A method and device for self-focusing missile-borne radar imaging

Abstract

The present application relates to a method and device for self-focusing missile-borne radar imaging, belonging to the field of image processing, and solving the problem that the dynamic characteristics of multiple targets in maritime interference scenarios reduce the focusing effect. The method includes: azimuthally superimposing echo data to obtain one-dimensional range image echo data; based on the amplitude and coordinates of the one-dimensional range image echo data, differentiating the coordinates corresponding to each peak in the one-dimensional range image echo data to obtain a differential result; based on the differential result, determining the target area corresponding to the observed target; wherein the target area is used to indicate a range coordinate set; based on the target area, determining whether the observed target is a non-rigid structure; based on whether the observed target is a non-rigid structure, performing local self-focusing on the echo data corresponding to the target area. The technical solution provided by the present application can improve the focusing effect of SAR images, thereby improving the accuracy of SAR imaging.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a method and device for self-focusing missile-borne radar imaging. Background Art

[0002] In order to improve the missile radar seeker's ability to counter passive interference, it is necessary to simultaneously integrate the target's one-dimensional high-resolution range image recognition and two-dimensional high-resolution SAR features, and use multiple dimensions of recognition information to improve the radar's detection and recognition capabilities for ship targets.

[0003] The probability of missile detection and recognition of a target is closely related to the focusing effect of SAR images. Due to the complex sea conditions at sea, the dynamic characteristics of multiple targets in the interference scene vary greatly, which seriously affects the focusing effect of SAR images.

[0004] However, existing recognition methods usually assume that the target is a rigid structure, which cannot eliminate the impact of the dynamic characteristics of multiple targets in maritime interference scenarios on focusing, reducing the focusing effect of SAR images and further reducing the accuracy of SAR imaging. Summary of the Invention

[0005] In view of the above analysis, the present application aims to propose a method and device for self-focusing of missile-borne radar imaging, so as to improve the focusing effect of SAR images and thereby improve the accuracy of SAR imaging.

[0006] The purpose of this application is mainly achieved through the following technical solutions:

[0007] In one aspect, the present application provides a method for self-focusing missile-borne radar imaging, comprising:

[0008] The echo data is superimposed in azimuth to obtain one-dimensional range image echo data;

[0009] Based on the amplitude and coordinates of the one-dimensional range image echo data, differentiating the coordinates corresponding to each peak in the one-dimensional range image echo data to obtain a differential result;

[0010] Determine a target area corresponding to the observed target based on the difference result; wherein the target area is used to indicate a range coordinate set;

[0011] Based on the target area, determining whether the observed target is a non-rigid structure;

[0012] Based on whether the observed target is a non-rigid structure, local self-focusing is performed on the echo data corresponding to the target area.

[0013] Furthermore, the step of differentiating the coordinates corresponding to each peak in the one-dimensional range image echo data based on the amplitude and coordinates of the one-dimensional range image echo data includes:

[0014] Arranging the coordinates corresponding to each of the peaks in descending order of amplitude;

[0015] The coordinates corresponding to the sorted peaks are differentiated pairwise.

[0016] Furthermore, the difference result includes: coordinate difference;

[0017] Determining the target area corresponding to the observed target based on the difference result includes:

[0018] When the coordinate difference is less than the area decision threshold, determining that the waveforms corresponding to the coordinate difference correspond to the same observation target;

[0019] The waveforms corresponding to the observed targets are arranged in ascending order of the coordinates to obtain the target area.

[0020] Furthermore, determining whether the observed target is a non-rigid structure based on the target area includes:

[0021] Determining, based on the target area, a duty cycle of the observed target, a number of sub-areas of the target area, and a size of the target area;

[0022] Whether the observed target is a rigid structure is determined according to the duty cycle, the number of sub-areas, and the size of the target area.

