows: Select the image point coordinate set of satellite image control point data in multi-view SAR imagery The coordinate values ​​in the image are compared with the satellite image control point data in the original SAR image coordinate set. The corresponding coordinate values ​​are then subtracted, as follows: ； Where α is the polynomial correction coefficient for the range direction and β is the polynomial correction coefficient for the azimuth direction.

4. The orthorectification method for SAR images based on DEM according to claim 1, characterized in that: In step S5, the image point coordinate set of the direct satellite image control point data in the multi-view SAR image The specific steps for orthorectifying SAR images are as follows: Selecting the image point coordinate set of direct satellite imagery control point data in multi-view SAR imagery The coordinate values ​​in the image are compared with the satellite image control point data in the original SAR image coordinate set. The corresponding coordinate values ​​are then subtracted, as follows: ； Where α` is the polynomial correction coefficient for the direct-form range direction, and β` is the polynomial correction coefficient for the direct-form azimuth direction.

5. The orthorectification method for SAR images based on DEM according to claim 1, characterized in that: The mean-averaging process in step S6 specifically includes the following steps: E1. Add the polynomial correction coefficients α' of the direct-form range direction to obtain the mean-valued polynomial correction coefficients γ of the range direction, denoted as: γ = ; E2. Add the correction coefficients β` of the direct-form azimuth polynomial and the correction coefficients β of the azimuth polynomial to obtain the correction coefficients ζ of the mean-valued azimuth polynomial. Let ζ = .

6. The orthorectification method for SAR images based on DEM according to claim 1, characterized in that: In step S2, the DEM model is used to simulate each data in step S1, and the coordinate relationship of the one-to-one mapping between the circumscribed rectangular digital elevation model RefDEM data, the DEM simulated SAR image, the DEM simulated SAR image and the multi-view SAR image radar coordinate space is obtained. The SAR simulated image consistent with the multi-view SAR image radar coordinate space is calculated. Then, grayscale matching is performed on the SAR simulated image and the multi-view SAR image to calculate the new coordinate relationship of the one-to-one mapping between the DEM simulated SAR image and the multi-view SAR image radar coordinate space.

7. The orthorectification method for SAR images based on DEM according to claim 6, characterized in that: The circumscribed rectangular digital elevation model (RefDEM) data includes the width, height, starting point geographic coordinates, latitudinal resolution, longitudinal resolution, and elevation value of the circumscribed rectangular digital elevation model (RefDEM) in the external reference DEM data. By acquiring the width nWidthDEM and height nHeightDEM of the circumscribed rectangular digital elevation model RefDEM, two real empty matrices of size (nHeightDEM, nWidthDEM) are created to store the circumscribed rectangular digital elevation model RefDEM data and the DEM-simulated SAR image data, respectively. A complex empty matrix of size (nHeightDEM, nWidthDEM) is created to store the coordinate relationship data of the one-to-one mapping between the DEM-simulated SAR image and the multi-view SAR image radar coordinate space.

8. The SAR image orthorectification method based on DEM according to claim 7, characterized in that: The coordinate relationship between the circumscribed rectangular digital elevation model (RefDEM) data and the multi-view SAR image radar coordinate space is calculated based on the rigorous SAR imaging model.

Images