A method and system for ambiguity suppression for SAR doppler centroid estimation

Abstract

The application relates to a SAR Doppler center estimation-oriented blur suppression method and system, which comprises the following steps: segmenting a SAR single-view complex image to obtain a plurality of sub-images; estimating an initial Doppler center frequency of each sub-image; estimating an azimuth blur ratio of each sub-image based on the initial Doppler center frequency of each sub-image and an amplitude image corresponding to each sub-image; constructing an azimuth blur suppression regularization function based on the azimuth blur ratio of each sub-image, calculating an optimal Doppler spectrum weighting coefficient based on the regularization function; suppressing the azimuth blur of each sub-image based on the optimal Doppler weighting coefficient, and estimating a final Doppler center frequency of each sub-image. The influence of the azimuth blur on the Doppler center estimation is effectively suppressed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of synthetic aperture radar signal and information processing, and particularly relates to a method and system for SAR Doppler center estimation oriented ambiguity suppression. BACKGROUND

[0002] Synthetic Aperture Radar (SAR) is an active microwave imaging sensor. Compared with traditional optical remote sensing and hyperspectral remote sensing, SAR has the ability of all-weather, all-day, large range, and high resolution imaging, and has become one of the main means of earth observation.

[0003] SAR Doppler center estimation is a prerequisite and necessary step for SAR imaging, and is also a core technology for applications such as ground moving target detection and sea current velocity estimation. Existing Doppler center estimation methods are mainly divided into two categories: frequency domain estimation and time domain estimation. The frequency domain estimation method correlates the estimation operator with the Doppler spectrum to find the zero crossing point to estimate the Doppler center; the time domain estimation method correlates the autocorrelation function of the time domain signal and the Doppler power spectrum to estimate the Doppler center by Wiener-Sinai theorem. However, these Doppler center estimation methods do not consider the azimuth ambiguity effect, resulting in a large deviation in the Doppler center estimation result in the area seriously affected by the azimuth ambiguity, which seriously affects SAR imaging and its applications.

[0004] Azimuth ambiguity is an inherent phenomenon in SAR images, which is caused by sampling the continuous Doppler spectrum with a limited sampling frequency, and is more common in spaceborne SAR images. For land, the strength of azimuth ambiguity is generally lower than that of real targets, so its characteristics are relatively unobvious; but for the weak scattering scene of the ocean, strong scattering points such as land and ships are easy to form false targets on the ocean background, i.e. "ghosts", which brings great trouble to SAR image applications such as ship detection.

[0005] To solve this problem, many scholars have carried out research on azimuth ambiguity and proposed a series of azimuth ambiguity suppression methods, including azimuth ambiguity suppression methods based on SAR single-view complex images, azimuth ambiguity suppression methods based on SAR amplitude images, and methods for azimuth ambiguity suppression using SAR multi-polarization data and multi-temporal data. However, these methods only focus on SAR amplitude images, and severely damage the phase information in the process of azimuth ambiguity suppression, resulting in that the single-view complex image based on ambiguity suppression cannot accurately estimate the Doppler center.

[0006] The above analysis shows that in order to accurately estimate the Doppler center frequency, the influence of azimuth ambiguity must be considered, and at the same time, the phase information must be protected to the greatest extent during azimuth ambiguity suppression. Obviously, the existing methods cannot meet the above two requirements at the same time. SUMMARY

[0007] In view of the above analysis, the embodiments of the present application aim to provide a blurring suppression method and system for SAR Doppler center estimation, to solve the problem that the single-view complex image after blurring suppression cannot accurately estimate the Doppler center due to the destruction of phase information in the prior art.

[0008] In one aspect, the embodiments of the present application provide a blurring suppression method for SAR Doppler center estimation, comprising the following steps:

[0009] Segmenting the SAR single-view complex image to obtain a plurality of sub-images; estimating the initial Doppler center frequency of each sub-image;

[0010] Estimating the azimuth blurring ratio of each sub-image based on the initial Doppler center frequency of each sub-image and the amplitude image corresponding to each sub-image;

[0011] Constructing an azimuth blurring suppression regularization function based on the azimuth blurring ratio of each sub-image, and calculating the optimal Doppler spectrum weighting coefficient based on the regularization function;

[0012] Suppressing the azimuth blurring of each sub-image based on the optimal Doppler weighting coefficient, and estimating the final Doppler center frequency of each sub-image.

