Antenna phase center track error compensation method based on image registration

Abstract

The invention discloses an antenna phase center track error compensation method based on image registration, and belongs to the field of SAR data processing. The method comprises the following steps: imaging SAR echo data and a corresponding antenna phase center track by using a video backward projection algorithm to obtain a sub-aperture SAR image; registering the SAR image frame sequence to obtain a transformation relation, taking the first frame as a reference during registration, and registering a subsequent frame image to the first frame; calculating a rotation angle between the image frames; performing low-order fitting on the mismatching amount; establishing a relationship between the SAR image frame sequence mismatch amount and the antenna phase center trajectory; and compensating the phase center track of the antenna by using the fitted mismatch amount. According to the method, motion error compensation is carried out by using the video SAR inter-frame mismatching amount, the mismatching amount of the video SAR image frame sequence is obtained through a registration algorithm, accurate compensation of the antenna phase center track is realized, finally, the SAR image imaging quality is effectively improved, and the method has relatively high robustness.

Description

Technical Field

[0001] The invention belongs to the field of SAR data processing, and in particular relates to an antenna phase center trajectory error compensation method based on image registration. Background Art

[0002] Compared with optical and infrared remote sensing equipment, synthetic aperture radar (SAR) can achieve reliable imaging of the observation area in optically unfriendly environments (clouds, smoke, rain, snow, optical low scattering, etc.), and has become one of the important technical means for current and future military reconnaissance, resource remote sensing, and terrain mapping. Video SAR has the advantages of traditional SAR and can further obtain time-sensitive features of the observation area. It has become one of the hot spots in radar imaging technology research in recent years.

[0003] Accurate antenna phase center trajectory is the basis for reliable and high-precision video SAR imaging. However, the accuracy of the antenna phase center trajectory is affected by factors such as the accuracy of the trajectory measurement module and the uncertainty of the carrier flight. In order to achieve high-precision imaging, it is necessary to perform motion compensation on its trajectory error. In terms of motion compensation, the main idea of ​​current research at home and abroad is to perform motion error compensation through echo analysis and phase error estimation. This type of method requires extracting prominent points in the scene for phase estimation. The algorithm structure is complex, the amount of calculation is large, the estimation accuracy is greatly affected by environmental factors, the reliability is low, and it cannot meet the needs of high-precision video SAR imaging processing. Therefore, it is of great significance to study new motion compensation methods. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the above-mentioned shortcomings and provide an antenna phase center trajectory error compensation method based on image registration. The method analyzes the correlation between the mismatch amount of the image frame sequence and the antenna phase center measurement error, uses the video SAR frame sequence to estimate the error parameters, and realizes antenna phase center trajectory error compensation.

[0005] The technical solution adopted by the present invention is:

[0006] A method for compensating antenna phase center trajectory error based on image registration is proposed. First, the video SAR sub-aperture images are registered to obtain the mismatch between the images. In order to reduce the error caused by the registration process, a low-order fitting is performed on the mismatch. According to the relationship between the antenna phase center trajectory error and the mismatch of the video SAR image frame sequence, the mismatch is used to compensate the trajectory error. The main steps are as follows:

[0007] S1, imaging the SAR echo data and its corresponding antenna phase center trajectory using a video backprojection algorithm to obtain a sub-aperture SAR image;

[0008] S2, registering the SAR image frame sequence to obtain the transformation relationship;

[0009] S3, calculating the rotation angle between image frames;

[0010] S4, low-order fitting of the mismatch;

[0011] S5, establishing the relationship between the mismatch amount of the SAR image frame sequence and the antenna phase center trajectory;

[0012] S6, using the fitted mismatch amount to compensate the antenna phase center trajectory.

[0013] Furthermore, step S2 further includes: registering the sub-aperture SAR images to obtain the transformation relationship {R n ,t n}, where R n represents the rotation matrix, t n represents the translation vector, which is expressed as follows:

[0014]

[0015] t n =(t x ,t y ,0) T

[0016] Among them, θ n represents the rotation angle around the z-axis; t x , t y They are expressed as the translation in the x and y directions respectively.

[0017] To avoid the accumulation of measurement errors, the first frame is used as a reference during registration, and subsequent frame images are registered to the first frame.

[0018] Further, step S3 further includes: calculating one of the mismatch amounts between the image frames in the following manner, namely, the rotation angle θ n :

[0019] θ n =arccosA 11

[0020] In the formula, A 11 Represents R n The first row and first column element in .