[0023] Furthermore, determining the number of sub-regions of the target region according to the target region includes:

[0024] Differences are made between adjacent distance coordinates corresponding to the peaks in the target area to obtain position difference values;

[0025] When the position difference value is not less than the non-rigid multi-target decision threshold, the number of sub-regions is increased by 1 to determine the number of sub-regions.

[0026] Furthermore, determining whether the observed target is a rigid structure according to the duty cycle, the number of sub-areas, and the size of the target area includes:

[0027] When the duty cycle is less than the non-rigid structure duty cycle threshold, the size of the target area is not greater than the non-rigid structure radial size threshold, and the number of sub-areas is less than the non-rigid structure target number threshold, the corresponding observation target is determined to be a non-rigid structure.

[0028] Furthermore, after determining that the corresponding observation target is a non-rigid structure, the method further includes:

[0029] When the duty cycle is greater than the rigid structure duty cycle threshold, or the size of the target area is greater than the rigid structure radial size threshold, or the number of sub-areas is greater than or equal to the rigid structure target number threshold, determining that the corresponding observation target is a rigid structure;

[0030] The remaining observation targets are determined to be non-rigid structures.

[0031] Furthermore, the performing local self-focusing on the echo data corresponding to the target area based on whether the observed target is a non-rigid structure includes:

[0032] When the observed target is a rigid structure, a prominent point in the target area is selected; based on the prominent point, the echo data corresponding to the target area is subjected to autofocus processing.

[0033] Furthermore, when the observed target is a non-rigid structure, the target area includes at least two sub-areas;

[0034] The performing local autofocusing on the echo data corresponding to the target area based on whether the observed target is a non-rigid structure includes:

[0035] Selecting a prominent point from each of the sub-regions;

[0036] The echo data corresponding to each of the sub-areas are auto-focused according to the highlighted points.

[0037] On the other hand, the present application also provides a device for self-focusing missile-borne radar imaging, comprising: a superposition module, a difference module, a classification module and a local self-focusing module;

[0038] The superposition module is used to perform azimuth superposition on the echo data to obtain one-dimensional range image echo data;

[0039] The difference module is used to differentiate the coordinates corresponding to each peak in the one-dimensional range image echo data based on the amplitude and coordinates of the one-dimensional range image echo data;

[0040] The classification module is used to determine a target area corresponding to the observed target based on the difference result, where the target area is used to indicate a range coordinate set; and based on the target area, determine whether the observed target is a non-rigid structure;

[0041] The local self-focusing module is used to perform local self-focusing on the echo data corresponding to the target area based on whether the observed target is a non-rigid structure.

[0042] Compared with the existing technology, this application can achieve at least one of the following technical effects:

[0043] 1. The observation targets are classified according to non-rigid structures and rigid structures, and the corresponding method is used for local self-focusing for each type of observation target, which improves the self-focusing effect of SAR as a whole.

[0044] 2. Based on the coordinates and amplitude of the echo data, differential analysis can quickly separate the waveforms corresponding to the mixed observation targets, provide technical support for the subsequent classification and judgment of non-rigid structures and rigid structures, and improve the calculation speed.

[0045] Other features and advantages of the present application will be described in the subsequent description, and some will become apparent from the description or be understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are only for the purpose of illustrating particular embodiments and are not to be considered as limiting the present application. Like reference symbols denote like components throughout the drawings.

[0047] Figure 1 A flowchart of a method for self-focusing missile-borne radar imaging provided in an embodiment of the present application;

[0048] Figure 2 A schematic structural diagram of a device for self-focusing missile-borne radar imaging provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The preferred embodiments of the present application are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of the present application and are used together with the embodiments of the present application to illustrate the principles of the present application, and are not used to limit the scope of the present application.

[0050] The key factor affecting SAR imaging clarity is focusing, which is significantly affected by the dynamic characteristics of the observed target. For example, photographers often ask their subjects to remain motionless when taking photos, as any movement can affect the quality of the photo, such as the appearance of ghosting. Similarly, unstable motion during SAR imaging can significantly affect the image quality.

