Information processing device

The information processing device enhances the accuracy and efficiency of buried object location using ground-penetrating radar by automating data analysis and integrating machine learning to compare radar results with existing drawings, thereby reducing excavation risks.

WO2026234752A1PCT designated stage Publication Date: 2026-11-12NT T INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2025/016777
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-05-07
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing methods for locating buried objects using ground penetrating radar are inefficient and prone to errors in position estimation, often requiring manual interpretation and leading to potential damage during excavation due to inaccurate or incomplete data analysis.

Method used

An information processing device that utilizes ground-penetrating radar data to automatically detect and estimate the position and type of buried objects, integrating machine learning to enhance accuracy and efficiency by comparing radar results with existing drawings and environmental data.

Benefits of technology

The device provides accurate and efficient estimation of buried object positions, reducing the risk of damage by automating the analysis of radar data and integrating confidence levels based on measurement parameters and drawing comparisons.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2025016777_12112026_PF_FP_ABST
    Figure JP2025016777_12112026_PF_FP_ABST
Patent Text Reader

Abstract

An information processing device 10 is provided with a control unit 11 that detects, from a search result image which is a bitmap image generated from a search result of an underground radar, a region reflecting the influence of a reflected wave from a buried object, acquires the type and certainty factor of the buried object on the basis of the extracted region, estimates the position and shape of the buried object from the search result image using a measurement parameter, compares the figure of the buried object shown by an existing drawing with the estimated position and size of the buried object to determine the weight of the certainty factor, and causes a display unit to associate and display an image showing the position and size of the buried object estimated on the basis of the search result image, and the type and certainty factor weighted by the weight of the buried object.
Need to check novelty before this filing date? Find Prior Art

Description

Information Processing Apparatus

[0001] The present disclosure relates to an information processing apparatus.

[0002] To supply lifelines such as electric power, communications, gas, and water and sewage systems, a huge number of pipelines, manholes, and the like are buried underground. Many of these buried structures were constructed during the period of high economic growth. The burial positions of the buried structures are documented in drawings and managed as relative positions from reference positions such as road boundaries at the time of construction. However, the position references for buried structures often differ depending on the burial period and the operator. In addition, the surface conditions of roads and other areas may have changed since the time of burial. For this reason, when excavation work is later carried out in the vicinity of buried structures, constructors conduct inquiries and consultations among operators in advance to confirm the consistency of multiple drawings, or spend a great deal of time conducting on-site surveys to grasp the positional relationship of the buried structures. If such advance preparations are not sufficiently performed, there is a risk of damage accidents to the buried structures during excavation.

[0003] Accordingly, there is a demand for a technique for efficiently measuring the position of buried pipelines with high accuracy.

[0004] Non-Patent Document 1 describes a method for exploring buried objects using ground penetrating radar.

[0005] Atsuo Chiga, "Principles and Characteristics of Ground Penetrating Radar and Points to Note for Proper Utilization", [online], Construction Machinery Construction, Vol.68, No.9, September 2016, p. 88-91, [Searched on April 1, 2025], Internet <URL: https: / / jcmanet.or.jp / bunken / kikanshi / 2016 / 09 / 088.pdf>

[0006] However, analysis of measurement results according to the configuration of Non-Patent Document 1 requires individual interpretation by workers. For this reason, the conventional configuration may have a risk of overlooking buried objects and variations in position estimation accuracy, or may take a great deal of time to analyze a large amount of data.

[0007] An object of the present disclosure is to efficiently estimate the position of a buried object with higher accuracy using the exploration result of ground penetrating radar.

[0008] According to this disclosure, the information processing device includes a control unit that detects a region reflecting the influence of reflected waves from a buried object from a search result image, which is a bitmap image generated from the results of a ground-penetrating radar survey; obtains the type and confidence level of the buried object based on the extracted region; estimates the position and shape of the buried object from the search result image using measurement parameters; compares the figure of the buried object shown in an existing drawing with the estimated position and size of the buried object to determine the weight of the confidence level; and displays on a display unit, in association with an image showing the position and size of the buried object estimated based on the search result image and the type of the buried object and the confidence level weighted by the weight.

[0009] According to one embodiment of the present disclosure, the location of buried objects can be estimated more accurately and efficiently using the search results of ground-penetrating radar.

[0010] This figure shows an example configuration of a measurement system according to one embodiment. This figure illustrates the overview of underground radar exploration by a measurement device. This block diagram shows an example of the hardware configuration of an information processing device. This block diagram shows an example of the functional configuration of an information processing device. This figure shows an example of an exploration result image. This figure illustrates the generation of an exploration result image. This figure illustrates the generation of an exploration result image. This figure shows an example configuration of the buried object detection unit in Figure 4. This figure shows an example of an exploration result image. This figure illustrates the learning of the detector in Figure 6. This figure shows an example configuration of the estimation unit in Figure 4. This figure illustrates the migration of exploration result images. This figure illustrates the continuity determination of multiple exploration result images by the continuity determination unit in Figure 4. This figure shows an example of pipeline specifications. This figure shows an example of an existing drawing. This figure shows an example of an image containing buried objects extracted from an existing drawing. This figure illustrates the quantitative evaluation of the consistency between the detection result and the existing drawing. This figure illustrates the quantitative evaluation of the consistency between the detection result and the existing drawing. This figure illustrates the correction of the detection result by the correction unit in Figure 4. This figure illustrates the correction of the detection result by the correction unit in Figure 4. This figure shows an example of an exploration result output from the result output unit in Figure 4. This figure shows an example of a case where the detection result and the existing drawing do not match. This figure shows an example of a case where the detection result and the existing drawing do not match. This figure shows an example of the exploration results output from the result output unit in Figure 4. This figure shows an example of the exploration results output from the result output unit. This is a flowchart showing an example of the operation of the information processing device.

