Signal processing device and signal processing method

Abstract

This signal processing device is configured to comprise: a frequency component calculation unit (1) that calculates a Doppler frequency component of an object included in an SAR image from image data indicating the SAR image, which is a synthetic aperture radar image; a frequency component division unit (2) that divides the Doppler frequency component calculated by the frequency component calculation unit (1) into a component in which the Doppler frequency component is positive and a component in which the Doppler frequency component is negative; and an image calculation unit (3) that calculates a first SAR image from the positive component after division by the frequency component division unit (2) and calculates a second SAR image from the negative component after division by the frequency component division unit (2). The signal processing device is also provided with a determination unit (4) that compares the first SAR image calculated by the image calculation unit (3) and the second SAR image calculated by the image calculation unit (3), and determines whether or not the object is a movement target, on the basis of the result of the comparison between the first SAR image and the second SAR image.

Description

Signal processing device and signal processing method

[0001] This disclosure relates to a signal processing device and a signal processing method.

[0002] There is a signal processing method for determining whether an object included in a synthetic aperture radar (SAR) image is a moving target. As an example of such a signal processing method, Non-Patent Document 1 discloses a signal processing method that performs batch processing to determine whether an object included in a SAR image is a moving target. Batch processing involves dividing a signal and processing multiple divided signals at once. Specifically, batch processing prepares multiple candidate moving speeds for each divided signal, estimates the moving speed candidate that corresponds to the moving speed of the moving target from among the multiple moving speed candidates, and then determines whether the object included in the SAR image is a moving target.

[0003] RK Raney, “Synthetic aperture imaging radar and moving tagets,” IEEE Trans. Aeronaut. Navig. Electron., vol. ANE-7, no. 3, pp. 499-505, May 1971, doi:10.1109 / TAES.1971.310292.

[0004] The signal processing method disclosed in Non-Patent Document 1 has the problem that it is necessary to perform batch processing to determine whether or not an object included in a SAR image is a moving target. Since the computational load of batch processing is generally high, it can be difficult to estimate the movement speed of each object by batch processing when multiple objects are included in a SAR image. For this reason, when multiple objects are included in a SAR image, it may not be possible to determine whether or not each object is a moving target.

[0005] This disclosure was made to solve the above-mentioned problems and aims to provide a signal processing device that can determine whether or not an object included in a SAR image is a moving target without performing batch processing.

[0006] The signal processing device according to this disclosure includes: a frequency component calculation unit that calculates the Doppler frequency components of an object included in a SAR image from image data showing a SAR image which is a synthetic aperture radar image; a frequency component splitting unit that splits the Doppler frequency components calculated by the frequency component calculation unit into a component with positive Doppler frequency components and a component with negative Doppler frequency components; and an image calculation unit that calculates a first SAR image from the positive components after splitting by the frequency component splitting unit and calculates a second SAR image from the negative components after splitting by the frequency component splitting unit. The signal processing device also includes a determination unit that compares the first SAR image calculated by the image calculation unit with the second SAR image calculated by the image calculation unit and determines whether or not the object is a moving target based on the comparison result between the first SAR image and the second SAR image.

[0007] According to this disclosure, it is possible to determine whether or not an object included in a SAR image is a moving target without performing batch processing.

[0008] This is a configuration diagram showing a signal processing device according to Embodiment 1. This is a hardware configuration diagram showing the hardware of the signal processing device according to Embodiment 1. This is a hardware configuration diagram of a computer when the signal processing device is implemented by software or firmware, etc. This is a flowchart showing a signal processing method, which is the processing procedure of the signal processing device. This is an explanatory diagram showing the Doppler frequency components of ground clutter included in a SAR image. This is an explanatory diagram showing the Doppler frequency components of a moving target included in a SAR image. This is a configuration diagram showing a signal processing device according to Embodiment 2. This is a hardware configuration diagram showing the hardware of the signal processing device according to Embodiment 2.

[0009] To provide a more detailed explanation of this disclosure, the forms for implementing this disclosure will be described below with reference to the attached drawings.

[0010] Embodiment 1. Figure 1 is a configuration diagram showing a signal processing device according to Embodiment 1. Figure 2 is a hardware configuration diagram showing the hardware of the signal processing device according to Embodiment 1. The signal processing device shown in Figure 1 comprises a frequency component calculation unit 1, a frequency component division unit 2, an image calculation unit 3, and a determination unit 4.

[0011] The frequency component calculation unit 1 is implemented, for example, by the frequency component calculation circuit 11 shown in Figure 2. The frequency component calculation unit 1 acquires image data showing a SAR image, which is a synthetic aperture radar image. From the image data showing the SAR image, the frequency component calculation unit 1 calculates the Doppler frequency components of objects contained in the SAR image. The frequency component calculation unit 1 outputs the Doppler frequency components of the objects to the frequency component division unit 2. Specifically, the frequency component calculation unit 1 converts the image data showing the SAR image into a signal in the frequency domain in the azimuth direction and outputs the signal in the frequency domain as the Doppler frequency components of the objects to the frequency component division unit 2.

