US20200408901A1

US20200408901A1 - Radar image processing device and radar image processing method

Info

Publication number: US20200408901A1
Application number: US17/015,836
Authority: US
Inventors: Yumiko Katayama; Kei Suwa
Original assignee: Mitsubishi Electric Corp
Current assignee: Mitsubishi Electric Corp
Priority date: 2018-03-30
Filing date: 2020-09-09
Publication date: 2020-12-31
Also published as: CA3095695A1; JP6599050B1; EP3757610B1; EP3757609B1; US11474236B2; EP3757609A1; EP3757610A1; US20200408902A1; EP3757610A4; WO2019187236A1; JPWO2019189765A1; JP6698964B2; CA3092620A1; CA3095695C; WO2019187061A1; CA3092620C; WO2019189765A1; JPWO2019187236A1; EP3757609A4

Abstract

A radar image processing device includes a phase difference calculating unit for calculating a phase difference, which is the difference between the phases, with respect to the radio wave receiving points different from each other, of each of a plurality of reflected signals present in one pixel, and the rotation amount calculating unit that calculates each of the phase rotation amounts in a plurality of pixels included in the second radar image from the respective phase differences, in which the difference calculating unit rotates the phases in the plurality of pixels included in the second radar image on the basis of the respective rotation amounts, and calculates a difference between pixel values of pixels at corresponding pixel positions among the plurality of pixels included in the first radar image and the plurality of pixels obtained by the phase rotation included in the second radar image.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application is a Continuation of PCT International Application No. PCT/JP2018/036184 filed on Sep. 28, 2018, which claims priority under 35 U.S.C. § 119(a) to Patent Application No. PCT/JP2018/013795 filed in Japan on Mar. 30, 2018, all of which are hereby expressly incorporated by reference into the present application.

TECHNICAL FIELD

The present invention relates to a radar image processing device and a radar image processing method for calculating differences between pixels included in a first radar image and pixels obtained by phase rotation included in a second radar image.

BACKGROUND ART

A tall building or the like may appear as a scatterer in a radar image acquired by a radar device.
The distance from a platform on which the radar device is mounted to a high position of the scatterer is shorter than that from the platform to a low position of the scatterer by the height of the scatterer.
When the distance from a platform to a high position of a scatterer is shorter than that to a low position of the scatterer, layover, which is a phenomenon that a signal reflected at the high position of the scatterer is displaced toward the platform, occurs.
When layover occurs, a signal reflected at a high position of a scatterer is displaced and thus overlaps with another reflected signal present at the position to which the reflected signal is displaced, which may result in presence of a plurality of reflected signals in one pixel in a radar image.
Non-patent Literature 1 mentioned below teaches a radar image processing device that calculate a difference between a pixel included in a first radar image and a pixel included in a second radar image.
By calculating the difference, the radar image processing device can suppress a reflected signal with a phase difference between the phase with respect to a first radio wave receiving point and the phase with respect to a second radio wave receiving point being zero among a plurality of reflected signals present in one pixel.
The first radio wave receiving point refers to the position of a platform when a first radar image is taken, and the second radio wave receiving point refers to the position of the platform when a second radar image is taken.

CITATION LIST

Non-Patent Literature

Non-patent Literature 1: D. L. Bickel, “A null-steering viewpoint of interferometric SAR,” IGARSS2000

SUMMARY OF INVENTION

Technical Problem

The radar image processing device of the related art can suppress a reflected signal with a phase difference between the phase with respect to a first radio wave receiving point and the phase with respect to a second radio wave receiving point being zero among a plurality of reflected signals present in one pixel.
As for a reflected signal that is scattered at the same height as the position where a reflected signal that can be suppressed is scattered among a plurality of reflected signals present in one pixel, however, the phase difference between the phase with respect to a first radio wave receiving point and the phase with respect to a second radio wave receiving point is not zero.
There has thus been a problem in that a reflected signal with a phase difference between the phase with respect to a first radio wave receiving point and the phase with respect to a second radio wave receiving point not being zero cannot be suppressed.
The present invention has been made to solve such problems as described above, and an object thereof is to provide a radar image processing device and a radar image processing method capable of also suppressing a reflected signal with the difference between phases with respect to radio wave receiving points different from each other not being zero.

Solution to Problem

A radar image processing device according to the present invention includes processing circuitry performing a process of: calculating a phase shift component in a first axis direction on a two-dimensional inclined surface included in the first radar image and the second radar image, the first axis being an axis of the inclined surface inclined with respect to a ground-range direction; calculating a phase on a surface parallel to the inclined surface with respect to the inclined surface; and calculating a phase difference in each of a plurality of reflected signals present in each of pixels included in first and second radar images, the phase difference being a difference between phases with respect to the respective radio wave receiving points from a phase shift component calculated and a phase calculated; calculating each of phase rotation amounts in a plurality of pixels included in the second radar image from each phase difference calculated; and rotating phases in a plurality of pixels included in the second radar image on a basis of the respective rotation amounts calculated, and calculating a difference between pixel values of pixels being complex numbers at corresponding pixel positions among a plurality of pixels included in the first radar image and a plurality of pixels resulting from phase rotation included in the second radar image.

Advantageous Effects of Invention

According to the present invention, a radar image processing device has a configuration including: processing circuitry performing a process of: calculating a phase shift component in a first axis direction on a two-dimensional inclined surface included in the first radar image and the second radar image, the first axis being an axis of the inclined surface inclined with respect to a ground-range direction; calculating a phase on a surface parallel to the inclined surface with respect to the inclined surface; and calculating a phase difference in each of a plurality of reflected signals present in each of pixels included in first and second radar images, the phase difference being a difference between phases with respect to the respective radio wave receiving points from a phase shift component calculated and a phase calculated; calculating each of phase rotation amounts in a plurality of pixels included in the second radar image from each phase difference calculated; and rotating phases in a plurality of pixels included in the second radar image on a basis of the respective rotation amounts calculated, and calculating a difference between pixel values of pixels being complex numbers at corresponding pixel positions among a plurality of pixels included in the first radar image and a plurality of pixels resulting from phase rotation included in the second radar image. The radar image processing device according to the present invention is therefore capable of also suppressing a reflected signal from a scatterer with the difference between phases with respect to the radio wave receiving points different from each other not being zero.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a configuration diagram illustrating a radar image processing device 10 according to a first embodiment.

FIG. 2 is a configuration diagram illustrating a phase processing unit 12 of the radar image processing device 10 according to the first embodiment.

FIG. 3 is a configuration diagram illustrating an image processing unit 13 of the radar image processing device 10 according to the first embodiment.

FIG. 4 is a hardware configuration diagram illustrating hardware of each of the phase processing unit 12 and the image processing unit 13.

FIG. 5 is a hardware configuration diagram of a computer in a case where the phase processing unit 12 and the image processing unit 13 are implemented by software, firmware, or the like.

FIG. 6 is a flowchart illustrating processing of the phase processing unit 12.

FIG. 7 is an explanatory diagram illustrating an inclined surface 51, a parallel surface 52, and imaging parameters.

FIG. 8 is an explanatory diagram illustrating the relation of a spacing Δsl of pixels in a slant-range direction, the range Sw of radar images (a first radar image, a second radar image) and the distance sl from a position in the slant-range direction corresponding to the center position of the radar image to the observation area.

FIG. 9 is a flowchart illustrating processing of the image processing unit 13.

FIG. 10 is an explanatory diagram illustrating suppression of reflected signals present in one pixel in a case where the phases in pixels included in the second radar image are not rotated by a phase rotating unit 33.

FIG. 11 is an explanatory diagram illustrating suppression of reflected signals present in one pixel in a case where the phases in pixels included in the second radar image are rotated by the phase rotating unit 33.

FIG. 12 is a configuration diagram illustrating an image processing unit 13 of a radar image processing device 10 according to a second embodiment.

FIG. 13 is a hardware configuration diagram illustrating hardware of each of a phase processing unit 12 and the image processing unit 13.

FIG. 14 is an explanatory diagram illustrating a plurality of reflected signals present in one pixel in a case where only two radar images are included in a radar image group 2.

FIG. 15 is an explanatory diagram illustrating a plurality of reflected signals present in one pixel in a case where two or more radar images are included in a radar image group 2.

FIG. 16 is a configuration diagram illustrating an image processing unit 13 of a radar image processing device 10 according to a third embodiment.

FIG. 17 is a hardware configuration diagram illustrating hardware of each of a phase processing unit 12 and the image processing unit 13.

FIG. 18 is a configuration diagram illustrating an image processing unit 13 of a radar image processing device 10 according to a fourth embodiment.

FIG. 19 is a hardware configuration diagram illustrating hardware of each of a phase processing unit 12 and the image processing unit 13.

FIG. 20 is a configuration diagram illustrating an image processing unit 13 of a radar image processing device 10 according to a fifth embodiment.

FIG. 21 is a hardware configuration diagram illustrating hardware of each of a phase processing unit 12 and the image processing unit 13.

DESCRIPTION OF EMBODIMENTS

Embodiments for carrying out the invention will now be described with reference to the accompanying drawings for more detailed explanation of the invention.

First Embodiment

FIG. 1 is a configuration diagram illustrating a radar image processing device 10 according to a first embodiment.
In FIG. 1, a radar 1 is a synthetic aperture radar (SAR), a real aperture radar, or the like, and is mounted on a platform for observing the Earth, etc. The radar 1 takes a radar image, and acquires parameters when taking the radar image. The platform can be a satellite, an aircraft, or the like.
The radar 1 images an observation area from a radio wave receiving point, and then images the observation area again when the platform is at a radio wave receiving point near the aforementioned radio wave receiving point.
In a case of repeat-pass imaging, when the platform is a satellite, the radar 1 images an observation area from a radio wave receiving point, the platform then orbits the Earth, and the radar 1 images the same observation area again to acquire a radar image when the platform has returned to a radio wave receiving point near the aforementioned radio wave receiving point. When the platform is an aircraft, the platform is flown to repeatedly pass the same path, and the radar 1 images one observation area when the platform is at substantially the same radio wave receiving points to acquire radar images.
In a case of single-pass imaging, a plurality of radars 1 are mounted on one platform, and the plurality of radars 1 image one observation area from a radio wave receiving point to acquire radar images. In this case, the plurality of the radars 1 are installed at different positions on the platform.
In addition, a plurality of radars 1 having equal imaging parameters such as a wavelength are mounted on different platforms from each other, and each of the plurality of radars 1 image one observation area from a radio wave receiving point to acquire radar images.
Thus, the radars 1 image the same observation area twice from the respective radio wave receiving points, which are different from each other, to acquire two radar images; a first radar image and a second radar image.
Hereinafter, the position of the platform when the first radar image is taken will be referred to as a first radio wave receiving point, and the position of the platform when the second radar image is taken will be referred to as a second radio wave receiving point.
The first radar image and the second radar image have an equal resolution. Thus, the pixel positions of a plurality of pixels included in a first radar image and those of a plurality of pixels included in a second radar image are expressed in the same manner by (pixel,line).
“pixel” is a variable representing the position of a pixel in a slant-range direction in each of a first radar image and a second radar image, and “line” is a variable representing the position of a pixel in an azimuth direction in each of a first radar image and a second radar image.
The radar 1 transmits a radar image group 2 including a first radar image and a second radar image to the radar image processing device 10.
The radar 1 transmits an imaging parameter group 3 including a first imaging parameter associated with the first radar image and a second imaging parameter associated with the second radar image to the radar image processing device 10.
The radar image group 2 is an image group including a first radar image and a second radar image.
The types of polarization used in imaging a first radar image and in imaging a second radar image are not limited, and each of a first radar image and a second radar image may thus be any of a single-polarization radar image, a dual-polarization radar image, and a quad-polarization radar image.
Each of a first radar image and a second radar image is a radar image showing intensity distribution of radio waves emitted by the radar 1, then reflected by an observation area, and received by the radar 1.
A plurality of pixels included in a first radar image and a plurality of pixels included in a second radar image each have a complex pixel value.
A complex pixel value includes information indicating the distance between the radar 1 and a scatterer present in the observation area, and also information indicating phase shift occurring when a radio wave emitted by the radar 1 is reflected by a scatterer. Hereinafter, a “pixel value” has a value of a complex number unless otherwise noted.
An imaging parameter group 3 is a parameter group including a first imaging parameter and a second imaging parameter.
The first imaging parameter includes position information on the orbit of the platform and sensor information when a first radar image is taken by the radar 1.
The second imaging parameter includes position information on the orbit of the platform and sensor information when a second radar image is taken by the radar 1.
The position information on the orbit is information indicating the latitude, the longitude, and the altitude of the platform when a first radar image or a second radar image is taken by the radar 1. Thus, the position information on the orbit is used as information indicating a first radio wave receiving point or a second radio wave receiving point.
The sensor information includes information indicating an off-nadir angle θ of the radar 1 when a first radar image or a second radar image is taken, information indicating a wavelength a, of a radio wave emitted from the radar 1, and information indicating an average R of distances from the radar 1 to an observation area.
The radar image processing device 10 includes a radar image acquiring unit 11, a phase processing unit 12, and an image processing unit 13.
The radar image acquiring unit 11 acquires each of a radar image group 2 and an imaging parameter group 3 transmitted from the radar 1.
The radar image acquiring unit 11 outputs the radar image group 2 to the image processing unit 13, and outputs the imaging parameter group 3 to the phase processing unit 12.
The phase processing unit 12 acquires the imaging parameter group 3 output from the radar image acquiring unit 11, and the inclination angle α of a two-dimensional inclined surface 51 with respect to a ground-range direction (see FIG. 7).
The phase processing unit 12 also acquires the distance α between the inclined surface 51 and a parallel surface 52 that is a surface parallel to the inclined surface 51 (see FIG. 7).
Details of the inclined surface 51 and the parallel surface 52 will be described later.
The phase processing unit 12 performs a process of calculating a phase shift component φ(x) in an x-axis (first axis) direction on the inclined surface 51 by using the first imaging parameter, the second imaging parameter, and the inclination angle α.
The phase processing unit 12 performs a process of calculating a phase φ(z₀) on the parallel surface 52 with respect to the inclined surface 51 by using the first imaging parameter, the second imaging parameter, the inclination angle α, and the distance z₀.
The phase processing unit 12 performs a process of calculating, in each of a plurality of reflected signals present in each of pixels included in the first and second radar images, a phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point and the phase with respect to the second radio wave receiving point.
The image processing unit 13 acquires the radar image group 2 output from the radar image acquiring unit 11, and each phase difference Δφ(x,z₀) output from the phase processing unit 12.
The image processing unit 13 performs a process of calculating each of phase rotation amounts exp[j·Δφ(x,z₀)] in a plurality of pixels included in the second radar image from each phase difference Δφ(x,z₀) output from the phase processing unit 12.
The image processing unit 13 performs a process of rotating the phases in the plurality of pixels included in the second radar image on the basis of the respective calculated rotation amounts exp[j·Δφ(x,z₀)].
The image processing unit 13 performs a process of calculating a difference between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the first radar image and among a plurality of pixels obtained by phase rotation included in the second radar image.
FIG. 2 is a configuration diagram illustrating the phase processing unit 12 of the radar image processing device 10 according to the first embodiment.
FIG. 3 is a configuration diagram illustrating the image processing unit 13 of the radar image processing device 10 according to the first embodiment.
FIG. 4 is a hardware configuration diagram illustrating hardware of each of the phase processing unit 12 and the image processing unit 13.
In FIG. 2, a phase shift component calculating unit 21 is implemented by a phase shift component calculating circuit 41 illustrated in FIG. 4, for example.
The phase shift component calculating unit 21 acquires the imaging parameter group 3 output from the radar image acquiring unit 11, and the inclination angle α.
The phase shift component calculating unit 21 performs the process of calculating the phase shift component φ(x) in the x-axis direction on the inclined surface 51 by using the first imaging parameter, the second imaging parameter, and the inclination angle α.
The phase shift component calculating unit 21 outputs the phase shift component φ(x) in the x-axis direction to a phase difference calculating unit 23.
A phase calculating unit 22 is implemented by a phase calculating circuit 42 illustrated in FIG. 4, for example.
The phase calculating unit 22 acquires the imaging parameter group 3 output from the radar image acquiring unit 11, the inclination angle α, and the distance z₀.
The phase calculating unit 22 performs the process of calculating the phase φ(z₀) on the parallel surface 52 with respect to the inclined surface 51 by using the first imaging parameter, the second imaging parameter, the inclination angle α, and the distance z₀.
The phase calculating unit 22 outputs the phase φ(z₀) to the phase difference calculating unit 23.
The phase difference calculating unit 23 is implemented by a phase difference calculating circuit 43 illustrated in FIG. 4, for example.
The phase difference calculating unit 23 performs the process of calculating, in each of a plurality of reflected signals present in each of pixels included in the first and second radar images, the phase difference Δφ(x,z₀) from the phase shift component φ(x) and the phase φ(z₀).
The phase difference Δφ(x,z₀) is a phase difference in each of the reflected signals, between the phase of the reflected signal with respect to the first radio wave receiving point and the phase of the reflected signal with respect to the second radio wave receiving point.
The phase difference calculating unit 23 outputs each phase difference Δφ(x,z₀) to the image processing unit 13.
In FIG. 3, a rotation amount calculating unit 31 is implemented by a rotation amount calculating circuit 44 illustrated in FIG. 4, for example.
The rotation amount calculating unit 31 performs the process of calculating each of phase rotation amounts exp[j·Δφ(x,z₀)] in a plurality of pixels included in the second radar image from each phase difference Δφ(x,z₀) output from the phase difference calculating unit 23.
The rotation amount calculating unit 31 outputs each rotation amount exp[j·Δφ(x,z₀)] to a phase rotating unit 33.
A difference calculating unit 32 includes the phase rotating unit 33 and a difference calculation processing unit 34.
The phase rotating unit 33 is implemented by a phase rotating circuit 45 illustrated in FIG. 4, for example.
The phase rotating unit 33 acquires the second radar image from the radar image group 2 output from the radar image acquiring unit 11.
The phase rotating unit 33 performs the process rotating the phases in the plurality of pixels included in the second radar image on the basis of the respective rotation amounts exp[j·Δφ(x,z₀)] output from the rotation amount calculating unit 31.
The phase rotating unit 33 outputs a second radar image including a plurality of pixels obtained by phase rotation to the difference calculation processing unit 34.
The difference calculation processing unit 34 is implemented by a difference calculation processing circuit 46 illustrated in FIG. 4, for example.
The difference calculation processing unit 34 acquires the first radar image from the radar image group 2 output from the radar image acquiring unit 11, and acquires the second radar image output from the phase rotating unit 33.
The difference calculation processing unit 34 performs the process of calculating a difference ΔS(pixel,line) between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the first radar image and among a plurality of pixels obtained by phase rotation included in the second radar image.
The difference ΔS(pixel,line) corresponds to a pixel of a suppressed image in which unnecessary reflected signals from the scatterer are suppressed.
The difference calculation processing unit 34 outputs the suppressed image including the respective differences Δs (pixel, line) to the outside of the unit.
In FIG. 2, it is assumed that each of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23, which are components of the phase processing unit 12, is implemented by such dedicated hardware as illustrated in FIG. 4.
In addition, in FIG. 3, it is assumed that each of the rotation amount calculating unit 31, the phase rotating unit 33, and the difference calculation processing unit 34, which are components of the image processing unit 13, is implemented by such dedicated hardware as illustrated in FIG. 4.
Specifically, the phase processing unit 12 and the image processing unit 13 are assumed to be implemented by the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, and the difference calculation processing circuit 46.
Note that each of the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, and the difference calculation processing circuit 46 may be a single circuit, a composite circuit, a programmed processor, a parallel-programmed processor, an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination thereof, for example.
The components of the phase processing unit 12 and the components of the image processing unit 13 are not limited to those implemented by dedicated hardware. The phase processing unit 12 and the image processing unit 13 may be implemented by software, firmware, or a combination of software and firmware.
The software or firmware is stored in a memory of a computer in the form of programs. The computer refers to hardware for executing programs, and may be a central processing unit (CPU), a central processor, a processing unit, a computing unit, a microprocessor, a microcomputer, a processor, or a digital signal processor (DSP), for example.
FIG. 5 is a hardware configuration diagram of a computer in a case where the phase processing unit 12 and the image processing unit 13 are implemented by software, firmware, or the like.
In the case where the phase processing unit 12 is implemented by software, firmware, or the like, programs for causing a computer to perform procedures of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23 are stored in a memory 61.
In addition, in the case where the image processing unit 13 is implemented by software, firmware, or the like, programs for causing a computer to perform procedures of the rotation amount calculating unit 31, the phase rotating unit 33, and the difference calculation processing unit 34 are stored in the memory 61.
A processor 62 of the computer thus executes the programs stored in the memory 61.
In addition, FIG. 4 illustrates an example in which each of the components of the phase processing unit 12 and the components of the image processing unit 13 is implemented by dedicated hardware, and FIG. 5 illustrates an example in which the phase processing unit 12 and the image processing unit 13 are implemented by software, firmware, or the like.
The implementations are not limited to the above, and some components of the phase processing unit 12 and some components of the image processing unit 13 may be implemented by dedicated hardware and others may be implemented by software, firmware, and the like, for example.
Next, the operation of the radar image processing device 10 illustrated in FIG. 1 will be explained.
The radar 1 transmits a radar image group 2 including a first radar image and a second radar image, and an imaging parameter group 3 including a first imaging parameter and a second imaging parameter to the radar image processing device 10.
The radar image acquiring unit 11 acquires each of the radar image group 2 and the imaging parameter group 3 transmitted from the radar 1.
The radar image acquiring unit 11 outputs the radar image group 2 to the image processing unit 13, and outputs the imaging parameter group 3 to the phase processing unit 12.
The pixel values of the pixels included in radar images (the first radar image, the second radar image) are complex numbers that are expressed as in the following formula (1).
S(pixel,line)=Av(pixel,line)exp[jΨ(pixel,line)] (1)
In the formula (1), Av(pixel,line) represents the amplitude of a pixel at a pixel position (pixel,line).
Ψ(pixel,line) represents the phase (argument) of a pixel at a pixel position (pixel,line).
j is a symbol representing an imaginary unit.
The phase processing unit 12 performs a process of calculating a phase difference Δφ(x,z₀).
FIG. 6 is a flowchart illustrating the processing of the phase processing unit 12.
The processing of the phase processing unit 12 will now be explained in detail with reference to FIG. 6.
The phase shift component calculating unit 21 acquires the imaging parameter group 3 output from the radar image acquiring unit 11, and the inclination angle α (step ST1 in FIG. 6).
The phase calculating unit 22 acquires the imaging parameter group 3 output from the radar image acquiring unit 11, the inclination angle α, and the distance z₀(step ST2 in FIG. 6).
The inclination angle α is a parameter set in advance by a user, and expressed as in FIG. 7, for example.
The distance z₀is a parameter set in advance by a user, and expressed as in FIG. 7, for example.
Each of the inclination angle α and the distance z₀may be provided to the phase calculating unit 22 by manual operation made by a user, or may be provided to the phase calculating unit 22 from an external device, which is not illustrated, for example.
FIG. 7 is an explanatory diagram illustrating the inclined surface 51, the parallel surface 52, and the imaging parameters.
In FIG. 7, the inclined surface 51 is a two-dimensional surface included in common in the first radar image and the second radar image.
The direction of the x axis, which is the first axis, of the inclined surface 51 is a direction inclined by the inclination angle α with respect to the ground-range direction, and the direction of a second axis of the inclined surface 51 is the azimuth direction (the depth direction from the sheet surface of FIG. 7).
The parallel surface 52 is a surface parallel to the inclined surface 51 and at a distance of z₀from the inclined surface 51.
In a case where the inclined surface 51 is a flat roof of a building built vertically on a horizontal ground surface, for example, the inclination angle α is set to 0 degrees.
In a case where the inclined surface 51 is a wall surface of a building built vertically on a horizontal ground surface, for example, inclination angle α is set to 90 degrees.
P₁represents the first radio wave receiving point, and P₂represents the second radio wave receiving point.
The first radio wave receiving point P₁is a center position on the orbit of the platform when the first radar image is taken, and the second radio wave receiving point P₂is a center position on the orbit of the platform when the second radar image is taken.
B_1,2represents a distance component, in a direction perpendicular to the direction (hereinafter referred to as a “slant-range direction”) of a radio wave emitted by the radar 1, of the distance between the first radio wave receiving point P₁and the second radio wave receiving point P₂.
θ is an off-nadir angle, which is an angle between a vertically downward direction from the platform and the slant-range direction.
R represents an average of the distance between the first radio wave receiving point P₁and the observation area and the distance between the second radio wave receiving point P₂and the observation area.
The distance component B_1,2, the off-nadir angle θ, and the average R of the distances are information included in the imaging parameters.
Sw represents a range of the first radar image and a range of the second radar image that capture an observation object.
The range Sw of the first radar image and the range Sw of the second radar image are the same to each other.
Herein, because the distance between the first radio wave receiving point P₁and the observation area and the distance between the second radio wave receiving point P₂and the observation area are long, the phase shift component calculating unit 21 assume that each of the off-nadir angle θ and the average R of the distances does not change.
Specifically, the off-nadir angle θ included in the first imaging parameter and the off-nadir angle θ included in the second imaging parameter are the same value.
In addition, the average R of the distances included in the first imaging parameter and the average R of the distances included in the second imaging parameter are the same value.
In addition, a pixel at a pixel position (pixel,line) among a plurality of pixels included in the first radar image and a pixel at a pixel position (pixel,line) among a plurality of pixels included in the second radar image are pixels at the same pixel position.
FIG. 8 is an explanatory diagram illustrating the relation of a spacing Δsl of pixels in the slant-range direction, the range Sw of radar images (the first radar image, the second radar image) and the distance sl from a position in the slant-range direction corresponding to the center position of the radar image to the observation area.
In FIG. 8, the distance from a position in the slant-range direction corresponding to a near range of the radar image to a position in the slant-range direction corresponding to the center position of the radar image is (Sw/2)·sin 0.
Thus, the distance sl is expressed as in the following formula (2).
$\begin{matrix} s l = Δ sl \times pixel - \frac{S w}{2} \sin θ & (2) \end{matrix}$
Each of the spacing Δsl and the range Sw of the radar images is information included in the imaging parameters.
In addition, the relation between a position x in the x-axis direction on the inclined surface 51 and the distance sl based on the center position of the radar image is expressed as in the following formula (3).
sl=x sin(θ−α) (3)
The following formula (4) is satisfied on the basis of the formula (2) and the formula (3).
$\begin{matrix} x \sin (θ - α) = Δ sl \times pixel - \frac{Sw}{2} \sin θ & (4) \end{matrix}$
The phase shift component calculating unit 21 calculates the position x on the inclined surface 51 corresponding to a pixel position “pixel” in the slant-range direction in the radar image by substituting the position “pixel” into the formula (4).
A plurality of reflected signals from scatterers are present in the pixel at the position “pixel” substituted into the formula (4).
The position “pixel” substituted into the formula (4) may be provided to the phase shift component calculating unit 21 by manual operation made by a user, or may be provided to the phase shift component calculating unit 21 from an external device, which is not illustrated, for example.
The phase shift component calculating unit 21 calculates the phase shift component φ(x) at the position x in the x-axis direction on the inclined surface 51 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, and an observation path parameter p (step ST3 in FIG. 6).
The observation path parameter p is a parameter indicating whether the observation path when the radar image is taken is repeat pass or single pass, which is p=2 in the case of repeat pass or p=1 in the case of single pass. The observation path parameter p may be provided to the phase shift component calculating unit 21 and the phase calculating unit 22 by manual operation made by a user, or may be provided to the phase shift component calculating unit 21 and the phase calculating unit 22 from an external device, which is not illustrated, for example.
The following formula (5) is a formula for calculating the phase shift component φ(x) used by the phase shift component calculating unit 21.
$\begin{matrix} φ (x) = (\frac{2 p π B_{1, 2} \cos (θ - α)}{λ R}) x & (5) \end{matrix}$
The phase shift component calculating unit 21 outputs the phase shift component φ(x) in the x-axis direction to the phase difference calculating unit 23.
The phase calculating unit 22 calculates the phase φ(z₀) on the parallel surface 52 with respect to the inclined surface 51 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, the distance z₀, and the observation path parameter p (step ST4 in FIG. 6).
The following formula (6) is a formula for calculating the phase φ(z₀) used by the phase calculating unit 22.
$\begin{matrix} ρ (z_{0}) = (\frac{2 p π B_{1, 2}}{λ R \sin (θ - α)}) z_{0} & (6) \end{matrix}$
The phase calculating unit 22 outputs the phase φ(z₀) to the phase difference calculating unit 23.
The phase difference calculating unit 23 calculates, in each of a plurality of reflected signals present in each of pixels included in the first and second radar images, the phase difference Δφ(x,z₀) by using the phase shift component φ(x) and the phase φ(z₀) (step ST5 in FIG. 6).
The phase difference Δφ(x,z₀) is a phase difference, in each of the reflected signals, between the phase of the reflected signal with respect to the first radio wave receiving point P₁and the phase of the reflected signal with respect to the second radio wave receiving point P₂.
The following formula (7) is a formula for calculating the phase difference Δφ(x,z₀) used by the phase difference calculating unit 23.
Δϕ(x,z ₀)=ϕ(x)+ρ(z ₀) (7)
The phase difference calculating unit 23 outputs each phase difference Δφ(x,z₀) to the image processing unit 13.
The image processing unit 13 performs a process of acquiring a suppressed image.
FIG. 9 is a flowchart illustrating the processing of the image processing unit 13.
The processing of the image processing unit 13 will now be explained in detail with reference to FIG. 9.
The rotation amount calculating unit 31 acquires each phase difference Δφ(x,z₀) output from the phase difference calculating unit 23.
The rotation amount calculating unit 31 calculates each of phase rotation amounts exp[j·Δφ(x,z₀)] in a plurality of pixels included in the second radar image from each phase difference Δφ(x,z₀) (step ST11 in FIG. 9).
The rotation amount calculating unit 31 outputs each rotation amount exp[j·Δφ(x,z₀)] to the phase rotating unit 33.
The phase rotating unit 33 acquires the second radar image from the radar image group 2 output from the radar image acquiring unit 11.
The phase rotating unit 33 performs the process rotating the phases in the plurality of pixels included in the second radar image on the basis of the respective rotation amounts exp[j·Δφ(x,z₀)] output from the rotation amount calculating unit 31 (step ST12 in FIG. 9).
The following formula (8) is a formula representing the process of rotating a phase performed by the phase rotating unit 33.
S ₂′(pixel,line)=S ₂(pixel,line)exp[jΔ(x,z ₀)] (8)
In the formula (8), S₂(pixel,line) represents the pixel value of a pixel included in the second radar image output from the radar image acquiring unit 11, and S₂′(pixel,line) represents the pixel value of a pixel included in the second radar image obtained by rotation of the phase in the pixel by the phase rotating unit 33.
The phase rotating unit 33 outputs a second radar image including a plurality of pixels obtained by phase rotation to the difference calculation processing unit 34.
The difference calculation processing unit 34 acquires the first radar image from the radar image group 2 output from the radar image acquiring unit 11, and acquires the second radar image including a plurality of pixels obtained by the phase rotation and output from the phase rotating unit 33.
The difference calculation processing unit 34 calculates a difference ΔS(pixel,line) between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the first radar image and among a plurality of pixels obtained by phase rotation included in the second radar image (step ST13 in FIG. 9).
The following formula (9) is a formula for calculating the difference ΔS(pixel,line) used by the difference calculation processing unit 34.
ΔS(pixel,line)=S ₁(pixel,line)−S ₂′(pixel,line) (9)
In the formula (9), S₁(pixel,line) represents the pixel value of a pixel included in the first radar image.
The difference calculation processing unit 34 outputs the suppressed image including the respective differences Δs (pixel, line) to the outside of the unit.
Here, FIG. 10 is an explanatory diagram illustrating suppression of reflected signals present in one pixel in a case where the phases in the pixels included in the second radar image are not rotated by the phase rotating unit 33.
In FIG. 10, regarding a reflected signal assigned with “1”, the distance from the scatterer that scatters the reflected signal to the first radio wave receiving point P₁and the distance from the scatterer that scatters the reflected signal to the second radio wave receiving point P₂are equal to each other. Thus, regarding the reflected signal assigned with “1”, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is zero.
Because the difference ΔS(pixel,line) for the reflected signal assigned with “1” is thus zero, the reflected signal assigned with “1” is suppressed.
Regarding a reflected signal assigned with “2”, the distance from the scatterer that scatters the reflected signal to the first radio wave receiving point P₁and the distance from the scatterer that scatters the reflected signal to the second radio wave receiving point P₂are not equal to each other. Thus, regarding the reflected signal assigned with “2”, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is other than zero.
Because the difference ΔS(pixel,line) for the reflected signal assigned with “2” is thus other than zero, the reflected signal assigned with “2” is not suppressed.
Regarding a reflected signal assigned with “3” as well, the distance from the scatterer that scatters the reflected signal to the first radio wave receiving point P₁and the distance from the scatterer that scatters the reflected signal to the second radio wave receiving point P₂are not equal to each other. Thus, regarding the reflected signal assigned with “3”, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is other than zero.
Because the difference ΔS(pixel,line) for the reflected signal assigned with “3” is thus other than zero, the reflected signal assigned with “3” is not suppressed.
FIG. 11 is an explanatory diagram illustrating suppression of reflected signals present in one pixel in a case where the phases in pixels included in the second radar image are rotated by the phase rotating unit 33.
Regarding a reflected signal assigned with “1”, as illustrated in FIG. 10, the distance from the scatterer that scatters the reflected signal to the first radio wave receiving point P₁and the distance from the scatterer that scatters the reflected signal to the second radio wave receiving point P₂are equal to each other. Thus, regarding the reflected signal assigned with “1”, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is zero, and the phase rotation amount exp[j·Δφ(x,z₀)] calculated by the rotation amount calculating unit 31 is zero.
Regarding the reflected signal assigned with “1”, because the phase rotation amount exp[j·Δφ(x,z₀)] is zero, the phase is not rotated by the phase rotating unit 33 as illustrated in FIGS. 10 and 11. Thus, because the phase difference Δφ(x,z₀) is still zero for the reflected signal assigned with “1”, the difference ΔS(pixel,line) is zero, and the reflected signal assigned with “1” is thus suppressed.
Regarding a reflected signal assigned with “2”, as illustrated in FIG. 10, the distance from the scatterer that scatters the reflected signal to the first radio wave receiving point P₁and the distance from the scatterer that scatters the reflected signal to the second radio wave receiving point P₂are not equal to each other. Thus, regarding the reflected signal assigned with “2”, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is other than zero. Thus, the phase rotation amount exp[j·Δφ(x,z₀)] calculated by the rotation amount calculating unit 31 is other than zero.
Regarding the reflected signal assigned with “2” resulting from phase rotation, as illustrated in FIG. 11, the distance to the first radio wave receiving point P₁and the distance to the second radio wave receiving point P₂are not equal to each other even after the rotation by the rotation amount exp[j·Δφ(x,z₀)] by the phase rotating unit 33. Thus, regarding the reflected signal assigned with “2” resulting from the phase rotation, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is other than zero.
Because the difference ΔS(pixel,line) for the reflected signal assigned with “2” resulting from the phase rotation is thus other than zero, the reflected signal assigned with “2” resulting from the phase rotation is not suppressed.
Regarding a reflected signal assigned with “3”, as illustrated in FIG. 10, the distance from the scatterer that scatters the reflected signal to the first radio wave receiving point P₁and the distance from the scatterer that scatters the reflected signal to the second radio wave receiving point P₂are not equal to each other. Thus, regarding the reflected signal assigned with “3”, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is other than zero. Thus, the phase rotation amount exp[j·Δφ(x,z₀)] calculated by the rotation amount calculating unit 31 is other than zero.
Regarding the reflected signal assigned with “3”, as illustrated in FIG. 11, the distance to the first radio wave receiving point P₁and the distance to the second radio wave receiving point P₂have become equal to each other as a result of the rotation by the rotation amount exp[j·Δφ(x,z₀)] by the phase rotating unit 33. Thus, regarding the reflected signal assigned with “3” resulting from the phase rotation, the phase difference Δφ(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂is zero.
Because the difference ΔS(pixel,line) for the reflected signal assigned with “3” resulting from the phase rotation is thus zero, the reflected signal assigned with “3” resulting from the phase rotation is suppressed.
In the first embodiment described above, the radar image processing device 10 has a configuration including the phase difference calculating unit 23 that calculates a phase difference, which is the difference between the phases with respect to the radio wave receiving points different from each other, of each of a plurality of reflected signals present in one pixel, and the rotation amount calculating unit 31 that calculates each of the phase rotation amounts in a plurality of pixels included in the second radar image from the respective phase differences, in which the difference calculating unit 32 rotates the phases in the plurality of pixels included in the second radar image on the basis of the respective rotation amounts, and calculates a difference between pixel values of pixels at corresponding pixel positions among the plurality of pixels included in the first radar image and the plurality of pixels obtained by the phase rotation included in the second radar image. The radar image processing device 10 is therefore capable of also suppressing a reflected signal with the difference between phases with respect to the radio wave receiving points different from each other not being zero.

