WO2022264342A1 - Depth estimation device, depth estimation method, and depth estimation program - Google Patents

Depth estimation device, depth estimation method, and depth estimation program Download PDF

Info

Publication number
WO2022264342A1
WO2022264342A1 PCT/JP2021/022937 JP2021022937W WO2022264342A1 WO 2022264342 A1 WO2022264342 A1 WO 2022264342A1 JP 2021022937 W JP2021022937 W JP 2021022937W WO 2022264342 A1 WO2022264342 A1 WO 2022264342A1
Authority
WO
WIPO (PCT)
Prior art keywords
coordinates
local
relative permittivity
point
buried object
Prior art date
Application number
PCT/JP2021/022937
Other languages
French (fr)
Japanese (ja)
Inventor
和彦 村崎
慎吾 安藤
潤 島村
Original Assignee
日本電信電話株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電信電話株式会社 filed Critical 日本電信電話株式会社
Priority to JP2023528867A priority Critical patent/JPWO2022264342A1/ja
Priority to PCT/JP2021/022937 priority patent/WO2022264342A1/en
Publication of WO2022264342A1 publication Critical patent/WO2022264342A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications

Definitions

  • the disclosed technique relates to a depth estimation device, a depth estimation method, and a depth estimation program.
  • Non-Patent Literature 1 discloses that machine learning can detect a location with a buried object reaction from ground penetrating radar data.
  • the propagation speed of the electromagnetic wave or the relative dielectric constant in the ground is used to know how deep the reaction site actually occurs in the ground. calculation is required. Since it is generally difficult to grasp the relative permittivity of the object to be measured in detail, either a rough estimate is used or an engineer sets the relative permittivity visually.
  • Non-Patent Document 2 discloses that the relative permittivity can be automatically estimated by evaluating the result of migration processing on GPR data using a numerical index. It has been shown that the relative permittivity can be estimated with particularly high accuracy in a low-noise environment.
  • Non-Patent Document 2 assumes that the entire measurement target has a uniform dielectric constant, and that the reaction of buried objects is small. In a general road environment, the properties of the soil are not uniform and many buried objects are mixed. It's becoming
  • An object of the present invention is to provide a depth estimation device, a depth estimation method, and a depth estimation program.
  • a first aspect of the present disclosure is a depth estimation device based on ground penetrating radar data, which is two-dimensional data obtained by measuring an embedded object buried in the ground, and a previously obtained range of relative permittivity.
  • a migration unit that performs a predetermined migration process while changing the relative dielectric constant according to the range, and obtains a migration result for each changed relative dielectric constant;
  • Each local point that is either a local maximum point or a local minimum point is extracted, and with respect to the local point, the relative permittivity in the migration result, the traveling direction of the antenna that measured the ground penetrating radar data an extremity information extraction unit for extracting three-dimensional data of coordinates and reflection time coordinates;
  • a buried object detection unit for detecting the coordinates of the buried object using a pre-learned detector; Refer to the coordinates of the local points of the three-dimensional data in the peripheral coordinates based on the relative permittivity corresponding to the local points, adopt the relative permittivity, the reflection time and and and a depth
  • a second aspect of the present disclosure is a depth estimation method based on ground penetrating radar data, which is two-dimensional data obtained by measuring an object buried underground, and a previously obtained range of relative permittivity. Then, a predetermined migration process is performed while changing the relative dielectric constant according to the range, the migration result for each changed relative dielectric constant is obtained, and from the migration result, the maximum point of the migrated value or the maximum point and Each of the local points that are any of the local minimum points is extracted, and with respect to the local points, the relative permittivity in the migration result, the coordinates of the traveling direction of the antenna that measured the ground penetrating radar data, and 3D data of the coordinates of the reflection time are extracted, the coordinates of the buried object are detected using a pre-trained detector, and the local coordinates of the 3D data in the peripheral coordinates based on the coordinates of the buried object are detected. referring to the coordinates of the point, adopting the dielectric constant corresponding to said local point, and estimating the depth of the
  • a third aspect of the present disclosure is a depth estimation program based on ground penetrating radar data, which is two-dimensional data obtained by measuring an embedded object buried in the ground, and a previously obtained range of relative permittivity. Then, a predetermined migration process is performed while changing the relative dielectric constant according to the range, the migration result for each changed relative dielectric constant is obtained, and from the migration result, the maximum point of the migrated value or the maximum point and Each of the local points that are any of the local minimum points is extracted, and with respect to the local points, the relative permittivity in the migration result, the coordinates of the traveling direction of the antenna that measured the ground penetrating radar data, and 3D data of the coordinates of the reflection time are extracted, the coordinates of the buried object are detected using a pre-trained detector, and the local coordinates of the 3D data in the peripheral coordinates based on the coordinates of the buried object are detected. referring to the coordinates of the point, adopting the dielectric constant corresponding to said local point, and estimating the
  • FIG. 2 is a block diagram showing the hardware configuration of a depth estimation device according to an embodiment of the present disclosure
  • FIG. 1 is a block diagram showing a functional configuration of a depth estimation device according to an embodiment of the present disclosure
  • FIG. It is an example of a processing result obtained by migration. It is an example of the migration result according to the dielectric constant. It is an example of three-dimensional data in the dielectric constant direction, the antenna traveling direction, and the reflection time direction.
  • FIG. 4 is a diagram showing a configuration related to an embedded object detection unit;
  • the depth to which each buried object is buried can be obtained in a realistic scene where the soil to be measured is not uniform and multiple buried objects are buried. is the subject.
  • an appropriate dielectric constant is estimated for each buried object. Assuming that the average relative permittivity of each buried object is constant and estimating it, we can obtain It realizes buried depth estimation in complex environment. Specifically, migration processing is applied to the entire data while changing the dielectric constant within an appropriate range, and the coordinate value and dielectric constant that maximize the migrated value are extracted. The relative permittivity obtained at the point of this maximum (maximum point) can be regarded as an appropriate average relative permittivity at that coordinate value. Next, the approximate position of the buried object is detected from the GPR data using a previously machine-learned buried object detector.
  • GPR data is two-dimensional data obtained by measuring a buried object buried in the ground.
  • GPR data is an example of ground penetrating radar data of this disclosure.
  • the method of the embodiment of the present disclosure it is possible to automatically estimate the depth of a buried object without requiring the know-how of an engineer for GPR data measured on a general road where the soil environment is not maintained. .
  • the buried object can be maintained and managed based on the three-dimensional position, contributing to a reduction in maintenance costs such as construction planning.
  • FIG. 1 is a block diagram showing the hardware configuration of the depth estimation device 100 according to the embodiment of the present disclosure.
  • the depth estimation device 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and communication It has an interface (I/F) 17 .
  • Each component is communicatively connected to each other via a bus 19 .
  • the CPU 11 is a central processing unit that executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 performs control of each configuration and various arithmetic processing according to programs stored in the ROM 12 or the storage 14 . In this embodiment, the ROM 12 or storage 14 stores a depth estimation program.
  • the ROM 12 stores various programs and various data.
  • the RAM 13 temporarily stores programs or data as a work area.
  • the storage 14 is configured by a storage device such as a HDD (Hard Disk Drive) or SSD (Solid State Drive), and stores various programs including an operating system and various data.
  • HDD Hard Disk Drive
  • SSD Solid State Drive
  • the input unit 15 includes a pointing device such as a mouse and a keyboard, and is used for various inputs.
  • the display unit 16 is, for example, a liquid crystal display, and displays various information.
  • the display unit 16 may employ a touch panel system and function as the input unit 15 .
  • the communication interface 17 is an interface for communicating with other devices such as terminals.
  • the communication uses, for example, a wired communication standard such as Ethernet (registered trademark) or FDDI, or a wireless communication standard such as 4G, 5G, or Wi-Fi (registered trademark).
  • FIG. 2 is a block diagram showing the functional configuration of the depth estimation device 100 according to the embodiment of the present disclosure.
  • Each functional configuration is realized by the CPU 11 reading a depth estimation program stored in the ROM 12 or the storage 14, developing it in the RAM 13, and executing it.
  • the depth estimation device 100 includes a detector storage unit 102, a migration unit 110, an extremity information extraction unit 112, a buried object detection unit 114, and a depth estimation unit 116. It is configured. Migration processing is applied to the input GPR data in the migration unit 110, and the local information extraction unit 112 calculates the maximum coordinate value and relative dielectric constant for the output.
  • the embedded object detection unit 114 applies an embedded object detector to the GPR data, specifies the reaction site of the embedded object, and detects the position.
  • the depth estimating unit 116 picks up local maximum points around the position of the buried object reaction site, and picks up the dielectric constant of the local maximum point that maximizes the result of migration. The depth is calculated using this dielectric constant and output as the depth of the buried object.
  • the migration unit 110 performs migration processing while changing the dielectric constant according to the range based on the GPR data and the range of the dielectric constant obtained in advance, and outputs the migration result for each changed dielectric constant. demand.
