A SAR sea surface imaging method based on radial velocity field

Abstract

The present invention discloses a SAR sea surface imaging method based on radial velocity field. By obtaining the radial velocity field based on SAR data, sea surface imaging is achieved, which can improve the sea surface texture focusing effect, thereby improving the interpretability of SAR sea surface images, and at the same time can improve the utilization rate of SAR ocean data, help improve the sea surface texture characteristics, and lay a foundation for subsequent target detection.

Description

Technical Field

[0001] The invention belongs to the technical field of synthetic aperture radar (SAR) signal processing, and in particular relates to a SAR sea surface imaging method based on radial velocity field. Background Art

[0002] Synthetic Aperture Radar (SAR) is a high-resolution imaging radar. In the range direction, it obtains a high range resolution by emitting broadband pulse signals. In the azimuth direction, SAR uses the movement of the radar platform to form a longer "aperture" through signal processing, thereby obtaining a high azimuth resolution. Compared with traditional optical sensors, SAR has the advantages of all-day, all-weather, high resolution, and a wide observation range. Ocean observation is one of the important application directions of SAR. SAR can detect a variety of ocean phenomena such as waves, internal waves, vortices, and fronts, can invert vector information such as wind fields and flow fields, and can identify and classify oil films and ships. SAR sea surface imaging is the basis of sea observation, but due to the overall random movement of the sea surface, it usually brings challenges to sea surface imaging.

[0003] At present, the commonly used SAR imaging algorithms are RD algorithm, CS algorithm, Omega-K algorithm and BP algorithm. SAR is mainly used to image stationary targets or scenes on the ground. When SAR is used to observe the sea, the random movement of the sea surface must be considered, otherwise the quality of the SAR image will be seriously reduced.

[0004] The radial velocity component generated by the random motion of the sea surface scattering points will cause the Doppler center shift, which will cause the position of the sea surface scattering points to shift. Due to the overall random motion of the sea surface, the offset of each scattering point is different, which will cause the scattering points at different positions to be accumulated together, resulting in the loss of some weak detail information in the SAR image, which brings difficulties to SAR image interpretation and target recognition.

[0005] In summary, most of the existing SAR sea surface imaging algorithms directly apply the land imaging algorithms without considering the azimuth offset caused by the radial velocity difference between different sea surface scattering points. This will lead to the loss of some sea surface texture information in the SAR image, seriously affecting SAR image interpretation and target detection. Summary of the invention

[0006] In view of this, the present invention provides a SAR sea surface imaging method based on radial velocity field, which can realize sea surface imaging based on SAR data.

[0007] The present invention provides a SAR sea surface imaging method based on radial velocity field, comprising:

[0008] Step 1: Generate and save a single-view complex image based on the raw data received by SAR;

[0009] Step 2: dividing the single-view complex image into multiple data sub-blocks, and estimating the Doppler center frequencies of all the data sub-blocks;

[0010] Step 3, calculating the radial velocity corresponding to each data sub-block, and forming a radial velocity field from the radial velocities of all data sub-blocks;

[0011] Step 4, performing azimuth Fourier transform on the single-view complex image to obtain range-Doppler domain data, and performing phase compensation on the range-Doppler domain data according to the radial velocity field;

[0012] Step 5: transform the compensated range-Doppler domain data into the two-dimensional time domain to complete the sea surface imaging.

[0013] Furthermore, in the process of generating and saving the single-view complex image in step 1, an unweighted matched filter is used to implement pulse compression.

[0014] Furthermore, the size of the data sub-block is 100m*100m.

[0015] Furthermore, the Doppler center frequency of the data sub-block is composed of a component generated by the oblique angle and a component generated by the moving target, and is expressed as: dc =f dc_θ +f dc_tar , where f dc_θ represents the Doppler center shift caused by the oblique angle, f dc_tar Indicates the Doppler center shift caused by a moving target.

[0016] Furthermore, the radial velocity corresponding to the data sub-block is calculated using the following formula:

[0017]

[0018] Among them, V r (i,j) represents the radial velocity of the (i,j)th data sub-block, and λ is the radar wavelength;

[0019] The radial velocity field is thus expressed as:

[0020]

[0021] Furthermore, in step 4, the compensation factor for performing phase compensation on the range-Doppler domain data according to the radial velocity field is: Where V is the radial velocity field, λ is the radar wavelength, j is the imaginary unit, and t is the slow time variable.