[0012] The frequency component splitting unit 2 is implemented, for example, by the frequency component splitting circuit 12 shown in Figure 2. The frequency component splitting unit 2 obtains the Doppler frequency components of an object from the frequency component calculation unit 1. The frequency component splitting unit 2 splits the Doppler frequency components of the object into a component with a positive Doppler frequency component and a component with a negative Doppler frequency component. The frequency component splitting unit 2 outputs the component with a positive Doppler frequency component and the component with a negative Doppler frequency component to the image calculation unit 3.

[0013] The image calculation unit 3 is implemented, for example, by the image calculation circuit 13 shown in Figure 2. The image calculation unit 3 includes a first inverse frequency conversion unit 3a and a second inverse frequency conversion unit 3b. The image calculation unit 3 obtains components with positive Doppler frequency components and components with negative Doppler frequency components from the frequency component division unit 2. The image calculation unit 3 calculates a first SAR image from the positive components and a second SAR image from the negative components. The image calculation unit 3 outputs the first SAR image and the second SAR image, respectively, to the determination unit 4.

[0014] The first inverse frequency conversion unit 3a acquires the positive Doppler frequency component from the frequency component division unit 2. The first inverse frequency conversion unit 3a converts the positive component into a time-domain signal and outputs the time-domain signal as the first SAR image to the determination unit 4. The second inverse frequency conversion unit 3b acquires the negative Doppler frequency component from the frequency component division unit 2. The second inverse frequency conversion unit 3b converts the negative component into a time-domain signal and outputs the time-domain signal as the second SAR image to the determination unit 4.

[0015] The determination unit 4 is implemented, for example, by the determination circuit 14 shown in Figure 2. The determination unit 4 includes an amplitude ratio map calculation unit 4a, an imbalance map calculation unit 4b, and a determination processing unit 4c. The determination unit 4 acquires the first SAR image and the second SAR image from the image calculation unit 3. The determination unit 4 compares the first SAR image and the second SAR image and determines whether or not an object included in the SAR image is a moving target based on the comparison result between the first SAR image and the second SAR image.

[0016] The amplitude ratio map calculation unit 4a acquires the first SAR image from the first inverse frequency conversion unit 3a and the second SAR image from the second inverse frequency conversion unit 3b. The amplitude ratio map calculation unit 4a calculates an amplitude ratio map of the first SAR image and the second SAR image. The amplitude ratio map calculation unit 4a outputs the amplitude ratio map to the imbalance map calculation unit 4b. The imbalance map calculation unit 4b acquires the amplitude ratio map from the amplitude ratio map calculation unit 4a. The imbalance map calculation unit 4b calculates an imbalance map from the amplitude ratio map. The imbalance map calculation unit 4b outputs the imbalance map to the determination processing unit 4c. The determination processing unit 4c acquires the imbalance map from the imbalance map calculation unit 4b. Based on the imbalance map, the determination processing unit 4c determines whether or not an object included in the SAR image is a moving target.

[0017] In Figure 1, the signal processing device is assumed to be implemented by dedicated hardware as shown in Figure 2, with each of its components being the frequency component calculation unit 1, frequency component division unit 2, image calculation unit 3, and determination unit 4. Specifically, the signal processing device is assumed to be implemented by a frequency component calculation circuit 11, a frequency component division circuit 12, an image calculation circuit 13, and a determination circuit 14. Each of the frequency component calculation circuit 11, frequency component division circuit 12, image calculation circuit 13, and determination circuit 14 can be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0018] The components of a signal processing device are not limited to those implemented by dedicated hardware; the signal processing device may also be implemented by software, firmware, or a combination of software and firmware. The software or firmware is stored as a program in the computer's memory. A computer refers to the hardware that executes programs, and includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor).

[0019] Figure 3 is a hardware configuration diagram of a computer when the signal processing device is implemented by software or firmware. When the signal processing device is implemented by software or firmware, programs that cause the computer to execute the respective processing procedures in the frequency component calculation unit 1, the frequency component division unit 2, the image calculation unit 3, and the determination unit 4 are stored in memory 21. The computer's processor 22 then executes the programs stored in memory 21.

[0020] Furthermore, Figure 2 shows an example in which each component of the signal processing unit is implemented by dedicated hardware, and Figure 3 shows an example in which the signal processing unit is implemented by software or firmware, etc. However, this is only one example, and some components of the signal processing unit may be implemented by dedicated hardware, while the remaining components may be implemented by software or firmware, etc.

[0021] Next, the operation of the signal processing device shown in Figure 1 will be explained. Figure 4 is a flowchart of the signal processing method, which is the processing procedure of the signal processing device. The frequency component calculation unit 1 acquires image data showing the SAR image s(az,r) from, for example, a SAR sensor mounted on a satellite. az is the azimuth bin, where 1 ≤ az ≤ Az. Az is an integer of 2 or more. r is the range bin, where 1 ≤ r ≤ R. R is an integer of 2 or more.