[0011] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In each drawing, parts having the same configuration or function are denoted by the same reference numerals. In the description of this embodiment, redundant descriptions of the same parts may be omitted or simplified as appropriate.

[0012] <Measurement System 1> Figure 1 shows an example of the configuration of measurement system 1 according to one embodiment. Measurement system 1 uses radar to explore the ground and detect buried objects. In this embodiment, buried objects are artificial structures including pipelines such as culverts and pipes (e.g., polyvinyl chloride pipes, sewer pipes, etc.), but may also include naturally occurring structures such as underground cavities and areas where the ground is loose. Measurement system 1 automatically estimates the location of buried objects from the exploration results in ground-penetrating radar exploration and determines the confidence level of the estimation based on measurement parameters and known data.

[0013] The measurement system 1 comprises an information processing device 10 and a measuring device 20. The measuring device 20 is a search radar that transmits electromagnetic waves into the ground, receives reflected waves, and acquires search results. The information processing device 10 is a computer that analyzes the search results acquired by the measuring device 20. The information processing device 10 and the measuring device 20 are connected to each other so as to be able to communicate with each other, but they do not have to be connected.

[0014] <Measuring Device 20> Figure 2 is a diagram illustrating the overview of ground radar exploration using the measuring device 20. The measuring device 20 comprises a housing 21, a gripping part 22, a transmitting and receiving antenna 23, and wheels 24. The housing 21 houses the transmitting and receiving antenna 23 and a device for controlling the transmitting and receiving antenna. The gripping part 22 is used by the user to grip the measuring device 20 when moving it. The transmitting and receiving antenna 23 is an antenna that transmits electromagnetic waves into the ground and receives reflected waves. The wheels 24 rotate on the ground surface G in response to the force applied to the measuring device 20 by gripping the gripping part 22, thereby moving the measuring device 20. The measuring device 20 may also detect the distance traveled by measuring the result of a sensor that records the rotation speed of the wheels 24.

[0015] In this configuration, the measuring device 20 moves along the ground surface G and emits electromagnetic waves 41 from the transmitting / receiving antenna 23 towards the ground. The emitted electromagnetic waves 41 are reflected by the interface of underground structures and buried objects 40 such as cavities (the surface of an object with a different dielectric constant from a medium such as soil and sand), and the reflected electromagnetic waves 42 (reflected waves) reach the ground surface G. The measuring device 20 receives these reflected electromagnetic waves 42 with the transmitting / receiving antenna 23 to investigate the location and depth of the buried objects 40. The investigation results are expressed as the strength of the received signal for each time it takes for the reflected wave to return and the radar's movement position.

[0016] The depth d to the buried object 40 is given by the speed v of electromagnetic waves traveling through the ground and the round-trip reflection time t, where d = vt / 2 = ε r It is expressed as ct / 2, where c is the speed of light in a vacuum, and ε r ε is the relative permittivity of the soil medium. r It is known that this varies depending on the type of medium, such as air, asphalt, concrete, and soil (sandy, loamy, clayey, etc.), as well as the amount of moisture (dry, wet). The time axis of the exploration results can be interpreted as distance (depth), and the exploration results reflect the cross-section of the trajectory the radar has traveled. However, the transmitting and receiving antenna 23 is not directional, and the electromagnetic waves 41 are radiated from the transmitting and receiving antenna 23 spreading out on a spherical surface. The transmitting and receiving antenna 23 receives electromagnetic waves 42 (reflected waves) not only from objects vertically below the measuring device 20, but also from buried objects 40 in the ground that are shifted horizontally. Therefore, the measuring system 1 migrates the image of the exploration results (exploration result image) as described later to obtain a cross-sectional image of the ground (see Figure 10).

[0017] In this embodiment, the user has a general understanding of the layout of buried objects 40, such as pipelines, from drawings or the like. The user uses this knowledge to explore by moving the measuring device 20 to measure multiple cross-sectional images of the buried objects 40 that are perpendicular to the longitudinal direction of the pipelines or other buried objects 40. For example, if the pipeline to be explored extends north-south and is located underground, the user may perform a radar survey by moving the measuring device 20 in the east-west direction and emitting electromagnetic waves 41 into the ground, thereby obtaining a cross-sectional image of the ground cut by a vertical plane perpendicular to the north-south direction. By performing such east-west scanning multiple times, shifted by a certain distance in the north-south direction, the user can obtain multiple cross-sectional images including the pipeline extending in the north-south direction.

[0018] <Hardware Configuration of Information Processing Device 10> Figure 3 is a block diagram showing an example of the hardware configuration of the information processing device 10. The information processing device 10 is one or a plurality of computer devices that can communicate with each other. Specifically, the information processing device 10 may be any general-purpose electronic device such as a PC (Personal Computer) or a tablet terminal, or it may be another dedicated electronic device. The information processing device 10 may be installed, for example, inside the housing 21 of the measuring device 20, or near the gripping part 22, or it may be installed in a different location from the measuring device 20 (for example, including a server facility). As shown in Figure 1, the information processing device 10 includes a control unit 11, a storage unit 12, a communication unit 13, an input unit 14, and an output unit 15.

[0019] The control unit 11 includes one or more processors. In one embodiment, the "processor" is a general-purpose processor or a dedicated processor specialized for a specific process, but is not limited to these. The control unit 11 is communicatively connected to each component constituting the information processing device 10 and controls the operation of the entire information processing device 10.