Second Embodiment

The first embodiment presents an example in which the radar image processing device 10 acquires a radar image group 2 including a first radar image and a second radar image, and outputs a suppressed image.
In a second embodiment, a radar image processing device 10 that acquires a radar image group 2 including two or more radar images capturing the same observation area taken from radio wave receiving points different from each other, and outputs a suppressed image will be described.
In the radar image processing device 10 of the second embodiment, the phase processing unit 12 and the image processing unit 13 perform processes on each combination of two radar images included in the radar image group 2. In this case, one radar image included in each combination will be referred to as a first radar image, and the other radar image included in the combination will be referred to as a second radar image.
Specifically, the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23 repeat the process of calculating the phase difference Δφ_i(x,z₀) until the process of calculating the phase difference Δφ_i(x,z₀) is completed for all of the combinations i of two radar images. Symbol i is a variable representing a combination of two radar images.
The rotation amount calculating unit 31, the phase rotating unit 33, and the difference calculation processing unit 34 repeat the process of calculating the difference ΔS_i(pixel,line) until the process of calculating the difference ΔS_i(pixel,line) is completed for all of the combinations i of two radar images.
The radar image processing device 10 in the second embodiment has a configuration as illustrated in FIG. 1, that is similar to the radar image processing device 10 of the first embodiment.
The phase processing unit 12 in the second embodiment has a configuration as illustrated in FIG. 2, that is similar to the phase processing unit 12 of the first embodiment.
Note that the radar image group 2 includes two or more radar images, and the imaging parameter group 3 includes two or more imaging parameters.
FIG. 12 is a configuration diagram illustrating an image processing unit 13 of the radar image processing device 10 according to the second embodiment.
FIG. 13 is a hardware configuration diagram illustrating hardware of each of the phase processing unit 12 and the image processing unit 13.
In FIGS. 12 and 13, reference numerals that are the same as those in FIGS. 3 and 4 represent the same or corresponding components, and the description thereof will thus not be repeated.
An image combining unit 35 is implemented by an image combining circuit 47 illustrated in FIG. 13, for example.
The image combining unit 35 acquires a weight parameter w_iused for generation of a suppressed image.
The image combining unit 35 performs a process of combining differences ΔS_i(pixel,line) at corresponding pixel positions among the respective differences calculated for the respective combinations i by the difference calculation processing unit 34 by using the weight parameter w_i.
The image combining unit 35 outputs a suppressed image including the respective differences S_sup(pixel, line) resulting from the combining to the outside of the unit.
In FIG. 2, it is assumed that each of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23, which are components of the phase processing unit 12, is implemented by such dedicated hardware as illustrated in FIG. 13.
In addition, in FIG. 12, it is assumed that each of the rotation amount calculating unit 31, the phase rotating unit 33, the difference calculation processing unit 34, and the image combining unit 35, which are components of the image processing unit 13, is implemented by such dedicated hardware as illustrated in FIG. 13.
Specifically, the phase processing unit 12 and the image processing unit 13 are assumed to be implemented by the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, the difference calculation processing circuit 46, and the image combining circuit 47.
Note that each of the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, the difference calculation processing circuit 46, and the image combining circuit 47 may be a single circuit, a composite circuit, a programmed processor, a parallel-programmed processor, an ASIC, an FPGA, or a combination thereof, for example.
The components of the phase processing unit 12 and the components of the image processing unit 13 are not limited to those implemented by dedicated hardware. The phase processing unit 12 and the image processing unit 13 may be implemented by software, firmware, or a combination of software and firmware.
In the case where the phase processing unit 12 is implemented by software, firmware, or the like, programs for causing a computer to perform procedures of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23 are stored in a memory 61 illustrated in FIG. 5.
In addition, in the case where the image processing unit 13 is implemented by software, firmware, or the like, programs for causing a computer to perform procedures of the rotation amount calculating unit 31, the phase rotating unit 33, the difference calculation processing unit 34, and the image combining unit 35 are stored in the memory 61.
A processor 62 of the computer thus executes the programs stored in the memory 61.
Next, the operation of the radar image processing device 10 will be explained.
The phase processing unit 12 performs a process of calculating the phase difference Δφ_i(x,z₀) for each combination i of two radar images among the two or more radar images included in the radar image group 2.
The phase shift component calculating unit 21 acquires a combination of two imaging parameters associated with the two radar images from the imaging parameter group 3 output from the radar image acquiring unit 11.
Herein, one radar image included in the combination i will be referred to as a first radar image, and the other radar image included in the combination i will be referred to as a second radar image.
A radio wave receiving point for a first radar image included in one combination and a radio wave receiving point for a first radar image included in another combination are different from each other. Herein, however, for convenience of explanation, both of such radio wave receiving points will be referred to as first radio wave receiving points P₁.
In addition, a radio wave receiving point for a second radar image included in one combination and a radio wave receiving point for a second radar image included in another combination are different from each other. Herein, however, for convenience of explanation, both of such radio wave receiving points will be referred to as second radio wave receiving points P₂.
An imaging parameter associated with the first radar image will be referred to as a first imaging parameter, and an imaging parameter associated with the second radar image will be referred to as a second imaging parameter.
In addition, the phase shift component calculating unit 21 acquires the inclination angle α.
The phase calculating unit 22 acquires the first imaging parameter, the second imaging parameter, the inclination angle α, and the distance z₀.
The phase shift component calculating unit 21 calculates the position x on the inclined surface 51 corresponding to a pixel position “pixel” in the slant-range direction in the radar image by substituting the position “pixel” into the formula (4).
The pixel at the position “pixel” substituted into the formula (4) is a pixel in which a plurality of reflected signals from scatterers are present.
The phase shift component calculating unit 21 calculates the phase shift component φ_i(x) in the x-axis direction on the inclined surface 51 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, and the observation path parameter p.
The following formula (10) is a formula for calculating the phase shift component φ_i(x) used by the phase shift component calculating unit 21.
$\begin{matrix} φ_{i} (x) = (\frac{2 p π B_{1 2} \cos (θ - α)}{λ R}) x & (10) \end{matrix}$
The phase shift component calculating unit 21 outputs the phase shift component φ_i(x) in the x-axis direction to the phase difference calculating unit 23.
The phase calculating unit 22 calculates the phase ρ_i(z₀) on the parallel surface 52 with respect to the inclined surface 51 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, the distance z₀, and the observation path parameter p. The same applies to the phase ρ_i(z₀) in any combination.
The following formula (11) is a formula for calculating the phase ρ_i(z₀) used by the phase calculating unit 22.
$\begin{matrix} ρ_{i} (z_{0}) = (\frac{2 p π B_{1, 2}}{λ R \sin (θ - α)}) z_{0} & (11) \end{matrix}$
The phase calculating unit 22 outputs the phase ρ_i(z₀) to the phase difference calculating unit 23.
The phase difference calculating unit 23 calculates, in each of a plurality of reflected signals present in one pixel in each combination i, a phase difference Δφ_i(x,z₀) between the phase with respect to the first radio wave receiving point P₁and the phase with respect to the second radio wave receiving point P₂by using the phase shift component φ_i(x) and the phase ρ_i(z₀).
The following formula (12) is a formula for calculating the phase difference Δφ_i(x,z₀) used by the phase difference calculating unit 23.
Δϕ_i(x,z ₀)=ϕ_i(x)+ρ_i(z ₀) (12)
The phase difference calculating unit 23 outputs each phase difference Δφ_i(x,z₀) to the image processing unit 13.
The rotation amount calculating unit 31 acquires each phase difference Δφ_i(x,z₀) output from the phase difference calculating unit 23.
The rotation amount calculating unit 31 calculates, for each combination i, each of phase rotation amounts exp[j·Δφ_i(x,z₀)] in a plurality of pixels included in the second radar image from each phase difference Δφ_i(x,z₀).
The rotation amount calculating unit 31 outputs each rotation amount exp[j·Δφ_i(x,z₀)] to the phase rotating unit 33.
The phase rotating unit 33 acquires the second radar image included in the combination i from the radar image group 2 output from the radar image acquiring unit 11.
The phase rotating unit 33 performs the process of rotating the phases in the plurality of pixels included in the acquired second radar image on the basis of the respective rotation amounts exp[j·Δφ_i(x,z₀)] output from the rotation amount calculating unit 31.
The following formula (13) is a formula representing the process of rotating a phase performed by the phase rotating unit 33.
S ₂′(pixel,line)=S ₂(pixel,line)exp[jΔϕ _i(x,z ₀)] (13)
The phase rotating unit 33 outputs a second radar image including a plurality of pixels obtained by phase rotation to the difference calculation processing unit 34.
The difference calculation processing unit 34 acquires the first radar image included in the combination i from the radar image group 2 output from the radar image acquiring unit 11, and acquires the second radar image including a plurality of pixels obtained by the phase rotation and output from the phase rotating unit 33.
The difference calculation processing unit 34 calculates the difference ΔS_i(pixel,line) between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the acquired first radar image and among a plurality of pixels obtained by phase rotation included in the acquired second radar image.
The following formula (14) is a formula for calculating the difference ΔS_i(pixel,line) used by the difference calculation processing unit 34.
ΔS _i(pixel,line)=S ₁(pixel,line)−S ₂′(pixel,line) (14)
The difference calculation processing unit 34 outputs each difference ΔS_i(pixel,line) to the image combining unit 35.
The rotation amount calculating unit 31, the phase rotating unit 33, and the difference calculation processing unit 34 repeat the process of calculating the difference ΔS_i(pixel,line) until the process of calculating the difference ΔS_i(pixel,line) is completed for all of the combinations i of two radar images.
The image combining unit 35 acquires a weight parameter w_iused for generation of a suppressed image.
The weight parameter w_imay be provided to the image combining unit 35 by manual operation made by a user, or may be provided to the image combining unit 35 from an external device, which is not illustrated.
The image combining unit 35 combines differences ΔS_i(pixel,line) at corresponding pixel positions among the respective differences calculated for the respective combinations i by the difference calculation processing unit 34 by using the weight parameter w_i.
The image combining unit 35 outputs a suppressed image including the respective differences S_sup(pixel, line) resulting from the combining to the outside of the unit.
For the method of combining the differences ΔS_i(pixel,line) in all the combination, a method of obtaining an arithmetic mean or a method of obtaining a geometric mean can be used.
In a case where the method of obtaining an arithmetic mean is used, the image combining unit 35 combines the differences ΔS_i(pixel,line) in all the combinations by the following formula (15).
$\begin{matrix} S_{\sup} (pixel, line) = \frac{1}{N} \sum_{i}^{N} w_{i} Δ S_{i} (pixel, line) & (15) \end{matrix}$
In a case where the method of obtaining a geometric mean is used, the image combining unit 35 combines the differences ΔS_i(pixel,line) in all the combinations by the following formula (16).
$\begin{matrix} S_{\sup} (pixel, line) = {\prod_{i}^{N} Δ {S_{i} (pixel, line)}^{w_{i}}}^{1 / N} & (16) \end{matrix}$
In the formulas (15) and (16), N represents the number of combinations of two radar images.
Here, FIG. 14 is an explanatory diagram illustrating a plurality of reflected signals present in one pixel in a case where only two radar images are included in the radar image group 2 like the radar image processing device 10 of the first embodiment.
In the case where only two radar images are included in the radar image group 2, a plurality of null points may be formed as a result of the process of calculating the differences ΔS_i(pixel, line) performed by the difference calculation processing unit 34 as illustrated in FIG. 14.
In the example of FIG. 14, null points are formed in all of a reflected signal assigned with “1”, a reflected signal assigned with “2”, and a reflected signal assigned with “3”.
Thus, in the example of FIG. 14, all of the reflected signal assigned with “1”, the reflected signal assigned with “2”, and the reflected signal assigned with “3” are suppressed.
FIG. 15 is an explanatory diagram illustrating a plurality of reflected signals present in one pixel in a case where two or more radar images are included in the radar image group 2 like the radar image processing device 10 of the second embodiment.
In FIG. 15, the number of radar images included in the radar image group 2 is M, and P_Mrepresents the position of the platform when an M-th radar image is taken.
Because the number of radar images included in the radar image group 2 is two or larger and the image combining unit 35 combines the differences ΔS_i(pixel,line) at corresponding pixel positions, the number of null points that are formed is reduced as compared with that in the case where the number of radar images is two.
In the example of FIG. 15, the number of null points that are formed is one, and no null point is formed in the reflected signal assigned with “2”.
In the second embodiment described above, the radar image processing device 10 has a configuration including the image combining unit 35 that combines differences ΔS_i(pixel,line) at corresponding pixel positions among the respective differences calculated for the respective combinations i by the difference calculation processing unit 34. The radar image processing device 10 is therefore capable of reducing the number of null points that are formed, which can prevent reflected signals that need to be maintained from being suppressed.

Third Embodiment

The second embodiment presents an example in which the radar image processing device 10 output the differences S_sup(pixel,line) obtained by the combining as a suppressed image.
In a third embodiment, a radar image processing device 10 that calculates an image in which a plurality of reflected signals present in one pixel are extracted from the differences S_sup(pixel,line) resulting from the combining by the image combining unit 35 will be described.
The radar image processing device 10 in the third embodiment has a configuration as illustrated in FIG. 1, that is similar to the radar image processing device 10 of the first or second embodiment.
The phase processing unit 12 in the third embodiment has a configuration as illustrated in FIG. 2, that is similar to the phase processing unit 12 of the first or second embodiment.
Note that the radar image group 2 includes two or more radar images, and the imaging parameter group 3 includes two or more imaging parameters.
FIG. 16 is a configuration diagram illustrating an image processing unit 13 of the radar image processing device 10 according to the third embodiment.
FIG. 17 is a hardware configuration diagram illustrating hardware of each of the phase processing unit 12 and the image processing unit 13.
In FIGS. 16 and 17, reference numerals that are the same as those in FIGS. 3, 4, 12, and 13 represent the same or corresponding components, and the description thereof will thus not be repeated.
An extraction image calculating unit 36 is implemented by an extraction image calculating circuit 48 illustrated in FIG. 17, for example.
The extraction image calculating unit 36 acquires the first radar image from the radar image group 2 output from the radar image acquiring unit 11, and acquires the respective differences S_sup(pixel,line) resulting from the combining output from the image combining unit 35.
The extraction image calculating unit 36 performs a process of calculating an image in which a plurality of reflected signals present in one pixel are extracted on the basis of the pixel values of a plurality of pixels included in the first radar image and the respective differences S_sup(pixel,line) resulting from the combining.
In FIG. 2, it is assumed that each of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23, which are components of the phase processing unit 12, is implemented by such dedicated hardware as illustrated in FIG. 17.
In addition, in FIG. 16, it is assumed that each of the rotation amount calculating unit 31, the phase rotating unit 33, the difference calculation processing unit 34, the image combining unit 35, and the extraction image calculating unit 36, which are components of the image processing unit 13, is implemented by such dedicated hardware as illustrated in FIG. 17.
Specifically, the phase processing unit 12 and the image processing unit 13 are assumed to be implemented by the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, the difference calculation processing circuit 46, the image combining circuit 47, and the extraction image calculating circuit 48.
The components of the phase processing unit 12 and the components of the image processing unit 13 are not limited to those implemented by dedicated hardware. The phase processing unit 12 and the image processing unit 13 may be implemented by software, firmware, or a combination of software and firmware.
Next, the operation of the radar image processing device 10 will be explained.
Note that the radar image processing device 10 is similar to the radar image processing device 10 of the second embodiment except that the extraction image calculating unit 36 is included, and thus, only the operation of the extraction image calculating unit 36 will be explained here.
The extraction image calculating unit 36 acquires the first radar image from the radar image group 2 output from the radar image acquiring unit 11, and acquires the respective differences S_sup(pixel,line) resulting from the combining output from the image combining unit 35.
The extraction image calculating unit 36 calculates a pixel value S_ext(pixel,line) of a pixel in which a plurality of reflected signals are present from the pixel values of a plurality of pixels included in the first radar image and the respective differences S_sup(pixel,line) resulting from the combining.
The following formula (17) is a formula for calculating the pixel value S_ext(pixel,line) used by the extraction image calculating unit 36.
S _ext(pixel,line)=S _i(pixel,line)/S _sup(pixel,line) (17)
The extraction image calculating unit 36 outputs, to the outside of the unit, an image including the pixel having the pixel value S_ext(pixel, line) as an image in which a plurality of reflected signal present in one pixel are extracted.
In the third embodiment described above, the radar image processing device 10 has a configuration including the extraction image calculating unit 36 that calculates an image in which a plurality of reflected signals present in one pixel are extracted on the basis of the pixel values of a plurality of pixels included in the first radar image and the respective differences S_sup(pixel,line) resulting from the combining. The radar image processing device 10 is therefore capable of outputting not only a suppressed image in which reflected signals are suppressed but also an extraction image in which reflected signals are extracted.

