A sea surface adaptive imaging method based on SAR subview analysis

Abstract

The application provides a sea surface adaptive imaging method based on SAR subview analysis, which comprises the following steps: transforming a SAR single-view complex image to a range frequency domain; according to a window function type added in an imaging direction, performing reverse correction of the window function weighting; dividing a Doppler bandwidth in the imaging direction into two parts, i.e. a left part and a right part; performing inverse Fourier transform on echoes containing only one half of the spectrum in the imaging direction to obtain left and right subaperture images; calculating a subview cross-correlation coefficient according to the left and right subaperture images; calculating a subview coherent entropy according to the left and right subaperture images; dividing a two-dimensional feature space composed of the subview cross-correlation coefficient and the subview coherent entropy into three regions, i.e. an interference region, a target region and a sea clutter region; segmenting pixels according to the divided regions; and calculating pixel amplitudes in each region. Through reprocessing of the SAR image, the application suppresses the imaging direction ambiguity and the sea clutter in the marine background SAR image, and improves the SAR imaging quality in the marine background.

Description

Technical Field

[0001] The present invention relates to the field of SAR image processing, and in particular to a sea surface adaptive imaging method based on SAR sub-view analysis. Background Art

[0002] Synthetic Aperture Radar (SAR) is a microwave active imaging radar that can achieve high two-dimensional resolution. The radar emits electromagnetic waves with a large time-bandwidth product and then achieves high range resolution through pulse compression. High azimuth resolution is achieved using the principle of synthetic aperture. With the improvement of resolution and the emergence of new imaging modes, the requirements for SAR imaging processing algorithms are becoming increasingly complex.

[0003] Classic SAR imaging algorithms mainly include RD algorithm and CS algorithm. RD algorithm is the first imaging processing algorithm developed for civilian spaceborne SAR, but it has two shortcomings: first, when using a longer kernel function to improve the accuracy of range migration correction, the amount of calculation is large; second, the dependence of quadratic range compression on azimuth frequency is difficult to solve, which limits its processing accuracy for certain large squint angles and long aperture SAR. CS algorithm avoids the interpolation operation in range migration correction. Figure 1 As shown in Figure 1, this algorithm, based on the scaling principle, achieves a scale transformation or translation of the chirp signal by frequency modulating it. Based on this principle, range migration correction with range transformation is achieved by replacing time-domain interpolation with phase multiplication. Furthermore, because it requires data processing in the two-dimensional frequency domain, CS also addresses the dependency of quadratic range compression on azimuth frequency. The CS algorithm is currently the most commonly used imaging algorithm for spaceborne SAR.

[0004] Currently, image quality blurring is a common problem in spaceborne synthetic aperture radar (SAR) images, with azimuth blurring being particularly prominent. When the blurring energy is high, a large number of false targets are generated, affecting the misjudgment of true targets in the image. Because the amplitude of sea surface images is much lower than that of land images, azimuth blurring has a more severe impact on marine applications than on land. This is particularly true in marine monitoring along coastal zones and at the sea-land interface. Targets with strong backscattering, such as coastal structures, aquaculture areas, and ships, generate blur signals that are still very strong relative to the ocean background, severely impacting the subsequent application of ocean background images. Furthermore, ocean background SAR images are often interfered with by sea clutter, which compromises image quality. Summary of the Invention

[0005] The purpose of the present invention is to propose a sea surface adaptive imaging method based on sub-view analysis.

[0006] The technical solution for achieving the purpose of the present invention is: a sea surface adaptive imaging method based on SAR sub-view analysis, comprising the following steps:

[0007] Step 1, sub-view separation: The SAR single-view complex image is transformed into the azimuth frequency domain. Based on the type of window function applied in the azimuth direction during imaging, a weighted inverse correction of the window function is performed. The azimuth Doppler bandwidth is then evenly divided into two parts, the left and right parts. The echo containing only half of the spectrum is subjected to an azimuth inverse Fourier transform to obtain the left and right sub-aperture images.

[0008] Step 2, sub-view cross-correlation coefficient calculation: Calculate the sub-view cross-correlation coefficient based on the left and right sub-aperture images;

[0009] Step 3, sub-view coherence entropy calculation: Calculate the sub-view coherence entropy based on the left and right sub-aperture images;

[0010] Step 4: Feature domain pixel segmentation: Divide the two-dimensional feature space composed of sub-view cross-correlation coefficient and sub-view coherence entropy into three regions: interference, target, and sea clutter, and segment the pixels according to the divided regions;

[0011] Step 5: Calculate the new pixel amplitude: Calculate the pixel amplitude of each sub-region.