[0051] Therefore, in the embodiment of the present application, a non-rigid structure target refers to an observation target whose motion state affects the imaging effect when there are multiple observation targets, which is a non-rigid structure, and vice versa. It should be noted that a non-rigid structure can be a collection of multiple observation objects, or it can be a single observation object. In actual scenarios, non-rigid structures are prone to appear in targets at sea. This is because targets at sea will move irregularly with the waves, especially spherical observation targets floating on the sea surface connected by ropes. Under the action of waves, the movement direction and speed of each spherical observation target are different. If SAR imaging images the above-mentioned spherical observation target during the image acquisition process, it will cause unclear images and even affect the subsequent positioning accuracy.

[0052] In order to solve the above technical problems, the present application provides a method for self-focusing missile-borne radar imaging, such as Figure 1 As shown, the following steps are included:

[0053] Step 1: Superimpose the echo data in azimuth to obtain one-dimensional range image echo data.

[0054] Echo data refers to the target echo data received by the missile-borne radar.

[0055] Preferably, in order to achieve a better image autofocusing effect, before performing step 1, the echo data may be subjected to an autofocusing process, and the echo data after the autofocusing process may be subjected to a target-based local autofocusing process using the method of this embodiment. Optionally, the autofocusing process of the echo data may include the following steps:

[0056] S1, performs range pulse compression on the echo signal of SAR radar.

[0057] S2, envelope alignment of the echo data after pulse compression.

[0058] S3, transform the envelope aligned data into the frequency domain for image autofocusing processing.

[0059] Step 2: Based on the amplitude and coordinates of the one-dimensional range image echo data, the coordinates corresponding to each peak in the one-dimensional range image echo data are differentiated.

[0060] In the embodiment of the present application, step 2 includes: arranging the coordinates corresponding to each peak in descending order of amplitude; and performing pairwise differentiation on the coordinates corresponding to each peak after sorting.

[0061] For echo signals, amplitude represents signal strength; that is, the larger the amplitude, the more likely the corresponding waveform represents the observed target. Because there are multiple targets, the waveforms of each target are intertwined in the coordinate system. Therefore, sorting by descending amplitude not only prevents loss of valid data but also facilitates the organization of echo data for subsequent differencing.

[0062] Step 3: Determine the target area corresponding to the observed target based on the difference results.

[0063] In the embodiment of the present application, the target area is used to indicate a set of range coordinates, that is, one observed target may correspond to multiple pulses. The set of range coordinates corresponding to these pulses is the target area.

[0064] In this embodiment of the present application, the difference result of step 2 includes a coordinate difference. When the coordinate difference is less than a region determination threshold, it is determined that the waveforms corresponding to the coordinate difference correspond to the same observed target. The region determination threshold is used to indicate whether the corresponding waveforms correspond to the same observed target, and its value can be determined based on the accuracy requirements.

[0065] For example, the coordinate corresponding to the peak of waveform A is a, and the coordinate corresponding to the peak of waveform B is b. If ba is less than the regional decision threshold, waveform A and waveform B correspond to the same observation target.

[0066] Afterwards, the waveforms corresponding to the observed targets are arranged in ascending order of coordinates to obtain the target area. The above method can conveniently separate the waveforms corresponding to the observed targets that are clustered together.

[0067] Step 4: Based on the target area, determine whether the observed target is a non-rigid structure.

[0068] In the embodiment of the present application, the duty cycle of the observed target, the number of sub-regions of the target region, and the size of the target region are determined according to the target region.

[0069] Specifically, the duty cycle = the total number of distance coordinates corresponding to the waveform / (the starting coordinate of the target area - the ending coordinate of the target area + 1).

[0070] The size of the target area = the starting coordinate of the target area - the ending coordinate of the target area.