[0022] Furthermore, in step 5, the method of transforming the compensated range-Doppler domain data into the two-dimensional time domain is: performing an inverse azimuth Fourier transform on the compensated range-Doppler domain data.

[0023] Beneficial effects:

[0024] The present invention achieves sea surface imaging by obtaining the radial velocity field based on SAR data, which can improve the focusing effect of sea surface texture, thereby improving the interpretability of SAR sea surface images, and at the same time improve the utilization rate of SAR ocean data, which is helpful to improve the sea surface texture characteristics and lay the foundation for subsequent target detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention provides a flow chart of a SAR sea surface imaging method based on radial velocity field.

[0026] Figure 2 A detailed flow chart of a SAR sea surface imaging method based on radial velocity field provided by the present invention.

[0027] Figure 3 Schematic diagram of a simulation point target used in the embodiment.

[0028] FIG. 4( a ) is a diagram showing an imaging result obtained by using the Omega-K algorithm in a SAR sea surface imaging method based on a radial velocity field provided by the present invention.

[0029] FIG. 4( b ) is an imaging result obtained by using a SAR sea surface imaging method based on radial velocity field provided by the present invention. DETAILED DESCRIPTION

[0030] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0031] Affected by the random motion of the sea surface, the radial velocities between different scattering points on the sea surface are different, which will lead to different Doppler center offsets caused by different scattering points during the imaging process, and then lead to differences in position offsets. Due to the differences between the position offsets of different scattering points, the use of traditional SAR imaging algorithms will cause aliasing of the target, resulting in the loss of some information.

[0032] The present invention provides a SAR sea surface imaging method based on radial velocity field, the process is as follows: Figure 1 As shown, the following steps are included:

[0033] Step 1: Generate and save a single-view complex image based on the original data received by SAR, where the data block size is M*N.

[0034] In order to accurately estimate the Doppler center frequency and obtain an accurate velocity field, the present invention uses an unweighted matched filter to complete the matched filtering process and realize pulse compression when generating a single-look complex image. The raw data received by the SAR is processed using SAR imaging algorithms such as RD and Omega-K to obtain a single-look complex (SLC).

[0035] Step 2: Divide the single-view complex image obtained in step 1 into multiple data sub-blocks to obtain m*n data sub-blocks, and estimate the Doppler center frequency of each data sub-block respectively.

[0036] The single-view complex image is divided into multiple sub-blocks to obtain m*n data sub-blocks, where the size of each sub-block is determined by the real data, and generally 100m*100m can be selected. Estimate the Doppler center frequency f of the m*n data sub-blocks dc , Doppler center frequency f dc It is composed of the component generated by the oblique angle and the component generated by the moving target, which can be expressed as:

[0037] f dc =f dc_θ +f dc_tar (1).

[0038] Among them, f dc_θ The Doppler center shift caused by the oblique angle is expressed as:

[0039]

[0040] V is the aircraft's flight speed, θ is the oblique angle, and λ is the radar wavelength; f dc_tar Indicates the Doppler center shift caused by a moving target.

[0041] Step 3: Calculate the radial velocities corresponding to the m*n data sub-blocks, and compose the corresponding radial velocities into a velocity field.

[0042] The radial velocity corresponding to the data sub-block is calculated using the following formula:

[0043]

[0044] Among them, V r (i, j) represents the radial velocity of the (i, j)th data sub-block, i is the lateral position coordinate of the data sub-block, and j is the longitudinal position coordinate of the data sub-block.

[0045] Therefore, the velocity field V composed of the radial velocities corresponding to the m*n data sub-blocks is expressed as:

[0046]

[0047] Step 4: Perform azimuth Fourier transform on the single-view complex image to obtain range-Doppler domain data, and perform phase compensation on the range-Doppler domain data according to the radial velocity field obtained in step 3.

[0048] The SLC data is transformed into the range-Doppler domain, and then phase compensation is performed according to the velocity field V. The compensation factor is expressed as:

[0049]

[0050] Where j is the imaginary unit and t is the slow time variable.

[0051] Step 5: transform the compensated range-Doppler domain data into the two-dimensional time domain to achieve sea surface imaging.

[0052] The compensated range-Doppler domain data is subjected to an inverse azimuth Fourier transform to obtain the final imaging result, i.e., the sea surface imaging.