[0022] The frequency component calculation unit 1 calculates the Doppler frequency component S(f,r) of an object contained in the SAR image s(az,r) from the image data representing the SAR image s(az,r) (step ST1 in Figure 4). Specifically, the frequency component calculation unit 1 converts the image data representing the SAR image s(az,r) into a frequency domain signal in the azimuth direction by, for example, performing an FFT (Fast Fourier Transform) on the image data representing the SAR image s(az,r) in the azimuth direction. The frequency component calculation unit 1 outputs the frequency domain signal as the Doppler frequency component S(f,r) of the object to the frequency component division unit 2. The Doppler frequency component S(f,r) of the object is two-dimensional data given by the range of Doppler frequency f being -PRF / 2 ≤ f ≤ PRF / 2. PRF is the pulse repetition frequency.

[0023] Assuming that the Doppler shift due to squint is corrected, if the object included in the SAR image s(az,r) is stationary ground clutter, no Doppler frequency shift occurs due to the object's movement, and therefore, as shown in Figure 5, the amplitude characteristics of the object's Doppler frequency component S(f,r) are symmetric. Ground clutter is an undetectable object in the SAR image s(az,r). On the other hand, if the object included in the SAR image s(az,r) is a moving target, a Doppler frequency shift occurs due to the object's movement, and therefore, as shown in Figure 6, the amplitude characteristics of the object's Doppler frequency component S(f,r) become asymmetric. In particular, if the size of the moving target is sufficiently larger than the wavelength of the radio waves used for observation, the scattering of the target may form a beam. In such cases, the asymmetry of the amplitude characteristics of the Doppler frequency component S(f,r) becomes greater. Figure 5 is an explanatory diagram showing the Doppler frequency component of ground clutter included in the SAR image. Figure 6 is an explanatory diagram showing the Doppler frequency component of a moving target included in the SAR image. In Figures 5 and 6, the horizontal axis represents the Doppler frequency, and the vertical axis represents the power of the Doppler frequency component.

[0024] The frequency component division unit 2 obtains the Doppler frequency component S(f,r) of the object from the frequency component calculation unit 1. The frequency component division unit 2 divides the Doppler frequency component S(f,r) of the object into components where the Doppler frequency component is positive S ＋ (f, r) and the negative component S of the Doppler frequency component － The frequency is divided into (f,r) and (step ST2 in Figure 4). Specifically, as shown in the following equations (1) to (2), the frequency component division unit 2 divides the Doppler frequency component S(f,r) into components S(f,r) and (f,r) with respect to the center frequency of the Doppler frequency f (0 Hz), where the Doppler frequency component is positive. ＋ (f, r) and the negative component S of the Doppler frequency component － The signal is divided into (f, r). The frequency component division unit 2 divides the Doppler frequency component into the positive component S. ＋ (f, r) and the Doppler frequency component is the negative component S －outputs (f, r) to the image calculation unit 3.

[0025]

[0026] When an object included in a SAR image s(az, r) is ground clutter, as described above, the amplitude characteristic of the Doppler frequency component S(f, r) of the object is symmetric. For this reason, as shown in Fig. 5, among the Doppler frequency components S(f, r) of the ground clutter included in the SAR image s(az, r), a component having a frequency equal to or higher than the center frequency (0 Hz) of the Doppler frequency f is a positive component S ＋ (f, r). Further, among the Doppler frequency components S(f, r) of the ground clutter included in the SAR image s(az, r), a component having a frequency lower than the center frequency (0 Hz) of the Doppler frequency f is a negative component S － (f, r). When an object included in the SAR image s(az, r) is a moving target, as described above, the amplitude characteristic of the Doppler frequency component S(f, r) of the object is asymmetric. For this reason, all of the Doppler frequency components S(f, r) of the moving target included in the SAR image s(az, r) are positive components S ＋ (f, r), or negative components S － (f, r). Fig. 6 shows an example in which all of the Doppler frequency components S(f, r) of the moving target included in the SAR image s(az, r) are negative components S － (f, r).

[0027] The image calculation unit 3 receives, from the frequency component dividing unit 2, a positive component S having a Doppler frequency component ＋ (f, r) and a negative component S having a Doppler frequency component － (f, r). The image calculation unit 3 obtains the positive component S ＋ (f, r) to obtain a first SAR image s ＋ (az, r), and calculates the second SAR image s from the negative component S － (az, r) (step ST3 in Fig. 4). Specifically, the first inverse frequency conversion unit 3a is, for example, configured to process the positive component S － ＋ ​By applying IFFT (Inverse Fast Fourier Transform) to (f,r) in the azimuth direction, the positive component S ＋ (f, r) is the first SAR image s ＋ Convert to (az,r). The first inverse frequency conversion unit 3a converts the first SAR image s ＋ (az, r) is output to the determination unit 4. The second inverse frequency conversion unit 3b, for example, the negative component S － By applying IFFT to (f,r) in the azimuth direction, the negative component S is obtained. － (f, r) is the second SAR image s － Convert to (az,r). The second inverse frequency conversion unit 3b converts the second SAR image s － (az, r) is output to the determination unit 4.