  • the range of the relative permittivity is, for example, if the target is general soil, the relative permittivity changes from about 2 to 30 depending on the degree of moisture.
  • the range of dielectric constants that the object can take is set in advance.
  • a specific dielectric constant is taken out from this set range at regular intervals, and a migration process is applied.
  • the migration process is a process of correcting the parabolic reflected waveform that appears due to the directivity of the antenna based on the relative permittivity so that only the reflection directly below the antenna can be captured.
  • Kirchhoff migration method there is a Kirchhoff migration method as a representative migration processing method, other migration methods may be used.
  • FIG. 3 shows an example of processing results obtained by Kirchhoff migration.
  • the local information extraction unit 112 extracts each local maximum point from the migration result, and for each of the extracted local maximum points, the relative permittivity of the migration result, the coordinates of the traveling direction of the antenna for which the GPR data was measured, and the reflection time. Extract the three-dimensional data of the coordinates.
  • a local maximum is an example of a local point of this disclosure.
  • the local information extraction unit 112 assumes that the reaction is maximum at the center point of the reflection location for the migration result, and extracts the point where the migrated value is maximum (maximum value) as the local maximum point.
  • the migration results corresponding to the changing relative permittivity obtained in FIG. 4 are accumulated to form three-dimensional data as shown in FIG.
  • the three-dimensional data values are obtained for each local maximum point in each of the dielectric constant direction, the antenna traveling direction, and the reflection time direction.
  • the relative permittivity direction the relative permittivity in the migration result with the relative permittivity changed is obtained, in the antenna traveling direction, the coordinates of the antenna are obtained, and in the reflection time advancing direction, the coordinates of the reflection time are obtained.
  • the coordinates that take the maximum value are calculated based on this three-dimensional data, and the dielectric constant that maximizes the aggregated reaction for each buried object is obtained.
  • Antenna coordinates and reflection times are obtained from GPR data. Since the obtained local maximum coordinates are three-dimensional coordinates (relative permittivity, antenna traveling direction coordinates, reflection time coordinates), three values are obtained for each local maximum point.
  • the maximum value should be used, but when the GPR data is represented by real waves, negative values are aggregated. Since extreme values also occur, local minimum values may also be used at the same time. If the GPR data were complex waves, only maxima would be obtained, and if the GPR data were real waves, maxima and minima would be obtained.
  • the buried object detection unit 114 uses the detector of the detector storage unit 102 to detect the coordinates of the buried object in the GPR data.
  • the detector is trained to detect the position of the buried object by applying a machine learning technique used for image recognition with the original GPR data as input.
  • the detector storage unit 102 stores detector parameters learned in advance by a machine learning technique.
  • the configuration related to the buried object detection unit 114 is shown in FIG. Since detector parameters learned in advance are required for detection, GPR data for learning and learning data of annotation data of the corresponding buried object positions are prepared in advance, and the detector learning unit 104 learns the learning data. Train the detector to minimize the error between the correct answer and the output. Regarding the machine learning model and learning method used for learning and detection, for example, high detection accuracy can be expected when learning by error backpropagation using a detection model based on a convolutional neural network as used in Non-Patent Document 1. . In addition, even if it is a method other than this, you may use it. By applying the detector thus obtained, it is possible to obtain coordinate values indicating the rough position of each buried object appearing in the GPR data.
  • the depth estimator 116 is composed of a dielectric constant estimator 120 and a depth calculator 122, as shown in FIG.
  • the depth estimating unit 116 refers to the coordinates of the maximum point in the peripheral coordinates based on the coordinates of the buried object, adopts the relative permittivity corresponding to the maximum point, and adopts the relative permittivity and Estimate the depth of the buried object based on the reflection time.
  • the depth estimating unit 116 receives as input the coordinates of the maximum point and its maximum value output from the extremity information extraction unit 112 and the coordinates of the buried object output from the buried object detection unit 114 .
  • Relative permittivity estimator 120 limits the search range for the maximum value of the migration result to the surrounding coordinates for the coordinates of each buried object.
  • Peripheral coordinates shall be defined within ⁇ T X meters in the antenna traveling direction and within ⁇ TT seconds in the reflection time direction based on the coordinates of the buried object.
  • Search for local maxima in peripheral coordinates Refer to the coordinates of the searched maximal point and extract the maximal value. The coordinate at which the maximum value is maximum among the extracted maximum values is referred to, and the dielectric constant at the maximum point corresponding to the coordinate is adopted as the dielectric constant corresponding to the buried object.
  • the GPR data is a wave of real numbers, since the maximum and minimum values are obtained, the maximum and minimum points are searched for in the peripheral coordinates.
  • the maximum value and the minimum value are determined, respectively, and the average value of the relative permittivity corresponding to the maximum point and the relative permittivity corresponding to the minimum point is adopted. Also, the corresponding reflection time is adopted in the three-dimensional data including the adopted dielectric constant.
  • the depth calculator 122 calculates the detailed depth of the buried object using the estimated dielectric constant corresponding to each buried object. Using the reflection time t and the dielectric constant ⁇ r, the depth d is obtained by the following equation (1). ... (1)
  • FIG. 8 is a flowchart showing the flow of depth estimation processing by the depth estimation device 100 according to the embodiment of the present disclosure.
  • Depth estimation processing is performed by the CPU 11 reading a depth estimation program from the ROM 12 or the storage 14, developing it in the RAM 13, and executing it. The following processes are executed by the CPU 11 functioning as each part of the depth estimation device 100 .
  • step S100 the CPU 11 performs the migration process while changing the relative permittivity according to the range based on the GPR data and the range of the relative permittivity obtained in advance. Seek results.
  • step S102 the CPU 11 extracts each local maximum point from the migration result.
  • step S104 the CPU 11 extracts the three-dimensional data of the relative permittivity of the migration result, the coordinates of the traveling direction of the antenna that measured the GPR data, and the coordinates of the reflection time for each of the extracted local maximum points.
  • step S106 the CPU 11 uses the detector of the detector storage unit 102 to detect the coordinates of the buried object in the GPR data.
  • step S108 the CPU 11 refers to the coordinates of the local maximum point in the peripheral coordinates based on the coordinates of the buried object, and adopts the dielectric constant corresponding to the local maximum point.
  • step S110 the CPU 11 estimates the depth of the buried object based on the adopted dielectric constant and reflection time.
  • the depth at which the buried object is buried can be obtained using the relative permittivity.
  • the technique of the present disclosure can be said to be a method of automatically estimating the dielectric constant using the migration results.
  • the technique of the present disclosure can be said to be a method of automatically estimating the dielectric constant using the migration results.
  • the depth estimation processing executed by the CPU reading the software (program) in the above embodiment may be executed by various processors other than the CPU.
  • the processor is a PLD (Programmable Logic Device) whose circuit configuration can be changed after manufacturing, such as an FPGA (Field-Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit) to execute specific processing.
  • a dedicated electric circuit or the like which is a processor having a specially designed circuit configuration, is exemplified.
  • the depth estimation process may be performed by one of these various processors, or by a combination of two or more processors of the same or different type (e.g., multiple FPGAs and a combination of CPU and FPGA). combination, etc.). More specifically, the hardware structure of these various processors is an electric circuit in which circuit elements such as semiconductor elements are combined.
  • the depth estimation program has been pre-stored (installed) in the storage 14, but the present invention is not limited to this.
  • Programs are stored in non-transitory storage media such as CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), and USB (Universal Serial Bus) memory.
  • CD-ROM Compact Disk Read Only Memory
  • DVD-ROM Digital Versatile Disk Read Only Memory
  • USB Universal Serial Bus
  • (Appendix 1) memory at least one processor connected to the memory; including The processor Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity.
  • a migration unit that performs while changing and obtains the migration result for each changed dielectric constant; From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, Extracting three-dimensional data of the coordinates of the direction of travel of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time, detecting the coordinates of the buried object using a pre-learned detector; referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and the reflection time, estimating the depth of the buried object;
  • a depth estimator configured to:
  • Appendix 2 A non-transitory storage medium storing a program executable by a computer to perform a depth estimation process, Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity.
  • a migration unit that performs while changing and obtains the migration result for each changed dielectric constant; From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, Extracting three-dimensional data of the coordinates of the direction of travel of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time, detecting the coordinates of the buried object using a pre-learned detector; referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and the reflection time, estimating the depth of the buried object; Non-transitory storage media.
  • Depth estimation device 100 Depth estimation device 102 Detector storage unit 104 Detector learning unit 110 Migration unit 112 Local information extraction unit 114 Buried object detection unit 116 Depth estimation unit 120 Relative permittivity estimation unit 122 Depth calculation unit