[0053] Example:

[0054] This embodiment adopts a SAR sea surface imaging method based on radial velocity field provided by the present invention. The detailed flow chart of the method is as follows: Figure 2 As shown, the simulation data is processed to verify the effectiveness of the method.

[0055] In this embodiment, four moving point target positions and velocity vector diagrams are set as follows: Figure 3 As shown in the figure, point target 1 and point target 2 are located in the same range gate, with different moving speeds and opposite directions; point target 3 and point target 4 are located in the same range gate, with the same moving speeds and opposite directions. Four point targets are used to simulate scattering points at four different positions on the sea surface. SAR echo signals are simulated according to the four point targets. The simulated SAR original echo signals are generated and saved using the Omega-K algorithm with a single-view complex image, where the data block size is 2048*2048 (number of points in the range direction M*number of points in the azimuth direction N).

[0056] The data block is divided into m*n sub-blocks. In this embodiment, m=11, n=2, the size of each sub-block is 100m*100m, and the Doppler center frequency of each sub-block is estimated by using the energy balance method.

[0057] In this embodiment, the simulated SAR system is in the positive side view working mode, and the oblique angle θ=0, and f can be calculated according to the formula dc_θ = 0, therefore, the Doppler center frequency is only caused by the radial motion of the scattering point, that is, f dc_tar =f dc .

[0058] The Doppler center shift f caused by the radial motion of the scattering point dc_tarSubstitute into formula (3), calculate the radial velocity corresponding to each data sub-block, and combine the corresponding radial velocities to form a velocity field V. In this embodiment, the simulation wavelength λ=0.2308.

[0059] In this embodiment, the single-view complex image is subjected to azimuth Fourier transform to obtain range-Doppler domain data, and the calculated radial velocity field V is substituted into formula (5) to perform phase compensation on the range-Doppler domain data.

[0060] Finally, the compensated range-Doppler domain data is transformed into the two-dimensional time domain to achieve sea surface imaging.

[0061] The results before and after the proposed method are compared. The imaging result obtained by the ωk imaging algorithm is shown in Figure 4(a). It can be found that due to the offset of the scattering point position caused by the target radial velocity, the scattering points of the same range gate overlap, and only two point targets are visible in the SAR image. The imaging result obtained by the SAR sea surface imaging method based on the radial velocity field is shown in Figure 4(b). After processing by this method, the four scattering points are separated. This method is of great significance for sea surface imaging and lays a foundation for subsequent SAR image interpretation and marine target detection.

[0062] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A SAR sea surface imaging method based on radial velocity field, characterized in that: include: Step 1: Generate and save a single-view complex image based on the raw data received by SAR; Step 2: dividing the single-view complex image into multiple data sub-blocks, and estimating the Doppler center frequencies of all the data sub-blocks; Step 3: Calculate the radial velocity corresponding to each data sub-block, and form a radial velocity field by the radial velocities of all data sub-blocks; Step 4, performing azimuth Fourier transform on the single-view complex image to obtain range-Doppler domain data, and performing phase compensation on the range-Doppler domain data according to the radial velocity field; Step 5: transform the compensated range-Doppler domain data into the two-dimensional time domain to complete the sea surface imaging; The Doppler center frequency of the data sub-block is composed of the component generated by the oblique angle and the component generated by the moving target, which is expressed as: f dc =f dc_θ +f dc_tar , where f dc_θ represents the Doppler center shift caused by the oblique angle, f dc_tar Indicates the Doppler center shift caused by a moving target; The radial velocity corresponding to the data sub-block is calculated using the following formula: Among them, V r (i,j) represents the radial velocity of the (i,j)th data sub-block, and λ is the radar wavelength; The radial velocity field is thus expressed as: The compensation factor for performing phase compensation on the range-Doppler domain data according to the radial velocity field in step 4 is: Where V is the radial velocity field, λ is the radar wavelength, j is the imaginary unit, and t is the slow time variable.

2. The SAR sea surface imaging method based on radial velocity field according to claim 1, characterized in that: In the process of generating and saving the single-view complex image in step 1, an unweighted matched filter is used to implement pulse compression.

3. The SAR sea surface imaging method based on radial velocity field according to claim 1, characterized in that: The size of the data sub-block is 100m*100m.

4. The SAR sea surface imaging method based on radial velocity field according to claim 1, characterized in that: The method of transforming the compensated range-Doppler domain data into the two-dimensional time domain in step 5 is: performing an inverse azimuth Fourier transform on the compensated range-Doppler domain data.

Images