[0028] The determination unit 4 receives the first SAR image s from the first inverse frequency conversion unit 3a. ＋ (az, r) is obtained, and the second SAR image s is obtained from the second inverse frequency conversion unit 3b. － The (az,r) is obtained. The determination unit 4 obtains the first SAR image s ＋ (az,r) and the second SAR image s － Compare with (az,r) and the first SAR image s ＋ (az,r) and the second SAR image s － Based on the comparison result with (az,r), it is determined whether or not the object contained in the SAR image s(az,r) is a moving target.

[0029] The following describes the determination process performed by the determination unit 4 in detail. The amplitude ratio map calculation unit 4a of the determination unit 4 receives the first SAR image s from the first inverse frequency conversion unit 3a. ＋ (az, r) is obtained, and the second SAR image s is obtained from the second inverse frequency conversion unit 3b. － The (az,r) is obtained. The amplitude ratio map calculation unit 4a calculates the first SAR image s as shown in the following equation (3). ＋ (az,r) and the second SAR image s － The amplitude ratio map m(az,r) is calculated with (az,r) (step ST4 in Figure 4). The amplitude ratio map calculation unit 4a outputs the amplitude ratio map m(az,r) to the imbalance map calculation unit 4b.

[0030] In equation (3), N is any real number. E(|s) ＋ (az,r)|) is the first SAR image s ＋ This is the expected value of (az, r). E(|s) － (az,r)|) is the second SAR image s － This is the expected value of (az, r).

[0031] The amplitude ratio map m(az,r) is obtained from the first SAR image s ＋ (az,r) and the second SAR image s － Rather than a simple amplitude ratio with (az, r), the first SAR image s ＋ (az,r) or second SAR image s － When either value of (az, r) is small, a correction is made so that the error is not exaggerated. If the object contained in the SAR image s(az, r) is ground clutter, the first SAR image s ＋ Pixel intensity of (az,r) and second SAR image s － Because the difference with the pixel intensity of (az,r) is small, the amplitude ratio map m(az,r) will be a value close to 1. On the other hand, if the object contained in the SAR image s(az,r) is a moving target, the first SAR image s ＋ Pixel intensity of (az,r) and second SAR image s － Because the difference between the pixel intensity (az, r) and the amplitude ratio map m(az, r) is large, the amplitude ratio map m(az, r) will be a value far from 1.

[0032] The imbalance map calculation unit 4b obtains the amplitude ratio map m(az,r) from the amplitude ratio map calculation unit 4a. The imbalance map calculation unit 4b calculates the imbalance map i(az,r) from the amplitude ratio map m(az,r) as shown in the following equation (6) (step ST5 in Figure 4). The imbalance map calculation unit 4b outputs the imbalance map i(az,r) to the determination processing unit 4c.

[0033]

[0034] If the object contained in the SAR image s(az,r) is ground clutter, then the first SAR image s＋ (az,r) and the second SAR image s － Because the imbalance with (az,r) is small, the value of the imbalance map i(az,r) is approximately 0. On the other hand, if the object contained in the SAR image s(az,r) is a moving target, the first SAR image s ＋ (az,r) and the second SAR image s － Because the imbalance with (az, r) is large, the value of the imbalance map i(az, r) will be greater than or equal to the threshold Th. The threshold Th is a value greater than 0.

[0035] The determination processing unit 4c obtains the imbalance map i(az,r) from the imbalance map calculation unit 4b. Based on the imbalance map i(az,r), the determination processing unit 4c determines whether or not an object included in the SAR image is a moving target. Specifically, the determination processing unit 4c compares the imbalance map i(az,r) with a threshold Th. The threshold Th may be stored in the internal memory of the determination processing unit 4c or may be provided from an external source. If the imbalance map i(az,r) is greater than or equal to the threshold Th (step ST6 in Figure 4: YES), the determination processing unit 4c determines that the object included in the SAR image s(az,r) is a moving target (step ST7 in Figure 4). If the imbalance map i(az,r) is less than the threshold Th (step ST6 in Figure 4: NO), the determination processing unit 4c determines that the object included in the SAR image s(az,r) is ground clutter (step ST8 in Figure 4).

[0036] The determination result of the determination processing unit 4c is output to, for example, a target tracking device (not shown). This enables the target tracking device to track the moving target. The determination result of the determination processing unit 4c is also output to, for example, a display device (not shown). This causes the display device to show whether or not there is a moving target. The user can confirm the presence or absence of a moving target by looking at the display on the display device.