[0013] Based on the further improvement of the above technical solution, the azimuth blurring suppression regularization function is:

[0014]

[0015] Wherein, λ represents a regularization parameter, and α L and α R represent the Doppler spectrum weighting coefficient, P(f) represents the Doppler power spectrum of the sub-image, AASR L and AASR R respectively represent the left azimuth blurring ratio and the right azimuth blurring ratio of the sub-image;

[0016] The Doppler spectrum weighting coefficient when the azimuth blurring suppression regularization function takes the maximum value is the optimal Doppler spectrum weighting coefficient of the sub-image.

[0017] Further, the initial Doppler center frequency of each sub-image is estimated by the following steps:

[0018] For each sub-image, a correlation function

[0019] The Doppler value corresponding to the zero point of the correlation function is the initial Doppler center frequency of the sub-image;

[0020] Wherein, P(f) represents the Doppler power spectrum of the sub-image, B(f) represents the Doppler center optimal estimation sub-module, Indicates the circular correlation.

[0021] Further, the Doppler center optimal estimation sub-module is:

[0022]

[0023] Wherein, A(f) is the antenna directivity diagram corresponding to the sub-image, A'(f) represents the derivative of A(f), and N(f) represents the noise power spectrum of the sub-image.

[0024] Further, the Doppler center optimal estimation sub-module is:

[0025]

[0026] Wherein, A(f) is the antenna directivity diagram corresponding to the sub-image, A'(f) represents the derivative of A(f).

[0027] Further, based on the initial Doppler center frequency of each sub-image and the amplitude image corresponding to each sub-image, the azimuth ambiguity ratio of each sub-image is estimated, comprising:

[0028] For each sub-image, a Wiener filter is constructed based on the initial Doppler center frequency of the sub-image, and the amplitude image corresponding to the sub-image is filtered;

[0029] The azimuth ambiguity ratio of each sub-image is estimated based on the amplitude image before and after filtering.

[0030] Further, the constructed Wiener filter is:

[0031]

[0032]

[0033]

[0034]

[0035] Wherein, H1(f) and H2(f) represent left ambiguity Wiener filter and right ambiguity Wiener filter respectively, P' m (f) represents the main signal normalized power spectrum, P' aL (f) represents the left ambiguity signal normalized power spectrum, P' aR (f) represents the right ambiguity signal normalized power spectrum, f dc Indicates the initial Doppler center frequency of the sub-image, W(f) represents the azimuth window function in the imaging process, G 2(·) represents the Doppler power spectrum, PRF represents the radar pulse repetition frequency, Ba represents the Doppler bandwidth, and f represents the azimuth frequency.

[0036] Further, the azimuth ambiguity ratio of each sub-image is estimated based on the amplitude image before and after filtering, comprising:

[0037] According to the power of the total signal being the sum of the main signal power and the ambiguous signal power, the following equation group is established:

[0038]

[0039] wherein,

[0040]

[0041]

[0042]

[0043] The azimuth ambiguity ratio of each sub-image is:

[0044]

[0045] wherein, P u represents the total signal power of the amplitude image before filtering, P z represents the total signal power of the amplitude image after filtering, Cm i represents the main signal power of the amplitude image after filtering by the i-th filter, Ca Li represents the left ambiguous signal power of the amplitude image after filtering by the i-th filter, Ca Ri represents the right ambiguous signal power of the amplitude image after filtering by the i-th filter, |H i (f)| represents the modulus of the i-th Wiener filter, AASR L and AASR R respectively represent the left azimuth ambiguity ratio and the right azimuth ambiguity ratio, P aL represents the left ambiguous signal power, P aR represents the right ambiguous signal power, P′ aL (f) represents the left ambiguous signal normalized power spectrum, P′ aR (f) represents the right ambiguous signal normalized power spectrum.