[0020] The storage unit 12 includes, for example, any storage module such as an HDD (Hard Disk Drive), SSD (Solid State Drive), ROM (Read-Only Memory), and RAM (Random Access Memory). The storage unit 12 may function as, for example, main memory, auxiliary memory, or cache memory. The storage unit 12 stores any information used in the operation of the information processing device 10. For example, the storage unit 12 may store system programs, application programs, and various information received by the communication unit 13. The storage unit 12 is not limited to those built into the information processing device 10, but may also be an external database or an external storage module.

[0021] For example, the memory unit 12 may hold a pre-trained model that, upon input of exploration result images acquired by the measuring device 20, outputs the detection result and confidence level of the buried object 40. The memory unit 12 may also hold parameters and functions for migrating exploration result images. The memory unit 12 may also hold specifications / design conditions for buried objects 40 such as pipelines. The memory unit 12 may also hold a pre-trained model that recognizes images of existing drawings and extracts the ground surface, road boundaries, buried objects 40, etc.

[0022] The communication unit 13 includes an optional communication module that can communicate with the measuring device 20 and other devices such as network storage using any communication technology. The communication unit 13 may further include a communication control module for controlling communication with other devices, and a storage module for storing communication data such as identification information necessary for communication with other devices.

[0023] The input unit 14 includes one or more input interfaces that receive user input operations and acquire input information based on user operations. For example, the input unit 14 may be, but is not limited to, physical keys, capacitive keys, a pointing device, a touchscreen integrated with the display of the output unit 15, or a microphone that accepts voice input.

[0024] The output unit 15 includes one or more output interfaces that output information to the user and notify the user. For example, the output unit 15 is a monitor (display) that outputs information as an image, or a speaker that outputs information as sound, but is not limited to these. Such a monitor may be, for example, a liquid crystal panel display or an organic EL (Electro Luminescence) display. At least one of the above-mentioned input unit 14 and output unit 15 may be configured integrally with the information processing device 10, or may be provided as a separate unit.

[0025] The information processing device 10 can also be realized by a computer and a program, and the program can be recorded on a recording medium or provided via a network. In other words, the functions of the information processing device 10 can be realized by executing a computer program (program) according to this embodiment on the processor included in the control unit 11. To put it another way, the functions of the information processing device 10 can be realized by software. The computer program causes the computer to execute the processing steps included in the operation of the information processing device 10, thereby realizing the functions corresponding to the processing of each step on the computer. In other words, the computer program is a program that causes the computer to function as the information processing device 10 according to this embodiment.

[0026] Computer programs can be recorded on computer-readable recording media. Examples of computer-readable recording media include magnetic recording devices, optical discs, magneto-optical recording media, or semiconductor memory. Programs can be distributed, for example, by selling, transferring, or leasing portable recording media such as DVDs (Digital Versatile Discs) or USB (Universal Serial Bus) memory containing the programs. Programs may also be distributed by storing them in server storage and transferring them from the server to other computers via a network. Programs may also be provided as program products.

[0027] A computer may, for example, temporarily store a program recorded on a portable storage medium or a program transferred from a server in its main memory. The computer may then read the program stored in the main memory with its processor and execute the processing according to the read program. The computer may also directly read a program from a portable storage medium and execute the processing according to the program. The computer may sequentially execute the processing according to the received program each time a program is transferred to it from a server. Such processing may also be performed by a so-called ASP-type service that does not transfer programs from the server to the computer, but realizes its function only through execution instructions and result acquisition. "ASP" is an abbreviation for Application Service Provider. A program includes information used for processing by an electronic computer that is equivalent to a program. For example, data that is not a direct instruction to the computer but has the nature of defining the computer's processing falls under "equivalent to a program".

[0028] Some or all of the functions of the information processing device 10 may be implemented by a dedicated circuit included in the control unit 11. In other words, some or all of the functions of the information processing device 10 may be implemented by hardware. Furthermore, the information processing device 10 may be implemented by a single computer or by the cooperation of multiple computers. For example, the information processing device 10 may be implemented by a server device installed in, for example, a data center or cloud computing system, and a terminal device that accesses the server device via a network.

[0029] In the configuration described above, the information processing device 10 detects the area of ​​the buried object 40 from the bitmap image generated from the exploration results of the measuring device 20 (ground-penetrating radar), and obtains the detected area of ​​the buried object 40 and the confidence level of detection. The information processing device 10 determines the position and size of the buried object 40 from the obtained area of ​​the buried object 40 using measurement parameters. The information processing device 10 determines that the buried object 40 with longitudinal continuity is a pipeline based on the position and size of the buried object 40 in multiple cross-sections and the specifications or design conditions of the buried pipeline. The information processing device 10 compares the position and size of the buried object 40 with the drawing data of the existing buried object 40 and weights the confidence level of detection. The information processing device 10 corrects the position and size of the buried object 40 using environmental information related to the soil or the drawing data of the existing buried object 40. The information processing device 10 presents the continuity determination result, the weighted confidence level of detection, and the corrected position and size of the buried object 40 in association with each other.

[0030] According to the information processing device 10, the location of the buried object 40 can be estimated from the ground-penetrating radar survey results, and the confidence level of the estimation can be determined by using known measurement parameters, specifications / design conditions, existing drawings, and environmental information. Therefore, even if existing drawings differ from one business operator to another, or if the ground surface G has changed since the buried object 40 was installed, the location of the buried object 40 can be estimated efficiently and with high accuracy.

[0031] <Functional Configuration of Information Processing Device 10> Next, the details of each functional element of the information processing device 10 will be explained.