Abstract

This invention is capable of using relative permittivity to determine the depth at which a buried object is buried.　In this depth estimation device, a migration unit carries out migration while changing a relative permittivity according to a relative permittivity range and determines migration results for each changed relative permittivity. An extremity information extraction unit extracts three-dimensional data for an extreme point. A buried object detection unit uses a detection device that has been trained in advance to detect the coordinates of a buried object. Referring to the coordinates of the extreme point in three-dimensional data for nearby coordinates based on the coordinates of the buried object, a depth estimation unit adopts the relative permittivity corresponding to the extreme point and estimates the depth of the buried object on the basis of the adopted relative permittivity and a reflection time.

Description

深さ推定装置、深さ推定方法、及び深さ推定プログラムDepth estimation device, depth estimation method, and depth estimation program
　開示の技術は、深さ推定装置、深さ推定方法、及び深さ推定プログラムに関する。 The disclosed technique relates to a depth estimation device, a depth estimation method, and a depth estimation program.
　　近年、画像又は音声などのデジタルデータを機械学習手法によって自動認識する技術が多数考案されているが、地中レーダデータについても同様の方法で学習及び認識が可能であることが知られている。非特許文献１では、機械学習によって地中レーダデータから埋設物反応のある箇所を検出できることが示されている。しかし、ＧＰＲデータ中の反応箇所を認識できていてもその反応箇所が実際にどの程度の深さの地中で起きているのかを知るには地中における電磁波の伝搬速度あるいは比誘電率を用いた計算が必要となる。一般的に計測対象の比誘電率を詳細に把握することは難しいため、おおまかな推定値を用いるか目視によって技術者が設定するといった対応が取られている。これに対して、非特許文献２では、ＧＰＲデータに対するマイグレーション処理の結果を数値的な指標によって評価することで比誘電率の自動推定ができることが示されている。ノイズの小さい環境においては特に高い精度で比誘電率を推定できることが示されている。 In recent years, many technologies have been devised to automatically recognize digital data such as images or sounds using machine learning methods, and it is known that ground penetrating radar data can also be learned and recognized in the same way. Non-Patent Literature 1 discloses that machine learning can detect a location with a buried object reaction from ground penetrating radar data. However, even if the reaction site can be recognized in the GPR data, the propagation speed of the electromagnetic wave or the relative dielectric constant in the ground is used to know how deep the reaction site actually occurs in the ground. calculation is required. Since it is generally difficult to grasp the relative permittivity of the object to be measured in detail, either a rough estimate is used or an engineer sets the relative permittivity visually. On the other hand, Non-Patent Document 2 discloses that the relative permittivity can be automatically estimated by evaluating the result of migration processing on GPR data using a numerical index. It has been shown that the relative permittivity can be estimated with particularly high accuracy in a low-noise environment.
　非特許文献２のアプローチは計測対象全体が一様な比誘電率を持つことを仮定しており、また埋設物反応も少数であることを想定している。一般的な路上環境では、土壌の性質が画一的でなくまた埋設物も多数混在しており、このような想定が成り立たないため、ＧＰＲデータからの埋設物深さの推定は未だ難しい問題となっている。 The approach of Non-Patent Document 2 assumes that the entire measurement target has a uniform dielectric constant, and that the reaction of buried objects is small. In a general road environment, the properties of the soil are not uniform and many buried objects are mixed. It's becoming
　開示の技術は、上記の点に鑑みてなされたものであり、比誘電率を用いて埋設物が埋設された深さを求めることができる。深さ推定装置、深さ推定方法、及び深さ推定プログラムを提供することを目的とする。 The disclosed technology has been made in view of the above points, and can determine the depth at which the buried object is buried using the dielectric constant. An object of the present invention is to provide a depth estimation device, a depth estimation method, and a depth estimation program.
　本開示の第１態様は、深さ推定装置であって、地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求めるマイグレーション部と、前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出する極所情報抽出部と、予め学習された検出器を用いて前記埋設物の座標を検出する埋設物検出部と、前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する深さ推定部と、を含む。 A first aspect of the present disclosure is a depth estimation device based on ground penetrating radar data, which is two-dimensional data obtained by measuring an embedded object buried in the ground, and a previously obtained range of relative permittivity. a migration unit that performs a predetermined migration process while changing the relative dielectric constant according to the range, and obtains a migration result for each changed relative dielectric constant; Each local point that is either a local maximum point or a local minimum point is extracted, and with respect to the local point, the relative permittivity in the migration result, the traveling direction of the antenna that measured the ground penetrating radar data an extremity information extraction unit for extracting three-dimensional data of coordinates and reflection time coordinates; a buried object detection unit for detecting the coordinates of the buried object using a pre-learned detector; Refer to the coordinates of the local points of the three-dimensional data in the peripheral coordinates based on the relative permittivity corresponding to the local points, adopt the relative permittivity, the reflection time and and a depth estimator that estimates the depth of the buried object based on.
　本開示の第２態様は、深さ推定方法であって、地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求め、前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出し、予め学習された検出器を用いて前記埋設物の座標を検出し、前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する、処理をコンピュータに実行させる。 A second aspect of the present disclosure is a depth estimation method based on ground penetrating radar data, which is two-dimensional data obtained by measuring an object buried underground, and a previously obtained range of relative permittivity. Then, a predetermined migration process is performed while changing the relative dielectric constant according to the range, the migration result for each changed relative dielectric constant is obtained, and from the migration result, the maximum point of the migrated value or the maximum point and Each of the local points that are any of the local minimum points is extracted, and with respect to the local points, the relative permittivity in the migration result, the coordinates of the traveling direction of the antenna that measured the ground penetrating radar data, and 3D data of the coordinates of the reflection time are extracted, the coordinates of the buried object are detected using a pre-trained detector, and the local coordinates of the 3D data in the peripheral coordinates based on the coordinates of the buried object are detected. referring to the coordinates of the point, adopting the dielectric constant corresponding to said local point, and estimating the depth of the buried object based on the adopted dielectric constant and said reflection time; run on the computer.
　本開示の第３態様は、深さ推定プログラムであって、地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求め、前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出し、予め学習された検出器を用いて前記埋設物の座標を検出し、前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する、処理をコンピュータに実行させる。 A third aspect of the present disclosure is a depth estimation program based on ground penetrating radar data, which is two-dimensional data obtained by measuring an embedded object buried in the ground, and a previously obtained range of relative permittivity. Then, a predetermined migration process is performed while changing the relative dielectric constant according to the range, the migration result for each changed relative dielectric constant is obtained, and from the migration result, the maximum point of the migrated value or the maximum point and Each of the local points that are any of the local minimum points is extracted, and with respect to the local points, the relative permittivity in the migration result, the coordinates of the traveling direction of the antenna that measured the ground penetrating radar data, and 3D data of the coordinates of the reflection time are extracted, the coordinates of the buried object are detected using a pre-trained detector, and the local coordinates of the 3D data in the peripheral coordinates based on the coordinates of the buried object are detected. referring to the coordinates of the point, adopting the dielectric constant corresponding to said local point, and estimating the depth of the buried object based on the adopted dielectric constant and said reflection time; run on the computer.
　開示の技術によれば、比誘電率を用いて埋設物が埋設された深さを求めることができる。 According to the disclosed technology, it is possible to obtain the depth of the buried object using the dielectric constant.
