Non-contact shape and dielectric constant measurement device, non-contact shape and dielectric constant measurement method, and non-contact shape and dielectric constant measurement program

Abstract

To enable a complex dielectric constant of a target object to be estimated under an ideal condition that does not include information such as antenna and surrounding environment.SOLUTION: A non-contact shape and dielectric constant measurement device acquires a scattering electric field scattering at a target of an electromagnetic wave transmitted toward the target from the electromagnetic wave received by a reception antenna as experimental data, calculates the scattering electric field in a simulation environment in which an experimental environment is simulated independently of the experimental data, obtains a correction coefficient from the experimental data obtained from the acquired scattering electric field of the target that has a known dielectric constant and the simulation data obtained by simulating the calculated scattering electric field of the known target, converts the experimental data obtained from the scattering electric field of the target that has an unknown dielectric constant into the simulation data of the target that has the unknown dielectric constant by using the correction coefficient, and obtains a complex dielectric constant distribution image with a CSI method by using the converted simulation data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a non-contact shape / dielectric constant measuring device, a non-contact shape / dielectric constant measuring method, and a non-contact shape / dielectric constant measuring program. [Background technology]

[0002] The total length of tunnels in Japan is over 23,000 km for railway tunnels and over 4,100 km for road tunnels, with expressways and other roads being several dozen times longer. Most of these were constructed rapidly during the period of high economic growth in the 1980s, and in recent years have become increasingly dilapidated, posing a high risk of the collapse of tunnel tops, bridge collapses, road subsidence, etc., due to earthquakes, etc. For this reason, there is a demand for technology that can screen the inside of roads, etc. quickly and with a high rate of defect identification.

[0003] Conventional hammering inspections and ultrasonic measurements are contact-type measurements performed by touching the walls of tunnels, roads, etc., making it difficult to thoroughly inspect large areas of tunnels, roads, etc. Furthermore, conventional hammering inspections and ultrasonic measurements are based on qualitative and subjective judgments, so the discrimination rate for cavities, corrosion, etc. from other parts is insufficient.

[0004] It is known that each target, such as concrete, cavity, water, rebar, rust, etc., has a characteristic complex permittivity value. Therefore, it is expected that the discrimination performance of non-contact shape and permittivity measurement equipment will be significantly improved by reconstructing the complex permittivity distribution with high accuracy.

[0005] Microwaves, with wavelengths of a few centimeters, can penetrate low-loss dielectrics such as concrete to a depth of several tens of centimeters, making them promising for non-destructive testing of aging transportation infrastructure. In other words, because spatial attenuation can be ignored in microwave testing, non-contact measurement from a distance of several meters is possible, and low-frequency microwaves of around 1 to 3 GHz can penetrate to a depth of up to about 1 m inside concrete. Therefore, by mounting the transceiver module on a vehicle and collecting scattered data while driving along roads or inside tunnels, it is possible to inspect large areas in a short period of time.

[0006] Microwave imaging methods are broadly divided into two types. One is the radar method, and the other is the tomography method. The radar method reproduces only the reflection intensity distribution of the target object, so it is difficult to quantitatively evaluate the electrical properties of the target. In particular, it is known that the complex dielectric constant of cavities, corrosion, etc. varies considerably depending on the type of concrete.

[0007] On the other hand, the tomography method can quantitatively image the complex permittivity of a target, and is therefore effective in identifying targets such as cavities and rebar corrosion. The tomography method is a method for achieving quantitative imaging of the complex permittivity by solving a domain integral equation derived from Maxwell's equations. However, existing tomography methods are constrained in that the observation direction is restricted because it is impossible to arrange elements to surround the target object. For this reason, in the nondestructive inspection model, the ill-posedness of the number of observation data being much smaller than the unknowns becomes evident, and sufficient accuracy cannot be obtained.

[0008] In other words, in the observation model for general non-destructive inspection of tunnels and road walls, the observation direction is limited, so it is necessary to solve an extremely ill-posed inverse problem. To solve this ill-posed inverse problem, it is necessary to narrow the region of interest (ROI) and significantly reduce the number of unknowns.

[0009] As described above, the tomography method is an effective means for imaging dielectric constant characteristics, but the problem of reconstructing the dielectric constant of a target using the tomography method is also called the inverse scattering problem, and requires solving a nonlinear and ill-posed problem. A nonlinear problem is one that includes multiple scattering, and an ill-posed problem is one in which the amount of data is far less than the number of unknowns. In nondestructive testing, the observation area is limited, so solving this inverse scattering problem has been extremely difficult.

[0010] Non-Patent Document 1 proposes a method that combines DBIM (Distorted Born Iterative Method), which is a tomography method, with a radar method, and shows its effectiveness for a numerical calculation model. However, the DBIM method requires repeated electromagnetic field analysis, which is computationally expensive, for optimization. In addition, it is necessary to provide all complex permittivity distributions of the antenna and the surrounding environment of the target in advance, which creates practical problems.

[0011] Furthermore, Patent Document 1 describes a non-contact dielectric constant measuring device that measures shape and the like in a non-contact manner by combining an RPM (Range Point Migration) method as a range point migration method and an elliptometry method.

[0012] The non-contact dielectric constant measuring device described in Patent Document 1 converts an ultra-wideband (UWB) signal into a pulsed electromagnetic wave, transmits it to a target, receives the reflected wave from the target, and displays an image formed at each scattering point in a different color based on the dielectric constant calculated from the reflection count ratio of two polarized waves of the received wave. For example, an image is generated in which the calculated dielectric constant is used to distinguish between areas estimated to be metal or other materials, areas estimated to be cavities, and areas assumed to be filled with water, each of which is classified in a different color. [Prior art documents] [Patent documents]

[0013] [Patent Document 1] JP 2019-148510 A [Non-patent literature]

[0014] [Non-Patent Document 1] S. Takahashi and S. Kidera, URSI AP-RASC2019, New Delhi, India, march, 2019 Summary of the Invention [Problem to be solved by the invention]

[0015] The CSI (Contrast Source Inversion) method, which is one of the tomography methods, can simultaneously optimize the total electric field and the complex permittivity distribution within the integral kernel (ROI) of the domain integral equation. The inventors have discovered a method for estimating the complex permittivity distribution of a target object by using the CSI method while minimizing the need for complex calculations known as forward problems.

[0016] On the other hand, the CSI method requires information on the complex permittivity of the antenna, background, and surrounding media. Therefore, in order to estimate the complex permittivity distribution from experimental data using the tomography method, a calibration method that bridges the gap between the real environment and the simulation environment is required. Therefore, the inventors first obtained the scattered electric field obtained in an experiment for a calibration target with a known permittivity distribution, and attempted to create a simulation environment that is close to the real environment using the known scattered electric field.

[0017] In the CSI method, a cost function is introduced for two domain integral equations called the state equation and the data equation. By introducing this cost function, it is possible to solve the optimization problem including the entire electric field inside the region of interest (ROI). This makes it possible to dramatically reduce the computational cost.

[0018] An object of the present invention is to provide a non-contact shape and dielectric constant measurement device, a non-contact shape and dielectric constant measurement method, and a non-contact shape and dielectric constant measurement program that are capable of estimating the complex dielectric constant of a target object under ideal conditions that do not include information about the antenna, surrounding environment, etc., by generating data through electromagnetic simulation using the above-mentioned CSI method. [Means for solving the problem]

[0019] The non-contact shape and dielectric constant measuring device of the present invention is a non-contact shape and dielectric constant measuring device that receives electromagnetic waves transmitted from a transmitting antenna toward a target whose dielectric constant is to be measured by a receiving antenna, and obtains the shape and dielectric constant of the target from the received electromagnetic waves.The device is equipped with a scattered electric field acquisition unit that acquires, from the electromagnetic waves received by the receiving antenna, the scattered electric field scattered by the target of the electromagnetic waves transmitted toward the target as experimental data, and an electromagnetic field calculation data output unit that calculates the scattered electric field in a simulation environment that simulates the experimental environment, separately from the experimental data. Furthermore, the non-contact shape and dielectric constant measuring device of the present invention includes a calibration coefficient calculation unit that obtains a correction coefficient from experimental data obtained from the scattered electric field of a target having a known dielectric constant obtained by the scattered electric field acquisition unit and simulation data that simulates the scattered electric field of the known target calculated by the electromagnetic field calculation data output unit, a simulation data conversion unit that converts the experimental data obtained from the scattered electric field of a target having an unknown dielectric constant obtained by the scattered electric field acquisition unit into simulation data of the target having an unknown dielectric constant by dividing the experimental data obtained from the scattered electric field of a target having an unknown dielectric constant obtained by the scattered electric field acquisition unit by the correction coefficient obtained by the calibration coefficient calculation unit, and a dielectric constant image acquisition unit that obtains a complex dielectric constant distribution image by the CSI method using the simulation data converted by the simulation data conversion unit.

[0020] The non-contact shape and dielectric constant measuring method of the present invention includes the following steps (a) to (j). (a) transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from a background medium having a known dielectric constant, not including a target, at a receiving antenna to obtain an experimental scattered electric field signal; (b) transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from an object including a target with a known dielectric constant in a background medium with a receiving antenna to obtain an experimental scattered electric field signal; (c) subtracting the scattered electric field signal from the experiment without the target from the scattered electric field signal from the experiment with the target to obtain a difference signal; (d) calculating, by simulation, the scattered electric field signal of a background medium having a known dielectric constant that does not include the target; (e) calculating, by simulation, a scattered electric field signal of an object including a target with a known dielectric constant in a background medium; (f) subtracting the simulated electromagnetic field signal not including the target from the simulated electromagnetic field signal including the target to obtain a difference signal; (g) dividing the experimental with-target and without-target difference signals by the simulated with-target and without-target difference signals to obtain a correction factor; (h) transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from an object including a target having an unknown dielectric constant with a receiving antenna to obtain an experimental scattered electric field signal; (i) converting experimental scattered electric field signals obtained from an object, including a target with an unknown dielectric constant, into simulated data using correction factors; (j) performing CSI using the converted simulation data to obtain a complex permittivity distribution of the object including the unknown target;

[0021] The non-contact shape / dielectric constant measurement program of the present invention is a program for causing a computer to execute the steps (a) to (j) of the non-contact shape / dielectric constant measurement method. Effect of the Invention

[0022] According to the present invention, particularly in non-destructive testing of concrete, the spatial distribution of the complex dielectric constant of a target object can be measured with high accuracy and high resolution, making it possible to objectively and quantitatively evaluate cavities, water content, siltation, corrosion, etc. Therefore, it is possible to provide highly accurate sensing information for analyzing quantitative image information relating to elucidating the internal structure of concrete, the progression of corrosion over time, changes in concrete strength, etc. [Brief description of the drawings]

[0023] [Figure 1] 1 is a block diagram showing the overall configuration of a non-contact shape and dielectric constant measuring device according to an embodiment of the present invention. [Diagram 2]This is a diagram schematically showing the state where radio waves transmitted from a transmitting antenna are reflected by a target object and received by a receiving antenna in a non-contact shape and dielectric constant measuring apparatus according to an embodiment of the present invention. [Diagram 3] This is a diagram for explaining the difference between an experimental environment and a simulation environment to which a non-contact shape and dielectric constant measuring apparatus according to an embodiment of the present invention is applied. [Figure 4] This is a diagram schematically showing a method of converting experimental data into simulation data in a non-contact shape and dielectric constant measuring apparatus according to an embodiment of the present invention. [Diagram 5] This is a diagram showing an example in an experimental model to which a non-contact shape and dielectric constant measuring apparatus according to an embodiment of the present invention is applied, where there is one cavity in concrete of a background medium with a known dielectric constant. [Figure 6] This is a diagram showing an example in an experimental model to which a non-contact shape and dielectric constant measuring apparatus according to an embodiment of the present invention is applied, where there are four cavities with unknown dielectric constants in a background medium with a known dielectric constant. [Figure 7] This is a flowchart showing the procedure of a non-contact shape and dielectric constant measuring method according to an embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0024] First, before explaining a non-contact shape and dielectric constant measuring apparatus which is an embodiment of the present invention, a CSI method which is a kind of inverse scattering analysis method applied to the non-contact shape and dielectric constant measuring apparatus of the present invention will be explained.

[0025] <Inverse Scattering Analysis Method Based on CSI> There are DBIM and CSI in the imaging method using a tomography method, but CSI is used in the present invention. Equation (1) is a Helmholtz type integral equation generally used for estimating a dielectric contrast function from scattered electric field data at each antenna element by a tomography method.

[0026]

Equation

[0027] Here, E s (ω,r t ,r r ) is the scattered electric field from the target, E t (ω,r t ,r r ) is the scattered electric field E from the target s (ω.r t ,r r ) and the incident electric field E i (ω;r t ,r r ) is the total electric field, ω is the angular frequency, k b is the wave number of the electromagnetic wave in the background medium. χ is a function called the contrast function (or objective function) for the background medium, and the following formula holds: χ(r')≡{ε(r')-ε b (r')} / ε b (r')

[0028] where ε(r') is the complex permittivity of the entire system including the target, ε b (r') is the complex dielectric constant of the background medium. b (ω,r t ,r ’ ) is the Green's function of the background medium, Ω D denotes the region of interest (ROI). r' is the position vector within the region of interest (r'∈Ω D ), r r is the position vector of the receiving antenna, r t is the position vector of the transmitting antenna.

[0029] CSI is a method to obtain the distribution of the complex permittivity and total electric field within the analysis domain by solving the Helmholtz integral equation shown in formula 1. Unlike DBIM, CSI has the advantage that it does not need to solve a forward problem. Therefore, it generally has a lower calculation cost than DBIM. In other words, CSI is a method to minimize a cost function determined by two equations: a data equation, which is a condition equation for the observed data, and a state equation, which is a condition equation for the total electric field within the region of interest (ROI).

[0030] Equation 2 is a data equation, and the position vector r is the observation domain (r∈Ω S ) in the region of interest ROI(r∈Ω). D ) where the observation area is the area surrounded by multiple transmitting and receiving antennas, and the observation area includes the ROI, which is the region of interest where the target exists. The only difference between formula 2 and formula 3 is the range in which the Helmholtz integral equation is implemented.

[0031] In CSI, by using Equation 2 and Equation 3, the total electric field E can be calculated without solving the forward problem using the FDTD method (Finite-difference time-domain method) or the like. t It is possible to estimate both the contrast function χ(r) and the contrast function χ(r). The FDTD method is a method for analytically determining the electromagnetic field from Maxwell's equations using finite differences, and is also known as the "finite difference time domain method" or "finite difference time domain method". The electromagnetic field value determined by the FDTD method is the true value.

[0032] As described later, a forward problem is a problem of finding true values ​​from the state equation for the electric field and the data equation for the measured data. In other words, a forward problem is a problem of finding what kind of simulation data can be obtained when the distribution of the dielectric constant is given. FDTD is one method for obtaining simulation data. Conversely, the inverse scattering problem is used to find the original dielectric constant when simulation data is given. When solving this inverse scattering problem, the FDTD method used to solve the forward problem is not used, so the amount of calculation is significantly reduced.

[0033] In Equation 2 and Equation 3, the unknowns are the contrast function χ(r) and the total electric field E t (r). On the other hand, the known value of the incident electric field E i (r), the wave number k of the electromagnetic wave in the background medium (e.g. concrete) b , measurement data G b (r) χ(r) E t (r), region of interest Ω D It is. Here, the contrast function χ(r) and the total electric field E t Since (r) appears in the form of a product, a contrast source w(r) is introduced as shown in equation 4.

[0034]

number

[0035]

number

[0036] Next, we define the contrast source w(r) expressed by Equation 4. The contrast source w(r) is defined by the contrast function χ(r) and the total electric field E t (r) are the values ​​set for the optimization variables.

number

[0037] Furthermore, using this contrast source w(r), a cost function F expressed by Equation 5 is introduced.

number

[0038] As shown in Equation 5, the cost function F is defined in CSI. Then, the contrast source w(r) and the total electric field E are calculated so that the cost function F is minimized. t (r) and χ(r) are restored. That is, in CSI, the direct total electric field E t (r) is not estimated, and after obtaining the contrast source w(r), the total electric field E is calculated from Eq. t (r) is calculated.

[0039] By using this cost function F(w(r),χ(r)), the total electric field E t (r)(r∈Ω D ) can be solved, which reduces computational costs.

[0040] In formula 5, Ω S is the observation domain, Ω D is the region of interest (ROI). The first term of the cost function F is the normalized value obtained by subtracting the right side from the left side of the data equation in Equation 2, and the second term is the normalized value obtained by multiplying both sides of the state equation in Equation 3 by the contrast function χ(r) and subtracting the left side from the right side. The first term of Eq. 5 is the observation domain Ω S The sum of all antenna elements is taken over the entire region of interest Ω D The symbol Σ indicates that it is the sum of all antennas. The numbers in the numerator and denominator indicate the square of the absolute value of a complex number, and are called norms.

[0041] <Overall configuration of non-contact shape and dielectric constant measurement device> Next, the configuration and functions of a non-contact shape and dielectric constant measuring apparatus 10 according to an embodiment of the present invention (hereinafter referred to as "this example") will be described with reference to Figs. FIG. 1 is a block diagram showing the overall configuration of a non-contact shape and dielectric constant measuring apparatus 10 of this embodiment. As shown in FIG. 1, the non-contact shape and dielectric constant measuring device 10 is a device that acquires the shape and dielectric constant of a target object at a relatively short distance of about 1 to 3 m using a radar with a center frequency of 3 GHz, a bandwidth of 1 GHz, and a wavelength of about 100 mm as a transmission signal.

[0042] The non-contact shape and dielectric constant measuring device 10 uses, for example, 25 antennas A1 to A25, including transmitting and receiving antennas, to supply microwaves to a nearby object and receive reflected light from the object. Of the 25 antennas, four antennas A1, A5, A21, and A25 serve as both transmitting and receiving antennas, and the remaining 21 antennas A2 to A4, A6 to A20, and A22 to A24 are receiving antennas.

[0043] The non-contact shape and dielectric constant measuring apparatus 10 includes a microwave transmission processing section 11, a microwave reception processing section 12, a baseband processing section 13, and a filter processing section . The non-contact shape and dielectric constant measuring device 10 also includes an experimental data differential signal acquiring unit 15, an electromagnetic field calculation data output unit 16, a simulation data differential signal acquiring unit 17, a calibration coefficient calculating unit 18 for calibration, a simulation data converting unit 19, a dielectric constant image acquiring unit 20 and an image output unit 21.

[0044] The microwave transmission processing unit 11 supplies the generated microwave signal to the transmission / reception antennas A1, A5, A21, and A25 via the transmission / reception switches SW1 to SW4. The transmission / reception antennas A1, A5, A21, and A25 transmit pulsed electromagnetic waves based on the supplied microwave signal. Here, the microwave signal has, for example, a center frequency of 3 GHz and a bandwidth of 1 GHz. By using the 3 GHz band, a radar with a center wavelength λ of approximately 100 mm is obtained. The microwave transmission processing unit 11 generates a microwave signal, which is a transmission signal, based on the signal supplied from the baseband processing unit 13.

[0045] Electromagnetic waves transmitted from the four transmitting and receiving antennas A1, A5, A21, and A25 are reflected from the target, and the reflected waves are received by all of the antennas A1 to A25. The received signals received by all of the antennas A1 to A25 are then supplied to the microwave receiving processor 12. The microwave receiving processor 12 demodulates the received signals and outputs baseband signals.

[0046] The baseband signal obtained by the microwave reception processing unit 12 is supplied to the baseband processing unit 13. The baseband processing unit 13 supplies the baseband signal to the microwave transmission processing unit 11 as described above, and also supplies a baseband reception signal obtained from the baseband signal to the filter processing unit 14. The filter processing unit 14 extracts a reflected wave component of the electromagnetic wave contained in the reception signal.

[0047] The microwave transmission processing unit 11, the microwave reception processing unit 12, the baseband processing unit 13, and the filter processing unit 14 obtain an electromagnetic field signal which is a component of a reflected wave only from a background medium such as concrete without a target in an experiment, and an electromagnetic field signal which is a component of a reflected wave from an object with a target in the background medium in the same experiment. These two electromagnetic field signals, one with no target and one with a target, are recorded in a storage unit (not shown) connected to the filter processing unit 14.

[0048] The above-described processing in the microwave transmission processing unit 11, microwave reception processing unit 12, baseband processing unit 13, and filter processing unit 14 are processing units for obtaining scattered electric fields from a target used in the above-described CSI method, and these processing units are collectively referred to as a "scattered electric field acquisition unit" here. Here, the scattered electric field acquisition unit is described as acquiring components of waves reflected from a target, but it may be considered that the same is applied to transmitted waves that penetrate a target.

[0049] Then, the filter processing unit 14 constituting the scattered electric field acquisition unit supplies the experimental electromagnetic field signal without a target and the experimental electromagnetic field signal with a target to the experimental data differential signal acquisition unit 15. The experimental data differential signal acquisition unit 15 outputs a differential signal between the signal with a target and the signal without a target.

[0050] Furthermore, the electromagnetic field calculation data output unit 16 converts the experimental data obtained by the microwave transmission processing unit 11, the microwave reception processing unit 12, the baseband processing unit 13, and the filter processing unit 14 described above into simulation data. The conversion of simulation data in this electromagnetic field calculation data output unit 16 will be described in detail with reference to Fig. 3, and the electromagnetic field calculation data output unit 16 converts the experimental data with and without a target obtained by an experiment into simulation data that models the transmitting and receiving antennas. Here too, the electromagnetic field calculation data output unit 16 converts both the experimental data without a target and the experimental data with a target into simulation data, and stores the simulation data in a storage unit (not shown) connected to the electromagnetic field calculation data output unit 16.

[0051] The electromagnetic field calculation data output unit 16 supplies the simulation data without a target and the simulation data with a target to a simulation data difference signal acquisition unit 17 . The simulation data difference signal acquisition unit 17 outputs a difference signal between the simulation data with and without a target. Then, the experimental data difference signal from the experimental data difference signal acquisition unit 15 and the simulation data difference signal from the simulation data difference signal acquisition unit 17 are supplied to a calibration coefficient calculation unit 18 .

[0052] Calibration coefficient calculation unit 18 calculates a correction coefficient for converting unknown experimental data into simulation data from the experimental data difference signal acquisition unit 15 and the simulation data difference signal generated by simulation data difference signal acquisition unit 17. This correction coefficient will be described later with reference to FIG.

[0053] The calibration coefficient calculation unit 18 supplies the correction coefficient obtained by calculation to the simulation data conversion unit 19. At this time, the simulation data conversion unit 19 is supplied with the electromagnetic field signal obtained by experiment, which is obtained by subtracting the electromagnetic field signal measured in the known background medium from the electromagnetic field signal of the experimental data including the unknown target, from the experimental data difference signal acquisition unit 15. The simulation data conversion unit 19 uses the correction coefficient calculated by the calibration coefficient calculation unit 18 to convert the experimental data for measuring the permittivity of the unknown target (target) supplied from the experimental data difference signal acquisition unit 15 into simulation data.

[0054] The simulation data conversion unit 19 supplies the converted simulation data to the permittivity image acquisition unit 20. The permittivity image acquisition unit 20 obtains a complex permittivity distribution image using the converted simulation data. At this time, the above-described CSI method is used. The complex permittivity distribution image acquired by the permittivity image acquisition unit 20 is supplied to the image output unit 21, and a color-displayed complex permittivity distribution image is output from the image output unit 21. Note that the image output unit 21 can also add the substance name estimated from the permittivity to the image in addition to the permittivity information acquired by the permittivity image acquisition unit 20. Alternatively, the image output unit 21 can also display an image indicating the target shape, distinguished by color coding or the like for each permittivity.

[0055] <Experimental Consideration of CSI> Among the data equations shown in Equation 2, the total electric field E of the position of the receiving antenna in the observation region Ω S (ω, r t (ω, r t , r r ) and the incident electric field E i (ω, r t , r r ) can be experimentally obtained, but in the region of interest ROI, the experimental environment must be considered in order to measure these values. That is, the Green's function G b (ω, r, r') and the total electric field E t (ω, r t , r) are realistically impossible to calculate.

[0056] Figure 2 shows the observation region Ω S In the case of transmitting antenna r t The radio waves transmitted from the r The total electric field E received at t (ω,r t ,r r ) and the region of interest (ROI) Ω D The total electric field E t (ω,r t , r') in FIG. r The total electric field E received at t (ω,r t ,r r ) can be obtained experimentally, but the total electric field E t (ω,r t ,r') cannot be measured or calculated precisely.

[0057] Therefore, a solution requires the conversion of experimental data into simulation data. FIG. 3 is a schematic diagram showing the difference between the actual experimental environment and the simulation environment. As shown in Figure 3, in the experimental environment, the antenna characteristics are real characteristics, so the characteristics of the transmitter and receiver require calibration. In addition, in the experimental environment, there is cable loss, and unwanted waves other than the target objects such as pillars and the floor and walls of the anechoic chamber are observed. Furthermore, in the experimental environment, it becomes necessary to solve a three-dimensional problem including vertical polarization.

[0058] In contrast, in a simulation environment, the antenna characteristics are theoretical characteristics, so there is no need to calibrate the characteristics of the transmitter and receiver. In addition, there is no need to consider cable loss, and the perfect absorbing boundary condition is applied, so it is assumed that there are no unwanted waves other than the target object. Furthermore, there is no need to consider vertical polarization, so it can be made a two-dimensional problem of TM waves only.

[0059] FIG. 4 is a schematic diagram showing the procedure for converting experimental data into simulation data. Here, the scattered electric field E S (ω) is the total electric field E of the target t (ω) to the incident electric field E i (ω) is subtracted. As shown in Fig. 4, first, two calibration targets with known permittivity data are prepared: one with a cavity in the background medium, concrete, and the other with no cavity (background medium). In Fig. 4, the known calibration target is E S,1 esp is shown as:

[0060] Then, by simulating using the FDTD method, the total electric field E t (ω) is obtained, and the incident electric field E i (ω) is obtained. The simulation results are shown as E S,1 sim It is expressed as: As already mentioned, FDTD is one method for obtaining simulation data, and there are other methods for obtaining simulation data, such as the method of moments and the multipole field method, in addition to FDTD.

[0061] Next, the correction coefficient ζ(ω) is calculated from Equation 6 using the experimental data and the simulation results using the FDTD method.

number

[0062] Next, Unknown Target E s,2 exp For example, an unknown target is prepared with cavities #1 to #4 in the background medium, concrete. The data measured in the experiment for this unknown target is E S,2 exp It is expressed as: And experimental data E, which measured the unknown target cavities #1 to #4 S,2 expBy correcting the coefficient ζ(ω) calculated by Equation 6, the simulation data E S,2 sim (See Equation 7). In FIG. 4, subscript number “1” is used for known calibration targets and simulation results, and subscript number “2” is used for unknown targets and simulation results.

[0063]

number

[0064] <Explanation of the experimental model> In the experiment conducted by the inventors, a dipole antenna with vertical linear polarization was used as the transmitting and receiving antenna. A VNA (Vector Network Analyzer) with an effective frequency band of 1.0 GHz to 2.6 GHz was used as the measuring device, and data was acquired at 5° intervals by rotating the target. Measurements and simulations were performed for the target object with no cavity, one cavity, and four cavities. In this experimental model, the electromagnetic field data was measured when the radio wave from the transmitting antenna penetrated the target object and reached the receiving antenna. t The electromagnetic field data of the radio waves reflected from the target object is sent to the receiving antenna r r This differs from the method used to measure

[0065] Figures 5(a) to (c) show the dielectric constant distribution of the reconstruction results (one target cavity) based on experimental data. In this experiment, the MR (Multiplicative Regularized)-CSI method was used. The frequency used was 1.98 GHz, and the number of iterations was 20,000. The long, thin bar on the right is a color bar that shows the difference in dielectric constant with color. As you go up the color bar, the colors become brighter, indicating a higher dielectric constant, and as you go down, the colors become darker, indicating a lower dielectric constant.

[0066] In Figure 5(a), the dark black area (low permittivity) around the circle surrounded by a square represents the estimated permittivity image of the air surrounding the concrete cylinder, and the large circular area in the center represents the estimated permittivity image of the cylindrical concrete. The small circular area in the center shows the estimated permittivity image of the cavity in the concrete, which is the background medium. As shown in Figure 5(a), the diameter of the cylindrical concrete used in the experiment is approximately 300 mm. Furthermore, in Figure 5(a), it can be seen that the estimated permittivity image of the cavity in the concrete, which is the background medium, is significantly smaller than that of the surrounding concrete.

[0067] Figures 5(b) and (c) show the reconstructed distribution of the cavity, enlarging the central cavity in Figure 5(a). Figure 5(b) shows the true value of the dielectric constant, and Figure 5(c) shows the reconstruction result by simulation. As can be seen from the reconstruction result in Figure 5(c), the dielectric constant estimation image shows a value close to ε=6, even though it is a cavity. The cause of the error from the true value of the reconstruction result in Figure 5(c) is thought to be due to the fact that the solution converged to a local optimum solution in the cost function optimization in CSI, nonlinear interactions between the target and unwanted reflectors, and correction errors in the calibration.

[0068] 6(a) to (c) show the dielectric constant distribution of the reconstruction results (four target cavities) based on the experimental data. Figure 6(a) shows an image of the estimated dielectric constant of the entire concrete, including cavities #1 to #4. In the experiment in Figure 6, the specified frequency was set to 2.21 GHz, but other than that, it was the same as the experiment with one cavity in Figure 5. Note that cavities #1 to #4 were artificially opened for this experiment, so all of them are cavities with a diameter of approximately 10 mm.

[0069] Figure 6(b) shows the true dielectric constant distribution of cavities #1 to #4 in Figure 6(a), and Figure 6(c) shows the reconstruction results, i.e., the calibration results, of cavities #1 to #4 by simulation. The positions and sizes of cavities #1 to #4 shown in Figures 6(b) and (c) are the same as those of cavities #1 to #4 shown in Figure 6(a), and are approximately 10 mm in both length and width on the xy plane.

[0070] <Flow of target shape estimation process and permittivity acquisition process> FIG. 7 is a flowchart showing a process flow for estimating the position of a target and obtaining a dielectric constant estimation image in the non-contact shape and dielectric constant measuring apparatus 10 of this embodiment. In the experimental model of the present invention, two materials with known dielectric constants are prepared: one is a background material with no target, e.g., only concrete, and the other is a background material such as concrete with a target, e.g., a cavity, in it.

[0071] First, an electromagnetic field signal for an object (concrete) without a target having a known dielectric constant is acquired (step S1) by the transmitting / receiving antenna, microwave receiving processor 12, baseband processor 13, and filter processor 14 shown in Fig. 1. Next, an electromagnetic field signal for an object having a target (e.g., a cavity) with a known dielectric constant in concrete is acquired (step S2). Steps S1 and S2 are experimental signal acquisition without a target and signal acquisition with a target.

[0072] A differential signal between the electromagnetic field signal without a target acquired in step S1 and the electromagnetic field signal with a target acquired in step S2 is calculated by the experimental data differential signal acquisition unit 15 shown in FIG. 1 (step S3). Here, the electromagnetic field signal without a target obtained by the experiment is E I,1 exp The experimental electromagnetic field signal E T,1 exp Then, in step S3, E S,1 exp =E T,1 exp -E I,1 exp This formula corresponds to the data equation shown in formula 2, and the scattered electric field shown in FIG.

[0073] Next, the target-free electromagnetic field signal acquired in step S1 is converted into simulation data in the electromagnetic field calculation data output unit 16 shown in Fig. 1. That is, a target-free signal is acquired by simulation (step S4). Next, the target-free electromagnetic field signal acquired in step S2 is also converted into simulation data by the electromagnetic field calculation data output unit 16. That is, a target-free signal is acquired by simulation (step S5). The simulation in the electromagnetic field calculation data output unit 16 shown in Fig. 1 is performed using the method described in Fig. 3.

[0074] The simulation signal without a target obtained in step S4 and the simulation signal with a target obtained in step S5 are sent to the simulation data difference signal acquisition unit 17 in FIG. 1, where a difference signal between the simulation signal without a target and the simulation signal with a target is calculated (step S6). Here, the simulation signal without a target is E I,1 sim Then, the simulation signal E T,1 sim Then, in step S6, E S,1 sim =E T,1 sim -E I,1 sim is calculated.

[0075] Next, a correction coefficient ζ(ω) is calculated from the ratio of the difference signal between the signal without a target and the signal with a target obtained from the experimental data in step S3 to the difference signal between the signal without a target and the signal with a target obtained from the simulation in step S6 (step S7). This correction coefficient ζ(ω) is calculated using Equation 6.

[0076] The processing procedure from steps S1 to S6 up to the calculation of the correction coefficient ζ(ω) in step S7 uses data without a target and data with a target whose dielectric constant is known. Next, the procedure for measuring the dielectric constant of an unknown target using the correction coefficient ζ(ω) calculated in step S7 will be described.

[0077] First, the electromagnetic field signal E for an unknown target (see, for example, Fig. 6(a)) is calculated from experimental data. T,2 exp (Step S8). This electromagnetic field signal E T,2 exp The measurement of the electromagnetic field signal E is performed by the scattered electric field acquisition unit, which is composed of the microwave reception processing unit 12, the baseband processing unit 13, and the filter processing unit 14 in FIG. 1, in the same manner as the measurement of a known target. Then, in step S8, the electromagnetic field signal E T,2 exp From the above, the electromagnetic field signal E of the background medium without the target obtained in step S1 I,1 exp The value E with S,2 exp That is, in step S8, E calculated by the experimental data difference signal acquisition unit 15 is S,2 exp =E T,2 exp -E I,1 exp will be output.

[0078] The scattered electric field signal E including the unknown target obtained in step S8 S,2 exp In the simulation data conversion unit 19 shown in FIG. S,2 sim In the conversion of the simulation data in step S9, the correction coefficient ζ(ω) obtained in step S7 is used as shown in Equation 7.

[0079] Next, the simulation data E calculated in step S9 S,2 sim(Step S10). The CSI is performed by the dielectric constant image acquisition unit 20 in FIG. By executing this CSI, the permittivity image acquiring unit 20 obtains the complex permittivity distribution χ(ω,r′) (step S11). The above is the procedure for measuring the complex permittivity distribution using the simulation data for the experimental data of the unknown target.

[0080] Although the embodiment of the present invention has been described in detail above, it goes without saying that the present invention is not limited to the embodiment described here, and includes other applications and modifications as long as they do not deviate from the gist of the present invention described in the claims. [Explanation of symbols]

[0081] 10...non-contact shape and dielectric constant measuring device, 11...microwave transmission processing unit, 12...microwave reception processing unit, 13...baseband processing unit, 14...filter processing unit, 15...experimental data difference signal acquisition unit, 16...electromagnetic field calculation data output unit, 17...simulation data difference signal acquisition unit, 18...calibration coefficient calculation unit, 19...simulation data conversion unit, 20...dielectric constant image acquisition unit, 21...image output unit

Claims

1. A non-contact shape and dielectric constant measuring device that transmits electromagnetic waves from a transmitting antenna toward a target whose dielectric constant is to be measured, receives electromagnetic waves reflected from the target or transmitted through the target with a receiving antenna, and obtains the shape and dielectric constant of the target from the received electromagnetic waves, a scattered electric field acquiring unit that acquires, as experimental data, an electric field scattered by a target of the electromagnetic wave transmitted toward the target from the electromagnetic wave received by the receiving antenna; an electromagnetic field calculation data output unit that calculates a scattered electric field in a simulation environment that simulates an experimental environment separately from the experimental data; a calibration coefficient calculation unit that obtains a correction coefficient from experimental data obtained from the scattered electric field of a target having a known dielectric constant acquired by the scattered electric field acquisition unit and simulation data obtained by simulating the scattered electric field of the known target calculated by the electromagnetic field calculation data output unit; a simulation data conversion unit that converts experimental data obtained from the scattered electric field of a target having an unknown dielectric constant, acquired by the scattered electric field acquisition unit, into simulation data of the target having an unknown dielectric constant by dividing the experimental data obtained from the scattered electric field of the target having an unknown dielectric constant by the correction coefficient acquired by the calibration coefficient calculation unit; a dielectric constant image acquisition unit that acquires a complex dielectric constant distribution image by a CSI method using the simulation data converted by the simulation data conversion unit. Non-contact shape and dielectric constant measuring device.

2. an experimental data difference signal acquisition unit that calculates a difference between the scattered electric field of a background medium having a known dielectric constant acquired by the scattered electric field acquisition unit and the scattered electric field from an object including a target having a known dielectric constant; a simulation data difference signal acquisition unit that calculates a difference between a scattered electric field of a background medium having a known dielectric constant and a scattered electric field from an object including a target having a known dielectric constant in a simulation environment obtained by simulating an experimental environment separately from the experimental data; The calibration coefficient calculation unit is supplied with a difference signal between experimental data with and without a target calculated by the experimental data difference signal acquisition unit, and a difference signal between simulation data with and without a target calculated by the simulation data difference signal acquisition unit. The non-contact shape and dielectric constant measuring device according to claim 1 .

3. In the CSI used in the dielectric constant image acquisition unit, the cost function F shown below is defined, and the contrast source w(r), the total electric field Et(r), and the contrast function χ(r) are restored so that the cost function F is minimized.

3. The non-contact shape and dielectric constant measuring device according to claim 1 or 2.

4. The contrast source w(r) is a function that is the product of the total electric field Et(r) and the contrast function χ(r), with both the contrast function χ(r) and the total electric field Et(r) being set as optimization variables. The non-contact shape and dielectric constant measuring device according to claim 3 .

5. The scattered electric field acquisition unit is a reception processing unit that processes electromagnetic waves received after being reflected from or transmitted through a target; a baseband processing unit that obtains a received signal component from the baseband signal obtained by the reception processing unit; a filter processing unit that extracts a reflected wave or a transmitted wave component from the received signal component obtained by the baseband processing unit.

3. The non-contact shape and dielectric constant measuring device according to claim 1 or 2.

6. A step of transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from a background medium having a known dielectric constant, not including a target, at a receiving antenna to obtain an experimental scattered electric field signal; A step of transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from an object including a target having a known dielectric constant in a background medium by a receiving antenna to obtain an experimental scattered electric field signal; subtracting the scattered electric field signal from an experiment that does not include the target from the scattered electric field signal from an experiment that includes the target to obtain a difference signal; calculating, by simulation, a scattered electric field signal of a background medium having a known dielectric constant and not including the target; calculating, by simulation, a scattered electric field signal of an object including a target with a known dielectric constant in a background medium; subtracting a simulated electromagnetic field signal that does not include the target from a simulated electromagnetic field signal that includes the target to obtain a difference signal; dividing the experimental target-without-target difference signal by the simulated target-without-target difference signal to obtain a correction factor; A step of transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from an object including a target having an unknown dielectric constant by a receiving antenna to obtain an experimental electromagnetic field signal; converting experimental electromagnetic field signals obtained from an object, including a target with an unknown dielectric constant, into simulation data using the correction factors; performing CSI using the converted simulation data to obtain a complex permittivity distribution of objects including unknown targets; A non-contact shape and dielectric constant measurement method comprising:

7. A procedure of transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from a background medium having a known dielectric constant, which does not include a target, to obtain an experimental electromagnetic field signal; A procedure of transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from an object including a target having a known dielectric constant in a background medium by a receiving antenna to obtain an experimental electromagnetic field signal; subtracting an electromagnetic field signal from an experiment that does not include the target from an electromagnetic field signal from an experiment that includes the target to obtain a difference signal; A procedure for calculating, by simulation, an electromagnetic field signal of a background medium having a known dielectric constant and not including a target; calculating, by simulation, an electromagnetic field signal of an object containing a target with a known dielectric constant in a background medium; a step of subtracting a simulated electromagnetic field signal not including the target from a simulated electromagnetic field signal including the target to obtain a difference signal; Dividing the experimental difference signal between with and without target by the simulated difference signal between with and without target to obtain a correction factor; A procedure for obtaining an experimental electromagnetic field signal by transmitting an electromagnetic wave from a transmitting antenna and receiving a reflected wave or a transmitted wave from an object including a target having an unknown dielectric constant with a receiving antenna; converting experimental electromagnetic field signals obtained from an object, including a target with an unknown dielectric constant, into simulation data using said correction factors; performing CSI using the converted simulation data to obtain a complex permittivity distribution of an object including an unknown target; A non-contact shape and dielectric constant measurement program that causes a computer to execute the above.

Images