[0032] (Overall Configuration) Figure 4 is a block diagram showing an example of the functional configuration of the information processing device 10. The information processing device 10 comprises an image generation unit 31, a buried object detection unit 32, an estimation unit 33, a continuity determination unit 34, a weighting processing unit 35, a correction unit 36, and a result output unit 37.

[0033] The image generation unit 31 receives the exploration results (matrix) from the measuring device 20 (ground-penetrating radar), adjusts the size, and converts it into a bitmap image as the exploration result image. The buried object detection unit 32 detects the area of ​​the buried object 40 from the generated exploration result image.

[0034] The estimation unit 33 receives measurement parameters as input and integrates the output from the buried object detection unit 32 with the measurement parameters to calculate the position and size of the buried object 40. Here, the measurement parameters are parameters used for on-site measurements, such as the relative permittivity of the ground around the exploration site, the scanning distance, and the distance to adjacent cross-sections. The information processing device 10 detects the buried object 40 from multiple exploration result images using the image generation unit 31, the buried object detection unit 32, and the estimation unit 33. As a result of detecting the buried object 40, the estimation unit 33 obtains the area occupied by the buried object 40 and the confidence level of the estimation.

[0035] The continuity determination unit 34 determines the continuity of the buried objects 40 based on the positional relationship of the buried objects 40 detected from multiple exploration result images and the specifications / design conditions of the buried objects 40, and determines that buried objects 40 with continuity in the longitudinal direction are pipelines. The specifications / design conditions of the buried objects 40 are, for example, the specifications and design conditions of the pipeline for each burial company. The specifications of the pipeline are, for example, the shape, outer diameter, etc. The design conditions of the pipeline are, for example, whether there are branches, the maximum intersection angle of curved sections, the minimum curve radius of curved sections, whether there are manholes at connection points, etc.

[0036] The weighting processing unit 35 compares the estimated results of the buried object 40 based on the ground-penetrating radar survey results with existing drawing data and weights the confidence levels included in the detection results. The existing drawings are, for example, plan views and cross-sectional views of existing buried objects 40.

[0037] The correction unit 36 ​​corrects the position and shape of the area of ​​the buried object 40 included in the detection result using existing drawings and environmental information (medium, moisture content, etc.). The result output unit 37 displays the detection result of the buried object 40 as visual information on the monitor of the output unit 15 or the like.

[0038] (Image generation unit 31) The image generation unit 31 adjusts the size of the exploration results acquired by the measuring device 20 (ground-penetrating radar) and converts them into an exploration result image, which is a bitmap image with a predetermined number of pixels (e.g., 256 x 256 pixels). Specifically, the image generation unit 31 acquires a matrix as the exploration result, in which the signal strength of the reflected wave is a component in the rows and columns corresponding to the distance traveled by the measuring device 20 and the time until the reflected wave is received. The image generation unit 31 generates an exploration result image (image 51) by associating the brightness corresponding to the signal strength of the reflected wave with the pixels corresponding to the distance traveled by the measuring device 20 and the time until the reflected wave is received.

[0039] Figure 5A shows an example of a survey result image. As shown in Figure 5A, the horizontal axis of the survey result image (image 51) indicates the scanning distance of the measuring device 20. The vertical axis indicates the time from when the electromagnetic wave 41 is transmitted from the transmitting / receiving antenna 23 until the reflected wave (electromagnetic wave 42) is received, and corresponds to the depth.

[0040] If the size of the bitmap image obtained from the matrix of exploration results does not match a predetermined number of pixels, the image generation unit 31 interpolates or divides the bitmap image by inserting black (luminance 0) pixels.

[0041] Figures 5B and 5C illustrate the generation of exploration result images. As shown in Figure 5B, if the matrix of exploration results (corresponding to image 52) is less than 256 × 256 pixels, the image generation unit 31 may insert black pixels (corresponding to image 53) and convert it into a 256 × 256 pixel bitmap image to generate the exploration result image (image 54). As shown in Figure 5C, if the matrix of exploration results (corresponding to image 55) exceeds 256 × 256 pixels, the image generation unit 31 may divide it into multiple 256 × 256 pixel bitmap images to generate the exploration result images (images 56, 57).

[0042] (Buried Object Detection Unit 32) The buried object detection unit 32 detects a buried object 40 from the exploration result image generated by the image generation unit 31. FIG. 6 is a diagram showing a configuration example of the buried object detection unit 32 in FIG. 4. As shown in FIG. 6, the buried object detection unit 32 includes a detector set 321 composed of a plurality of detectors 322. Each of the plurality of detectors 322 detects the region of the buried object 40 from the input exploration result image.

[0043] As described above, since the transmitting and receiving antenna 23 performs exploration using the electromagnetic wave 41 spreading on a spherical surface, the exploration result image exhibits image characteristics different from the underground cross section. FIG. 7 is a diagram showing an example (image 71) of the exploration result image. As shown in FIG. 7, in the exploration result image (image 71), the reflected wave from the vinyl chloride pipe 72 exhibits a hyperbolic image characteristic as shown in region 74. The underground cavity and significant looseness 73 exhibit image characteristics as shown in region 75. In this way, the exploration result image exhibits different image characteristics depending on the type of the underground buried object 40. Therefore, the detector set 321 labels the exploration result image according to the type of the buried object 40 and performs machine learning, thereby making it possible to determine the type of the buried object 40 from the image characteristics represented by the exploration result image.