[0037] In the above-described embodiment 1, the signal processing device is configured to include: a frequency component calculation unit 1 that calculates the Doppler frequency component of an object contained in a SAR image from image data showing a SAR image which is a synthetic aperture radar image; a frequency component splitting unit 2 that splits the Doppler frequency component calculated by the frequency component calculation unit 1 into a component with a positive Doppler frequency component and a component with a negative Doppler frequency component; and an image calculation unit 3 that calculates a first SAR image from the positive component after splitting by the frequency component splitting unit 2 and calculates a second SAR image from the negative component after splitting by the frequency component splitting unit 2. The signal processing device also includes a determination unit 4 that compares the first SAR image calculated by the image calculation unit 3 with the second SAR image calculated by the image calculation unit 3 and determines whether or not the object is a moving target based on the comparison result between the first SAR image and the second SAR image. Therefore, the signal processing device can determine whether or not an object contained in a SAR image is a moving target without performing batch processing.

[0038] The batch processing disclosed in Non-Patent Document 1 prepares multiple candidate speeds for the moving target present in the divided signal and estimates the speed candidate that corresponds to the moving target's speed from among the multiple speed candidates. Because the computational load of batch processing is generally high, when multiple moving targets are present in the divided signal, it can be difficult to estimate the speed of each moving target using such batch processing. For this reason, when multiple objects are included in the SAR image, it may not be possible to determine whether each object is a moving target using such batch processing. The signal processing device shown in Figure 1, when multiple objects are included in the SAR image, can determine whether each object is a moving target by focusing on each object and comparing the first SAR image with the second SAR image for each object.

[0039] In the signal processing device shown in Figure 1, the frequency component calculation unit 1 converts the image data representing the SAR image s(az,r) into a signal in the frequency domain in the azimuth direction by performing an FFT (Fast Fast Transform) on the image data representing the SAR image s(az,r) in the azimuth direction. However, this is just one example, and the frequency component calculation unit 1 may, for example, convert the image data representing the SAR image s(az,r) into a signal in the frequency domain in the azimuth direction by performing a DFT (Discrete Fourier Transformation) on the image data representing the SAR image s(az,r) in the azimuth direction.

[0040] In the signal processing device shown in Figure 1, the first inverse frequency conversion unit 3a converts the positive component S ＋ By performing IFFT on (f,r) in the azimuth direction, the positive component S is obtained. ＋ (f, r) is the first SAR image s ＋ The second inverse frequency conversion unit 3b converts the negative component S to (az, r). － By applying IFFT to (f,r) in the azimuth direction, the negative component S is obtained. － (f, r) is the second SAR image s － It is being converted to (az, r). However, this is just one example, and the first inverse frequency conversion unit 3a converts, for example, the positive component S ＋ By performing IDFT (Inverse Discrete Fourier Transformation) on (f,r) in the azimuth direction, the positive component S ＋ (f, r) is the first SAR image s ＋ The second inverse frequency conversion unit 3b converts the negative component S to (az, r). － By applying IDFT to (f,r) in the azimuth direction, the negative component S is obtained. － (f, r) is the second SAR image s － Alternatively, you could convert it to (az, r).

[0041] In the signal processing device shown in Figure 1, the determination unit 4 determines the first SAR image s ＋ (az,r) and the second SAR image s －Based on the imbalance with (az,r), it is determined whether or not the object is a moving target. However, this is just one example, and the determination unit 4 also determines, for example, the first SAR image s ＋ (az,r) and the second SAR image s － The similarity to (az, r) may be calculated, and based on the similarity, it may be determined whether or not the object is a moving target.

[0042] Embodiment 2. In Embodiment 2, the positive component S ＋ Among the signals contained in (f, r), a signal reduction process is performed to reduce signals with smaller amplitudes by a larger amount, and the negative component S － This section describes a signal processing device equipped with a signal reduction processing unit 5 that performs signal reduction processing, which reduces signals with smaller amplitudes among the signals contained in (f, r) by a larger amount.

[0043] Figure 7 is a configuration diagram showing a signal processing device according to Embodiment 2. In Figure 7, the same reference numerals as in Figure 1 indicate the same or corresponding parts, so a detailed explanation is omitted. Figure 8 is a hardware configuration diagram showing the hardware of the signal processing device according to Embodiment 2. In Figure 8, the same reference numerals as in Figure 2 indicate the same or corresponding parts, so a detailed explanation is omitted. The signal processing device shown in Figure 7 includes a frequency component calculation unit 1, a frequency component division unit 2, a signal reduction processing unit 5, an image calculation unit 6, and a determination unit 4.

[0044] The signal reduction processing unit 5 is implemented, for example, by the signal reduction processing circuit 15 shown in Figure 8. The signal reduction processing unit 5 includes a first weighting unit 5a and a second weighting unit 5b. The signal reduction processing unit 5 obtains the Doppler frequency component S(f,r) from the frequency component calculation unit 1. The signal reduction processing unit 5 obtains the component S from the frequency component division unit 2 which is a positive Doppler frequency component. ＋ (f, r) and the Doppler frequency component is the negative component S － (f, r) is obtained. The signal reduction processing unit 5 reduces the positive component S ＋ Among the signals contained in (f, r), a signal reduction process is performed to reduce signals with smaller amplitudes by a larger amount, and the negative component S －The signal reduction processing unit 5 reduces the signal with smaller amplitudes within the signal contained in (f, r) by a larger amount. ＋ (f, r) and the negative component S' after signal reduction processing － The values ​​of (f) and (r) are output to the image calculation unit 6.

[0045] The first weighting unit 5a obtains the Doppler frequency component S(f,r) from the frequency component calculation unit 1. The first weighting unit 5a obtains the component S from the frequency component division unit 2 where the Doppler frequency component is positive. ＋ The first weighting unit 5a, as a signal reduction process, uses the expected value E(|S(f,r)|) of the Doppler frequency component S(f,r) to determine the positive component S of the Doppler frequency component. ＋ Weighting is applied to (f, r). The first weighting unit 5a reduces the positive component after weighting to the positive component S' after signal reduction processing. ＋ The output (f, r) is sent to the image calculation unit 6.

[0046] The second weighting unit 5b obtains the Doppler frequency component S(f,r) from the frequency component calculation unit 1. The second weighting unit 5b obtains the negative component S from the frequency component division unit 2. － The second weighting unit 5b performs a signal reduction process using the expected value E(|S(f,r)|) of the Doppler frequency component S(f,r), and the negative component S － Weighting is applied to (f, r). The second weighting unit 5b reduces the negative component after weighting to the negative component S' after signal reduction processing. － The output (f, r) is sent to the image calculation unit 6.

[0047] The image calculation unit 6 is implemented, for example, by the image calculation circuit 16 shown in Figure 8. The image calculation unit 6 includes a first inverse frequency conversion unit 6a and a second inverse frequency conversion unit 6b. The image calculation unit 6 receives the positive component S' after signal reduction processing from the signal reduction processing unit 5. ＋ (f, r) and the negative component S' after signal reduction processing － The (f, r) is obtained. The image calculation unit 6 obtains the positive component S' after signal reduction processing.＋ from (f, r), a first SAR image s ＋ (az, r) is calculated, and the negative component S' after signal reduction processing － from (f, r), a second SAR image s － (az, r) is calculated. The image calculation unit 6 calculates the first SAR image s ＋ (az, r) and the second SAR image s － outputs each of (az, r) to the determination unit 4.

[0048] The first inverse frequency transform unit 6a receives, from the first weighting unit 5a, the positive component S' after signal reduction processing ＋ (f, r) is acquired. The first inverse frequency transform unit 6a processes the positive component S' ＋ (f, r) into a time-domain signal, and converts the time-domain signal into the first SAR image s ＋ (az, r) and outputs it to the determination unit 4. The second inverse frequency transform unit 6b receives, from the second weighting unit 5b, the negative component S' after signal reduction processing － (f, r) is acquired. The second inverse frequency transform unit 6b processes the negative component S' － (f, r) into a time-domain signal, and converts the time-domain signal into the second SAR image s － (az, r) and outputs it to the determination unit 4.

[0049] In FIG. 7, it is assumed that each of the frequency component calculation unit 1, frequency component division unit 2, signal reduction processing unit 5, image calculation unit 6, and determination unit 4, which are constituent elements of the signal processing apparatus, is implemented by dedicated hardware as shown in FIG. 8. That is, it is assumed that the signal processing apparatus is implemented by the frequency component calculation circuit 11, frequency component division circuit 12, signal reduction processing circuit 15, image calculation circuit 16, and determination circuit 14. Each of the frequency component calculation circuit 11, frequency component division circuit 12, signal reduction processing circuit 15, image calculation circuit 16, and determination circuit 14 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0050] Components of the signal processing device are not limited to those implemented by dedicated hardware, and the signal processing device may be implemented by software, firmware, or a combination of software and firmware. When the signal processing device is implemented by software, firmware, or the like, a program for causing a computer to execute respective processing procedures in the frequency component calculation unit 1, the frequency component dividing unit 2, the signal reduction processing unit 5, the image calculation unit 6, and the determination unit 4 is stored in the memory 21 illustrated in FIG. 3. Then, the processor 22 illustrated in FIG. 3 executes the program stored in the memory 21.

[0051] Further, FIG. 8 illustrates an example in which each component of the signal processing device is implemented by dedicated hardware, and FIG. 3 illustrates an example in which the signal processing device is implemented by software, firmware, or the like. However, this is merely an example, and some components of the signal processing device may be implemented by dedicated hardware, and the remaining components may be implemented by software, firmware, or the like.