[0021] Furthermore, step S4 also includes: due to the defocus phenomenon of some images or the influence of the registration algorithm itself, the estimated mismatch between image frames will have errors, and a low-order fitting operation needs to be performed on the rotation angle and the translation amount to obtain the rotation angle after fitting. and translation

[0022] Furthermore, step S6 also includes: using the mismatch amount after fitting, that is, the rotation angle and translation Construct transformation relationship Compensate the antenna phase center trajectory, expressed as:

[0023]

[0024] in, represents the trajectory of the antenna phase center after compensation at time n; Represents the antenna phase center trajectory measured at time n.

[0025] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of an antenna phase center trajectory error compensation method based on image registration described in the present invention.

[0026] The beneficial effects of the present invention include:

[0027] By analyzing the correlation between the mismatch of the video SAR frame sequence and the measurement error of the antenna phase center, the present invention uses the mismatch between video SAR frames to perform motion error compensation, obtains the mismatch of the video SAR image frame sequence through the registration algorithm, and realizes accurate compensation of the antenna phase center trajectory, and finally effectively improves the imaging quality of SAR images, with strong robustness. At present, the existing technology mainly performs motion error compensation through echo analysis and phase error estimation, and needs to extract the prominent points in the scene for phase estimation. The algorithm structure is complex, the amount of calculation is large, the estimation accuracy is greatly affected by environmental factors, the reliability is low, and it cannot meet the needs of high-precision imaging processing of video SAR. Compared with the existing technology, the present invention only uses the amplitude information of the image, without using the phase information, and has the advantages of strong robustness and low computational complexity. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The present invention is a flowchart of the method.

[0029] Figure 2 The figure is a comparison of SAR imaging results after compensating the antenna phase center trajectory with that without compensation, where a is the image after compensating the antenna phase center trajectory, and b is the image without compensating the antenna phase center trajectory. DETAILED DESCRIPTION

[0030] To facilitate those skilled in the art to understand the technical content of the present invention, the present invention is further explained below in conjunction with specific embodiments and drawings.

[0031] Example:

[0032] like Figure 1 As shown, a method for compensating antenna phase center trajectory error based on image registration includes:

[0033] S1: Select the W-band SAR raw echo data and the corresponding antenna phase center trajectory, use the video backprojection algorithm for imaging, set the resolution to 0.2m×0.2m, and obtain a series of sub-aperture SAR images.

[0034] S2: Use the scale-invariant feature transformation method to detect image feature points, then use the Euclidean distance to match feature points, and use the fast sample consistency algorithm to eliminate all incorrectly matched feature points, and calculate the transformation relationship between images {R n ,t n}. R n represents the rotation matrix, t n represents the translation vector, which is expressed as follows:

[0035]

[0036] t n =(t x ,t y ,t z ) T

[0037] Among them, φ n represents the rotation angle around the x-axis; ψ n Represents the rotation angle around the y-axis; θ n represents the rotation angle around the z-axis; t x , t y , t z They are expressed as the translation in the x, y, and z directions respectively.

[0038] Since the images are registered on a two-dimensional plane, φ n =ψ n =0,t z = 0, so {R n ,t n} can be simplified to:

[0039]

[0040] t n =(t x ,t y ,0) T

[0041] To avoid the accumulation of measurement errors, the first frame is used as a reference during registration, and subsequent frame images are registered to the first frame.

[0042] S3: Calculate the rotation angle θ between image frames as follows n :

[0043] θ n =arccosA 11

[0044] In the formula, A 11 Represents R n The first row and first column element in .

[0045] S4: Due to the defocus phenomenon of some images or the influence of the registration algorithm itself, there will be errors in the mismatch between the estimated image frames. The least squares method is used to perform low-order fitting operations on the rotation angle and translation, and the rotation angle after fitting can be obtained. and translation

[0046] S5: Establish the relationship between the antenna phase center trajectory and the SAR image frame sequence mismatch. The specific process is as follows:

[0047] The relationship between the measured antenna phase center trajectory and the true trajectory is expressed as:

[0048]

[0049] Among them, p apc (n) represents the actual antenna phase center trajectory at time n; represents the antenna phase center trajectory measured at time n; L n [·] denotes a time-invariant linear transformation.

[0050] The distance history R(n,q) of the scattering point q(x,y,0) is expressed as:

[0051] R(n;q)=‖p apc (n)-q(x,y,0)||2

[0052] The compensated historical distance of the pixel point q'(x,y,0) in the imaging space It is expressed as:

[0053]

[0054] In general, for any transformation matrix L n [·] and vector x, ||x||2≠||L n [x]||2. However, in a very small time interval (such as a sub-aperture time), L n [·] is approximately time-invariant, so ||x||2=||L n[x]||2, which is an isometric transformation, such as translation, rotation and mirror transformation, etc., in this case:

[0055]

[0056] in, Represents the transformation matrix L n The inverse matrix of [·].

[0057] If and only if When , the scattering point q(x,y,0) can be projected to the pixel point q'(x,y,0) in the imaging space. Then the relationship between the theoretical pixel point (without antenna phase center trajectory error) and the actual pixel point (with antenna phase center trajectory error) projected by the scattering point can be established:

[0058] q'(x,y,0)=L n -1 [q(x,y,0)]

[0059] Since the trajectory error of the antenna phase center reflects the rotation and translation errors between the sub-aperture images, the time-invariant isometric transformation L between the sub-aperture images is measured. n [·], the actual antenna phase center trajectory can be obtained by Rebuild.

[0060] S6:L n [·] That is, by θ n ,t x ,t y The rotation and translation transformation relationship formed by step S4 is used to fit the rotation angle between the sub-aperture images. and translation Construct transformation relationship Compensate the antenna phase center trajectory, expressed as:

[0061]

[0062] in, represents the phase center trajectory of the antenna after compensation.

[0063] like Figure 2 As shown, Figure 2 In a, the image features after antenna phase center trajectory compensation are clearly visible, the object is well focused, and the shadow of the moving target is clearly visible. Figure 2 In b, when the antenna phase center trajectory is not compensated, the object in the image is severely defocused and the object features are not recognizable. It can be seen that the present invention can effectively compensate for the antenna phase center trajectory error, thereby improving the imaging quality of video SAR images.

[0064] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. An antenna phase center trajectory error compensation method based on image registration, characterized in that: The following steps are involved: S1, imaging the SAR echo data and its corresponding antenna phase center trajectory using a video backprojection algorithm to obtain a sub-aperture SAR image; S2, registering the SAR image frame sequence to obtain the transformation relationship, taking the first frame as a reference during registration, and registering the subsequent frame images to the first frame; S3, calculating the rotation angle between image frames; S4, low-order fitting of the mismatch; S5, establishing the relationship between the mismatch amount of the SAR image frame sequence and the antenna phase center trajectory; S6, using the fitted mismatch amount to compensate the antenna phase center trajectory.

2. The method according to claim 1, characterized in that The step S2 further comprises: Register the sub-aperture SAR images and obtain the transformation relationship between image frames {R n ,t n }, where R n represents the rotation matrix, t n represents the translation vector, which is expressed as follows: t n =(t x ,t y ,0) T Where: θ n represents the rotation angle around the z-axis; t x , t y They are expressed as the translation in the x and y directions respectively.

3. The method according to claim 1, characterized in that The step S3 further comprises: Calculate the mismatch between image frames, which includes the rotation angle θ n : i n =arccosA 11 In the formula, A 11 Represents R n The first row and first column element in .

4. The method according to claim 1, characterized in that The step S4 further comprises: The rotation angle and translation amount are fitted at a low order to obtain the rotation angle after fitting. and translation 5. The method according to claim 1, characterized in that The step S5 further comprises: The relationship between the antenna phase center trajectory and the SAR image frame sequence mismatch is established. The specific process is as follows: The relationship between the measured antenna phase center trajectory and the true trajectory is expressed as: Where: p apc (n) represents the actual antenna phase center at time n; represents the antenna phase center measured at time n; L n [·] denotes a time-invariant linear transformation. The distance history R(n,q) of the scattering point q(x,y,0) is expressed as: R(n;q)=||p apc (n)-q(x,y,0)||2 The compensated historical distance of the pixel point q'(x,y,0) in the imaging space It is expressed as: After isometric transformation: In the formula, Denotes the transformation matrix L n The inverse matrix of [·]; If and only if When , the scattering point q(x,y,0) can be projected to the pixel point q'(x,y,0) in the imaging space; let The relationship between the theoretical pixel point without antenna phase center trajectory error and the actual pixel point with antenna phase center trajectory error is established, which is expressed as: q'(x,y,0)=L n -1 [q(x,y,0)] The time-invariant isometric transformation L between sub-aperture images is measured n [·], the true antenna phase center trajectory is obtained through Rebuild.

6. The method according to any one of claims 1 to 5, characterized in that: The step S6 further comprises: The mismatch after fitting is the rotation angle and translation Construct transformation relationship Compensate the antenna phase center trajectory, expressed as: Where: represents the trajectory of the antenna phase center after compensation at time n; Represents the antenna phase center trajectory measured at time n.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the antenna phase center trajectory error compensation method based on image registration as described in any one of claims 1 to 6 are implemented.

Images