Fourth Embodiment

In a fourth embodiment, a radar image processing device 10 that calculates, as an interference phase Δγ_C1,C2(pixel,line), the phase at each pixel position from the difference ΔS_C1(pixel,line) at each pixel position in a first combination C1 and the difference ΔS_C2(pixel,line) at each pixel position in a second combination C2 will be described.
The radar image processing device 10 in the fourth embodiment has a configuration as illustrated in FIG. 1, that is similar to the radar image processing device 10 of the first, second, or third embodiment.
The phase processing unit 12 in the fourth embodiment has a configuration as illustrated in FIG. 2, that is similar to the phase processing unit 12 of the first, second, or third embodiment.
Note that, the radar image group 2 includes three or more radar images capturing the same observation area taken from radio wave receiving points different from each other, and the imaging parameter group 3 includes three or more imaging parameters.
FIG. 18 is a configuration diagram illustrating the image processing unit 13 of the radar image processing device 10 according to the fourth embodiment.
FIG. 19 is a hardware configuration diagram illustrating hardware of each of the phase processing unit 12 and the image processing unit 13.
In FIGS. 18 and 19, reference numerals that are the same as those in FIGS. 3, 4, 12, 13, 16, and 17 represent the same or corresponding components, and the description thereof will thus not be repeated.
In the radar image processing device 10 of the fourth embodiment, a combination of any two radar images included in the radar image group 2 will be referred to as a first combination C1. In addition, a combination of any two radar images included in the radar image group 2 will be referred to as a second combination C2.
The two radar images included in the first combination C1 and the two radar images included in the second combination C2 are different from each other. Note that, one of the two radar images included in the first combination C1 may be the same as any one of the two radar images included in the second combination C2.
In the radar image processing device 10 of the fourth embodiment, the difference calculation processing unit 34 calculates the difference ΔS_C1(pixel,line) at each pixel position in the first combination C1, and the difference ΔS_C2(pixel,line) at each pixel position in the second combination C2.
The interference phase calculating unit 37 is implemented by an interference phase calculating circuit 49 illustrated in FIG. 19, for example.
The interference phase calculating unit 37 acquires the differences ΔS_C1(pixel,line) at the respective pixel positions calculated for the first combination C1, and the differences ΔS_C2(pixel,line) at the respective pixel positions calculated for the second combination C2 by the difference calculation processing unit 14.
The interference phase calculating unit 37 calculates, as interference phases Δγ_C1,C2(pixel,line), the phases at the respective pixel positions from the differences ΔS_C1(pixel,line) and the differences ΔS_C2(pixel,line).
In FIG. 2, it is assumed that each of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23, which are components of the phase processing unit 12, is implemented by such dedicated hardware as illustrated in FIG. 19.
In addition, in FIG. 18, it is assumed that each of the rotation amount calculating unit 31, the phase rotating unit 33, the difference calculation processing unit 34, and the interference phase calculating unit 37, which are components of the image processing unit 13, is implemented by such dedicated hardware as illustrated in FIG. 19.
Specifically, the phase processing unit 12 and the image processing unit 13 are assumed to be implemented by the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, the difference calculation processing circuit 46, and the interference phase calculating circuit 49.
The components of the phase processing unit 12 and the components of the image processing unit 13 are not limited to those implemented by dedicated hardware. The phase processing unit 12 and the image processing unit 13 may be implemented by software, firmware, or a combination of software and firmware.
Next, the operation of the radar image processing device 10 will be explained.
The phase processing unit 12 performs a process of calculating the phase differences Δφ_C1(x,z₀) for the first combination C1, and a process of calculating the phase differences Δφ_C2(x,z₀) for the second combination C2.
The process of calculating a phase difference performed by the phase processing unit 12 will now be explained in detail.
First, the phase shift component calculating unit 21 acquires a combination of two imaging parameters associated with the two radar images included in the first combination C1 from the imaging parameter group 3 output from the radar image acquiring unit 11.
Herein, one radar image included in the first combination C1 will be referred to as a first radar image, and the other radar image included in the first combination C1 will be referred to as a second radar image.
In addition, an imaging parameter associated with the first radar image will be referred to as a first imaging parameter, and an imaging parameter associated with the second radar image will be referred to as a second imaging parameter.
In addition, the phase shift component calculating unit 21 acquires the inclination angle α.
The phase calculating unit 22 acquires the first imaging parameter, the second imaging parameter, the inclination angle α, and the distance z₀.
The phase shift component calculating unit 21 calculates the position x on the inclined surface 51 corresponding to a pixel position “pixel” in the slant-range direction in the radar image by substituting the position “pixel” into the formula (4).
The phase shift component calculating unit 21 calculates the phase shift component φ_C1(x) in the x-axis direction on the inclined surface 51 for the first combination C1 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, and the observation path parameter p.
The following formula (18) is a formula for calculating the phase shift component Δφ_i(x) used by the phase shift component calculating unit 21.
$\begin{matrix} φ_{C 1} (x) = (\frac{2 p π B_{1, 2} \cos (θ - α)}{λ R}) x & (18) \end{matrix}$
The phase shift component calculating unit 21 outputs the phase shift component Δφ_i(x) in the x-axis direction to the phase difference calculating unit 23.
Subsequently, the phase shift component calculating unit 21 acquires a combination of two imaging parameters associated with the two radar images included in the second combination C2 from the imaging parameter group 3 output from the radar image acquiring unit 11.
Herein, one radar image included in the second combination C2 will be referred to as a first radar image, and the other radar image included in the second combination C2 will be referred to as a second radar image.
In addition, an imaging parameter associated with the first radar image will be referred to as a first imaging parameter, and an imaging parameter associated with the second radar image will be referred to as a second imaging parameter.
In addition, the phase shift component calculating unit 21 acquires the inclination angle α.
The phase calculating unit 22 acquires the first imaging parameter, the second imaging parameter, the inclination angle α, and the distance z₀.
The phase shift component calculating unit 21 calculates the position x on the inclined surface 51 corresponding to a pixel position “pixel” in the slant-range direction in the radar image by substituting the position “pixel” into the formula (4).
The phase shift component calculating unit 21 calculates the phase shift component φ_C2(x) in the x-axis direction on the inclined surface 51 for the second combination C2 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, and the observation path parameter p.
The following formula (19) is a formula for calculating the phase shift component φ_C2(x) used by the phase shift component calculating unit 21.
$\begin{matrix} φ_{C 2} (x) = (\frac{2 p π B_{1, 2} \cos (θ - α)}{λ R}) x & (19) \end{matrix}$
The phase shift component calculating unit 21 outputs the phase shift component φ_C2(x) in the x-axis direction to the phase difference calculating unit 23.
The phase calculating unit 22 calculates the phases ρ_C1(z₀) and ρ_C2(z₀) on the parallel surface 52 with respect to the inclined surface 51 by using the distance component B_1,2, the off-nadir angle θ, the average R of the distances, the wavelength λ of the emitted radio wave, the inclination angle α, the distance z₀, and the observation path parameter p.
The following formula (20) is a formula for calculating the phases ρ_C1(z₀) and ρ_C2(z₀) used by the phase calculating unit 22.
$\begin{matrix} ρ_{C 1} (z_{0}) = ρ_{C 2} (z_{0}) = (\frac{2 p π B_{1, 2}}{λ R \sin (θ - α)}) z_{0} & (20) \end{matrix}$
The phase calculating unit 22 outputs the phases ρ_C1(z₀) and ρ_C2(z₀) to the phase difference calculating unit 23.
The phase difference calculating unit 23 acquires the respective phase shift components φ_C2(x) and Δφ_i(x) output from the phase shift component calculating unit 21, and acquires the respective phases ρ_C1(z₀) and ρ_C2(z₀) output from the phase calculating unit 22.
The phase difference calculating unit 23 calculates, for the first combination C1, the phase difference Δφ_C1(x,z₀) in each of a plurality of reflected signals present in one pixel by using the phase shift component Δφ_i(x) and the phase ρ_C1(z₀).
The following formula (21) is a formula for calculating the phase difference Δφ_C1(x,z₀) used by the phase difference calculating unit 23.
Δϕ_C1(x,z ₀)=ϕ_C1(x)+ρ_C1(z ₀) (21)
The phase difference calculating unit 23 outputs each phase difference Δφ_C1(x,z₀) to the image processing unit 13.
Subsequently, the phase difference calculating unit 23 calculates, for the second combination C2, the phase difference Δφ_C2(x,z₀) in each of a plurality of reflected signals present in one pixel by using the phase shift component φ_C2(x) and the phase ρ_C2(z₀).
The following formula (22) is a formula for calculating the phase difference Δφ_C2(x,z₀) used by the phase difference calculating unit 23.
Δϕ_C2(x,z ₀)=ϕ_C2(x)ρ_C2(z ₀) (22)
The phase difference calculating unit 23 outputs each phase difference Δφ_C2(x,z₀) to the image processing unit 13.
The rotation amount calculating unit 31 acquires the respective phase differences Δφ_C1(x,z₀) and Δφ_C2(x,z₀) output from the phase difference calculating unit 23.
The rotation amount calculating unit 31 calculates, for the first combination C1, each of phase rotation amounts exp[j·Δφ_C1(x,z₀)] in a plurality of pixels included in the second radar image from each phase difference Δφ_C1(x,z₀).
The rotation amount calculating unit 31 outputs each rotation amount exp[j·Δφ_C1(x,z₀)] to the phase rotating unit 33.
Subsequently, the rotation amount calculating unit 31 calculates, for the second combination C2, each of phase rotation amounts exp[j·Δφ_C2(x,z₀)] in a plurality of pixels included in the second radar image from each phase difference Δφ_C2(x,z₀).
The rotation amount calculating unit 31 outputs each rotation amount exp[j·Aφ_C2(x,z₀)] to the phase rotating unit 33.
The phase rotating unit 33 first acquires the second radar image included in the first combination C1 from the radar image group 2 output from the radar image acquiring unit 11.
The phase rotating unit 33 performs the process of rotating the phases in the plurality of pixels included in the acquired second radar image on the basis of the respective rotation amounts exp[j·Δφ_C1(x,z₀)] output from the rotation amount calculating unit 31.
The following formula (23) is a formula representing the process of rotating a phase performed by the phase rotating unit 33.
S ₂′(pixel,line)=S ₂(pixel,line)exp[jΔϕ _C1(x,z ₀)] (23)
The phase rotating unit 33 outputs a second radar image including a plurality of pixels obtained by phase rotation to the difference calculation processing unit 34.
Subsequently, the phase rotating unit 33 acquires the second radar image included in the second combination C2 from the radar image group 2 output from the radar image acquiring unit 11.
The phase rotating unit 33 performs the process of rotating the phases in the plurality of pixels included in the acquired second radar image on the basis of the respective rotation amounts exp[j·Δφ_C2(x,z₀)] output from the rotation amount calculating unit 31.
The following formula (24) is a formula representing the process of rotating a phase performed by the phase rotating unit 33.
S ₂′(pixel,line)=S ₂(pixel,line)exp[jΔϕ _C2(x,z ₀)] (24)
The phase rotating unit 33 outputs a second radar image including a plurality of pixels obtained by phase rotation to the difference calculation processing unit 34.
The difference calculation processing unit 34 acquires the first radar image included in the first combination C1 from the radar image group 2 output from the radar image acquiring unit 11.
The difference calculation processing unit 34 also acquires the second radar image including a plurality of pixels obtained by the phase rotation on the first combination C1 output from the phase rotating unit 33.
The difference calculation processing unit 34 calculates the difference ΔS_C1(pixel,line) between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the acquired first radar image and among a plurality of pixels obtained by phase rotation included in the acquired second radar image.
The following formula (25) is a formula for calculating the difference ΔS_C1(pixel,line) used by the difference calculation processing unit 34.
ΔS _C1(pixel,line)=S ₁(pixel,line)−S ₂′(pixel,line) (25)
The difference calculation processing unit 34 outputs each difference ΔS_C1(pixel,line) to the interference phase calculating unit 37.
Subsequently, the difference calculation processing unit 34 acquires the first radar image included in the second combination C2 from the radar image group 2 output from the radar image acquiring unit 11.
The difference calculation processing unit 34 also acquires the second radar image including a plurality of pixels obtained by the phase rotation on the second combination C2 output from the phase rotating unit 33.
The difference calculation processing unit 34 calculates the difference ΔS_C2(pixel,line) between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the acquired first radar image and among a plurality of pixels obtained by phase rotation included in the acquired second radar image.
The following formula (26) is a formula for calculating the difference ΔS_C2(pixel,line) used by the difference calculation processing unit 34.
ΔS _C2(pixel,line)=S ₁(pixel,line)−S ₂′(pixel,line) (26)
The difference calculation processing unit 34 outputs each difference ΔS_C2(pixel,line) to the interference phase calculating unit 37.
The interference phase calculating unit 37 acquires the differences ΔS_C1(pixel,line) at the respective pixel positions calculated for the first combination C1 by the difference calculation processing unit 14.
The interference phase calculating unit 37 also acquires the differences ΔS_C2(pixel,line) at the respective pixel positions calculated for the second combination C2 by the difference calculation processing unit 14.
The interference phase calculating unit 37 calculates, as interference phases Δγ_C1,C2(pixel,line), the phases at the respective pixel positions from the differences ΔS_C1(pixel,line) and the differences ΔS_C2(pixel,line) by using the following formula (27) or formula (28).
Δγ_C1,C2=<(ΔS _C1 /ΔS _C2) (27)
Δγ_C1,C2 =<ΔS _C1 −<ΔS _C2 (28)
In the formula (27) and the formula (28), < is a symbol representing the argument of a complex number.
The interference phases Δγ_C1,C2(pixel,line) are the phases of only the reflected signals remaining without being suppressed among a plurality of reflected signals present in one pixel.
The interference phase calculating unit 37 outputs the interference phases Δγ_C1,C2(pixel,line) to the outside of the unit.
For example, when signals reflected by a ground surface and signals reflected by the roof of a building are present in one pixel, the signals reflected by the ground surface are suppressed by the phase processing unit 12 and the image processing unit 13, and only the signals reflected by the roof of the building remain. Thus, the interference phases Δγ_C1,C2(pixel,line) are calculated as the phases of the signals reflected by the roof of the building.
In the fourth embodiment described above, the radar image processing device 10 has a configuration including the interference phase calculating unit 37 that calculates, as an interference phase Δγ_C1,C2(pixel,line), the phase at each pixel position from the difference ΔS_C1(pixel,line) in the first combination C1 and the difference ΔS_C2(pixel,line) in the second combination C2. The radar image processing device 10 is therefore capable of obtaining the phases of reflected signals in a state in which unnecessary reflected signals from a scatterer are suppressed.

Fifth Embodiment

In a fifth embodiment, a radar image processing device 10 that estimates the position of a scatterer present in an observation area by using interference phases Δγ_C1,C2(pixel,line) calculated by the interference phase calculating unit 37 will be described.
The radar image processing device 10 has a configuration as illustrated in FIG. 1, that is similar to that in the first embodiment.
The phase processing unit 12 has a configuration as illustrated in FIG. 2, that is similar to that in the first, second, or third embodiment.
FIG. 20 is a configuration diagram illustrating the image processing unit 13 of the radar image processing device 10 according to the fifth embodiment.
FIG. 21 is a hardware configuration diagram illustrating hardware of each of the phase processing unit 12 and the image processing unit 13.
In FIGS. 20 and 21, reference numerals that are the same as those in FIGS. 3, 4, 12, 13, 16 to 19 represent the same or corresponding components, and the description thereof will thus not be repeated.
A position estimating unit 38 is implemented by a position estimating circuit 50 illustrated in FIG. 21, for example.
The position estimating unit 38 estimate the position of a scatterer present in an observation area by using the interference phases Δγ_C1,C2(pixel,line) calculated by the interference phase calculating unit 37.
In FIG. 2, it is assumed that each of the phase shift component calculating unit 21, the phase calculating unit 22, and the phase difference calculating unit 23, which are components of the phase processing unit 12, is implemented by such dedicated hardware as illustrated in FIG. 21.
In addition, in FIG. 20, it is assumed that each of the rotation amount calculating unit 31, the phase rotating unit 33, the difference calculation processing unit 34, the interference phase calculating unit 37, and the position estimating unit 38, which are components of the image processing unit 13, is implemented by such dedicated hardware as illustrated in FIG. 21.
Specifically, the phase processing unit 12 and the image processing unit 13 are assumed to be implemented by the phase shift component calculating circuit 41, the phase calculating circuit 42, the phase difference calculating circuit 43, the rotation amount calculating circuit 44, the phase rotating circuit 45, the difference calculation processing circuit 46, the interference phase calculating circuit 49, and the position estimating circuit 50.
The components of the phase processing unit 12 and the components of the image processing unit 13 are not limited to those implemented by dedicated hardware. The phase processing unit 12 and the image processing unit 13 may be implemented by software, firmware, or a combination of software and firmware.
Next, the operation of the radar image processing device 10 will be explained.
Note that the radar image processing device 10 is similar to the radar image processing device 10 of the fourth embodiment except that the position estimating unit 38 is included, and thus, only the operation of the position estimating unit 38 will be explained here.
Herein, for convenience of explanation, a radio wave receiving point for a first radar image included in a first combination C1 will be referred to as a radio wave receiving point P_a, and a radio wave receiving point for a second radar image included in the first combination C1 will be referred to as a radio wave receiving point P_b.
In addition, a radio wave receiving point for a first radar image included in a second combination C2 will be referred to as a radio wave receiving point P_c, and a radio wave receiving point for a second radar image included in the second combination C2 will be referred to as a radio wave receiving point P_d.
The position estimating unit 38 acquires the interference phases Δγ_C1,C2(pixel,line) output from the interference phase calculating unit 37.
The position estimating unit 38 also acquires the respective phase differences Δφ_C1(x,z₀) and Δφ_C2(x,z₀) output from the phase difference calculating unit 23.
The position estimating unit 38 estimates the position z-hat of a scatterer present in an observation area by using the interference phases Δγ_C1,C2(pixel,line) and the respective phase differences Δφ_C1(x,z₀) and Δϕ_C2(x,z₀) output from the phase difference calculating unit 23, as expressed in the following formula (29).
Because the symbol “{circumflex over ( )}” cannot be typed above the character “z” in the description due to electronic filing, it is described in such a manner as z-hat herein.
$\begin{matrix} \hat{z} = R \sin (θ - α) [\frac{λ (B_{a, c} + B_{b, d})}{2 p π} {2 ∠ Δ γ_{C 1, C 2} - {Δφ}_{C 1} (x, z_{0}) + Δ φ_{C 2} (x, z_{0})} - \frac{\cos (θ - α) x}{R}] & (29) \end{matrix}$
In the formula (29), B_a,crepresents a distance component, in a direction perpendicular to the slant-range direction, of the distance between the radio wave receiving point P_aand the radio wave receiving point P_c.
B_b,drepresents a distance component, in a direction perpendicular to the slant-range direction, of the distance between the radio wave receiving point P_band the radio wave receiving point P_d.
R represents an average of the distances between each of the radio wave receiving point P_a, the radio wave receiving point P_b, the radio wave receiving point P_c, and the radio wave receiving point P_dand the observation area.
The distance component B_a,c, the distance component B_b,d, the off-nadir angle θ, and the average R of the distances are information included in the imaging parameter.
Symbol x represents the position on the inclined surface 51 associated with the position “pixel”, and is output from the phase shift component calculating unit 21.
The position z-hat of the scatterer is the distance (height) in a z-axis direction from the inclined surface 51 to a signal reflecting surface of the scatterer.
The position estimating unit 38 outputs the estimated position z-hat of the scatterer to the outside of the unit.
Herein, the two radar images included in the first combination C1 and the two radar images included in the second combination C2 are different from each other. The combinations, however, are not limited thereto, and one of the two radar images included in the first combination C1 may be the same as one of the two radar images included in the second combination C2.
For example, the second radar image included in the first combination C1 and the second radar image included in the second combination C2 may be the same radar image.
In the case where the second radar image included in the first combination C1 and the second radar image included in the second combination C2 may be the same radar image, B_b,d=0 is obtained, and the formula (29) used for estimation of the position z-hat is simplified as in the following formula (30).
$\begin{matrix} \hat{z} = R \sin (θ - α) [\frac{λ B_{a, c}}{2 p π} {2 ∠Δ γ_{C 1, C 2} - {Δφ}_{C 1} (x, z_{0}) + {Δφ}_{C 2} (x, z_{0})} - \frac{\cos (θ - α) x}{R}] & (30) \end{matrix}$
In the fifth embodiment described above, the radar image processing device 10 has a configuration including the position estimating unit 38 that estimates the position z-hat of a scatterer present in an observation area by using the interference phases Δγ_C1,C2(pixel,line) calculated by the interference phase calculating unit 37. The radar image processing device 10 is therefore capable of obtaining the position of a scatterer present in an observation area.
Note that the embodiments of the present invention can be freely combined, any components in the embodiments can be modified, and any components in the embodiments can be omitted within the scope of the invention of the present application.

INDUSTRIAL APPLICABILITY

The present invention is suitable for a radar image processing device and a radar image processing method that calculate differences between a plurality of pixels included in a first radar image and a plurality of pixels obtained by phase rotation included in a second radar image.

REFERENCE SIGNS LIST

- 1: radar, 2: radar image group, 3: imaging parameter group, 10: radar image processing device, 11: radar image acquiring unit, 12: phase processing unit, 13: image processing unit, 21: phase shift component calculating unit, 22: phase calculating unit, 23: phase difference calculating unit, 31: rotation amount calculating unit, 32: difference calculating unit, 33: phase rotating unit, 34: difference calculation processing unit, 35: image combining unit, 36: extraction image calculating unit, 37: interference phase calculating unit, 38: position estimating unit, 41: phase shift component calculating circuit, 42: phase calculating circuit, 43: phase difference calculating circuit, 44: rotation amount calculating circuit, 45: phase rotating circuit, 46: difference calculation processing circuit, 47: image combining circuit, 48: extraction image calculating circuit, 49: interference phase calculating circuit, 50: position estimating circuit, 51: inclined surface, 52: parallel surface, 61: memory, 62: processor

Claims

1. A radar image processing device, comprising:

processing circuitry performing a process of:

calculating a phase shift component in a first axis direction on a two-dimensional inclined surface included in the first radar image and the second radar image, the first axis being an axis of the inclined surface inclined with respect to a ground-range direction;

calculating a phase on a surface parallel to the inclined surface with respect to the inclined surface; and

calculating a phase difference in each of a plurality of reflected signals present in each of pixels included in first and second radar images, the phase difference being a difference between phases with respect to the respective radio wave receiving points from a phase shift component calculated and a phase calculated;

calculating each of phase rotation amounts in a plurality of pixels included in the second radar image from each phase difference calculated; and

rotating phases in a plurality of pixels included in the second radar image on a basis of the respective rotation amounts calculated, and calculating a difference between pixel values of pixels being complex numbers at corresponding pixel positions among a plurality of pixels included in the first radar image and a plurality of pixels resulting from phase rotation included in the second radar image.

2. The radar image processing device according to claim 1, wherein

a radar image group includes two or more radar images capturing one observation area from radio wave receiving points different from each other,

the process calculates the phase shift component in the first axis direction on the inclined surface for each combination of two radar images included in the radar image group, one of the radar images included in each combination being the first radar image, the other of the radar images included in each combination being the second radar image,

the process calculates, for each combination of two radar images, the phase difference in each of the plurality of reflected signals from the phase shift components calculated for each combination and the phases calculated,

the process calculates, for each combination of two radar images, each of the phase rotation amounts of the plurality of pixels included in the second radar image from the respective phase differences calculated for each combination,

the process rotates, for each combination of two radar images, the phases of the plurality of pixels included in the second radar image on a basis of the respective rotation amounts calculated for each combination, and calculates a difference between pixel values of pixels at corresponding pixel positions among the plurality of pixels included in the first radar image and the plurality of pixels resulting from the phase rotation included in the second radar image, and

the radar image processing device includes

combining differences at corresponding pixels position among the respective differences calculated for each combination.

3. The radar image processing device according to claim 2, further comprising:

calculating an image in which the plurality of reflected signals are extracted, from pixel values of the plurality of pixels included in the first radar image and the respective differences obtained by the combining.

4. The radar image processing device according to claim 1, wherein

a first combination includes any two radar images among three or more radar images capturing one observation area from radio wave receiving points different from each other, and a second combination includes two radar images among the three or more radar images, at least one of the two radar images in the second combination being different from the two radar images included in the first combination,

the process calculates, for each of the first and second combination, the phase shift component in the first axis direction on the inclined surface, one of radar images included in each combination being the first radar image, the other of the radar images included in each combination being the second radar image,

the process calculates, for each of the first and second combinations, the phase difference in each of the plurality of reflected signals from the phase shift components calculated for each combination and the phases calculated,

the process calculates, for each of the first and second combinations, each of the phase rotation amounts of the plurality of pixels included in the second radar image from the respective phase differences calculated for each combination,

the processs rotates, for each of the first and second combinations, the phases of the plurality of pixels included in the second radar image on a basis of the respective rotation amounts calculated for each combination, and calculates a difference between pixel values of pixels at corresponding pixel positions among the plurality of pixels included in the first radar image and the plurality of pixels resulting from the phase rotation included in the second radar image, and

calculating, as interference phases, phases at respective pixel positions from the differences at the respective pixel positions calculated for the first combination and the differences at the respective pixel positions calculated for the second combination.

5. The radar image processing device according to claim 4, further comprising:

estimating a position of a scatterer present in the observation area by using the interference phases calculated.

6. A radar image processing device comprising:

processing circuitry performing a process of:

calculating a phase difference in each of a plurality of reflected signals present in each of pixels included in first and second radar images, the first radar images capturing an observation area from a first radio wave receiving points and the second radar images capturing an observation area same as the observation area from a second radio wave receiving points, the phase difference being a difference between a phase with respect to the first radio wave receiving point and a phase with respect to the second radio wave receiving point, with respect to a two-dimensional surface inclined by an inclination angle α against a ground-range direction;

rotating phases in a plurality of pixels included in the second radar image on a basis of the respective rotation amounts calculated, and calculating a difference between pixel values of pixels at corresponding pixel positions among a plurality of pixels included in the first radar image and a plurality of pixels resulting from phase rotation included in the second radar image.

7. The radar image processing method, comprising: