Inverse synthetic aperture radar image synthesis apparatus, inverse synthetic aperture radar image synthesis method, and inverse synthetic aperture radar image synthesis program

Abstract

To provide an inverse synthetic aperture radar image synthesis device which can sharpen an ISAR image by using a plurality of ISAR images.SOLUTION: An inverse synthetic aperture radar image synthesis device 20 comprises: an inspection unit 26 which executes processing of calculating the mutual correlation coefficients of a correlation window and an inspection range by superimposing a center pixel of the set correlation window on any one pixel within the set inspection range over all the pixels within the inspection range; and a determination unit 27 which determines that the position of an imaging object indicated by coordinates of a pixel of a slave calculated from the position of the correlation window that gives the maximum mutual correlation coefficient in the plurality of calculated mutual correlation coefficients is the same as the position of the imaging object indicated by coordinates of a pixel with the maximum luminance value within the range equivalent to the extracted range in a master.SELECTED DRAWING: Figure 14

Description

Technical Field

[0001] The present invention relates to an inverse synthetic aperture radar image synthesizing apparatus, an inverse synthetic aperture radar image synthesizing method, and an inverse synthetic aperture radar image synthesizing program.

Background Art

[0002] An ISAR (Inverse Synthetic Aperture Radar) image indicates a difference in Doppler frequency caused by a difference in the angular velocity of each point of an imaging target due to the shaking of the imaging target. Therefore, ISAR images often become unclear due to changes in angular velocity or changes in the axis of rotation.

[0003] When sharpening an ISAR image while maintaining the integration time, which is the acquisition time of ISAR data used for generating the ISAR image, a method of performing autofocus processing or the like can be considered. However, even if autofocus processing or the like is performed, there is a possibility that the above autofocus processing or the like may not converge due to changes in angular velocity or changes in the axis of rotation included in the ISAR image.

[0004] If ISAR data with a short integration time is used, there is a high possibility of obtaining a clear ISAR image. However, if ISAR data with a short integration time is used, the signal of the imaging target may be buried in noise, and the signal-to-noise ratio (S / N) may decrease.

[0005] Also, Patent Document 1 describes a radar apparatus and a signal processing apparatus that can increase the resolution of an ISAR image.

[0006] Also, Patent Document 2 describes a radar signal processing apparatus that can more accurately identify an object.

[0007] Also, Patent Document 3 describes a radar apparatus that can obtain an accurate target image with few pixel dropouts regardless of the structure and posture of the target by making it possible to overlap ISAR images for each time using the characteristic points of the target.

[0008] Furthermore, Patent Document 4 describes a time-series image processing device that can accurately identify the boundaries of a target in order to reliably acquire the target when using images created from radar echoes from a radar target. [Prior art documents] [Patent Documents]

[0009] [Patent Document 1] Japanese Patent Publication No. 2021-006779 [Patent Document 2] Japanese Patent Publication No. 2019-219339 [Patent Document 3] Japanese Patent Publication No. 2009-162611 [Patent Document 4] Japanese Patent Publication No. 2002-365363 [Overview of the project] [Problems that the invention aims to solve]

[0010] As described above, ISAR images show the intensity of Doppler frequencies caused by the shaking of the imaged object, and therefore often become blurry due to changes in angular velocity or rotation axis. Thus, establishing a method for sharpening ISAR images is a challenge when using them.

[0011] One possible method for sharpening ISAR images is to simply add multiple ISAR images together. This method involves selecting pixels from each ISAR image that represent the same location of the imaged object, and then transforming the coordinates of these selected pixels so that they overlap. After transforming the coordinates, the ISAR images are then combined by overlapping the selected pixels.

[0012] However, because ISAR images capture the shaking of the object being imaged, the appearance of the image can differ depending on the time, even at the same location. Therefore, it is difficult to select pixels that indicate the same location of the object from multiple ISAR images.

[0013] Furthermore, when overlaying aligned ISAR images, if the complex numerical deviations of the pixels are not the same, the brightness may be reduced. To avoid this reduction in brightness, it is necessary to match the deviations of each pixel between the ISAR images.

[0014] However, matching the polarization angle based on pixels where noise is displayed reduces the signal-to-noise ratio (S / N). Therefore, when overlaying ISAR images, it is important to adjust the polarization angle pixel by pixel between the ISAR images so that only the signal components reinforce each other.

[0015] Patent documents 1 to 4 do not describe selecting pixels from multiple ISAR images that indicate the same position of the object being imaged, nor do they describe adjusting the polarity of each pixel between ISAR images so that only the signal components reinforce each other.

[0016] Therefore, the present invention aims to provide an inverse synthetic aperture radar image synthesis apparatus, an inverse synthetic aperture radar image synthesis method, and an inverse synthetic aperture radar image synthesis program that can enhance ISAR images using multiple ISAR images. [Means for solving the problem]

[0017] The inverse synthetic aperture radar image synthesis apparatus according to the present invention includes: a conversion unit that converts an ISAR image into a real number image by setting the absolute value of each pixel of the ISAR image to the brightness value of each pixel; a generation unit that generates a single real number image by adding a plurality of real number images obtained by converting a plurality of ISAR images of a predetermined size and the same imaging target, each captured at different times, so that pixels with the same coordinates overlap; an extraction unit that extracts a range that satisfies a predetermined first condition from the single generated real number image; a first setting unit that sets a correlation window in the master, which is one of the plurality of ISAR images, that includes a range corresponding to the extracted range in the master; and a plurality of ISAR images The system is characterized by comprising: a second setting unit that sets a range corresponding to the set correlation window in a slave ISAR image other than the master as the inspection range for the slave; an inspection unit that overlays the center pixel of the set correlation window onto any one pixel within the set inspection range and calculates the cross-correlation coefficient between the correlation window and the inspection range for all pixels within that inspection range; and a determination unit that determines that the position of the imaging target indicated by the coordinates of the slave pixel calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the calculated multiple cross-correlation coefficients is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master.

[0018] The inverse synthetic aperture radar image synthesis method according to the present invention converts an ISAR image into a real number image by setting the absolute value of each pixel of the ISAR image to the brightness value of each pixel; generates a single real number image by adding together multiple real number images obtained from multiple ISAR images of a predetermined size and the same imaging target, each captured at different times, so that pixels with the same coordinates overlap; extracts a range that satisfies a predetermined first condition from the generated single real number image; sets a correlation window in the master, which is one of the multiple ISAR images, that includes the range corresponding to the extracted range in the master; and in the multiple ISAR images... The process involves setting an inspection range for a slave ISAR image other than the master image, where the range corresponding to the set correlation window in the slave image is set as the inspection range. The process of overlaying the center pixel of the set correlation window onto any one pixel within the set inspection range and calculating the cross-correlation coefficient between that correlation window and the inspection range is performed for all pixels within that inspection range. The system then determines that the position of the imaging target indicated by the coordinates of the slave pixel, calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the calculated multiple cross-correlation coefficients, is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master image.

[0019] The inverse synthetic aperture radar image synthesis program according to the present invention causes a computer to perform a conversion process of converting an ISAR image into a real number image by using the absolute value of each pixel of the ISAR image as the luminance value of each pixel, a generation process of generating one real number image by adding a plurality of real number images obtained by converting a plurality of ISAR images, each having a predetermined size and taken at different times of the same imaging target, so that pixels with the same coordinates overlap each other, an extraction process of extracting a range that satisfies a predetermined first condition from the generated one real number image, a first setting process of setting a correlation window including a range corresponding to the extracted range in a master, which is one of the plurality of ISAR images, in the master, a second setting process of setting, in a slave, which is an ISAR image other than the master among the plurality of ISAR images, a range corresponding to the set correlation window as an inspection range in the slave, an inspection process of calculating the cross-correlation coefficient between the set correlation window and the inspection range by overlapping the center pixel of the set correlation window with any one pixel within the set inspection range and executing this process for all pixels within the inspection range, and a determination process of determining that the position of the imaging target indicated by the coordinates of the pixels of the slave calculated from the position of the correlation window giving the maximum cross-correlation coefficient among the calculated plurality of cross-correlation coefficients is the same as the position of the imaging target indicated by the coordinates of the pixel having the maximum luminance value within the range corresponding to the extracted range in the master.

Advantages of the Invention

[0020] According to the present invention, an ISAR image can be sharpened using a plurality of ISAR images.

Brief Description of the Drawings

[0021] [Figure 1] It is a block diagram showing a configuration example of an inverse synthetic aperture radar image synthesis apparatus according to an embodiment of the present invention. [Figure 2] It is a block diagram showing a configuration example of the same position selection unit 130. [Figure 3] It is an explanatory diagram showing an example of an ISAR image after a plurality of ISAR images are simply added. [Figure 4] It is an explanatory diagram showing an example of a master. [Figure 5] This is an explanatory diagram showing other examples of master data. [Figure 6] This is an explanatory diagram illustrating an example of a master and a slave. [Figure 7] This is an explanatory diagram showing other examples of master and slave relationships. [Figure 8] This is a block diagram showing an example configuration of the angular alignment processing unit 150. [Figure 9] This is an explanatory diagram showing an example of a complex ratio map generated by the coefficient map generation unit 152. [Figure 10] This flowchart shows the operation of the ISAR image sharpening process by the inverse synthetic aperture radar image synthesis device 100 of this embodiment. [Figure 11] This flowchart shows the operation of the same-position selection process by the same-position selection unit 130 in this embodiment. [Figure 12] This flowchart shows the operation of the angle alignment process by the angle alignment processing unit 150 of this embodiment. [Figure 13] This is an explanatory diagram showing an example of the hardware configuration of the inverse synthetic aperture radar image synthesis apparatus 100 according to the present invention. [Figure 14] This is a block diagram illustrating the outline of the inverse synthetic aperture radar image synthesis apparatus according to the present invention. [Modes for carrying out the invention]

[0022] [Explanation of the structure] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Figure 1 is a block diagram showing an example configuration of an inverse synthetic aperture radar image synthesis apparatus according to an embodiment of the present invention.

[0023] The inverse synthetic aperture radar image synthesis apparatus 100 of this embodiment improves the signal-to-noise ratio of ISAR images by superimposing ISAR images with short integration times. Furthermore, the inverse synthetic aperture radar image synthesis apparatus 100 selects the same position of the imaging target in each ISAR image using an ISAR image with a long integration time that contains information for all time points. In addition, the inverse synthetic aperture radar image synthesis apparatus 100 matches only the declination angle of pixels where the signal component is considered dominant between the ISAR images.

[0024] The inverse synthetic aperture radar image synthesis apparatus 100 shown in Figure 1 comprises an integration time division unit 110, an ISAR imaging unit 120, a same-position selection unit 130, a position alignment processing unit 140, a declination angle alignment processing unit 150, and an imaging unit 160.

[0025] The integration time division unit 110 has the function of converting a single ISAR data (long-time ISAR data), in which the received signals are arranged in a time series, into multiple ISAR data (short-time ISAR data) by dividing it into predetermined time intervals. The long-time ISAR data corresponds to the data of the image target acquired by ISAR.

[0026] The ISAR imaging unit 120 has the function of converting ISAR data into ISAR images. In the example shown in Figure 1, the ISAR imaging unit 120 converts multiple short-term ISAR data into multiple short-term ISAR images.

[0027] Short-time ISAR images are ISAR images with a short integration time. Furthermore, the transformed short-time ISAR image is a complex number image. Multiple short-time ISAR images correspond to images of the same subject, each of a predetermined size, captured at different time points.

[0028] The same-position selection unit 130 has the function of selecting the same position of the imaging target from multiple short-time ISAR images (hereinafter referred to as multiple ISAR images). Figure 2 is a block diagram showing an example of the configuration of the same-position selection unit 130.

[0029] The same position selection unit 130 shown in Figure 2 includes a real number conversion unit 131, an addition unit 132, a range extraction unit 133, a correlation window setting unit 134, a correlation test range setting unit 135, and a correlation test unit 136.

[0030] The real number conversion unit 131 converts each of the multiple input ISAR images into real number images. The real number conversion unit 131 converts the images into real number images by taking the absolute value of each pixel in the short-time ISAR image, which represents a complex number, as the value (luminance) of each pixel.

[0031] The addition unit 132 adds together pixels with the same coordinates in the multiple converted real number images. Note that the addition process by the addition unit 132 does not perform alignment. Through the addition process, the addition unit 132 generates a single real number image of a predetermined size based on the multiple real number images.

[0032] The range extraction unit 133 generates a histogram of the luminance values ​​of the generated real image. Next, the range extraction unit 133 determines the lower a% and upper b% luminance ranges from the generated histogram. Then, the range extraction unit 133 extracts all pixels from the generated real image that represent the luminance values ​​included in the determined luminance ranges. Note that a and b are arbitrary values ​​between 0 and 100. Also, a and b do not have to be equal.

[0033] Next, the range extraction unit 133 extracts from the generated real number image a range of size d pixels × d pixels, in which each range contains at least p pixels representing luminance values ​​included in the lower a% and upper b% luminance ranges (a predetermined first condition described later). Note that d and p are arbitrary positive integers.

[0034] However, the range extraction unit 133 excludes the 2W pixels at both ends of the x-axis and y-axis directions of the ISAR image from the extracted range, so that the cross-correlation coefficient can be calculated using a range of size (2W+1) pixels × (2W+1) pixels. Note that W is a parameter used to make the number of pixels in the vertical and horizontal directions of the correlation window, which will be described later, odd.

[0035] The maximum number of ranges to be extracted is N. N is any positive integer, for example, 10. The range extraction unit 133 extracts ranges such that no pixels are common to any of them. Furthermore, the range extraction unit 133 calculates N coordinates (x) formed by the average values ​​of the x and y coordinates within the N extracted ranges. q , y q Extract the data (q=1, 2, ..., N) in such a way that the correlation coefficient is minimized.

[0036] Figure 3 is an explanatory diagram showing an example of an ISAR image after multiple ISAR images have been simply added together. The adder 132 generates the ISAR image shown in Figure 3 by simply adding together multiple ISAR images with short integrated times that have been reconstructed.

[0037] Next, the range extraction unit 133 extracts a range in the ISAR image shown in Figure 3 that includes pixels with high brightness values ​​and pixels with low brightness values, as described above. The rectangle in the ISAR image shown in Figure 3 represents the range extracted by the range extraction unit 133.

[0038] If there are multiple independent ranges to be extracted, the range extraction unit 133 will extract up to N ranges. The method of counting the ranges to be extracted is arbitrary. For example, the range extraction unit 133 may extract up to the Nth range, counting from the top left of the ISAR image.

[0039] Furthermore, the range extraction unit 133 may count the ranges extracted by other methods. Also, the range extraction unit 133 may extract ranges by any method that uniquely extracts N ranges.

[0040] Although the range extraction unit 133 uses an ISAR image obtained by simply adding multiple ISAR images using the addition unit 132, it may also use an ISAR image obtained by imaging long-term ISAR data input to the integral time division unit 110.

[0041] The correlation window setting unit 134 defines, for example, the ISAR image with the earliest acquisition time among multiple ISAR images input to the same position selection unit 130 as the master. In this embodiment, among multiple ISAR images, the one image that serves as the reference for alignment is called the master, and all other images are called slaves.

[0042] Next, the correlation window setting unit 134 selects the coordinates (x0, y0) of the pixel with the maximum brightness value from the range on the master corresponding to the range extracted by the range extraction unit 133.

[0043] Furthermore, if there are multiple pixels with the highest brightness value, the correlation window setting unit 134 may use x0 as the average of the x-coordinates of the multiple pixels rounded to the nearest whole number, and y0 as the average of the y-coordinates of the multiple pixels rounded to the nearest whole number. Note that the coordinates (x0, y0) calculated as described above may not be the coordinates of the pixel with the highest brightness value.

[0044] Furthermore, if there are multiple pixels with the maximum brightness value, the correlation window setting unit 134 may select the coordinate (x0, y0) that minimizes (x+y). Also, if there are multiple pixels with the maximum brightness value, the correlation window setting unit 134 may select the coordinate (x0, y0) of one of the multiple pixels. The correlation window setting unit 134 may select the coordinate (x0, y0) by any method that allows for the unique selection of the coordinate (x0, y0) from among multiple coordinates.

[0045] Figure 4 is an explanatory diagram showing an example of a master image. The correlation window setting unit 134 selects the coordinates (x0, y0) of the pixel with the maximum brightness value from the range corresponding to the range extracted by the range extraction unit 133 for the master image, which is one of the regenerated ISAR images.

[0046] The rectangle in the master shown in Figure 4 represents the range corresponding to the range extracted by the range extraction unit 133. The white circle in the master shown in Figure 4 represents the coordinates (x0, y0) selected by the correlation window setting unit 134.

[0047] Next, the correlation window setting unit 134 selects a range of (2W+1) pixels × (2W+1) pixels that includes the selected coordinates (x0, y0) of the master, and sets the selected range as the correlation window.

[0048] Furthermore, the correlation window setting unit 134 defines the set of brightness values ​​M of pixels with coordinates (x, y) included in the correlation window in the master as M = {M(x, y) | Brightness values ​​of pixels at coordinates (x, y), x = x0-W, x0-W+1, ..., x0+W, y = y0-W, y0-W+1, ..., y0+W}.

[0049] Figure 5 is an explanatory diagram showing another example of the master. The correlation window setting unit 134 sets a rectangular range as the correlation window, centered on the selected coordinates and encompassing the range corresponding to the range extracted by the range extraction unit 133. The dashed rectangle shown in Figure 5 represents the correlation window set by the correlation window setting unit 134.

[0050] The correlation test range setting unit 135 selects a range on the slave that corresponds to the correlation window in the master, with a size of (2W+1) pixels × (2W+1) pixels, and sets the selected range as the correlation test range.

[0051] Furthermore, the correlation testing range setting unit 135 sets the set of brightness values ​​S of pixels whose coordinates (x, y) are included in the correlation testing range of the i-th slave. i S i ={S i Let (x, y) be the luminance value of the pixel at coordinate (x, y), where x = x0 - W, x0 - W + 1, ..., x0 + W, and y = y0 - W, y0 - W + 1, ..., y0 + W (i = 1, 2, ..., m-1). Note that m-1 is the number of slaves, and i is the slave identifier.

[0052] The correlation testing unit 136 is M and S i +(j, k)={S i The cross-correlation coefficient of (x+j, y+k)|x=x0-W, x0-W+1, ..., x0+W, y=y0-W, y0-W+1, ..., y0+W} is calculated according to equation (1) below.

[0053]

number

[0054] In equation (1), n ​​represents the number of pixels in the correlation window. The correlation testing unit 136 calculates the cross-correlation coefficients for each of j = -W, -W+1, ..., +W and k = -W, -W+1, ..., +W. That is, the correlation testing unit 136 calculates the total cross-correlation coefficients as (2W+1). 2 Calculate the number.

[0055] Figure 6 is an explanatory diagram showing an example of a master and a slave. The correlation check range setting unit 135 sets the correlation check range for each slave, which is an ISAR image other than the regenerated master, to a range corresponding to the correlation window set in the master. The solid rectangle in the slave shown in Figure 6 represents the correlation check range set by the correlation check range setting unit 135.

[0056] The correlation testing unit 136 performs a correlation test by overlaying the center pixel of the correlation window in the master onto any one pixel within the correlation testing range in the slave. That is, the correlation testing unit 136 calculates the cross-correlation coefficient between the moved correlation window and the correlation testing range. The dashed rectangle in the slave shown in Figure 6 represents the correlation window in the master. The correlation testing unit 136 performs the correlation test on all pixels within the correlation testing range in the slave.

[0057] The correlation testing unit 136 uses the (j, k) coordinates on the i-th slave that are greater than or equal to the threshold H and give the maximum cross-correlation coefficient to determine that the position of the imaging target indicated by the coordinates (x0, y0) on the master are the same. Note that the coordinates (x0-j, y0-k) on the slave are also coordinates calculated from the position of the correlation window that gives the maximum cross-correlation coefficient.

[0058] Furthermore, the threshold H is any real number between 0 and 1, but it is preferable that it is close to 1. The correlation testing unit 136 performs the above identical position selection process over i=1, 2, ..., m-1, that is, for all slaves. Also, if there are no coordinates that give a cross-correlation coefficient of threshold H or higher, the correlation testing unit 136 stops the process of selecting identical positions.

[0059] Figure 7 is an explanatory diagram showing another example of a master and a slave. The correlation testing unit 136 selects from the slave the coordinate that gives the largest cross-correlation coefficient above a threshold among the tested cross-correlation coefficients. The correlation testing unit 136 determines that the position indicated by the selected coordinate in the slave is the same as the position indicated by the coordinate (x0, y0) in the master that is superimposed.

[0060] The white circles in the slave shown in Figure 7 represent the coordinates selected by the correlation inspection unit 136. Furthermore, the white circles in the master shown in Figure 7 and the white circles in the slave shown in Figure 7 represent the same position on the imaging target.

[0061] The same-position selection unit 130 repeatedly (i.e., N times) executes the process from selecting the coordinates of the pixel with the maximum brightness value to selecting the same position for each range extracted by the range extraction unit 133.

[0062] The alignment processing unit 140 has the function of performing alignment between multiple ISAR images using the same position selected by the same position selection unit 130. Specifically, the alignment processing unit 140 aligns the positions of the master and slave so that pixels with coordinates indicating the position of the same imaging target overlap.

[0063] For example, the alignment processing unit 140 calculates a general affine transformation formula based on the selected identical positions. The alignment processing unit 140 then performs alignment using the calculated affine transformation formula.

[0064] The polarization adjustment processing unit 150 has the function of adjusting the polarization angles of multiple ISAR images. Figure 8 is a block diagram showing an example of the configuration of the polarization adjustment processing unit 150.

[0065] The angle adjustment processing unit 150 shown in Figure 8 includes a region selection unit 151, a coefficient map generation unit 152, and an angle conversion unit 153.

[0066] The region selection unit 151 generates a histogram of the brightness values ​​of each pixel in the master. Next, the region selection unit 151 selects a region of pixels in the generated histogram that exhibit brightness values ​​greater than or equal to a threshold c% (a predetermined second condition described later).

[0067] The coefficient map generation unit 152 calculates the complex number ratio between each pixel of a slave, whose pixel value is a complex number, and the corresponding pixel of a master, for each pixel in the region selected by the region selection unit 151. Next, the coefficient map generation unit 152 normalizes the calculated complex number ratios.

[0068] For example, the pixel value of a pixel at coordinates (x, y) in the master is given by the following equation (2).

[0069]

number

[0070] Furthermore, the pixel value of the pixel at coordinate (x, y) of the slave is given by the following equation (3).

[0071]

number

[0072] The coefficient map generation unit 152 can obtain the normalized complex ratio using the following equation (4).

[0073]

number

[0074] Next, the coefficient map generation unit 152 generates a map of normalized complex ratios for the pixels in the region selected by the region selection unit 151. The coefficient map generation unit 152 stores a 1 in the map location corresponding to pixels outside the region selected by the region selection unit 151. The coefficient map generation unit 152 generates a map of normalized complex ratios for each slave.

[0075] Figure 9 is an explanatory diagram showing an example of a complex ratio map generated by the coefficient map generation unit 152. In the complex ratio map shown in Figure 9, the areas indicated by thick lines correspond to the pixel regions selected by the region selection unit 151.

[0076] As shown in Figure 9, the region indicated by the thick line stores the complex ratio obtained by equation (4). Also, as shown in Figure 9, the region outside the area indicated by the thick line stores the value 1.

[0077] The angle conversion unit 153 multiplies the slave pixel by pixel with the corresponding complex ratio map generated by the coefficient map generation unit 152. The angle conversion unit 153 multiplies the slave pixel by pixel with the complex ratio map for each slave pixel.

[0078] In other words, the polarization adjustment processing unit 150 can adjust the polarization of each pixel value for each pixel where the signal component is dominant, for multiple ISAR images showing complex numbers that have been aligned. As described above, the polarization adjustment processing unit 150 can adjust the polarization so that only the signal components reinforce each other.

[0079] The declination adjustment processing unit 150 does not necessarily have to be provided in the inverse synthetic aperture radar image synthesis device 100. If the declination adjustment processing unit 150 is not provided, the alignment processing unit 140 converts each of the multiple ISAR images (complex numbers) after alignment into real number images and inputs the converted multiple real number images to the imaging unit 160.

[0080] If the angle-shifting adjustment processing unit 150 is not provided, the overall processing load of the inverse synthetic aperture radar image synthesis device 100 is reduced. However, the degree of improvement in the signal-to-noise ratio (S / N) of the final ISAR image will be lower compared to when angle-shifting adjustment processing is performed. Nevertheless, the S / N of the final ISAR image will be higher than that of the short-time ISAR image.

[0081] Furthermore, the polarization adjustment processing unit 150 performs a process to adjust the polarization only for pixels where the signal component is considered dominant by setting a threshold, but the polarization may also be adjusted for all pixels using a general maximum ratio blending method.

[0082] When adjusting the deflection angle using the maximum ratio synthesis method, the deflection angle adjustment processing unit 150 repeatedly updates the coefficient map and investigates the S / N ratio so that the S / N ratio of the ISAR image after the multiple ISAR images have been added is maximized. In other words, although the computational cost of the deflection angle adjustment process is large, the S / N ratio of the ISAR image after the multiple ISAR images have been added is maximized.

[0083] The imaging unit 160 has the function of generating a single ISAR image by adding together multiple ISAR images whose position and declination have been aligned. In other words, the imaging unit 160 adds together a master and a slave whose position and declination have been aligned.

[0084] Next, the imaging unit 160 generates a single real-valued image by using the absolute value of the complex number of each pixel in the generated ISAR image as the pixel value. Alternatively, the imaging unit 160 may perform a process in the real-valued conversion to use a function value that increases monotonically with respect to the absolute value, such as the intensity (square of the absolute value) or the logarithm of the intensity, as the pixel value.

[0085] As described above, the real number conversion unit 131 of this embodiment converts the ISAR image into a real number image by setting the absolute value of each pixel in the ISAR image to the brightness value of each pixel. The addition unit 132 generates a single real number image by adding together multiple real number images, each obtained by converting multiple ISAR images of a predetermined size and the same object captured at different times, such that pixels with the same coordinates overlap.

[0086] Furthermore, the range extraction unit 133 of this embodiment extracts a range that satisfies a predetermined first condition from a single generated real number image. The correlation window setting unit 134 sets a correlation window in the master, which is one of the multiple ISAR images, that includes the range corresponding to the extracted range in the master. The correlation inspection range setting unit 135 sets a range in the slave, which is an ISAR image other than the master among the multiple ISAR images, that corresponds to the set correlation window in the slave, as the inspection range for the slave.

[0087] Furthermore, the correlation inspection unit 136 of this embodiment performs a process for all pixels within the inspection range by superimposing the center pixel of the set correlation window onto any one pixel within the set inspection range and calculating the cross-correlation coefficient between the correlation window and the inspection range. The correlation inspection unit 136 also determines that the position of the imaging target indicated by the coordinates of the slave pixel calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the multiple calculated cross-correlation coefficients is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master.

[0088] Furthermore, the alignment processing unit 140 of this embodiment aligns the positions of the master and the slave so that pixels with coordinates indicating the position of the same imaging target overlap. In addition, the imaging unit 160 of this embodiment adds the positions of the aligned master and the slave.

[0089] Furthermore, the region selection unit 151 of this embodiment selects a region of pixels from the master whose brightness value satisfies a predetermined second condition. The coefficient map generation unit 152 calculates a normalized ratio between the complex number represented by the master's pixels and the complex number represented by the corresponding pixels of the slave, for each pixel in the selected region.

[0090] Furthermore, the angle conversion unit 153 of this embodiment matches the angle of the master and the angle of the slave by multiplying the complex number indicated by the slave pixel corresponding to the pixel in the selected region by the corresponding calculated ratio, and then multiplying the complex number indicated by the slave pixel corresponding to the pixel other than the pixel in the selected region by 1, and performing this process for each slave pixel.

[0091] The imaging unit 160 in this embodiment may add the master and slave, whose positions and declination angles have been aligned.

[0092] Furthermore, the integration time division unit 110 of this embodiment generates multiple ISAR data by dividing the ISAR data, which is the data of the imaging target acquired by ISAR, into predetermined time intervals. The ISAR imaging unit 120 generates multiple ISAR images by imaging each of the generated ISAR data.

[0093] With the above configuration, the inverse synthetic aperture radar image synthesis apparatus 100 of this embodiment can sharpen ISAR images using multiple ISAR images.

[0094] Images captured by inverse synthetic aperture radar (ISAR) show the intensity of Doppler frequencies generated at each point of the object due to its movement. Therefore, the longer the integration time, the more likely the ISAR image is to become blurry due to changes in angular velocity and rotation axis.

[0095] Furthermore, a shorter integration time increases the likelihood of obtaining a clearer ISAR image. However, this may result in the signal of the target being imaged being buried in noise, leading to a lower signal-to-noise ratio (S / N).

[0096] The inverse synthetic aperture radar image synthesis apparatus 100 of this embodiment generates a clear ISAR image with a high signal-to-noise ratio by dividing the integration time and adding together multiple ISAR images with different acquisition times and short integration times.

[0097] When adding multiple ISAR images taken at different times, it is necessary to consider that the coordinates representing the same position of the object in the ISAR images will differ at each time point due to the object's movement, and that even at the same position, the appearance will differ. Therefore, in order to maintain image clarity, it is necessary to align the multiple ISAR images so that the same position of the object overlaps when the multiple ISAR images are added together.

[0098] Furthermore, when adding multiple ISAR images after alignment, it is necessary to match the polarization angle for each pixel to prevent signal attenuation. However, if the polarization angle is matched based on pixels where noise is displayed, the signal-to-noise ratio (S / N) will decrease.

[0099] The inverse synthetic aperture radar image synthesis device 100 of this embodiment finds a position from multiple ISAR images where the change in appearance over time is relatively small, and selects the same position of the object to be imaged. Next, based on the selected same position, the inverse synthetic aperture radar image synthesis device 100 aligns the position of the entire complex number image across the complex number images by applying a general affine transform to each complex number image in which each pixel value is represented by a complex number.

[0100] Furthermore, the inverse synthetic aperture radar image synthesizer 100 adjusts the deflection angle for each pixel in each complex number image, but only for pixels where the signal component is dominant. By adding together multiple complex number images that have undergone the above processing, the inverse synthetic aperture radar image synthesizer 100 can obtain a clear ISAR image with a high signal-to-noise ratio.

[0101] [Explanation of operation] The operation of the inverse synthetic aperture radar image synthesis device 100 of this embodiment to sharpen ISAR images will be described below with reference to Figure 10. Figure 10 is a flowchart showing the operation of the ISAR image sharpening process by the inverse synthetic aperture radar image synthesis device 100 of this embodiment.

[0102] First, long-term ISAR data is input to the integration time division unit 110 of the inverse synthetic aperture radar image synthesis device 100 (step S1100).

[0103] Next, the integration time division unit 110 generates multiple short-time ISAR data by dividing the input long-time ISAR data into predetermined time intervals (step S1200).

[0104] Next, the ISAR imaging unit 120 generates multiple ISAR images based on the multiple short-term ISAR data that have been generated (step S1300).

[0105] Next, the same-position selection unit 130 executes the same-position selection process (step S1400).

[0106] Next, the alignment processing unit 140 aligns the position of the entire image among multiple ISAR images (step S1500).

[0107] Next, the angle adjustment processing unit 150 performs the angle adjustment process (step S1600).

[0108] Next, the imaging unit 160 adds together multiple ISAR images whose position and signal component deviation angles match (step S1700).

[0109] Next, the imaging unit 160 converts the generated complex number ISAR image into an image representing a real number (step S1800). After the conversion, the inverse synthetic aperture radar image synthesizer 100 terminates the ISAR image sharpening process.

[0110] Next, the same-position selection process of step S1400, which is a sub-process constituting the ISAR image sharpening process shown in Figure 10, will be explained with reference to Figure 11. Figure 11 is a flowchart showing the operation of the same-position selection process by the same-position selection unit 130 in this embodiment.

[0111] First, the real number conversion unit 131 of the same position selection unit 130 converts multiple ISAR images into multiple real number images (step S1401).

[0112] Next, the addition unit 132 adds together pixels with the same coordinates in the multiple converted real number images (step S1402).

[0113] Next, the range extraction unit 133 extracts up to N ranges that satisfy a predetermined first condition from a single ISAR image representing the generated real numbers (step S1403). Then, the range extraction unit 133 enters a range loop (step S1404).

[0114] The correlation window setting unit 134 selects one of the extracted ranges that has not yet been aligned. Next, the correlation window setting unit 134 selects the coordinates of the pixel with the maximum brightness value from the range on the master corresponding to the extracted range (step S1405).

[0115] Next, the correlation window setting unit 134 selects a range that includes the selected coordinates of the master and sets the selected range as a correlation window on the master (step S1406). Then, the correlation window setting unit 134 enters the slave loop (step S1407).

[0116] The correlation window setting unit 134 selects one slave from among the slaves that has not yet been aligned with the selected range. Next, the correlation check range setting unit 135 sets the range on the slave that corresponds to the correlation window in the master as the correlation check range (step S1408).

[0117] Next, the correlation testing unit 136 calculates multiple cross-correlation coefficients between the correlation window and the correlation testing range in the slave while moving the correlation window in the master (step S1409).

[0118] Next, the correlation inspection unit 136 selects from the slave the same position as the position indicated by the coordinates of the pixel with the highest brightness value in the master, based on the multiple cross-correlation coefficients that have been calculated (step S1410).

[0119] The same-position selection unit 130 repeatedly executes steps S1408 to S1410 until all slaves are aligned within the selected range. When all slaves are aligned, the same-position selection unit 130 exits the slave loop (step S1411).

[0120] The same-position selection unit 130 repeatedly executes the processes in steps S1405 to S1411 until alignment is performed for all extracted ranges. When alignment is performed for all extracted ranges, the same-position selection unit 130 exits the range loop (step S1412). After exiting the range loop, the same-position selection unit 130 returns to the ISAR image sharpening process shown in Figure 10.

[0121] Next, the declination adjustment process, which is a sub-process of step S1600 that constitutes the ISAR image sharpening process shown in Figure 10, will be explained with reference to Figure 12. Figure 12 is a flowchart showing the operation of the declination adjustment process by the declination adjustment processing unit 150 of this embodiment.

[0122] First, the region selection unit 151 of the declination adjustment processing unit 150 selects a region of pixels from the master whose brightness value satisfies a predetermined second condition (step S1601).

[0123] Next, the coefficient map generation unit 152 calculates the complex number ratio between each pixel of the selected region and the corresponding pixel of the master, for each slave whose pixel value is a complex number, and normalizes it (step S1602).

[0124] Next, the coefficient map generation unit 152 generates a map of normalized complex ratios for the pixels in the selected region for each slave (step S1603). The coefficient map generation unit 152 stores a 1 in the map locations corresponding to pixels outside the selected region.

[0125] Next, the angle conversion unit 153 multiplies the slave pixel by pixel with the corresponding complex ratio map that was generated (step S1604). After multiplication, the angle adjustment processing unit 150 returns to the ISAR image sharpening process shown in Figure 10.

[0126] [Explanation of effects] ISAR images show differences in Doppler frequencies caused by differences in angular velocity at each point of the imaged object due to shaking of the object. Therefore, ISAR images tend to become blurrier as the integration time increases due to changes in angular velocity and rotation axis.

[0127] Furthermore, a shorter integration time increases the likelihood of obtaining a clearer ISAR image. However, this may result in the signal of the target being imaged being buried in noise, leading to a lower signal-to-noise ratio (S / N).

[0128] In this embodiment, the inverse synthetic aperture radar image synthesis apparatus 100 has an integration time division unit 110 that divides the integration time of an ISAR image, and an imaging unit 160 that adds up multiple ISAR images with different acquisition times and short integration times, thereby generating a clear ISAR image with a high signal-to-noise ratio.

[0129] Furthermore, due to shaking of the object being imaged, the coordinates of the same position of the object in the ISAR image differ depending on the time, and even at the same position, the appearance differs depending on the time. Therefore, when overlaying ISAR images, it is necessary to align the ISAR images so that the same position of the object is aligned.

[0130] In this embodiment, the same-position selection unit 130, the alignment processing unit 140, and the imaging unit 160 perform alignment so that the same positions of the imaging target overlap, and then superimpose multiple ISAR images with short integration times. Therefore, the inverse synthetic aperture radar image synthesis device 100 can generate a clear ISAR image with a high signal-to-noise ratio in a shorter time compared to manually selecting the same position from each ISAR image.

[0131] Furthermore, when adding the ISAR images after alignment, it is necessary to match the polarization angle for each pixel to prevent signal attenuation. However, matching the polarization angle based on pixels that show noise will reduce the signal-to-noise ratio.

[0132] In this embodiment, the declination adjustment processing unit 150 and the imaging unit 160 match the declination angles of all pixels to be superimposed when the signal component of the pixels to be superimposed is considered dominant, and then superimpose multiple ISAR images. Therefore, the inverse synthetic aperture radar image synthesis device 100 can generate an ISAR image with improved signal-to-noise ratio compared to a short-time ISAR image.

[0133] The following describes a specific example of the hardware configuration of the inverse synthetic aperture radar image synthesis apparatus 100 of this embodiment. Figure 13 is an explanatory diagram showing an example of the hardware configuration of the inverse synthetic aperture radar image synthesis apparatus 100 according to the present invention.

[0134] The inverse synthetic aperture radar image synthesis apparatus 100 shown in Figure 13 includes a CPU (Central Processing Unit) 11, a main memory unit 12, a communication unit 13, and an auxiliary memory unit 14. It also includes an input unit 15 for user operation and an output unit 16 for presenting processing results or the progress of processing to the user.

[0135] The inverse synthetic aperture radar image synthesis device 100 is implemented by software, with the CPU 11 shown in Figure 13 executing a program that provides the functions of each component.

[0136] In other words, the CPU 11 loads the program stored in the auxiliary memory 14 into the main memory 12 and executes it, thereby controlling the operation of the inverse synthetic aperture radar image synthesis device 100, and each function is realized by software.

[0137] Note that the inverse synthetic aperture radar image synthesis device 100 shown in Figure 13 may include a DSP (Digital Signal Processor) instead of a CPU 11. Alternatively, the inverse synthetic aperture radar image synthesis device 100 shown in Figure 13 may include both a CPU 11 and a DSP.

[0138] The main memory unit 12 is used as a data working area and a temporary data storage area. The main memory unit 12 is, for example, RAM (Random Access Memory).

[0139] The communication unit 13 has the function of inputting and outputting data to and from peripheral devices via a wired network or a wireless network (information and communication network).

[0140] The auxiliary storage unit 14 is a tangible storage medium that is not temporary. Examples of tangible storage media that are not temporary include magnetic disks, magneto-optical disks, CD-ROMs (Compact Disk Read Only Memory), DVD-ROMs (Digital Versatile Disk Read Only Memory), and semiconductor memory.

[0141] The input unit 15 has the function of inputting data and processing commands. The input unit 15 is, for example, an input device such as a keyboard, mouse, or touch panel.

[0142] The output unit 16 has the function of outputting data. The output unit 16 is, for example, a display device such as a liquid crystal display, a touch panel, or a printing device such as a printer.

[0143] Furthermore, as shown in Figure 13, in the inverse synthetic aperture radar image synthesis device 100, each component is connected to the system bus 17.

[0144] In the inverse synthetic aperture radar image synthesis device 100, the auxiliary storage unit 14 stores programs for implementing the integration time division unit 110, the ISAR imaging unit 120, the same position selection unit 130, the alignment processing unit 140, the declination angle adjustment processing unit 150, and the imaging unit 160.

[0145] Furthermore, the inverse synthetic aperture radar image synthesis device 100 may have a circuit implemented that includes hardware components such as an LSI (Large Scale Integration) that perform the functions shown in Figure 1.

[0146] Furthermore, the inverse synthetic aperture radar image synthesis apparatus 100 may be implemented using hardware that does not include computer functions using elements such as a CPU. For example, some or all of the components may be implemented by general-purpose circuits, dedicated circuits, processors, etc., or combinations thereof. These may be made up of a single chip (for example, the LSI described above) or by multiple chips connected via a bus. Some or all of the components may be implemented by a combination of the circuits etc. described above and a program.

[0147] Furthermore, some or all of the components of the inverse synthetic aperture radar image synthesis apparatus 100 may consist of one or more information processing devices equipped with a calculation unit and a storage unit.

[0148] If some or all of the components are implemented by multiple information processing devices or circuits, these devices may be centrally located or distributed. For example, the information processing devices or circuits may be implemented in a form where each is connected via a communication network, such as a client-server system or a cloud computing system.

[0149] Next, an overview of the present invention will be described. Figure 14 is a block diagram illustrating the overview of the inverse synthetic aperture radar image synthesis apparatus according to the present invention. The inverse synthetic aperture radar image synthesis apparatus 20 according to the present invention includes a conversion unit 21 (e.g., real number conversion unit 131) that converts an ISAR image into a real number image by setting the absolute value of each pixel of the ISAR image to the brightness value of each pixel; a generation unit 22 (e.g., addition unit 132) that generates a single real number image by adding a plurality of real number images obtained by converting a plurality of ISAR images of a predetermined size and the same imaging target, each captured at different times, so that pixels with the same coordinates overlap; an extraction unit 23 (e.g., range extraction unit 133) that extracts a range that satisfies a predetermined first condition from the single generated real number image; and a first setting unit 24 (e.g., correlation window setting unit 134) that sets a correlation window in the master that includes a range corresponding to the extracted range in the master, which is one of the plurality of ISAR images. The system includes a second setting unit 25 (e.g., a correlation inspection range setting unit 135) that sets a range corresponding to a set correlation window in a slave ISAR image other than the master among multiple ISAR images as an inspection range for the slave; an inspection unit 26 (e.g., a correlation inspection unit 136) that overlays the center pixel of the set correlation window onto any one pixel within the set inspection range and calculates the cross-correlation coefficient between the correlation window and the inspection range, performing this process for all pixels within the inspection range; and a determination unit 27 (e.g., a correlation inspection unit 136) that determines that the position of the imaging target indicated by the coordinates of the slave pixel calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the calculated multiple cross-correlation coefficients is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master.

[0150] Furthermore, the inverse synthetic aperture radar image synthesis apparatus 20 may include an alignment unit (for example, an alignment processing unit 140) that aligns the positions of the master and the slave so that pixels with coordinates indicating the position of the same imaging target overlap.

[0151] Furthermore, the inverse synthetic aperture radar image synthesis device 20 may also include an addition unit (for example, an imaging unit 160) that adds the aligned master and slave images.

[0152] With such a configuration, the inverse synthetic aperture radar image synthesis system can enhance ISAR images using multiple ISAR images.

[0153] Furthermore, the inverse synthetic aperture radar image synthesis apparatus 20 may also include a selection unit (e.g., region selection unit 151) that selects a region of pixels from the master whose brightness value satisfies a predetermined second condition; a calculation unit (e.g., coefficient map generation unit 152) that calculates a normalized ratio between the complex number shown by the master pixel and the complex number shown by the corresponding slave pixel for each pixel in the selected region; and a deflection angle matching unit (e.g., deflection angle conversion unit 153) that matches the deflection angle of the master with that of the slave by multiplying the complex number shown by the slave pixel corresponding to the pixel in the selected region by the calculated ratio, and then multiplying the complex number shown by the slave pixel corresponding to the pixel other than the pixel in the selected region by 1 for each slave pixel.

[0154] Furthermore, the addition unit may add a master and a slave whose positions and angles are aligned.

[0155] With such a configuration, the inverse synthetic aperture radar image synthesis system can further improve the signal-to-noise ratio of ISAR images.

[0156] Furthermore, the inverse synthetic aperture radar image synthesis device 20 may also include a division unit (e.g., an integral time division unit 110) that generates multiple ISAR data by dividing the ISAR data, which is the data of the imaging target acquired by ISAR, into predetermined time intervals, and an imaging unit (e.g., an ISAR imaging unit 120) that generates multiple ISAR images by imaging each of the generated multiple ISAR data.

[0157] With such a configuration, the inverse synthetic aperture radar image synthesis system can use multiple, sharper ISAR images. [Explanation of Symbols]

[0158] 11 CPU 12 Main memory 13 Communications Department 14 Auxiliary storage 15 Input section 16 Output section 17 System bus 20, 100 Inverse Synthetic Aperture Radar Image Synthesis System 21 Conversion section 22 Generation part 23 Extraction part 24. First Setting Section 25. Second Setting Section 26. Inspection Department 27 Judgment section 110 Integral time division section 120 ISAR Imaging Unit 130 Same position selection section 131 Real Number Conversion Section 132 Addition section 133 Range extraction section 134 Correlation Window Setting Section 135 Correlation test range setting unit 136 Correlation Testing Department 140 Alignment Processing Unit 150 Angle of deviation adjustment processing unit 151 Area Selection Section 152 Coefficient Map Generation Unit 153 Angle Conversion Unit 160 Image Processing Unit

Claims

1. A conversion unit that converts an ISAR (Inverse Synthetic Aperture Radar) image into a real number image by using the absolute value of each pixel of the ISAR image as the brightness value of each pixel, A generation unit generates a single real number image by adding together multiple real number images, each obtained by converting multiple ISAR images of the same object with a predetermined size and captured at different times, such that pixels with the same coordinates overlap. An extraction unit that extracts a range that satisfies a predetermined first condition from a single generated real number image, A first setting unit sets a correlation window in the master, which is one of the plurality of ISAR images, that includes a range corresponding to the extracted range in the master. A second setting unit sets the range corresponding to the set correlation window in the slave, which is an ISAR image other than the master among the plurality of ISAR images, as the inspection range for the slave. An inspection unit performs a process for all pixels within a set inspection range, overlaying the center pixel of a set correlation window onto any one pixel within the set inspection range and calculating the cross-correlation coefficient between the correlation window and the inspection range. The system includes a determination unit that determines whether the position of the imaging target indicated by the coordinates of the slave's pixels, calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the calculated multiple cross-correlation coefficients, is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master. An inverse synthetic aperture radar image synthesis device characterized by the following features.

2. It includes an alignment unit that aligns the master and slave positions so that pixels with coordinates indicating the same imaging target overlap. The inverse synthetic aperture radar image synthesis apparatus according to claim 1.

3. It includes an adder that adds the aligned master and slave. The inverse synthetic aperture radar image synthesis apparatus according to claim 2.

4. A selection unit that selects a region of pixels from a master whose brightness value satisfies a predetermined second condition, A calculation unit calculates a normalized ratio between the complex number represented by the pixel of the master and the complex number represented by the corresponding pixel of the slave, for each pixel in the selected region. The system includes an angle adjustment unit that adjusts the angle of the master and the angle of the slave by multiplying the complex number represented by the slave pixel corresponding to the pixel in the selected region by a calculated ratio, and then multiplying the complex number represented by the slave pixel corresponding to a pixel other than the pixel in the selected region by 1, for each slave pixel. The inverse synthetic aperture radar image synthesis apparatus according to claim 3.

5. The addition unit adds the master and slave, which have been aligned in terms of position and angle. The inverse synthetic aperture radar image synthesis apparatus according to claim 4.

6. A splitting unit that generates multiple ISAR data by dividing the ISAR data, which is the image target data acquired by ISAR, into predetermined time intervals, It comprises an imaging unit that generates multiple ISAR images by imaging each of the generated multiple ISAR data. An inverse synthetic aperture radar image synthesis apparatus according to any one of claims 1 to 5.

7. The ISAR image is converted into a real-valued image by taking the absolute value of each pixel in the ISAR image as the brightness value of each pixel. Multiple ISAR images, each of a predetermined size and taken at different times from the same target, are transformed into multiple real number images. These images are then added together so that pixels with the same coordinates overlap, thereby generating a single real number image. From the generated real number image, extract the range that satisfies a predetermined first condition, A correlation window is set in the master, which is one of the multiple ISAR images, that includes a range corresponding to the extracted range in the master. The range corresponding to the set correlation window in the slave, which is an ISAR image other than the master among the plurality of ISAR images, is set as the inspection range for the slave. The process of overlaying the center pixel of the set correlation window onto any one pixel within the set inspection range and calculating the cross-correlation coefficient between the correlation window and the inspection range is performed for all pixels within the inspection range. It is determined that the position of the imaging target indicated by the coordinates of the slave pixel, calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the multiple cross-correlation coefficients calculated, is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master. A method for synthesizing inverse aperture radar images, characterized by the following features.

8. Align the master and slave positions so that pixels with coordinates indicating the same target location overlap. The inverse synthetic aperture radar image synthesis method according to claim 7.

9. On the computer, A conversion process that converts an ISAR image into a real number image by using the absolute value of each pixel in the ISAR image as the brightness value of each pixel. A generation process that generates a single real number image by adding together multiple real number images, each obtained by converting multiple ISAR images of the same object with a predetermined size and captured at different times, so that pixels with the same coordinates overlap. Extraction process to extract a range that satisfies a predetermined first condition from a single generated real number image. A first setting process sets a correlation window in the master, which is one of the multiple ISAR images, that includes a range corresponding to the extracted range in the master. A second setting process sets the range corresponding to the set correlation window in the slave, which is an ISAR image other than the master among the plurality of ISAR images, as the inspection range for the slave. An inspection process that involves overlaying the center pixel of a set correlation window onto any one pixel within a set inspection range and calculating the cross-correlation coefficient between the correlation window and the inspection range, and performing this process for all pixels within the inspection range, and A determination process that determines that the position of the imaging target indicated by the coordinates of the slave's pixels, calculated from the position of the correlation window that gives the largest cross-correlation coefficient among the multiple cross-correlation coefficients calculated, is the same as the position of the imaging target indicated by the coordinates of the pixel with the largest brightness value within the range corresponding to the extracted range in the master. A program for inverse aperture radar image synthesis to perform the following operation.

10. On the computer, The system performs alignment processing to match the positions of the master and slave so that pixels with coordinates indicating the same target location overlap. The inverse aperture radar image synthesis program according to claim 9.

Images