[0071] The method for determining the number of target sub-regions is as follows: the adjacent distance coordinates corresponding to the peaks in the target area are differentiated pairwise to obtain position difference values; when there is a position difference value that is not less than the non-rigid multi-target judgment threshold value, the number of sub-regions is increased by 1. As mentioned above, the non-rigid structure can be multiple observation objects, so the target area represents all observation objects, and the sub-region corresponds to a single observation object. The non-rigid multi-target judgment threshold value is used to characterize whether adjacent waveforms belong to the same observation object. It should be noted that, for the convenience of calculation, this application sets the initial number of sub-regions to 1, that is, the target area contains at least one sub-region.

[0072] Whether the observed target is a rigid structure is determined based on the duty cycle, the number of sub-regions, and the size of the target area.

[0073] In the embodiment of the present application, in order to improve the calculation accuracy, three criteria are set:

[0074] Criterion 1: When the duty cycle is less than the non-rigid structure duty cycle threshold, the size of the target area is not greater than the non-rigid structure radial size threshold, and the number of sub-areas is less than the non-rigid structure target number threshold, the corresponding observed target is determined to be a non-rigid structure.

[0075] Criterion 2: When the duty cycle is greater than the rigid structure duty cycle threshold, or the target area size is greater than the rigid structure radial size threshold, or the number of sub-areas is greater than or equal to the rigid structure target number threshold, the corresponding observed target is determined to be a rigid structure;

[0076] Criterion 3: When the observed target does not meet Criterion 1 and Criterion 2, the corresponding observed target is determined to be a non-rigid structure.

[0077] It should be noted that the non-rigid structure duty cycle threshold + rigid structure duty cycle threshold, the non-rigid structure radial size threshold + rigid structure radial size threshold, and the non-rigid structure target number threshold + rigid structure target number threshold do not represent the entire domain of coordinate values. This means that rigid and non-rigid structures are not complementary, but rather partially overlap. Therefore, when determining non-rigid structures, criterion 1 is first used to identify some non-rigid structure observation targets to prevent non-rigid observation targets from being misidentified as rigid structures. Then, criterion 2 is used to identify rigid structure observation targets. Finally, the remaining observation targets are identified as non-rigid structure observation targets. This approach maximizes the distinction between non-rigid and rigid observation targets.

[0078] Step 5: Based on whether the observed target is a non-rigid structure, local autofocusing is performed on the echo data corresponding to the target area.

[0079] In the embodiment of the present application, local self-focusing is divided into two cases:

[0080] When the observed target is a rigid structure, a prominent point in the target area is selected;

[0081] The echo data corresponding to the target area is processed by autofocus based on the highlighted points.

[0082] As previously mentioned, in the embodiments of the present application, the target region includes at least one subregion. Since a target region with only one subregion corresponds to a rigid structure, when the observed target is a non-rigid structure, the target region includes at least two subregions. In this case, a distinct point is selected from each subregion, and autofocusing is performed on the echo data corresponding to each subregion based on the distinct point.

[0083] In summary, by classifying the observed targets and combining them with local autofocusing, the autofocusing effect of SAR can be enhanced, and the negative effects of autofocusing caused by the inconsistent movement directions and speeds of multiple targets can be minimized.

[0084] In addition, the present embodiment also provides a specific implementation of steps S1-S3:

[0085] The specific implementation of step S1 is:

[0086] The range pulse compression is subjected to range FFT, transformed into the azimuth frequency domain, multiplied by the range pulse compression reference function and the bullet velocity compensation coefficient, and then subjected to range IFFT, transformed into the azimuth time domain;

[0087]

[0088] f r is the distance frequency variable, f c is the carrier frequency, t m Azimuth time coordinate

[0089]

[0090]

[0091] in:

[0092] Install the rotation matrix [Δθ], [Δβ] and [θ], respectively [β] has the same form, where φ is the heading angle, θ is the pitch angle, γ is the roll angle, Δφ is the installation heading angle, Δθ is the installation pitch angle, Δγ is the installation roll angle; α is the antenna scanning pointing azimuth angle, β is the antenna scanning pointing pitch angle, v = (v 北 ,v 天 ,v 东 ) T is the carrier velocity, PRT is the radar repetition rate, and the subscripts "North, Sky, East" represent the direction of movement.

[0093] The specific implementation of step S2 is:

[0094] 1) The distance offset of each echo is initialized to 0;

[0095] 2) The sum of all range images is taken as the target vector;

[0096] 3) Perform cross-correlation calculations on each echo and the target vector using a correlation function, and the maximum value position is the distance offset of each echo;

[0097] 4) transforming each echo into the range frequency domain and performing range alignment using the range offset obtained in 3);

[0098] 5) Sum all range images of the echo after range alignment and calculate the entropy;

[0099] 6) Continue with steps 2) to 5) and compare the entropy obtained in this iteration with the entropy obtained in the previous iteration. Then determine whether the entropy has decreased. If so, perform 5 iterations; otherwise, skip the iteration.

[0100] 7) Output the echo after distance alignment.

[0101] The specific implementation of step S3 is:

[0102] The method used by PGA is conventional phase gradient PGA. In the image domain, the Doppler channel with the maximum energy is determined. In order to eliminate the phase error introduced by Doppler, it is moved to the point where Doppler is 0 through circular shift. Then the image domain data is transformed into the data domain, adjacent pulses are conjugate multiplied, the phase error between pulses is extracted, and the obtained phase errors are accumulated to obtain the phase difference relative to the first pulse. After that, the image domain is windowed and iterated continuously.

[0103] 1) Transform the envelope aligned data into the azimuth frequency domain to obtain image domain data;

[0104] 2) Initialize the window length to the number of azimuth points and then gradually reduce it;

[0105] 3) Square the amplitude of the image domain and perform azimuth superposition to obtain one-dimensional distance energy. Then sort the energy from large to small and record the corresponding coordinates. Select the prominent points that are greater than the contrast threshold according to the contrast principle.

[0106] 4) Perform the following operations on the selected highlight points:

[0107] a. Select the azimuth frequency domain data of the distance unit where the highlighted point is located;

[0108] b. Find the maximum position of the azimuth data of the range unit;

[0109] c. By cyclically shifting the maximum value of the azimuth data of the range unit to the Doppler frequency zero point, eliminating the rotational phase component at that point;

[0110] d. The data is then windowed and transformed into the azimuth time domain;

[0111] e. Conjugate multiply the adjacent pulse trains at all highlighted points, then perform distance superposition and take their phases, which is the phase difference between the adjacent pulse trains. Accumulate these phase differences to obtain the phase difference that needs to be compensated for each pulse, based on the first pulse.

[0112] 5) Perform phase error compensation on the image data and then transform it into the azimuth frequency domain to obtain image domain data;

[0113] 6) The window length is halved, and then operations 3) to 5) are repeated 5 times to output the image.

[0114] The embodiment of the present application provides a device for self-focusing missile-borne radar imaging, such as Figure 2 As shown, it includes: a superposition module 201, a difference module 202, a classification module 203 and a local self-focusing module 204;

[0115] The superposition module 201 is used to perform azimuth superposition on the echo data to obtain one-dimensional range image echo data;

[0116] The difference module 202 is used to differentiate the coordinates corresponding to each peak in the one-dimensional range image echo data based on the amplitude and coordinates of the one-dimensional range image echo data;

[0117] The classification module 203 is used to determine the target area corresponding to the observed target based on the difference result, and the target area is used to indicate the range coordinate set; based on the target area, it is determined whether the observed target is a non-rigid structure;

[0118] The local self-focusing module 204 is used to perform local self-focusing on the echo data corresponding to the target area based on whether the observed target is a non-rigid structure.

[0119] The specific manner in which each module is executed in this embodiment is the same as that in the method embodiment.

[0120] The above is only a preferred specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any changes or replacements that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed in this application should be covered by the scope of protection of the present application.

Claims

1. A method for self-focusing missile-borne radar imaging, characterized in that: include: The echo data is superimposed in azimuth to obtain one-dimensional range image echo data; Based on the amplitude and coordinates of the one-dimensional range image echo data, differentiating the coordinates corresponding to each peak in the one-dimensional range image echo data to obtain a differential result; Determine a target area corresponding to the observed target based on the difference result; wherein the target area is used to indicate a range coordinate set; Based on the target area, determining whether the observed target is a non-rigid structure; Based on whether the observed target is a non-rigid structure, locally autofocusing the echo data corresponding to the target area; The observation target whose motion state affects the imaging effect is a non-rigid structure, and vice versa, it is a rigid structure; when the observation target is a rigid structure, a prominent point in the target area is selected; and the echo data corresponding to the target area are autofocused based on the prominent point; When the observed target is a non-rigid structure, the target area includes at least two sub-areas; a distinctive point is selected from each of the sub-areas; and the echo data corresponding to each of the sub-areas are self-focused according to the distinctive point.

2. The method according to claim 1, characterized in that The step of differentiating the coordinates corresponding to each peak in the one-dimensional range image echo data based on the amplitude and coordinates of the one-dimensional range image echo data includes: Arranging the coordinates corresponding to each of the peaks in descending order of amplitude; The coordinates corresponding to the sorted peaks are differentiated pairwise.

3. The method according to claim 1, characterized in that The difference result includes: coordinate difference; Determining the target area corresponding to the observed target based on the difference result includes: When the coordinate difference is less than the area decision threshold, determining that the waveforms corresponding to the coordinate difference correspond to the same observation target; The waveforms corresponding to the observed targets are arranged in ascending order of the coordinates to obtain the target area.

4. The method according to claim 1, wherein The determining, based on the target area, whether the observed target is a non-rigid structure includes: Determining, based on the target area, a duty cycle of the observed target, a number of sub-areas of the target area, and a size of the target area; Whether the observed target is a rigid structure is determined according to the duty cycle, the number of sub-areas, and the size of the target area.

5. The method according to claim 4, characterized in that The step of determining the number of sub-regions of the target region according to the target region includes: Differences are made between adjacent distance coordinates corresponding to the peaks in the target area to obtain position difference values; When the position difference value is not less than the non-rigid multi-target decision threshold, the number of sub-regions is increased by 1 to determine the number of sub-regions.

6. The method according to claim 4, characterized in that The determining, based on the duty cycle, the number of sub-regions, and the size of the target region, whether the observed target is a rigid structure includes: When the duty cycle is less than the non-rigid structure duty cycle threshold, the size of the target area is not greater than the non-rigid structure radial size threshold, and the number of sub-areas is less than the non-rigid structure target number threshold, the corresponding observation target is determined to be a non-rigid structure.

7. The method according to claim 6, characterized in that After determining that the corresponding observation target is a non-rigid structure, the method further includes: When the duty cycle is greater than the rigid structure duty cycle threshold, or the size of the target area is greater than the rigid structure radial size threshold, or the number of sub-areas is greater than or equal to the rigid structure target number threshold, determining that the corresponding observation target is a rigid structure; The remaining observation targets are determined to be non-rigid structures.

8. A device for self-focusing missile-borne radar imaging, implementing the method according to any one of claims 1 to 7, characterized in that: include: Superposition module, difference module, classification module and local autofocus module; The superposition module is used to perform azimuth superposition on the echo data to obtain one-dimensional range image echo data; The difference module is used to differentiate the coordinates corresponding to each peak in the one-dimensional range image echo data based on the amplitude and coordinates of the one-dimensional range image echo data; The classification module is used to determine a target area corresponding to the observed target based on the difference result, and the target area is used to indicate a range coordinate set; Based on the target area, determining whether the observed target is a non-rigid structure; The local self-focusing module is used to perform local self-focusing on the echo data corresponding to the target area based on whether the observed target is a non-rigid structure.

Images