[0012] A sea surface adaptive imaging system based on SAR sub-view analysis realizes sea surface adaptive imaging by utilizing the sea surface adaptive imaging method based on SAR sub-view analysis.

[0013] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for sea surface adaptive imaging based on SAR sub-view analysis is utilized to implement sea surface adaptive imaging.

[0014] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the sea surface adaptive imaging method based on SAR sub-view analysis is utilized to realize sea surface adaptive imaging.

[0015] Compared with the existing technology, the present invention has the following significant advantages: by constructing a new feature space, it can achieve effective separation of interference signals, target signals and sea clutter signals, adopt a combination of fixed thresholds and adaptive thresholds to divide pixel attributes in the feature space, and innovatively adopt different calculation methods to reconstruct the image based on the principle of not destroying the integrity of the image, which can effectively suppress interference and sea clutter and improve the signal-to-clutter ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is the basic flow chart of the CS algorithm.

[0017] Figure 2 This is a flow chart of the sea surface adaptive imaging method based on SAR sub-view analysis of the present invention.

[0018] Figure 3 It is a schematic diagram of feature domain segmentation.

[0019] Figure 4 It is a comparison diagram of CS imaging results and processing results of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0021] The basic process of the sea surface adaptive imaging method based on SAR sub-view analysis of the present invention is as follows: Figure 2 As shown in the figure, it mainly includes five steps: sub-view separation, sub-view mutual correlation coefficient calculation, sub-view coherence entropy calculation, feature domain pixel segmentation, and new pixel amplitude calculation. The specific contents are as follows:

[0022] Step 1: Separate sub-views

[0023] First, the high-resolution SAR single-view complex image is transformed into the azimuth frequency domain. Then, the weighted inverse correction of the window function is performed according to the type of window function added in azimuth during imaging. Then, the azimuth Doppler bandwidth is evenly divided into two parts, the left and right parts. Finally, the echo containing only half of the spectrum is subjected to azimuth inverse Fourier transform to obtain the left and right sub-aperture images s1 and s2, respectively.

[0024] Step 2: Calculate the sub-view correlation coefficient

[0025] The mutual correlation coefficient between sub-views is calculated as follows:

[0026]

[0027] in,* T Indicates conjugate transpose; <> indicates spatial averaging, where the average is calculated for a 5*5 pixel area centered on the pixel point; || indicates taking the absolute value.

[0028] Step 3: Calculate sub-view coherence entropy

[0029] Assume there are n subviews s=[s1,s2,…,s n ] T , whose covariance matrix λ i is the ith eigenvalue of [s]. Define the sub-view coherence entropy

[0030]

[0031] in When n=2, H c =-(p1log2p1+p2log2p2).

[0032] Step 4: Feature domain pixel segmentation

[0033] The two-dimensional feature space composed of the sub-view mutual correlation coefficient and the sub-view coherence entropy is divided into three regions (R1: interference, R2: target, R3: sea clutter), and the pixels are segmented according to the divided regions:

[0034] When H c When ≤0.5, the pixel is divided into R1;

[0035] When ρ c >0.01max(ρ c ), H c When >0.5, the pixels are divided into R2;

[0036] When ρ c ≤0.01max(ρ c ), H c When >0.5, the pixel is divided into R3;

[0037] Among them, max() means taking the maximum value.

[0038] Step 5: Calculate the new pixel amplitude

[0039] The pixel amplitude of each partition is calculated according to the following formula:

[0040] R1: A a =2·min{|SL1|,|SL2|}

[0041] R2: A t =|SL1|+|SL2|

[0042] R3:A s =|SL1+SL2|

[0043] Among them, A a 、A t 、A s They represent the interference pixel amplitude, target pixel amplitude, and sea clutter pixel amplitude respectively. min{} represents the minimum value. SL1 and SL2 represent the values ​​of the pixel in the left and right sub-aperture images respectively.

[0044] The present invention also proposes a sea surface adaptive imaging system based on SAR sub-view analysis, and implements sea surface adaptive imaging by utilizing the sea surface adaptive imaging method based on SAR sub-view analysis.

[0045] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for sea surface adaptive imaging based on SAR sub-view analysis is utilized to implement sea surface adaptive imaging.

[0046] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the sea surface adaptive imaging method based on SAR sub-view analysis is utilized to realize sea surface adaptive imaging.

[0047] In summary, the present invention suppresses the azimuth ambiguity and sea clutter in the ocean background SAR image by reprocessing the SAR image, thereby improving the SAR imaging quality under the ocean background.

[0048] Example

[0049] In order to verify the effectiveness of the solution of the present invention, the following experiment was conducted.

[0050] Two slices were cut out from the CS imaging results of the ultra-fine strip ocean scene of the GF-3 satellite ( Figure 4 In (a) and (c), the slice size is 1200 x 1200 pixels. The pixel spacing in the range and azimuth directions is 1.124m and 1.713m respectively. The slice contains the ship, azimuth ambiguity and sea clutter signals. After processing by the method of the present invention, the following is obtained: Figure 4 Results of (b) and (d). From the visual effect, the azimuth ambiguity signal is suppressed and the ship target is more prominent.

[0051] To quantitatively demonstrate the advantages of the present invention, the strongest scattering points (S1, S2, S3, S4) of each ship, the strongest points of the point-like azimuth ambiguity signal (A1, A2, A4, A5), and the distributed azimuth ambiguity signal region A3 were selected from two slices. A relatively uniform sea clutter region was also selected in each slice. Table 1 shows the changes in the signal-to-clutter ratio before and after processing, where the regional power is averaged. The signal-to-clutter ratio results show that the application of the present invention significantly suppresses the azimuth ambiguity signal, while the signal-to-clutter ratio of the ship signal remains unchanged or slightly better than that of the CS imaging results.

[0052] Table 1 Signal-to-noise ratio before and after processing

[0053]

[0054]

[0055] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0056] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A sea surface adaptive imaging method based on SAR subview analysis, characterized in that: The steps include: Step 1, sub-view separation: The SAR single-view complex image is transformed into the azimuth frequency domain. Based on the type of window function applied in the azimuth direction during imaging, a weighted inverse correction of the window function is performed. The azimuth Doppler bandwidth is then evenly divided into two parts, the left and right parts. The echo containing only half of the spectrum is subjected to an azimuth inverse Fourier transform to obtain the left and right sub-aperture images. Step 2, sub-view cross-correlation coefficient calculation: Calculate the sub-view cross-correlation coefficient based on the left and right sub-aperture images; Step 3, sub-view coherence entropy calculation: Calculate the sub-view coherence entropy based on the left and right sub-aperture images; Step 4: Feature domain pixel segmentation: Divide the two-dimensional feature space composed of sub-view cross-correlation coefficient and sub-view coherence entropy into three regions: interference, target, and sea clutter, and segment the pixels according to the divided regions; Step 5, new pixel amplitude calculation: calculate the pixel amplitude of each sub-region; in, Step 3: Calculate the sub-view coherence entropy. The specific method is: Assume there are n subviews s=[s1,s2,…,s n ] T , whose covariance matrix λ i is the i-th eigenvalue of [s], and the sub-view coherence entropy H is defined as c for: in, When n=2, H c =-(p1log2p1+p2log2p2); Step 4: feature domain pixel segmentation. The specific method is as follows: The two-dimensional feature space composed of sub-view mutual correlation coefficient and sub-view coherence entropy is divided into three regions, namely R1: interference, R2: target, R3: sea clutter, and the pixels are segmented according to the divided regions; When H c When ≤0.5, the pixel is divided into R1; When ρ c >0.01max(ρ c ), H c When >0.5, the pixel is divided into R2; When ρ c ≤0.01max(ρ c ), H c When >0.5, the pixel is divided into R3; Among them, ρ c represents the cross-correlation coefficient between the left and right sub-aperture images, H c Represents the coherence entropy of the left and right sub-aperture images, and max() represents the maximum value; Step 5: Calculate the new pixel amplitude. Calculate the pixel amplitude of each partition according to the following formula: R1:A a =2 min{|SL1|,|SL2|} R2:A t =|SL1|+|SL2| R3:A s =|SL1+SL2| Among them, A a 、A t 、A s They represent the interference pixel amplitude, target pixel amplitude, and sea clutter pixel amplitude respectively. min{} represents the minimum value. SL1 and SL2 represent the values ​​of the pixel in the left and right sub-aperture images respectively.

2. The sea surface adaptive imaging method based on SAR subview analysis according to claim 1, characterized in that: Step 2: Calculate the sub-view correlation coefficient as follows: Among them, s1 and s2 represent the left and right sub-aperture images, ρ c represents the cross-correlation coefficient between the left and right sub-aperture images, * T represents conjugate transpose; < > represents spatial averaging, and || represents taking the absolute value.

3. A sea surface adaptive imaging system based on SAR sub-view analysis, characterized in that: The sea surface adaptive imaging method based on SAR sub-view analysis according to any one of claims 1 to 2 is used to realize sea surface adaptive imaging.

4. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for sea surface adaptive imaging based on SAR sub-view analysis according to any one of claims 1 to 2 is used to implement sea surface adaptive imaging.

5. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for sea surface adaptive imaging based on SAR sub-view analysis according to any one of claims 1 to 2 is used to implement sea surface adaptive imaging.

Images