[0044] Specifically, each detector 322 may be an AI (Artificial Intelligence) model that is provided for each buried object 40 to be detected, and is pre-trained to output the confidence that the corresponding target buried object 40 exists in the exploration result image. For example, the detector 322a may output the confidence that a pipe-shaped pipeline exists in the exploration result image. The detector 322b may output the confidence that a pipeline with a rectangular cross section exists in the exploration result image. The detector 322c may output the confidence that a cavity exists in the exploration result image. The confidence is an index indicating the probability that the result estimated by the detector 322, which is an AI model constructed by machine learning, after processing input data is considered to be correct. The confidence may take a larger value when the identity between the learned data and the newly input data is high, and a smaller value when the identity is low.

[0045] FIG. 8 is a diagram for explaining the learning of the detector 322 in FIG. 6. Each detector 322 can be constructed by a learner 323. The learner 323 is constructed using a convolutional neural network. The learner 323 inputs an image in which a label with a luminance of 1 is assigned to a region 77 reflecting the influence of reflected waves from the buried object 40 in the exploration result image (image 76), and performs learning using the convolutional neural network. In the example of FIG. 8, the region 77 of the exploration result image (image 76) corresponds to the region reflecting the influence of the reflected waves from the buried object 40. Therefore, the learner 323 performs learning using the image 76 that is provided with a label (image 78) having a luminance of 1 in the same region 79 as the region 77 and a luminance of 0 (black) in the region outside the region 79. Since the exploration result images (image 76) and the labels (image 78) are used for learning in the convolutional neural network of the learner 323, it is necessary to prepare several tens or more of the exploration result images (image 76).

[0046] The detector set 321 includes the plurality of detectors 322 learned and constructed by the learner 323 in this manner. Upon acquiring the exploration result image from the image generation unit 31, the detector set 321 inputs the exploration result image to each of the plurality of detectors 322. Each detector 322 outputs, as a detection result, the region of the buried object 40 to be surveyed in the exploration result image, and the confidence that the buried object 40 is included in the exploration result image. The buried object detection unit 32 may output, as a detection result, the region in the exploration result image and the confidence of each buried object 40 whose confidence is greater than a predetermined threshold. In the example of FIG. 6, the buried object detection unit 32 detects, from the exploration result image (image 61) input from the image generation unit 31, a region 62 of an underdrain (confidence 0.3) and a region 63 of a sewer pipe (confidence 0.4).

[0047] (Estimation Unit 33) The estimation unit 33 estimates the position and size of the buried object 40 detected by the buried object detection unit 32. Figure 9 shows an example of the configuration of the estimation unit 33 in Figure 4. The estimation unit 33 calculates the position and size of the buried object 40 based on the output from the buried object detection unit 32 and the parameters used during measurement (scanning distance, exploration depth). The scanning distance may be calculated based on the number of rotations of the wheels 24 of the measuring device 20. The exploration depth may be calculated based on the time from when the electromagnetic wave 41 is transmitted from the transmitting / receiving antenna 23 until the reflected electromagnetic wave 42 is received. The estimation unit 33 includes a position estimation unit 331 that estimates the position of the buried object 40 and a size estimation unit 332 that estimates the size.

[0048] The position estimation unit 331 can determine the positional relationship for each pixel relative to the exploration start point by associating the scanning distance in the horizontal direction and the exploration depth in the vertical direction of the exploration result image, which is a bitmap image. When estimating the position of a pipe from the region 63 of the exploration result image that reflects the reflected waves of the pipe, the position estimation unit 331 may estimate the position of the pipe using the position of the pixels at the vertices of the hyperbola, which is an image feature of the pipe.

[0049] The size estimation unit 332 may estimate the size of the buried object 40 by performing migration on the exploration result image. Figure 10 is a diagram illustrating the migration of the exploration result image. Migration is a process that converts the hyperbolic shape resulting from conducting an exploration by transmitting and receiving non-directional electromagnetic waves from the transmitting and receiving antenna 23 into the shape of the original reflector. As shown in Figure 10, the reflected waves (electromagnetic waves 42) from the buried object 40 spread radially. Therefore, when the measuring device 20 passes over the buried object 40 in a straight line, the time t from when the transmitting and receiving antenna 23 transmits electromagnetic waves 41 until it receives the signal 45 of the reflected waves (electromagnetic waves 42) has a hyperbolic shape as shown in graph 46 with respect to the position of the measuring device 20. Migration restores the shape of the buried object 40 from the exploration result image that reflects the shape of graph 46.

[0050] The size estimation unit 332 can estimate the size of the buried object 40 using any migration method. For example, the size estimation unit 332 may use Kirchhoff migration (directly superimposing time waveforms), f-k migration (using Fourier transform), or reverse time migration (using the difference method).

[0051] In Figure 9, a region 82 containing a buried object with a rectangular cross-section is estimated from region 62 in the exploration result image (image 61). A region 83 containing a pipe-shaped buried object is estimated from region 63. In the example in Figure 9, the estimation unit 33 outputs an image 81 as the estimation result, to which regions 82, 83, operating distance, and exploration depth have been added to the exploration result image (image 61).

[0052] (Continuity Determination Unit 34) As described above, the user uses information such as existing drawings to explore by moving the measuring device 20 to measure multiple cross-sectional images of buried objects 40 that are perpendicular to the longitudinal direction of the buried object 40, such as a pipeline. The continuity determination unit 34 performs a continuity determination process to determine whether the images 81 estimated from the multiple exploration result images are images of the same buried object 40, such as a pipeline, that are continuous.

[0053] Figure 11A is a diagram illustrating the continuity determination of multiple exploration result images by the continuity determination unit 34 in Figure 4. In the example in Figure 11A, the three images 81a, 81b, and 81c are estimation results output sequentially from the estimation unit 33 for mutually adjacent vertical planes. Images 81a and 81b correspond to cross-sections separated by a distance of 85. Images 81b and 81c correspond to cross-sections separated by a distance of 86. Image 81a includes regions 82a and 83a. Image 81b includes regions 82b and 83b. Image 81c includes regions 82c and 83c.