本開示の実施形態の深さ推定装置のハードウェア構成を示すブロック図である。2 is a block diagram showing the hardware configuration of a depth estimation device according to an embodiment of the present disclosure; FIG. 本開示の実施形態の深さ推定装置の機能的な構成を示すブロック図である。1 is a block diagram showing a functional configuration of a depth estimation device according to an embodiment of the present disclosure; FIG. マイグレーションによって得られる処理結果の例である。It is an example of a processing result obtained by migration. 比誘電率に応じたマイグレーション結果の例である。It is an example of the migration result according to the dielectric constant. 比誘電率方向、アンテナ進行方向、及び反射時間方向の３次元データの例である。It is an example of three-dimensional data in the dielectric constant direction, the antenna traveling direction, and the reflection time direction. 埋設物検出部に係る構成を示す図である。FIG. 4 is a diagram showing a configuration related to an embedded object detection unit; 深さ推定部に係る構成を示す図である。FIG. 4 is a diagram showing the configuration of a depth estimation unit; 本開示の実施形態の深さ推定装置による深さ推定処理の流れを示すフローチャートである。4 is a flowchart showing the flow of depth estimation processing by the depth estimation device according to the embodiment of the present disclosure;
　以下、開示の技術の実施形態の一例を、図面を参照しつつ説明する。なお、各図面において同一又は等価な構成要素及び部分には同一の参照符号を付与している。また、図面の寸法比率は、説明の都合上誇張されており、実際の比率とは異なる場合がある。 An example of an embodiment of the disclosed technology will be described below with reference to the drawings. In each drawing, the same or equivalent components and portions are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for convenience of explanation, and may differ from the actual ratios.
　本開示の実施形態において説明する技術では、計測対象の土壌が画一的でなく、また埋設物が複数埋まっているような現実的なシーンにおいて、各埋設物が埋設された深さを求めることを課題とする。 In the technology described in the embodiment of the present disclosure, the depth to which each buried object is buried can be obtained in a realistic scene where the soil to be measured is not uniform and multiple buried objects are buried. is the subject.
　本開示の実施形態では、計測されたデータ全体に対して画一的な比誘電率を推定するのではなく、埋設物毎に適切な比誘電率を推定する。埋設物ごとにそこに至るまでの平均的な比誘電率を一定とみなし、これを推定することによって、土質の変化や埋設物の影響によって複雑に変化する比誘電率を直接考慮することなく、複雑環境での埋設された深さの推定を実現する。具体的には、適当な範囲で比誘電率を変えながらデータ全体にマイグレーション処理を適用し、マイグレーションした値が極大となるような座標値及び比誘電率を抽出する。この極大を取る点（極大点）について得られる比誘電率が、その座標値での適切な平均比誘電率とみなすことができる。次に、予め機械学習された埋設物の検出器を用いてＧＰＲデータから埋設物のおおよその位置を検出する。検出された埋設物の位置の周辺で得られている極大点を用いて、マイグレーション値が最大となる比誘電率を採用する。このようにして各埋設物について得られた比誘電率と反射時間を用いて埋設物の深さを算出する。なお、ＧＰＲデータは、地中に埋設された埋設物を計測した二次元データである。ＧＰＲデータが、本開示の地中レーダデータの一例である。 In the embodiment of the present disclosure, instead of estimating a uniform dielectric constant for all measured data, an appropriate dielectric constant is estimated for each buried object. Assuming that the average relative permittivity of each buried object is constant and estimating it, we can obtain It realizes buried depth estimation in complex environment. Specifically, migration processing is applied to the entire data while changing the dielectric constant within an appropriate range, and the coordinate value and dielectric constant that maximize the migrated value are extracted. The relative permittivity obtained at the point of this maximum (maximum point) can be regarded as an appropriate average relative permittivity at that coordinate value. Next, the approximate position of the buried object is detected from the GPR data using a previously machine-learned buried object detector. Using the maximum points obtained around the position of the detected buried object, the dielectric constant with the maximum migration value is adopted. The depth of the buried object is calculated using the dielectric constant and the reflection time obtained for each buried object in this manner. GPR data is two-dimensional data obtained by measuring a buried object buried in the ground. GPR data is an example of ground penetrating radar data of this disclosure.
　本開示の実施形態の手法によって、土壌環境が整備されていない一般的な路上で計測されたＧＰＲデータに対して、技術者のノウハウを必要とせず自動で埋設物の深さ推定が可能となる。これによって埋設物を３次元位置に基づいて維持管理することができ、工事計画等メンテナンスコストの低減に寄与する。 According to the method of the embodiment of the present disclosure, it is possible to automatically estimate the depth of a buried object without requiring the know-how of an engineer for GPR data measured on a general road where the soil environment is not maintained. . As a result, the buried object can be maintained and managed based on the three-dimensional position, contributing to a reduction in maintenance costs such as construction planning.
　以下、本実施形態の構成について説明する。 The configuration of this embodiment will be described below.
　図１は、本開示の実施形態の深さ推定装置１００のハードウェア構成を示すブロック図である。 FIG. 1 is a block diagram showing the hardware configuration of the depth estimation device 100 according to the embodiment of the present disclosure.
　図１に示すように、深さ推定装置１００は、ＣＰＵ（Ｃｅｎｔｒａｌ　Ｐｒｏｃｅｓｓｉｎｇ　Ｕｎｉｔ）１１、ＲＯＭ（Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ）１２、ＲＡＭ（Ｒａｎｄｏｍ　Ａｃｃｅｓｓ　Ｍｅｍｏｒｙ）１３、ストレージ１４、入力部１５、表示部１６及び通信インタフェース（Ｉ／Ｆ）１７を有する。各構成は、バス１９を介して相互に通信可能に接続されている。 As shown in FIG. 1, the depth estimation device 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and communication It has an interface (I/F) 17 . Each component is communicatively connected to each other via a bus 19 .
　ＣＰＵ１１は、中央演算処理ユニットであり、各種プログラムを実行したり、各部を制御したりする。すなわち、ＣＰＵ１１は、ＲＯＭ１２又はストレージ１４からプログラムを読み出し、ＲＡＭ１３を作業領域としてプログラムを実行する。ＣＰＵ１１は、ＲＯＭ１２又はストレージ１４に記憶されているプログラムに従って、上記各構成の制御及び各種の演算処理を行う。本実施形態では、ＲＯＭ１２又はストレージ１４には、深さ推定プログラムが格納されている。 The CPU 11 is a central processing unit that executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 performs control of each configuration and various arithmetic processing according to programs stored in the ROM 12 or the storage 14 . In this embodiment, the ROM 12 or storage 14 stores a depth estimation program.
　ＲＯＭ１２は、各種プログラム及び各種データを格納する。ＲＡＭ１３は、作業領域として一時的にプログラム又はデータを記憶する。ストレージ１４は、ＨＤＤ（Ｈａｒｄ　Ｄｉｓｋ　Ｄｒｉｖｅ）又はＳＳＤ（Ｓｏｌｉｄ　Ｓｔａｔｅ　Ｄｒｉｖｅ）等の記憶装置により構成され、オペレーティングシステムを含む各種プログラム、及び各種データを格納する。 The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a work area. The storage 14 is configured by a storage device such as a HDD (Hard Disk Drive) or SSD (Solid State Drive), and stores various programs including an operating system and various data.
　入力部１５は、マウス等のポインティングデバイス、及びキーボードを含み、各種の入力を行うために使用される。 The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used for various inputs.
　表示部１６は、例えば、液晶ディスプレイであり、各種の情報を表示する。表示部１６は、タッチパネル方式を採用して、入力部１５として機能してもよい。 The display unit 16 is, for example, a liquid crystal display, and displays various information. The display unit 16 may employ a touch panel system and function as the input unit 15 .
　通信インタフェース１７は、端末等の他の機器と通信するためのインタフェースである。当該通信には、例えば、イーサネット（登録商標）若しくはＦＤＤＩ等の有線通信の規格、又は、４Ｇ、５Ｇ、若しくはＷｉ－Ｆｉ（登録商標）等の無線通信の規格が用いられる。 The communication interface 17 is an interface for communicating with other devices such as terminals. The communication uses, for example, a wired communication standard such as Ethernet (registered trademark) or FDDI, or a wireless communication standard such as 4G, 5G, or Wi-Fi (registered trademark).
　次に、深さ推定装置１００の各機能構成について説明する。図２は、本開示の実施形態の深さ推定装置１００の機能的な構成を示すブロック図である。各機能構成は、ＣＰＵ１１がＲＯＭ１２又はストレージ１４に記憶された深さ推定プログラムを読み出し、ＲＡＭ１３に展開して実行することにより実現される。 Next, each functional configuration of the depth estimation device 100 will be described. FIG. 2 is a block diagram showing the functional configuration of the depth estimation device 100 according to the embodiment of the present disclosure. Each functional configuration is realized by the CPU 11 reading a depth estimation program stored in the ROM 12 or the storage 14, developing it in the RAM 13, and executing it.
　図２に示すように、深さ推定装置１００は、検出器記憶部１０２と、マイグレーション部１１０と、極所情報抽出部１１２と、埋設物検出部１１４と、深さ推定部１１６とを含んで構成されている。マイグレーション部１１０において入力されたＧＰＲデータにマイグレーション処理を適用し、その出力に対して極所情報抽出部１１２において極大となる座標値、及び比誘電率を算出する。一方で埋設物検出部１１４ではＧＰＲデータに対して埋設物検出器を適用し、埋設物の反応箇所を特定して位置を検出する。最後に深さ推定部１１６では、埋設物反応箇所の位置の周辺に絞って極大点を取り出し、マイグレーション結果が最大となるような極大点の比誘電率を取り出す。この比誘電率を用いて深さを算出し、埋設物の深さとして出力する。 As shown in FIG. 2, the depth estimation device 100 includes a detector storage unit 102, a migration unit 110, an extremity information extraction unit 112, a buried object detection unit 114, and a depth estimation unit 116. It is configured. Migration processing is applied to the input GPR data in the migration unit 110, and the local information extraction unit 112 calculates the maximum coordinate value and relative dielectric constant for the output. On the other hand, the embedded object detection unit 114 applies an embedded object detector to the GPR data, specifies the reaction site of the embedded object, and detects the position. Finally, the depth estimating unit 116 picks up local maximum points around the position of the buried object reaction site, and picks up the dielectric constant of the local maximum point that maximizes the result of migration. The depth is calculated using this dielectric constant and output as the depth of the buried object.
　マイグレーション部１１０は、ＧＰＲデータと、予め求められた比誘電率の範囲とに基づいて、マイグレーション処理を範囲に応じて比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求める。 The migration unit 110 performs migration processing while changing the dielectric constant according to the range based on the GPR data and the range of the dielectric constant obtained in advance, and outputs the migration result for each changed dielectric constant. demand.