[0046] Further, the azimuth ambiguity of each sub-image is suppressed based on the optimal Doppler weighting coefficient, comprising:

[0047] The Doppler power spectrum of the sub-image is subjected to azimuth ambiguity suppression by using the formula

[0048] wherein, P(f) represents the Doppler power spectrum of the sub-image,​ and denotes the optimal Doppler spectrum weighting coefficient, denotes the Doppler power spectrum after azimuth ambiguity suppression, and denotes multiplication in the phase domain.

[0049] In another aspect, the embodiment of the present application provides an ambiguity suppression system for SAR Doppler center estimation, comprising the following modules:

[0050] An initial Doppler center estimation module is configured to segment a SAR single-aperture complex image to obtain a plurality of sub-images, and estimate an initial Doppler center frequency of each sub-image.

[0051] An azimuth ambiguity ratio estimation module is configured to estimate an azimuth ambiguity ratio of each sub-image based on the initial Doppler center frequency of each sub-image and an amplitude image corresponding to each sub-image.

[0052] A Doppler spectrum weighting coefficient calculation module is configured to construct an azimuth ambiguity suppression regularization function based on the azimuth ambiguity ratio of each sub-image, and calculate an optimal Doppler spectrum weighting coefficient based on the regularization function.

[0053] A final Doppler center estimation module is configured to suppress the azimuth ambiguity of each sub-image based on the optimal Doppler weighting coefficient, and estimate a final Doppler center frequency of each sub-image.

[0054] Compared with the prior art, the present application segments a SAR single-aperture complex image into a plurality of sub-images, estimates an azimuth ambiguity ratio of each sub-image based on the initial Doppler center frequency of each sub-image and an amplitude image corresponding to each sub-image, calculates an optimal Doppler weighting coefficient of the sub-image based on the azimuth ambiguity ratio, and suppresses the ambiguity of the sub-image (complex image), thereby realizing ambiguity suppression without destroying the phase information and improving the accuracy of Doppler center frequency estimation.

[0055] The above technical solutions can be combined with each other in the present application to realize more preferred combination solutions. Other features and advantages of the present application will be described in the subsequent description, and some advantages will become apparent from the description, or will be understood by implementing the present application. The purposes and other advantages of the present application can be realized and obtained from the contents specifically indicated in the description and the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0056] The accompanying drawings are included to provide a further understanding of the application, and are incorporated in and constitute a part of this specification, illustrate embodiments of the application, and together with the description serve to explain the principles of the application. In the drawings:

[0057] Figure 1 A flowchart of the ambiguity suppression method for SAR Doppler center estimation according to the embodiment of the present application;

[0058] Figure 2 A SAR image azimuth ambiguity diagram for an embodiment of the present application;

[0059] Figure 3 An amplitude diagram for SAR single-view complex image generation of an embodiment of the present application;

[0060] Figure 4 A Doppler center frequency estimation result diagram before SAR single-view complex image azimuth ambiguity suppression of an embodiment of the present application;

[0061] Figure 5 A SAR image azimuth ambiguity ratio estimation result diagram of an embodiment of the present application;

[0062] Figure 6 A Doppler center frequency estimation result diagram after SAR single-view complex image azimuth ambiguity suppression of an embodiment of the present application;

[0063] Figure 7 A block diagram of an ambiguity suppression system for SAR Doppler center estimation of an embodiment of the present application. DETAILED DESCRIPTION

[0064] The preferred embodiments of the present application will be described in detail below with reference to the drawings, which form a part of this application. The drawings and the associated descriptions are provided to illustrate the embodiments of the present application and to explain the principles of the present application, and are not intended to limit the scope of the present application.

[0065] The existing azimuth ambiguity suppression methods, including the azimuth ambiguity suppression method based on SAR single-view complex image, the azimuth ambiguity suppression method based on SAR amplitude diagram, and the azimuth ambiguity suppression method using SAR multi-polarization data and multi-temporal data, only focus on the SAR amplitude diagram, and the amplitude diagram is obtained by taking the modulus of the single-view complex image, so that the phase information is seriously damaged in the process of azimuth ambiguity suppression, resulting in that the Doppler center frequency cannot be accurately estimated based on the single-view complex image after ambiguity suppression.

[0066] One specific embodiment of the present application discloses a SAR Doppler center estimation-oriented ambiguity suppression method, as shown in Figure 1 which includes the following steps:

[0067] S1, performing segmentation on a SAR single-view complex image to obtain a plurality of sub-images; and estimating an initial Doppler center frequency of each sub-image;

[0068] S2, estimating an azimuth ambiguity ratio of each sub-image based on the initial Doppler center frequency of each sub-image and an amplitude image corresponding to each sub-image;

[0069] S3, constructing an azimuth blurring suppression regularization function based on the azimuth blurring ratio of each sub-image, and calculating an optimal Doppler spectrum weighting coefficient based on the regularization function;

[0070] S4, suppressing the azimuth blurring of each sub-image based on the optimal Doppler weighting coefficient, and estimating a final Doppler center frequency of each sub-image.

[0071] In implementation, the SAR single-view complex image can be obtained by a SAR imaging algorithm such as RD (Range Doppler) or CS (Chirp Scaling), or can be downloaded from a public data website.

[0072] For example, the single-view complex image of GF-3 satellite can be selected for Doppler center estimation. Figure 3 As shown in FIG. 1, a standard strip SAR single-view complex image is shown. As can be seen from the figure, the upper half of the image is mainly sea area, and the lower half is an island. The island forms a significant azimuth blurring in the sea area of the SAR image, that is, the "ghost" of the island. In addition, a small island in the upper half of the image also has a significant azimuth blurring phenomenon in the sea.

[0073] Because the motion speeds of scatterers in different regions are different, the corresponding Doppler center frequencies are different. In addition, the azimuth blurring ratios of different regions are also different. Therefore, in implementation, the SAR single-view complex image can be divided into M*N sub-images in the azimuth direction and the range direction, and each sub-image has the same size, that is, the number of division in the azimuth direction is M, and the number of division in the range direction is N, wherein M and N are positive integers not less than 3. Thus, the center frequency estimation is more accurate.

[0074] If the number of divided sub-images is large, the accuracy of Doppler center estimation can be reduced. If the number of divided sub-images is small, the spatial resolution of Doppler center estimation can be reduced. Therefore, in implementation, the sizes of M and N can be set according to the specific requirements of accuracy and spatial resolution.

[0075] The present application divides the SAR single-view complex image into multiple sub-images, estimates the azimuth blurring ratio of each sub-image based on the initial Doppler center frequency of each sub-image and the amplitude image corresponding to each sub-image, calculates the optimal Doppler weighting coefficient of each sub-image based on the azimuth blurring ratio, and thus suppresses the blurring of the sub-image (complex image), thereby realizing blurring suppression without destroying the phase information and improving the accuracy of Doppler center frequency estimation.

[0076] Specifically, the initial Doppler center frequency of each sub-image is estimated by the following steps in step S1:

[0077] S11, for each sub-image, constructing a correlation function Wherein, P(f) represents the Doppler power spectrum of the sub-image, B(f) represents a Doppler center optimal estimation sub-image, The circular correlation is represented.

[0078] It should be noted that P(f) is the Doppler power spectrum calculated by Fourier transform of the sub-image. That is, the sub-image is transformed to the range-Doppler domain, and the Doppler power spectrum is calculated.

[0079] In implementation, since the Doppler power spectrum P(f) is an even function, the zero-crossing point of the result of the correlation of the odd function and the even function represents the center position of the even function, that is, the Doppler center. Therefore, the Doppler center optimal estimation sub-image B(f) should be an odd function.

[0080] Specifically, the Doppler center optimal estimation sub-image can be:

[0081]

[0082] Wherein, A(f) is the antenna directivity pattern corresponding to the sub-image, A'(f) represents the derivative of A(f), and N(f) represents the noise power spectrum of the sub-image.

[0083] In implementation, if the accurate system noise is not provided in the spaceborne SAR data product, since the system noise is usually Gaussian white noise, the influence on the Doppler estimation is very small, and therefore, the Doppler center optimal estimation sub-image is:

[0084]

[0085] Wherein, A(f) is the antenna directivity pattern corresponding to the sub-image, and A'(f) represents the derivative of A(f).