[0054] The continuity determination unit 34 combines the detection results obtained at each cross-section from the image generation unit 31, the buried object detection unit 32, and the estimation unit 33, and compares this with the pipeline specifications / design conditions pre-stored in the memory unit 12. Based on the comparison, the continuity determination unit 34 estimates whether the detected buried object 40 is a pipeline defined in the specifications / design conditions. Figure 11B shows an example of pipeline specifications. If the combined detection results obtained at each cross-section from the estimation unit 33 match the pipeline specifications / design conditions, the continuity determination unit 34 may determine that the buried object 40, which has continuity in the longitudinal direction, is a pipeline. For example, the continuity determination unit 34 may use the specifications / design conditions shown in Figure 11B to determine whether each detected buried object 40 is likely to be a communication pipe or an electrical pipe.

[0055] (Weighting Processing Unit 35) The weighting processing unit 35 extracts the ground surface, road boundaries, buried objects 40, etc. from the existing drawings and matches them with the image (for example, image 81b) that is closest in location from among the estimation results of multiple cross-sections (81a, 81b, 81c). Subsequently, the weighting processing unit 35 overlays the existing drawings and the estimation results (image 81b) and weights the confidence level of the detection results in the buried object detection unit 32 according to the weight corresponding to the overlapping area of ​​the detected objects. This enables quantitative evaluation of the consistency of the results for the buried objects 40 with the existing drawings.

[0056] Figure 12A shows an example of an existing drawing 91. The existing drawing 91 includes information such as a figure 92 of a culvert which is an underground structure 40, a figure 93 of a sewer pipe, attribute information 94 and 95 of figures 92 and 93, horizontal distance 97 from a reference point 96, and vertical distance 98. The weighting processing unit 35 extracts figures 92 and 93 that show the shape of the underground structure 40, the ground surface 112, the road boundary 111, etc. from such an existing drawing 91.

[0057] Figure 12B shows an example of an image containing buried objects 40 extracted from an existing drawing. Image 98 includes figures 92 and 93 showing the shape of the buried objects 40, a road boundary 111, and the ground surface 112. The weighting processing unit 35 may use a known image recognition AI (for example, a pre-trained model) to extract the ground surface, road boundary, buried objects 40, etc. from the existing drawing 91.

[0058] The weighting processing unit 35 matches each of the multiple cross-sectional estimation results (81a, 81b, 81c) with the ground surface, road boundary, buried objects 40, etc. shown in image 98, and extracts the estimation result (for example, image 81b) in which the cross-sectional location is closest. The weighting processing unit 35 overlays the existing drawing with the exploration results and determines the weight of the exploration results based on the overlapping area of ​​the detected objects.

[0059] The drawings of the buried objects 40 were created at the time of construction (for example, during the period of rapid economic growth), and there may be cases where changes in the ground surface due to road widening, etc., have not been reflected. The contents of the existing drawings may deviate from the actual site. If the results of AI image analysis of the existing drawings are used as is, there is a risk of leading to incorrect evaluation and judgment. Therefore, the weighting processing unit 35 performs a quantitative evaluation of the consistency between the detection results of the buried object detection unit 32 and the existing drawings. For example, the weighting processing unit 35 may calculate IoU (Intersection over Union) between the existing drawings and the detection results to determine the weight of the confidence level of the detection. For two regions P and Q, IoU is given as the ratio of the area of ​​the overlapping region of P and Q (AND region of P and Q) to the area of ​​the integrated region of P and Q (OR region of P and Q).

[0060] Figures 13A and 13B illustrate the quantitative evaluation of the consistency between the detection results and existing drawings. In Figures 13A and 13B, the regions 101 and 102 of the buried objects (corresponding to regions 82 and 83 in Figure 9) output from the continuity determination unit 34 are superimposed on the image 98 of Figure 12B. In Figure 13A, the IoU between the region occupied by figure 92 and region 101 is 0.5. The IoU between the region occupied by figure 93 and region 102 is 0.7. On the other hand, in Figure 13B, there is no overlap between the regions occupied by figures 92 and 93 and regions 101 and 102, and the IoU is 0 in both cases. In the case of Figure 13A, it can be said that the detection results match the existing drawings, indicating high reliability of the detection results. On the other hand, in the case of Figure 13B, the existing drawings may differ significantly from the current situation, so caution is necessary during design and construction.

[0061] (Correction Unit 36) After weighting is performed in the weighting processing unit 35, the correction unit 36 ​​corrects the estimation results from existing drawings and environmental information (type of medium, moisture content, etc.) to improve the estimation accuracy of the position and shape of the buried object 40.

[0062] The estimated position of the buried object 40 changes depending on the medium and moisture content in the ground. Therefore, the correction unit 36 ​​acquires information on the medium and moisture content in the ground from the environmental information and drawing information of the measurement site, and adjusts the relative permittivity ε according to the current conditions of the measurement site. r The position of the buried object 40 is corrected by modifying the data. However, if the position estimation unit 331 is performing position detection using the relative permittivity that reflects the environmental information of the location, the correction unit 36 ​​does not need to correct the position of the buried object 40.

[0063] Furthermore, the shape of the buried object 40 obtained by radar survey may be difficult to restore, such as the height (length in the depth direction) of a rectangular buried object, which may be difficult to restore even after migration processing. Therefore, the correction unit 36 ​​may obtain the shape of the buried object 40 (for example, figures 92 and 93 in Figure 12B) from the information of the existing drawing and correct the shape of the buried object 40 in the estimated image by replacing it with the figure from the existing drawing.