　比誘電率の範囲とは、例えば対象が一般的な土であればその湿潤度合いによって比誘電率は２～３０程度で変化する。このように対象が取りうる比誘電率の範囲を予め設定する。この設定された範囲から比誘電率を一定間隔で取り出し、マイグレーション処理を適用する。マイグレーション処理とは、アンテナの指向性によって現れる放物線状の反射波形を比誘電率に基づいて修正し、アンテナ直下の反射のみを捉えられるよう変換する処理である。代表的なマイグレーション処理手法としてＫｉｒｃｈｈｏｆｆマイグレーション法があるが、これ以外のマイグレーション手法を用いてもよい。Ｋｉｒｃｈｈｏｆｆマイグレーションによって得られる処理結果の例を図３に示す。元のＧＰＲデータでは反射反応のある箇所において放物線状の帯が現れているが、マイグレーション処理によって１点の強い反応に集約されていることがわかる。このマイグレーション処理を比誘電率を変えて適用すると、図３のように比誘電率に応じて多数のマイグレーション結果が得られる。ここで、各埋設物の反応について正しい比誘電率を用いてマイグレーションを行った際に最も集約度合が高まると仮定すると、マイグレーション結果の値を比較することで比誘電率を推定することができる。 The range of the relative permittivity is, for example, if the target is general soil, the relative permittivity changes from about 2 to 30 depending on the degree of moisture. In this way, the range of dielectric constants that the object can take is set in advance. A specific dielectric constant is taken out from this set range at regular intervals, and a migration process is applied. The migration process is a process of correcting the parabolic reflected waveform that appears due to the directivity of the antenna based on the relative permittivity so that only the reflection directly below the antenna can be captured. Although there is a Kirchhoff migration method as a representative migration processing method, other migration methods may be used. FIG. 3 shows an example of processing results obtained by Kirchhoff migration. In the original GPR data, a parabolic band appears at a location where there is a reflex response, but it can be seen that the migration processing has concentrated it into a single strong response. When this migration treatment is applied while changing the relative dielectric constant, many migration results are obtained according to the relative dielectric constant as shown in FIG. Here, assuming that the degree of aggregation is highest when migration is performed using the correct relative permittivity for the reaction of each buried object, the relative permittivity can be estimated by comparing the values of the migration results.
　極所情報抽出部１１２は、マイグレーション結果から極大点の各々を抽出し、抽出した極大点の各々に関し、マイグレーション結果の比誘電率、ＧＰＲデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出する。極大点が、本開示の極所的な点の一例である。 The local information extraction unit 112 extracts each local maximum point from the migration result, and for each of the extracted local maximum points, the relative permittivity of the migration result, the coordinates of the traveling direction of the antenna for which the GPR data was measured, and the reflection time. Extract the three-dimensional data of the coordinates. A local maximum is an example of a local point of this disclosure.
　極所情報抽出部１１２では、マイグレーション結果に対して反射箇所の中心点で反応が極大となることを仮定し、マイグレーションした値が極大（極大値）になる点を極大点として抽出する。抽出した極大点の各々について、図４で得られているような変化する比誘電率に対応するマイグレーション結果を積み重ねて、図５のような３次元データとして構成する。３次元データは、各極大点について、比誘電率方向、アンテナ進行方向、及び反射時間方向のそれぞれについて値が得られる。比誘電率方向には比誘電率を変化させたマイグレーション結果における比誘電率、アンテナ進行方向にはアンテナの座標、反射時間進行方向には反射時間の座標が得られる。後述する処理で、この３次元データを元に極大値を取る座標を算出することで、個別の埋設物に対して集約された反応が最大となる比誘電率を求める。アンテナの座標、及び反射時間はＧＰＲデータより得る。得られる極大座標は３次元座標となるため（比誘電率、アンテナ進行方向の座標、反射時間の座標）の３つの値が各極大点について得られる。ここで、元のＧＰＲデータから反射波の強度が得られている場合には極大値を用いればよいが、ＧＰＲデータが実数の波で表現されている場合には、負の値が集約された極値も生じるため極小値も同時に用いてもよい。ＧＰＲデータが複素数の波であった場合は極大点のみが得られ、ＧＰＲデータが実数の波であった場合は極大点及び極小点が得られる。 The local information extraction unit 112 assumes that the reaction is maximum at the center point of the reflection location for the migration result, and extracts the point where the migrated value is maximum (maximum value) as the local maximum point. For each of the extracted local maximum points, the migration results corresponding to the changing relative permittivity obtained in FIG. 4 are accumulated to form three-dimensional data as shown in FIG. As for the three-dimensional data, values are obtained for each local maximum point in each of the dielectric constant direction, the antenna traveling direction, and the reflection time direction. In the relative permittivity direction, the relative permittivity in the migration result with the relative permittivity changed is obtained, in the antenna traveling direction, the coordinates of the antenna are obtained, and in the reflection time advancing direction, the coordinates of the reflection time are obtained. In a process described later, the coordinates that take the maximum value are calculated based on this three-dimensional data, and the dielectric constant that maximizes the aggregated reaction for each buried object is obtained. Antenna coordinates and reflection times are obtained from GPR data. Since the obtained local maximum coordinates are three-dimensional coordinates (relative permittivity, antenna traveling direction coordinates, reflection time coordinates), three values are obtained for each local maximum point. Here, when the intensity of the reflected wave is obtained from the original GPR data, the maximum value should be used, but when the GPR data is represented by real waves, negative values are aggregated. Since extreme values also occur, local minimum values may also be used at the same time. If the GPR data were complex waves, only maxima would be obtained, and if the GPR data were real waves, maxima and minima would be obtained.
　埋設物検出部１１４は、検出器記憶部１０２の検出器を用いて、ＧＰＲデータの埋設物の座標を検出する。検出器は、元のＧＰＲデータを入力として画像認識に用いられる機械学習手法を応用して埋設物の位置を検出するように学習されている。検出器記憶部１０２には、予め機械学習手法により学習された検出器のパラメータが格納されている。 The buried object detection unit 114 uses the detector of the detector storage unit 102 to detect the coordinates of the buried object in the GPR data. The detector is trained to detect the position of the buried object by applying a machine learning technique used for image recognition with the original GPR data as input. The detector storage unit 102 stores detector parameters learned in advance by a machine learning technique.
　埋設物検出部１１４に係る構成を図６に示す。検出には予め学習された検出器パラメータが必要となるため、事前に学習用のＧＰＲデータと対応する埋設物の位置のアノテーションデータの学習データを用意し、検出器学習部１０４により、学習データの正解と出力の誤差が最小化されるよう検出器を学習する。学習及び検出に用いる機械学習モデル及び学習手法については、例えば、非特許文献１で用いられているような畳み込みニューラルネットワークに基づく検出モデルを用いて誤差逆伝播法によって学習すると高い検出精度が期待できる。なお、これ以外の手法であっても用いてもよい。このようにして得られた検出器を適用することで、ＧＰＲデータに現れている各埋設物のおおまかな位置を示す座標値を得ることができる。 The configuration related to the buried object detection unit 114 is shown in FIG. Since detector parameters learned in advance are required for detection, GPR data for learning and learning data of annotation data of the corresponding buried object positions are prepared in advance, and the detector learning unit 104 learns the learning data. Train the detector to minimize the error between the correct answer and the output. Regarding the machine learning model and learning method used for learning and detection, for example, high detection accuracy can be expected when learning by error backpropagation using a detection model based on a convolutional neural network as used in Non-Patent Document 1. . In addition, even if it is a method other than this, you may use it. By applying the detector thus obtained, it is possible to obtain coordinate values indicating the rough position of each buried object appearing in the GPR data.
　深さ推定部１１６は、図７に示すように、比誘電率推定部１２０と、深さ算出部１２２とによって構成される。深さ推定部１１６は、各部の処理により、埋設物の座標に基づく周辺座標における極大点の座標を参照して、極大点に対応する比誘電率を採用し、採用された比誘電率と、反射時間とに基づいて、埋設物の深さを推定する。 The depth estimator 116 is composed of a dielectric constant estimator 120 and a depth calculator 122, as shown in FIG. The depth estimating unit 116 refers to the coordinates of the maximum point in the peripheral coordinates based on the coordinates of the buried object, adopts the relative permittivity corresponding to the maximum point, and adopts the relative permittivity and Estimate the depth of the buried object based on the reflection time.
　深さ推定部１１６では、入力として極所情報抽出部１１２から出力される極大点の座標及びその極大値と、埋設物検出部１１４から出力される埋設物の座標が与えられる。 The depth estimating unit 116 receives as input the coordinates of the maximum point and its maximum value output from the extremity information extraction unit 112 and the coordinates of the buried object output from the buried object detection unit 114 .
　比誘電率推定部１２０は、各埋設物の座標について、マイグレーション結果の極大値を探索する範囲をその周辺座標に限定する。周辺座標は、埋設物の座標を基準として、アンテナ進行方向に±ＴＸメートル以内、反射時間方向に±ＴＴ秒以内に定めた領域とする。周辺座標において極大点を探索する。探索した極大点の座標を参照し、その極大値を取り出す。取り出された極大値の中で極大値が最大となる座標を参照し、当該座標に対応する極大点の比誘電率を埋設物に対応する比誘電率として採用する。なお、ＧＰＲデータが実数の波である場合には、極大値と極小値が得られているため、周辺座標において極大点及び極小点を探索する。それぞれ最大値、最小値を求め、極大点に対応する比誘電率と、極小点に対応する比誘電率との平均値を採用する。また、採用された比誘電率を含む３次元データにおいて、対応する反射時間を採用する。 Relative permittivity estimator 120 limits the search range for the maximum value of the migration result to the surrounding coordinates for the coordinates of each buried object. Peripheral coordinates shall be defined within ± T X meters in the antenna traveling direction and within ±TT seconds in the reflection time direction based on the coordinates of the buried object. Search for local maxima in peripheral coordinates. Refer to the coordinates of the searched maximal point and extract the maximal value. The coordinate at which the maximum value is maximum among the extracted maximum values is referred to, and the dielectric constant at the maximum point corresponding to the coordinate is adopted as the dielectric constant corresponding to the buried object. When the GPR data is a wave of real numbers, since the maximum and minimum values are obtained, the maximum and minimum points are searched for in the peripheral coordinates. The maximum value and the minimum value are determined, respectively, and the average value of the relative permittivity corresponding to the maximum point and the relative permittivity corresponding to the minimum point is adopted. Also, the corresponding reflection time is adopted in the three-dimensional data including the adopted dielectric constant.