[0086] After the optimal estimation sub-image is obtained, the initial Doppler center frequency is estimated by correlating the Doppler power spectrum and the optimal estimation sub-image.

[0087] The SAR image of Figure 3 is segmented, and the initial Doppler center frequency is estimated according to steps S11-S12, and the estimation result is as shown in Figure 4 .

[0088] S12, the Doppler value corresponding to the zero point of the correlation function is the initial Doppler center frequency of the sub-image;

[0089] Specifically, D(f) is the result of the correlation of the two functions, and the value thereof represents the difference between the Doppler spectrum energy on both sides of the estimated Doppler center. Since the Doppler power spectrum is even symmetric, the Doppler value corresponding to the zero point of the function is the optimal Doppler estimation value, that is, the initial Doppler center frequency.

[0090] After obtaining the initial Doppler center frequency of each sub-image, an azimuth ambiguity ratio of each sub-image is estimated based on the initial Doppler center frequency of each sub-image and an amplitude image corresponding to each sub-image. Specifically, step S2 comprises:

[0091] S21, for each sub-image, a Wiener filter is constructed based on the initial Doppler center frequency of the sub-image, and an amplitude image corresponding to the sub-image is filtered;

[0092] A Wiener filter is a linear filter with the least square as the optimal criterion. Under certain constraints, the square of the difference between its output and a given function (usually referred to as the expected output) reaches the minimum, which can finally be changed into a solution problem of a Toeplitz equation through mathematical operation. Wiener filtering is a method of filtering signals mixed with noise by using the correlation characteristics and spectral characteristics of stationary random processes.

[0093] In implementation, first, left and right Wiener filters H1(f) and H2(f) are constructed, and the constructed Wiener filter is:

[0094]

[0095] wherein H1(f) and H2(f) represent left and right ambiguity Wiener filters respectively. P' m (f) represents a normalized power spectrum of a main signal, P' aL (f) represents a normalized power spectrum of a left ambiguity signal, P' aR (f) represents a normalized power spectrum of a right ambiguity signal.

[0096]

[0097]

[0098]

[0099] wherein f dc represents the initial Doppler center frequency of the sub-image, W(f) represents an azimuth window function in the imaging process, G 2 (·) represents a Doppler power spectrum, PRF represents a radar pulse repetition frequency, Ba represents a Doppler bandwidth, and f represents an azimuth frequency.

[0100] It should be noted that G 2 (·) represents a theoretical Doppler power spectrum, which can be obtained according to a two-way antenna directivity pattern. For the case where the antenna directivity pattern is unknown, the following formula can be used for equivalence:

[0101]

[0102] wherein f represents the azimuth frequency, sinc(·) represents the sinc function, and Ba represents the Doppler bandwidth.

[0103] Different sub-images have different degrees of blurring, and the estimated Doppler center frequency deviation of a SAR image blurring serious area is larger. In order to further accurately estimate the azimuth blurring ratio of each sub-image, when constructing the Wiener filter, for the sub-image with serious blurring, the initial Doppler center frequency in the formulas (5) and (6) for calculating the normalized power spectrum of the left and right blurred signals can be replaced by the initial Doppler center frequency of the sub-image with lower blurring degree closest to it.

[0104] S22, estimating the azimuth blurring ratio of each sub-image based on the amplitude images before and after filtering.

[0105] After the left / right Wiener filter is constructed, the amplitude image corresponding to the sub-image is filtered, and the azimuth blurring ratio of each sub-image is estimated based on the amplitude images before and after filtering.