[0064] Figures 14A and 14B illustrate the correction of the estimation results by the correction unit 36 ​​in Figure 4. In Figure 14A, regions 101 and 102 show the shape of the buried object 40 in the image of the estimation result before correction. In Figure 14B, regions 103 and 104 show the shape of the buried object 40 in the image of the estimation result after correction. By performing such correction, the accuracy of estimating the position and shape of the buried object 40 can be improved.

[0065] (Result Output Unit 37) The result output unit 37 visually displays the estimation results along with the confidence level on a monitor (output unit 15), etc. Figure 15 shows an example of the estimation results output from the result output unit 37 in Figure 4.

[0066] In Figure 15, regions 103 and 104 showing the estimated results of the buried object 40 are displayed superimposed on the exploration result image (Image 81) and the existing drawing 91. Regions 103 and 104 showing the estimated results of the buried object 40 are accompanied by displays 105 to 108 related to each region. Display 105 indicates the type of region 103 ("culvert") and the weighted confidence level ("0.7"). Display 106 indicates the type of region 104 ("sewer pipe") and the weighted confidence level ("0.6"). Here, the weighted confidence level is the value obtained by multiplying the confidence level output by the buried object detection unit 32 by the weight output by the weighting processing unit 35. Displays 107 and 108 display attribute information 94 and 95 extracted from the existing drawing 91.

[0067] The existing drawing 91, the exploration result image (image 81), and the estimation result may be displayed as an image with different layers superimposed. The result output unit 37 may be configured to arbitrarily switch the display / hide status of each layer in response to user instructions.

[0068] Furthermore, the result output unit 37 may output an alert from the output unit 15 to notify the user if a buried object 40 shown in the existing drawing is not detected from the radar scan results, or if a buried object 40 not shown in the existing drawing is detected from the radar scan results. Figures 16A and 16B show examples of cases where the detection results do not match the existing drawing. Figure 16A shows an example where, although a figure 121 exists on the existing drawing, the corresponding buried object 40 is not detected from the radar scan results (there is a possibility of overlooking it). In such cases, the result output unit 37 may output an overlooked alert by image or sound. Figure 16B shows an example where a buried object 40 not shown in the existing drawing is detected as a detection result 122 from the radar scan results (there is an unknown buried object). In such cases, the result output unit 37 may output an unknown buried object alert by image or sound.

[0069] The result output unit 37 may determine whether the existing drawings and the exploration results are consistent by comparing the IoU of both with a predetermined threshold. For example, if the IoU is smaller than the threshold, the result output unit 37 may output an alert indicating that the existing drawings and the exploration results are inconsistent. The information processing device 10 may accept the setting of this threshold from the user.

[0070] In this manner, the results output unit 37 issues an alert notification if the exploration results do not match the existing drawings. This reduces accidents caused by human error and unidentified buried objects. The alert notification may also be performed by the weighting processing unit 35.

[0071] Furthermore, the result output unit 37 may display the estimation results as a three-dimensional image on a monitor (output unit 15). Figures 17A and 17B show examples of exploration results output from the result output unit 37 in Figure 4. Figure 17A shows estimation results for multiple adjacent cross-sections (images 81a, 81b, 81c). In this case, the result output unit 37 may generate a three-dimensional image 125 of the buried object 40 from the regions 103a, 103b, and 103c included in images 81a, 81b, and 81c. The result output unit 37 may also generate a three-dimensional image 126 of the buried object 40 from the regions 104a, 104b, and 104c included in images 81a, 81b, and 81c.

[0072] The result output unit 37 may change the orientation and size of the three-dimensional images 125 and 126 and display them according to the user's instructions. In this way, by displaying the exploration results in three dimensions, the result output unit 37 allows the user to visually recognize the continuity and also facilitates comparison with plan views and the like.

[0073] <Example of Operation> Figure 18 is a flowchart showing an example of operation of the information processing device 10. The operation of the information processing device 10 described with reference to Figure 18 may correspond to one of the information processing methods. The operation of each step in Figure 18 may be performed based on control by the control unit 11 of the information processing device 10.

[0074] In step S1, the control unit 11 acquires the ground-penetrating radar survey results. Specifically, the control unit 11 may acquire the survey result image by processing to realize the function of the image generation unit 31.

[0075] In step S2, the control unit 11 detects the buried object 40 from the exploration result image. Specifically, the control unit 11 may detect the buried object 40 by processing to realize the function of the buried object detection unit 32.

[0076] In step S3, the control unit 11 estimates the location and size of the buried object 40 detected in step S2. Specifically, the control unit 11 may estimate the location and size of the buried object 40 by processing to realize the function of the estimation unit 33.

[0077] In step S4, the control unit 11 determines whether or not there are any unmeasured adjacent cross-sections. If there are any unmeasured adjacent cross-sections (YES in S4), the control unit 11 returns to step S1 and obtains the exploration results by measuring the adjacent cross-sections. If there are no unmeasured adjacent cross-sections (NO in S4), the control unit 11 proceeds to S5.

[0078] In step S5, the control unit 11 determines the continuity of the multiple cross-sectional images. Specifically, the control unit 11 may determine the continuity of the multiple cross-sectional images by performing processing to realize the function of the continuity determination unit 34.

[0079] In step S6, the control unit 11 weights the confidence level of the buried object estimation. Specifically, the control unit 11 may weight the confidence level of the buried object estimation by processing to realize the function of the weighting processing unit 35.

[0080] In step S7, the control unit 11 corrects the estimation result of the buried object 40. Specifically, the control unit 11 may correct the estimation result of the buried object 40 by processing to realize the function of the correction unit 36.

[0081] In step S8, the control unit 11 outputs the estimation result of the buried object 40. Specifically, the control unit 11 may output the estimation result of the buried object 40 by processing to realize the function of the result output unit 37.

[0082] This disclosure is not limited to the embodiments described above. For example, multiple blocks shown in the block diagram may be combined, or a single block may be divided. Multiple steps shown in the flowchart may be performed in parallel or in a different order, depending on the processing capacity of the device performing each step, or as necessary, instead of being performed in chronological order as described. Other modifications are possible without departing from the spirit of this disclosure.

[0083] For example, the information processing device 10 may perform migration on the exploration result image before detecting the buried object 40.

[0084] The following additional information is disclosed regarding the embodiments described above.

[0085] [Note 1] An information processing device comprising a control unit that detects a region reflecting the influence of reflected waves from a buried object from a search result image, which is a bitmap image generated from the results of a ground-penetrating radar; obtains the type and confidence level of the buried object based on the extracted region; estimates the position and shape of the buried object from the search result image using measurement parameters; determines the weight of the confidence level by comparing the figure of the buried object shown in an existing drawing with the estimated position and size of the buried object; and displays on a display unit, in association with an image showing the position and size of the buried object estimated based on the search result image and the confidence level weighted by the type of buried object and the weight.

[0086] [Addendum 2] The information processing apparatus according to claim 1, wherein the control unit corrects the size of the buried object estimated based on the exploration result image according to the shape of the buried object shown in the existing drawing.

[0087] [Addendum 3] The information processing apparatus according to claim 1, wherein the control unit corrects the position of the buried object estimated based on the exploration result image based on the relative permittivity obtained based on the type of soil medium and moisture content of the ground-penetrating radar exploration.

[0088] [Addendum 4] The information processing apparatus according to any one of claims 1 to 3, wherein the control unit notifies the user of an alert if the shape of the buried object shown in the existing drawing does not match the estimated location and size of the buried object.

[0089] [Appendix 5] An information processing method performed by an information processing device, comprising: detecting a region that reflects the influence of reflected waves from a buried object from a search result image, which is a bitmap image generated from the results of a ground-penetrating radar; obtaining the type and confidence level of the buried object based on the extracted region; estimating the position and shape of the buried object from the search result image using measurement parameters; determining the weight of the confidence level by comparing the figure of the buried object shown in an existing drawing with the estimated position and size of the buried object; and displaying on a display unit an image showing the position and size of the buried object estimated based on the search result image, and the confidence level weighted by the type of buried object and the weight.

[0090] [Addendum 6] A program that causes a computer to execute the following steps: a procedure for detecting a region that reflects the influence of reflected waves from a buried object from a search result image, which is a bitmap image generated from the results of a ground-penetrating radar; a procedure for obtaining the type and confidence level of the buried object based on the extracted region; a procedure for estimating the position and shape of the buried object from the search result image using measurement parameters; a procedure for determining the weight of the confidence level by comparing the figure of the buried object shown in an existing drawing with the estimated position and size of the buried object; and a procedure for displaying on a display unit an image showing the position and size of the buried object estimated based on the search result image, and the confidence level weighted by the type of buried object and the weight.

[0091] 1: Measurement system 10: Information processing device 11: Control unit 12: Memory unit 13: Communication unit 14: Input unit 15: Output unit 20: Measurement device 21: Housing 22: Gripping unit 23: Transmitting / receiving antenna 24: Wheel 31: Image generation unit 32: Buried object detection unit 321: Detector set 322: Detector 323: Learner 33: Estimation unit 331: Position estimation unit 332: Size estimation unit 34: Continuity determination unit 35: Weighting processing unit 36: Correction unit 37: Result output unit 40: Buried object 41, 42: Electromagnetic wave 45: Waveform 46: Graph 51-57: Image 61: Image 62, 63: Region 71: Image 72: Pipeline 73: Cavity and looseness 74, 75: Region 76: Image 77: Region 78: Image 79: Area 81: Image 82, 83: Area 85: Distance 86: Distance 91: Image 92, 93: Shape 94, 95: Attribute Information 96: Reference Point 97: Horizontal Distance 98: Vertical Distance 98: Image 101-104: Detection Results 105-108: Image 111: Road Boundary 112: Ground Surface 121: Shape 122: Detection Results 125, 126: 3D Image G: Ground Surface

Claims

1. An information processing device comprising a control unit that detects a region reflecting the influence of reflected waves from a buried object from a search result image, which is a bitmap image generated from the results of a ground-penetrating radar; obtains the type and confidence level of the buried object based on the extracted region; estimates the position and shape of the buried object from the search result image using measurement parameters; determines the weight of the confidence level by comparing the figure of the buried object shown in an existing drawing with the estimated position and size of the buried object; and displays on a display unit, in association with an image showing the position and size of the buried object estimated based on the search result image and the confidence level weighted by the type of buried object and the weight.

2. The information processing apparatus according to claim 1, wherein the control unit corrects the size of the buried object estimated based on the exploration result image according to the shape of the buried object shown in the existing drawing.

3. The information processing apparatus according to claim 1, wherein the control unit corrects the position of the buried object estimated based on the exploration result image based on the relative permittivity obtained based on the type of soil medium and moisture content of the ground-penetrating radar exploration.

4. The information processing apparatus according to any one of claims 1 to 3, wherein the control unit notifies the user of an alert if the shape of the buried object shown in the existing drawing does not match the estimated location and size of the buried object.