　深さ算出部１２２は、各埋設物に対応する推定された比誘電率を用いて詳細な埋設物の深さを算出する。反射時間ｔと比誘電率εｒを用いて以下の（１）式によって深さｄが求められる。
Figure JPOXMLDOC01-appb-M000001
                            ・・・（１） The depth calculator 122 calculates the detailed depth of the buried object using the estimated dielectric constant corresponding to each buried object. Using the reflection time t and the dielectric constant εr, the depth d is obtained by the following equation (1).
Figure JPOXMLDOC01-appb-M000001
... (1)
　次に、深さ推定装置１００の作用について説明する。 Next, the operation of the depth estimation device 100 will be described.
　図８は、本開示の実施形態の深さ推定装置１００による深さ推定処理の流れを示すフローチャートである。ＣＰＵ１１がＲＯＭ１２又はストレージ１４から深さ推定プログラムを読み出して、ＲＡＭ１３に展開して実行することにより、深さ推定処理が行なわれる。ＣＰＵ１１が深さ推定装置１００の各部として機能することにより、以下の処理を実行させる。 FIG. 8 is a flowchart showing the flow of depth estimation processing by the depth estimation device 100 according to the embodiment of the present disclosure. Depth estimation processing is performed by the CPU 11 reading a depth estimation program from the ROM 12 or the storage 14, developing it in the RAM 13, and executing it. The following processes are executed by the CPU 11 functioning as each part of the depth estimation device 100 .
　ステップＳ１００において、ＣＰＵ１１は、ＧＰＲデータと、予め求められた比誘電率の範囲とに基づいて、マイグレーション処理を範囲に応じて比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求める。 In step S100, the CPU 11 performs the migration process while changing the relative permittivity according to the range based on the GPR data and the range of the relative permittivity obtained in advance. Seek results.
　ステップＳ１０２において、ＣＰＵ１１は、マイグレーション結果から極大点の各々を抽出する。 In step S102, the CPU 11 extracts each local maximum point from the migration result.
　ステップＳ１０４において、ＣＰＵ１１は、抽出した極大点の各々に関し、マイグレーション結果の比誘電率、ＧＰＲデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出する。 In step S104, the CPU 11 extracts the three-dimensional data of the relative permittivity of the migration result, the coordinates of the traveling direction of the antenna that measured the GPR data, and the coordinates of the reflection time for each of the extracted local maximum points.
　ステップＳ１０６において、ＣＰＵ１１は、検出器記憶部１０２の検出器を用いて、ＧＰＲデータの埋設物の座標を検出する。 In step S106, the CPU 11 uses the detector of the detector storage unit 102 to detect the coordinates of the buried object in the GPR data.
　ステップＳ１０８において、ＣＰＵ１１は、埋設物の座標に基づく周辺座標における極大点の座標を参照して、極大点に対応する比誘電率を採用する。 In step S108, the CPU 11 refers to the coordinates of the local maximum point in the peripheral coordinates based on the coordinates of the buried object, and adopts the dielectric constant corresponding to the local maximum point.
　ステップＳ１１０において、ＣＰＵ１１は、採用された比誘電率と、反射時間とに基づいて、埋設物の深さを推定する In step S110, the CPU 11 estimates the depth of the buried object based on the adopted dielectric constant and reflection time.
　以上説明したように本実施形態の深さ推定装置１００によれば、比誘電率を用いて埋設物が埋設された深さを求めることができる。 As described above, according to the depth estimation device 100 of the present embodiment, the depth at which the buried object is buried can be obtained using the relative permittivity.
　また、本開示の技術は、マイグレーション結果を用いて自動的に比誘電率を推定する手法といえる。マイグレーション結果を積み重ねて極大の座標を得ることによって、反射箇所によって異なる最適な比誘電率を推定できる。これを埋設物の自動検出結果と統合することで、埋設物ごとにその座標に応じて適切な比誘電率を推定することが可能である。 In addition, the technique of the present disclosure can be said to be a method of automatically estimating the dielectric constant using the migration results. By accumulating the migration results and obtaining the coordinates of the maximum, it is possible to estimate the optimum relative permittivity that differs depending on the reflection point. By integrating this with the results of automatic detection of buried objects, it is possible to estimate the appropriate dielectric constant for each buried object according to its coordinates.
　なお、上記実施形態でＣＰＵがソフトウェア（プログラム）を読み込んで実行した深さ推定処理を、ＣＰＵ以外の各種のプロセッサが実行してもよい。この場合のプロセッサとしては、ＦＰＧＡ（Ｆｉｅｌｄ－Ｐｒｏｇｒａｍｍａｂｌｅ　Ｇａｔｅ　Ａｒｒａｙ）等の製造後に回路構成を変更可能なＰＬＤ（Ｐｒｏｇｒａｍｍａｂｌｅ　Ｌｏｇｉｃ　Ｄｅｖｉｃｅ）、及びＡＳＩＣ（Ａｐｐｌｉｃａｔｉｏｎ　Ｓｐｅｃｉｆｉｃ　Ｉｎｔｅｇｒａｔｅｄ　Ｃｉｒｃｕｉｔ）等の特定の処理を実行させるために専用に設計された回路構成を有するプロセッサである専用電気回路等が例示される。また、深さ推定処理を、これらの各種のプロセッサのうちの１つで実行してもよいし、同種又は異種の２つ以上のプロセッサの組み合わせ（例えば、複数のＦＰＧＡ、及びＣＰＵとＦＰＧＡとの組み合わせ等）で実行してもよい。また、これらの各種のプロセッサのハードウェア的な構造は、より具体的には、半導体素子等の回路素子を組み合わせた電気回路である。 Note that the depth estimation processing executed by the CPU reading the software (program) in the above embodiment may be executed by various processors other than the CPU. In this case, the processor is a PLD (Programmable Logic Device) whose circuit configuration can be changed after manufacturing, such as an FPGA (Field-Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit) to execute specific processing. A dedicated electric circuit or the like, which is a processor having a specially designed circuit configuration, is exemplified. Also, the depth estimation process may be performed by one of these various processors, or by a combination of two or more processors of the same or different type (e.g., multiple FPGAs and a combination of CPU and FPGA). combination, etc.). More specifically, the hardware structure of these various processors is an electric circuit in which circuit elements such as semiconductor elements are combined.
　また、上記実施形態では、深さ推定プログラムがストレージ１４に予め記憶（インストール）されている態様を説明したが、これに限定されない。プログラムは、ＣＤ－ＲＯＭ（Ｃｏｍｐａｃｔ　Ｄｉｓｋ　Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ）、ＤＶＤ－ＲＯＭ（Ｄｉｇｉｔａｌ　Ｖｅｒｓａｔｉｌｅ　Ｄｉｓｋ　Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ）、及びＵＳＢ（Ｕｎｉｖｅｒｓａｌ　Ｓｅｒｉａｌ　Ｂｕｓ）メモリ等の非一時的（ｎｏｎ－ｔｒａｎｓｉｔｏｒｙ）記憶媒体に記憶された形態で提供されてもよい。また、プログラムは、ネットワークを介して外部装置からダウンロードされる形態としてもよい。 Also, in the above embodiment, the depth estimation program has been pre-stored (installed) in the storage 14, but the present invention is not limited to this. Programs are stored in non-transitory storage media such as CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk Read Only Memory), and USB (Universal Serial Bus) memory. may be provided in the form Also, the program may be downloaded from an external device via a network.
　以上の実施形態に関し、更に以下の付記を開示する。 Regarding the above embodiments, the following additional remarks are disclosed.
　（付記項１）
　メモリと、
　前記メモリに接続された少なくとも１つのプロセッサと、
　を含み、
　前記プロセッサは、
　地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求めるマイグレーション部と、
　前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出し、
　予め学習された検出器を用いて前記埋設物の座標を検出し、
　前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する、
　ように構成されている深さ推定装置。 (Appendix 1)
memory;
at least one processor connected to the memory;
including
The processor
Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity. a migration unit that performs while changing and obtains the migration result for each changed dielectric constant;
From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, Extracting three-dimensional data of the coordinates of the direction of travel of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time,
detecting the coordinates of the buried object using a pre-learned detector;
referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and the reflection time, estimating the depth of the buried object;
A depth estimator configured to:
　（付記項２）
　深さ推定処理を実行するようにコンピュータによって実行可能なプログラムを記憶した非一時的記憶媒体であって、
　地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求めるマイグレーション部と、
　前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出し、
　予め学習された検出器を用いて前記埋設物の座標を検出し、
　前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する、
　非一時的記憶媒体。 (Appendix 2)
A non-transitory storage medium storing a program executable by a computer to perform a depth estimation process,
Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity. a migration unit that performs while changing and obtains the migration result for each changed dielectric constant;
From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, Extracting three-dimensional data of the coordinates of the direction of travel of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time,
detecting the coordinates of the buried object using a pre-learned detector;
referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and the reflection time, estimating the depth of the buried object;
Non-transitory storage media.
１００ 深さ推定装置
１０２ 検出器記憶部
１０４ 検出器学習部
１１０ マイグレーション部
１１２ 極所情報抽出部
１１４ 埋設物検出部
１１６ 深さ推定部
１２０ 比誘電率推定部
１２２ 深さ算出部 100 Depth estimation device 102 Detector storage unit 104 Detector learning unit 110 Migration unit 112 Local information extraction unit 114 Buried object detection unit 116 Depth estimation unit 120 Relative permittivity estimation unit 122 Depth calculation unit