[0052] Next, an operation of the signal processing device illustrated in FIG. 7 will be described. However, configurations except for the signal reduction processing unit 5 and the image calculation unit 6 are the same as those of the signal processing device illustrated in FIG. 1. Therefore, only operations of the signal reduction processing unit 5 and the image calculation unit 6 will be described herein.

[0053] A first weighting unit 5a of the signal reduction processing unit 5 acquires a Doppler frequency component S(f,r) from the frequency component calculation unit 1. The first weighting unit 5a acquires a positive Doppler frequency component S ＋ In order to reduce noise included in (f,r), as shown in the following formula (7), weighting is performed on the positive component S of Doppler frequency components by using an expected value E(|S(f,r)|) of the Doppler frequency component S(f,r) ＋ (f,r). The first weighting unit 5a outputs the weighted positive component as a positive component S' after signal reduction processing ＋ (f,r) to the image calculation unit 6.

[0054] In equation (7), M is any real number. E(|S(f,r)|) is the expected value of the Doppler frequency component S(f,r).

[0055] The second weighting unit 5b of the signal reduction processing unit 5 obtains the Doppler frequency component S(f,r) from the frequency component calculation unit 1. The second weighting unit 5b obtains the negative component S from the frequency component division unit 2. － The (f, r) is obtained. The second weighting unit 5b assigns a negative component S to the Doppler frequency component. － To reduce the noise contained in (f,r), the expected value E(|S(f,r)|) of the Doppler frequency component S(f,r) is used as shown in equation (9) below, so that the Doppler frequency component is a negative component S － Weighting is applied to (f, r). The second weighting unit 5b reduces the negative component after weighting to the negative component S' after signal reduction processing. － The output (f, r) is sent to the image calculation unit 6.

[0056]

[0057] The weighting in the first weighting unit 5a and the second weighting unit 5b significantly reduces signals with small amplitudes included in the Doppler frequency component, while reducing signals with large amplitudes included in the Doppler frequency component is less significant. Since the moving target is unevenly distributed on the Doppler frequency spectrum, it remains even after such weighting, and unwanted signals, which are noise, are significantly reduced.

[0058] The image calculation unit 6 receives the positive component S' after signal reduction processing from the signal reduction processing unit 5. ＋ (f, r) and the negative component S' after signal reduction processing － The (f, r) is obtained. The image calculation unit 6 obtains the positive component S' after signal reduction processing. ＋ (f, r) to the first SAR image s ＋ (az, r) is calculated, and the negative component S' after signal reduction processing is obtained. － (f, r) to the second SAR image s － Calculate (az, r).

[0059] Specifically, the first inverse frequency conversion unit 6a converts, for example, the positive component S' after signal reduction processing. ＋ By applying IFFT to (f,r) in the azimuth direction, the positive component S' is obtained. ＋ (f, r) is the first SAR image s ＋ Convert to (az,r). The first inverse frequency conversion unit 6a converts the first SAR image s ＋ (az, r) is output to the determination unit 4. The second inverse frequency conversion unit 6b, for example, the negative component S' after signal reduction processing. － By applying IFFT to (f,r) in the azimuth direction, the negative component S' is obtained. － (f, r) is the second SAR image s － Convert to (az,r). The second inverse frequency conversion unit 6b converts the second SAR image s － (az, r) is output to the determination unit 4.

[0060] In the above embodiment 2, the signal processing device shown in Figure 7 is configured to include a signal reduction processing unit 5 that reduces signals with smaller amplitudes more significantly among the signals included in the positive components after division by the frequency component division unit 2, and also reduces signals with smaller amplitudes more significantly among the signals included in the negative components after division by the frequency component division unit 2. The image calculation unit 6 of the signal processing device shown in Figure 7 calculates a first SAR image from the positive components after signal reduction processing by the signal reduction processing unit 5, and calculates a second SAR image from the negative components after signal reduction processing by the signal reduction processing unit 5. Therefore, the signal processing device shown in Figure 7 can improve the accuracy of determining moving targets compared to the signal processing device shown in Figure 1.

[0061] In the second embodiment, the signal reduction processing unit 5 is configured to perform a signal reduction process that uses the expected value of the Doppler frequency component calculated by the frequency component calculation unit 1 to weight the positive component after division by the frequency component division unit 2, and a signal reduction process that uses the expected value of the Doppler frequency component to weight the negative component after division by the frequency component division unit 2. This is how the signal processing device shown in Figure 7 is configured. Therefore, the signal processing device shown in Figure 7 can reduce the noise contained in both the positive and negative components after division.

[0062] Furthermore, this disclosure allows for free combination of each embodiment, modification of any component in each embodiment, or omission of any component in each embodiment.

[0063] The signal processing apparatus and signal processing method according to this disclosure include: a frequency component calculation unit that calculates the Doppler frequency component of an object contained in a SAR image from image data showing a SAR image which is a synthetic aperture radar image; a frequency component splitting unit that splits the Doppler frequency component calculated by the frequency component calculation unit into a component with a positive Doppler frequency component and a component with a negative Doppler frequency component; and an image calculation unit that calculates a first SAR image from the positive component after splitting by the frequency component splitting unit and calculates a second SAR image from the negative component after splitting by the frequency component splitting unit. Furthermore, the signal processing apparatus includes a determination unit that compares the first SAR image calculated by the image calculation unit with the second SAR image calculated by the image calculation unit and determines whether or not the object is a moving target based on the comparison result between the first SAR image and the second SAR image, and can determine whether or not an object contained in a SAR image is a moving target without performing batch processing, making it suitable for signal processing apparatus and signal processing method.

[0064] 1. Frequency component calculation unit, 2. Frequency component division unit, 3. Image calculation unit, 3a. First inverse frequency conversion unit, 3b. Second inverse frequency conversion unit, 4. Judgment unit, 4a. Amplitude ratio map calculation unit, 4b. Imbalance map calculation unit, 4c. Judgment processing unit, 5. Signal reduction processing unit, 5a. First weighting unit, 5b. Second weighting unit, 6. Image calculation unit, 6a. First inverse frequency conversion unit, 6b. Second inverse frequency conversion unit, 11. Frequency component calculation circuit, 12. Frequency component division circuit, 13. Image calculation circuit, 14. Judgment circuit, 15. Signal reduction processing circuit, 16. Image calculation circuit, 21. Memory, 22. Processor.

Claims

1. A signal processing device comprising: a frequency component calculation unit that calculates the Doppler frequency component of an object included in a SAR image from image data showing a SAR image which is a synthetic aperture radar image; a frequency component splitting unit that splits the Doppler frequency component calculated by the frequency component calculation unit into a component in which the Doppler frequency component is positive and a component in which the Doppler frequency component is negative; an image calculation unit that calculates a first SAR image from the positive component after splitting by the frequency component splitting unit and calculates a second SAR image from the negative component after splitting by the frequency component splitting unit; and a determination unit that compares the first SAR image calculated by the image calculation unit and the second SAR image calculated by the image calculation unit and determines whether or not the object is a moving target based on the comparison result between the first SAR image and the second SAR image.

2. The signal processing apparatus according to claim 1, comprising a signal reduction processing unit that reduces signals with smaller amplitudes more significantly among the signals contained in the positive components after division by the frequency component division unit, and a signal reduction processing unit that reduces signals with smaller amplitudes more significantly among the signals contained in the negative components after division by the frequency component division unit, wherein the image calculation unit calculates the first SAR image from the positive components after signal reduction processing by the signal reduction processing unit, and calculates the second SAR image from the negative components after signal reduction processing by the signal reduction processing unit.

3. The signal reduction processing unit is characterized in that, as a signal reduction process, it uses the expected value of the Doppler frequency component calculated by the frequency component calculation unit to weight the positive component after division by the frequency component division unit, and as a signal reduction process, it uses the expected value of the Doppler frequency component to weight the negative component after division by the frequency component division unit.

4. The signal processing apparatus according to any one of claims 1 to 3, characterized in that the frequency component calculation unit converts image data showing the SAR image into a signal in the frequency domain in the azimuth direction, and outputs the signal in the frequency domain to the frequency component division unit as the Doppler frequency component of the object.

5. The signal processing apparatus according to any one of claims 1 to 4, characterized in that the image calculation unit converts the positive components after division by the frequency component division unit into time-domain signals, outputs the time-domain signals as the first SAR image to the determination unit, and converts the negative components after division by the frequency component division unit into time-domain signals, and outputs the time-domain signals as the second SAR image to the determination unit.

6. The signal processing apparatus according to any one of claims 1 to 5, wherein the determination unit comprises: an amplitude ratio map calculation unit that calculates an amplitude ratio map of a first SAR image calculated by the image calculation unit and a second SAR image calculated by the image calculation unit; an imbalance map calculation unit that calculates an imbalance map from the amplitude ratio map calculated by the amplitude ratio map calculation unit; and a determination processing unit that determines whether or not the object is a moving target based on the imbalance map calculated by the imbalance map calculation unit.

7. A signal processing method comprising: a frequency component calculation unit calculating the Doppler frequency components of an object included in a SAR image from image data showing a SAR image which is a synthetic aperture radar image; a frequency component splitting unit splitting the Doppler frequency components calculated by the frequency component calculation unit into a component with positive Doppler frequency components and a component with negative Doppler frequency components; an image calculation unit calculating a first SAR image from the positive component after splitting by the frequency component splitting unit and calculating a second SAR image from the negative component after splitting by the frequency component splitting unit; and a determination unit comparing the first SAR image calculated by the image calculation unit with the second SAR image calculated by the image calculation unit, and determining whether the object is a moving target based on the comparison result between the first SAR image and the second SAR image.

Images