[0106] Specifically, the step S22 comprises:

[0107] The following equation set is established according to the total signal power being the sum of the main signal power and the blurred signal power:

[0108]

[0109] wherein,

[0110]

[0111]

[0112]

[0113] Then, the azimuth blurring ratio of each sub-image is:

[0114]

[0115] wherein P u represents the total signal power before filtering of the amplitude image, P z represents the total signal power after filtering of the amplitude image, Cm i represents the main signal power after filtering of the amplitude image by the i th filter, Ca Li represents the left blurred signal power after filtering of the amplitude image by the i th filter, Ca Ri represents the right blurred signal power after filtering of the amplitude image by the i th filter, |H i (f)| represents the modulus of the i th Wiener filter, AASR L and AASR R respectively represent the left azimuth blurring ratio and the right azimuth blurring ratio, PaL denotes the left blur signal power, P aR denotes the right blur signal power, P' aL (f) denotes the left blur signal normalized power spectrum, P' aR (f) denotes the right blur signal normalized power spectrum.

[0116] According to step S2, the azimuth blur ratio estimation is performed on the amplitude map corresponding to the SAR single-view complex image of Figure 3 The estimation result is shown in FIG. 3. Figure 6

[0117] After obtaining the azimuth blur ratio of each sub-image, the azimuth blur suppression regularization function is constructed based on the azimuth blur ratio of each sub-image, and the optimal Doppler spectrum weighting coefficient is calculated based on the regularization function.

[0118] Specifically, the azimuth blur suppression regularization function constructed in step S3 is as follows:

[0119]

[0120] wherein λ denotes a regularization parameter, α L and α R denote the Doppler spectrum weighting coefficient, P(f) denotes the Doppler power spectrum of the sub-image, AASR L and AASR R denote the left azimuth blur ratio and the right azimuth blur ratio of the sub-image respectively.

[0121] That is, the local Doppler power spectrum P(f) of each sub-image block of the SAR single-view complex image calculated in the range-Doppler domain is combined with the azimuth blur ratio AASR L and AASR R of each sub-image, and the azimuth blur suppression regularization function corresponding to the sub-image block is established according to formula (13).

[0122] wherein the value of λ can be a fixed value or a variable parameter. is the optimal Doppler spectrum weighting coefficient to be estimated for the sub-image.

[0123] The Doppler spectrum weighting coefficient when the azimuth blur suppression regularization function takes the maximum value is the optimal Doppler spectrum weighting coefficient of the sub-image.

[0124] In implementation, the Newton iteration method can be used to obtain the Doppler spectrum weighting coefficient when the regularization function takes the maximum value as the optimal Doppler spectrum weighting coefficient.

[0125] After obtaining the optimal Doppler weighting coefficient of each sub-image, the azimuth blur of each sub-image is suppressed based on the optimal Doppler weighting coefficient. Specifically, step S4 includes: ​

[0126] Using formula Suppress orientation blur in the Doppler power spectrum of the sub-image;

[0127] Where P(f) represents the Doppler power spectrum of the sub-image. and Represents the optimal Doppler spectrum weighting coefficients. This represents the Doppler power spectrum after azimuth ambiguity suppression, and * indicates positional multiplication.

[0128] After suppressing the orientation blur of each sub-image using the optimal Doppler weighting coefficients, the Doppler center frequency of each sub-image can be re-estimated using steps S11-S12, resulting in the final Doppler center frequency of each sub-image. The correlation function is... The Doppler value corresponding to the zero point of the correlation function is the final Doppler center frequency.

[0129] The Doppler center frequency of the SAR single-look complex image after orientation blur suppression is re-estimated, and the estimation results are as follows. Figure 6 As shown.

[0130] By Figure 6 The Doppler center frequency estimation results after azimuth ambiguity ratio suppression are compared with Figure 4 By comparing the Doppler center frequency estimation results before azimuth ambiguity suppression, it can be found that... Figure 4 The mid-Doppler center estimation results are severely affected by azimuth ambiguity, and the Doppler center frequency estimation results are inconsistent with... Figure 5 The azimuth ambiguity estimation results are similar, and the Doppler center estimation results have large deviations in areas with large azimuth ambiguity. However, the Doppler center estimation results after azimuth ambiguity suppression have better continuity and are closer to the actual situation, thus proving the effectiveness of the proposed method.

[0131] Compared with the prior art, the present invention has the following beneficial effects:

[0132] 1. This invention starts from SAR imaging characteristics, fully considers the impact of azimuth ambiguity on SAR Doppler center frequency estimation, and effectively suppresses azimuth ambiguity by preserving phase information, thereby improving the accuracy of Doppler center frequency estimation.

[0133] 2. Starting from the range Doppler domain, this invention makes full use of the local azimuth ambiguity ratio, effectively suppressing the influence of azimuth ambiguity on the estimation of Doppler center frequency while preserving the Doppler spectrum to the maximum extent.

[0134] A specific embodiment of the present invention discloses a fuzziness suppression system for SAR Doppler center estimation, such as... Figure 7 As shown, it includes the following modules:

[0135] an initial Doppler center estimation module configured to segment the SAR single-view complex image to obtain a plurality of sub-images, and estimate an initial Doppler center frequency of each sub-image;

[0136] an azimuth blurring ratio estimation module configured to estimate an azimuth blurring ratio of each sub-image based on the initial Doppler center frequency of each sub-image and an amplitude image corresponding to each sub-image;

[0137] a Doppler spectrum weighting coefficient calculation module configured to construct an azimuth blurring suppression regularization function based on the azimuth blurring ratio of each sub-image, and calculate an optimal Doppler spectrum weighting coefficient based on the regularization function;

[0138] a final Doppler center estimation module configured to suppress the azimuth blurring of each sub-image based on the optimal Doppler weighting coefficient, and estimate a final Doppler center frequency of each sub-image.

[0139] The method embodiments and the system embodiments described above are based on the same principle, and can achieve the same technical effects by mutual reference. For the specific implementation process, please refer to the foregoing embodiments, which will not be described here again.

[0140] Those skilled in the art can understand that all or part of the processes of the above-mentioned embodiments can be completed by a computer program instructing related hardware. The program can be stored in a computer readable storage medium, such as a magnetic disk, an optical disk, a read-only memory or a random access memory.

[0141] The above description is only the preferred embodiments of the present application, but the protection scope of the present application is not limited to this. Any changes or replacements within the technical scope disclosed by the present application can be easily thought by those skilled in the art, and should be covered within the protection scope of the present application.

Claims

1. A fuzziness suppression method for SAR Doppler center estimation, characterized in that, Includes the following steps: The SAR single-look complex image is segmented to obtain multiple sub-images; Estimate the initial Doppler center frequency for each sub-image; The azimuth blur ratio of each sub-image is estimated based on the initial Doppler center frequency of each sub-image and the amplitude image corresponding to each sub-image; An orientation blur suppression regularization function is constructed based on the orientation blur ratio of each sub-image, and the optimal Doppler spectrum weighting coefficient is calculated based on the regularization function. Based on the optimal Doppler weighting coefficients, the orientation blur of each sub-image is suppressed, and the final Doppler center frequency of each sub-image is estimated. The azimuth ambiguity suppression regularization function is: ， in, Represents the regularization parameter. and Represents the weighting coefficients of the Doppler spectrum. The Doppler power spectrum of the sub-image is represented. and These represent the left and right blur ratios of the sub-image, respectively; The Doppler spectral weighting coefficient that maximizes the orientation blur suppression regularization function is the optimal Doppler spectral weighting coefficient for the sub-image. The initial Doppler center frequency of each sub-image is estimated using the following steps: For each sub-image, construct the relevant function. ; The Doppler value corresponding to the zero of the correlation function is the initial Doppler center frequency of the sub-image; in, This represents the Doppler power spectrum of the sub-image. This represents the optimal estimator for the Doppler center. Indicates circumference correlation; The optimal Doppler center estimator is: , in, This is the antenna radiation pattern corresponding to the sub-image. express The derivative, This represents the noise power spectrum of the sub-image.

2. The ambiguity suppression method for SAR Doppler center estimation according to claim 1, characterized in that, The optimal Doppler center estimator is: ， in, This is the antenna radiation pattern corresponding to the sub-image. express The derivative of .

3. The ambiguity suppression method for SAR Doppler center estimation according to claim 1, characterized in that, The azimuth blur ratio of each sub-image is estimated based on the initial Doppler center frequency of each sub-image and the amplitude image corresponding to each sub-image, including: For each sub-image, a Wiener filter is constructed based on the initial Doppler center frequency of the sub-image, and the amplitude image corresponding to the sub-image is filtered. The azimuth blur ratio of each sub-image is estimated based on the amplitude images before and after filtering.

4. The ambiguity suppression method for SAR Doppler center estimation according to claim 3, characterized in that, The constructed Wiener filter is as follows: ， ， ， ， ， in, and These represent the left fuzzy Wiener filter and the right fuzzy Wiener filter, respectively. This represents the normalized power spectrum of the main signal. This represents the normalized power spectrum of the left-fuzzy signal. This represents the normalized power spectrum of the right-fuzzy signal. This represents the initial Doppler center frequency of the sub-image. This represents the azimuth window function during the imaging process. Represents the Doppler power spectrum. Indicates the radar pulse repetition frequency. Indicates the Doppler bandwidth. Indicates the azimuth frequency.

5. The ambiguity suppression method for SAR Doppler center estimation according to claim 3, characterized in that, The azimuth blur ratio of each sub-image is estimated based on the amplitude images before and after filtering, including: Based on the sum of the power of the total signal and the power of the main signal and the power of the fuzzy signal, the following system of equations is established: , in, , , , The orientation blur ratio of each sub-image is: , in, This represents the total signal power before amplitude image filtering. This represents the total signal power after amplitude image filtering. This represents the power of the main signal after the amplitude image is filtered by the i-th filter. This represents the power of the left blurred signal after the amplitude image is filtered by the i-th filter. This represents the power of the right-blurred signal after the amplitude image is filtered by the i-th filter. This represents the magnitude of the i-th Wiener filter. and These represent the blur ratios in the left and right directions, respectively. Indicates the power of the left fuzzy signal. Indicates the power of the right-hand fuzzy signal. This represents the normalized power spectrum of the left-fuzzy signal. Normalized power spectrum of right-fuzzy signal.

6. The ambiguity suppression method for SAR Doppler center estimation according to claim 1, characterized in that, Based on the optimal Doppler weighting coefficients, orientation blur of each sub-image is suppressed, including: Using formula Suppress orientation blur in the Doppler power spectrum of the sub-image; in, This represents the Doppler power spectrum of the sub-image. and Represents the optimal Doppler spectrum weighting coefficients. This represents the Doppler power spectrum after azimuth ambiguity suppression. This indicates digit-wise multiplication.

7. A fuzziness suppression system for SAR Doppler center estimation, characterized in that, Includes the following modules: The initial Doppler center estimation module is used to segment the SAR single-look complex image to obtain multiple sub-images; Estimate the initial Doppler center frequency for each sub-image; The azimuth blur ratio estimation module is used to estimate the azimuth blur ratio of each sub-image based on the initial Doppler center frequency of each sub-image and the amplitude image corresponding to each sub-image. The Doppler spectrum weighting coefficient calculation module is used to construct an azimuth blur suppression regularization function based on the azimuth blur ratio of each sub-image, and calculate the optimal Doppler spectrum weighting coefficient based on the regularization function; The final Doppler center estimation module is used to suppress the orientation blur of each sub-image based on the optimal Doppler weighting coefficients and estimate the final Doppler center frequency of each sub-image. The azimuth ambiguity suppression regularization function is: ， in, Represents the regularization parameter. and Represents the weighting coefficients of the Doppler spectrum. The Doppler power spectrum of the sub-image is represented. and These represent the left and right blur ratios of the sub-image, respectively; The Doppler spectral weighting coefficient that maximizes the orientation blur suppression regularization function is the optimal Doppler spectral weighting coefficient for the sub-image. The initial Doppler center frequency of each sub-image is estimated using the following steps: For each sub-image, construct the relevant function. ; The Doppler value corresponding to the zero of the correlation function is the initial Doppler center frequency of the sub-image; in, This represents the Doppler power spectrum of the sub-image. This represents the optimal estimator for the Doppler center. Indicates circumference correlation; The optimal Doppler center estimator is: , in, This is the antenna radiation pattern corresponding to the sub-image. express The derivative, This represents the noise power spectrum of the sub-image.

Images