Claims (5)

  1. 　地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求めるマイグレーション部と、
    　前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出する極所情報抽出部と、
    　予め学習された検出器を用いて前記埋設物の座標を検出する埋設物検出部と、
    　前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する深さ推定部と、
    　を含む深さ推定装置。     Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity. a migration unit that performs while changing and obtains the migration result for each changed dielectric constant;
    From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, a local information extraction unit that extracts three-dimensional data of the coordinates of the traveling direction of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time;
    an embedded object detection unit that detects the coordinates of the embedded object using a pre-learned detector;
    referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and a depth estimating unit that estimates the depth of the buried object based on the reflection time;
    Depth estimator including.
  2. 　前記深さ推定部は、
    　前記極所的な点について、極大点を抽出している場合には、
    　前記周辺座標において探索された極大点に基づいて、前記マイグレーションした値が最大となる極大点に対応する前記比誘電率を採用し、
    　前記極所的な点について、極大点及び極小点を抽出している場合には、
    　前記周辺座標において探索された極大点及び極小点に基づいて、前記マイグレーションした値が最大となる極大点に対応する前記比誘電率と、最小となる極小点に対応する前記比誘電率との平均の比誘電率を採用する請求項１に記載の深さ推定装置。     The depth estimation unit
    For the local points, if the local maximum point is extracted,
    Based on the maximum point searched for in the peripheral coordinates, adopting the relative permittivity corresponding to the maximum point where the migrated value is the maximum,
    For the local points, when extracting local maximum points and local minimum points,
    Based on the maximum and minimum points searched for in the peripheral coordinates, the average of the relative permittivity corresponding to the maximum point at which the migrated value is maximum and the relative permittivity corresponding to the minimum point at which the migrated value is minimum 2. The depth estimator of claim 1, wherein the relative permittivity of .
  3. 　前記周辺座標は、前記埋設物の座標を基準として、アンテナの進行方向、及び反射時間の方向により定めた領域として、前記マイグレーションした値を探索する範囲とする請求項１又は請求項２に記載の深さ推定装置。 3. The peripheral coordinates according to claim 1 or 2, wherein the peripheral coordinates are defined by the traveling direction of the antenna and the direction of the reflection time with reference to the coordinates of the buried object, and are defined as a range in which the migrated values are searched. Depth estimator.
  4. 　地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求め、
    　前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出し、
    　予め学習された検出器を用いて前記埋設物の座標を検出し、
    　前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する、
    　処理をコンピュータに実行させる深さ推定方法。     Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity. is performed while changing, and the migration result for each changed dielectric constant is obtained,
    From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, Extracting three-dimensional data of the coordinates of the direction of travel of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time,
    detecting the coordinates of the buried object using a pre-learned detector;
    referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and the reflection time, estimating the depth of the buried object;
    A depth estimation method that lets the computer do the work.
  5. 　地中に埋設された埋設物を計測した二次元データである地中レーダデータと、予め求められた比誘電率の範囲とに基づいて、所定のマイグレーション処理を前記範囲に応じて前記比誘電率を変化させながら行い、変化させた比誘電率ごとのマイグレーション結果を求め、
    　前記マイグレーション結果から、マイグレーションした値の極大点又は極大点及び極小点の何れかである極所的な点の各々を抽出し、当該極所的な点に関し、前記マイグレーション結果における比誘電率、前記地中レーダデータを計測したアンテナの進行方向の座標、及び反射時間の座標の３次元データを抽出し、
    　予め学習された検出器を用いて前記埋設物の座標を検出し、
    　前記埋設物の座標に基づく周辺座標における前記３次元データの前記極所的な点の座標を参照して、前記極所的な点に対応する比誘電率を採用し、採用された比誘電率と、前記反射時間とに基づいて、埋設物の深さを推定する、
    　処理をコンピュータに実行させる深さ推定プログラム。     Based on the ground penetrating radar data, which is two-dimensional data obtained by measuring the buried object buried in the ground, and the range of relative permittivity obtained in advance, a predetermined migration process is performed according to the range to change the relative permittivity. is performed while changing, and the migration result for each changed dielectric constant is obtained,
    From the migration result, extract each local point that is either a local maximum point or a local maximum point and a local minimum point of the migrated value, and with respect to the local point, relative permittivity in the migration result, Extracting three-dimensional data of the coordinates of the direction of travel of the antenna that measured the ground penetrating radar data and the coordinates of the reflection time,
    detecting the coordinates of the buried object using a pre-learned detector;
    referring to the coordinates of the local point of the three-dimensional data in the peripheral coordinates based on the coordinates of the buried object, adopting the relative permittivity corresponding to the local point, and adopting the relative permittivity and the reflection time, estimating the depth of the buried object;
    A depth estimation program that lets a computer do the work.
PCT/JP2021/022937 2021-06-16 2021-06-16 Depth estimation device, depth estimation method, and depth estimation program WO2022264342A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2023528867A JPWO2022264342A1 (en) 2021-06-16 2021-06-16
PCT/JP2021/022937 WO2022264342A1 (en) 2021-06-16 2021-06-16 Depth estimation device, depth estimation method, and depth estimation program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/022937 WO2022264342A1 (en) 2021-06-16 2021-06-16 Depth estimation device, depth estimation method, and depth estimation program

Publications (1)

Publication Number Publication Date
WO2022264342A1 true WO2022264342A1 (en) 2022-12-22

Family

ID=84527319

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/022937 WO2022264342A1 (en) 2021-06-16 2021-06-16 Depth estimation device, depth estimation method, and depth estimation program

Country Status (2)

Country Link
JP (1) JPWO2022264342A1 (en)
WO (1) WO2022264342A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05232220A (en) * 1992-02-20 1993-09-07 Osaka Gas Co Ltd Method and apparatus for measurement of specific permittivity and probing apparatus of buried object
JPH07270528A (en) * 1994-03-28 1995-10-20 Osaka Gas Co Ltd Method and device for measuring specific inductive capacity and equipment of probing buried object
JPH11271440A (en) * 1998-03-25 1999-10-08 Osaka Gas Co Ltd Method and apparatus for three dimensional ditection

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05232220A (en) * 1992-02-20 1993-09-07 Osaka Gas Co Ltd Method and apparatus for measurement of specific permittivity and probing apparatus of buried object
JPH07270528A (en) * 1994-03-28 1995-10-20 Osaka Gas Co Ltd Method and device for measuring specific inductive capacity and equipment of probing buried object
JPH11271440A (en) * 1998-03-25 1999-10-08 Osaka Gas Co Ltd Method and apparatus for three dimensional ditection

Also Published As

Publication number Publication date
JPWO2022264342A1 (en) 2022-12-22

Similar Documents

Publication Publication Date Title
KR101829560B1 (en) Method, system and computer device for predicting a capacity based on kalman filter
JP6239664B2 (en) Ambient environment estimation apparatus and ambient environment estimation method
KR101912195B1 (en) Method and device for real-time target location and map creation
JP6456141B2 (en) Generating map data
CN110501010A (en) Determine position of the mobile device in geographic area
CN103217673B (en) CFAR detecting method under inhomogeneous Weibull clutter background
EP3693759B1 (en) System and method for tracking motion of target in indoor environment
CN109828302B (en) Seismic source positioning method and device based on multiple vibration sensors
CN104180799A (en) Robot localization method based on self-adaptive Monte Carlo localization method
KR101960280B1 (en) Method for modifying magnetic field map, terminal and server for modifying magnetic field
Zeng et al. An elevation difference model for building height extraction from stereo-image-derived DSMs
WO2022264342A1 (en) Depth estimation device, depth estimation method, and depth estimation program
KR101354522B1 (en) Retrival method of high resolution wind fields of multiple-doppler radar using by variational and fgat method
Georgy et al. Unconstrained underwater multi-target tracking in passive sonar systems using two-stage PF-based technique
JP5765163B2 (en) Self-position estimation apparatus, method, and program
Choi et al. Correlation-based scan matching using ultrasonic sensors for EKF localization
KR101480834B1 (en) Target motion analysis method using target classification and ray tracing of underwater sound energy
Lubczonek Analysis of accuracy of surveillance radar image overlay by using georeferencing method
JP2022547580A (en) Improved navigation and positioning using surface-seeking radar and deep learning
CN113868807A (en) Three-dimensional environment analysis method and apparatus, computer storage medium, and wireless sensor system
US11686837B2 (en) Apparatus and method for determining kinetic information
He et al. Stereo radargrammetry in south-east Asia using TerraSAR-X stripmap data
CN112781591A (en) Robot positioning method and device, computer readable storage medium and robot
JP2020008310A (en) Object detecting device
KR102346970B1 (en) Lof-based senior loitering and missing children prevention systems and method

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21946019

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2023528867

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE