JP7145249B2 - Buried object discrimination system and buried object discrimination method - Google Patents

Description

The present invention relates to an embedded object identification system for identifying an embedded object.

Buried pipelines such as gas and water pipes are regularly replaced or new pipelines are installed to maintain the infrastructure. For new construction or replacement of pipelines, it is necessary to grasp not only the position of the target pipeline but also the laying status of nearby pipelines.

As background art in this technical field, there are JP-A-2015-90345 (Patent Document 1) and JP-A-2017-40547 (Patent Document 2). Japanese Patent Application Laid-Open No. 2015-90345 discloses three-dimensional position measuring means including at least a GPS receiver and an inertial measuring instrument, road surface condition measuring means including at least a camera and a laser scanner, and a road surface including an underground radar antenna. A mobile body equipped with an under investigation means, a three-dimensional position measurement means, a road surface condition measurement means, and an under-road investigation means, and a computer that analyzes each output data of the three-dimensional position measurement means, the road surface condition measurement means, and the under-road investigation means. and a three-dimensional position coordinate, which is output data of the three-dimensional position measuring means, using time obtained from a clock or GPS in the three-dimensional road surface diagnosis system, Time information is assigned to each of the road surface condition camera image and road surface condition 3D point cloud data, which are the output data of the road surface condition measuring means, and the depth direction information, which is the output data of the under road surface exploration means, and the attached time information is used. synchronizing the three-dimensional position coordinates, road surface condition camera image, road surface condition three-dimensional point group data, and depth direction information, and determining the relative positional relationship between the GPS receiver, camera, laser scanner, ground penetrating radar antenna, and each of the moving objects. and based on the attitude of the moving body obtained from the inertial measuring instrument, each 3D position coordinate is calculated, and 3D position coordinates are assigned to the road surface state camera image, road surface state 3D point cloud data, and depth direction information. , a three-dimensional under-road diagnosis system characterized in that the road surface condition camera image and the three-dimensional point cloud data of the road surface condition outputted by the road surface condition measuring means and the depth direction information outputted by the under-road investigation means can be integrated. (see claim 1).

Further, in JP-A-2017-40547, a first setting step of setting a first scanning range of a high-frequency electromagnetic wave radar probe with respect to a reference position, and the high-frequency electromagnetic wave radar probe a first scanning step of scanning in a first scanning range; and first positional information of a first buried object at a first depth to be searched based on analysis of data obtained in the first scanning step , a first display step of displaying the position of the first buried object based on the first position information, and a low-frequency electromagnetic wave radar probe with respect to the reference position a second setting step of setting two scanning positions; a second scanning step of causing the low-frequency electromagnetic wave radar probe to scan at the set second scanning position; a second analysis step of obtaining second position information of a second buried object at a second depth to be searched, which is deeper than the first depth to be searched, based on the analysis of the data obtained; and a second display step of displaying the position of the second buried object based on the positional information of (see claim 1).

JP 2015-90345 A JP 2017-40547 A

Conventionally, the locations of pipelines have been determined mainly using drawings, and the locations of pipelines and nearby buried objects have been investigated by test excavation based on radar surveys and the like. However, sometimes the drawings at the time of laying pipelines are not updated to the latest state, and the actual situation may differ from the drawings. In addition, grasping buried objects by radar exploration requires judgment by a skilled person's visual observation of radar responses.

In recent years, the number of skilled workers has been on the decline, and there is a growing need for a technique that allows even non-skilled workers to determine underground burial conditions. In addition, since the response of the exploration radar changes depending on external factors such as the soil quality and pavement conditions of the target area, it is difficult for even skilled workers to maintain a constant quality judgment. For this reason, the judgment of buried objects by skilled workers requires mutual checks and consultations by multiple people, and may not be fully satisfactory in terms of time, cost, and reliability of results.

Visual confirmation of radar response by skilled workers often uses an image expressed by changes in the brightness of reflected waves called B-mode survey data. A number of assistive techniques have been proposed to help locate radar response features in images of B-mode survey data. However, although conventional assistive technology can support the detection of reaction positions in images of B-mode exploration data, it is difficult to distinguish between types of buried objects (buried pipes or cavities) and to determine the size and position of buried objects. is difficult.

In addition, if trial drilling, which is a preliminary survey for construction work, is carried out based on inaccurate information, it is necessary to excavate a large number of locations, which results in a longer construction period and an increase in costs. In addition, there is a problem of high construction risks such as pipeline damage accidents and construction method changes due to unintended buried objects, etc.

Therefore, there is a demand for a technique for accurately judging the buried object without the work of a skilled person.

A representative example of the invention disclosed in the present application is as follows. That is, a buried object discriminating system for generating information on a buried object using an underground exploration image includes an arithmetic unit for executing predetermined arithmetic processing and a storage unit for storing data referred to in the arithmetic processing. an image generation unit that synthesizes B-mode exploration data and aperture synthesis exploration data to generate an underground exploration image; Having an embedded object identification unit for deriving embedded object information including position and size, an embedded object information unit for storing the derived buried object information, and an output unit for outputting the derived buried object information characterized by

ADVANTAGE OF THE INVENTION According to one aspect of the present invention, the situation of a buried object can be determined from the results of radar exploration without the need for skilled workers to perform operations. Problems, configurations, and effects other than those described above will be clarified by the following description of the embodiments.

It is a figure which shows the structure of a buried object discrimination|determination system. FIG. 4 is a diagram showing an example of teacher data input to the buried object learning model; 1 is a diagram showing a physical configuration of an embedded object discrimination system; FIG. 4 is a flowchart of processing executed by a validity determination unit; FIG. 4 is a diagram showing an example of an image input/output by an image generator; FIG. 4 is a diagram showing an example of an embedded object display screen displayed on a display terminal; FIG. 11 is a diagram showing an example of another buried object display screen displayed on the display terminal;

FIG. 1 is a diagram showing the configuration of an embedded object identification system 100 according to an embodiment of the present invention.

The buried object discrimination system 100 of this embodiment includes an image generation unit 110, a buried object identification unit 120, a buried object learning model 130, a validity determination unit 140, a positioning information integration unit 150, a buried object information unit 160, and a 2D/3D map. It has a display unit 170 . The buried object identification system 100 acquires exploration data from the exploration vehicle 200 . The buried object discrimination system 100 is also connected to a map information section 300 and acquires map information. Furthermore, the buried object discrimination system 100 is connected to the display terminal 400 and outputs the execution result of the program to the display terminal 400 .

The image generation unit 110 acquires the B-mode exploration data and the synthetic aperture exploration data from the exploration vehicle 200 , and outputs three-dimensional data obtained by synthesizing the B-mode exploration data and the synthetic aperture exploration data to the buried object identification unit 120 . For synthesis of the exploration data, for example, the B-mode exploration data is red, the aperture synthesis exploration data is blue, each exploration data is arranged in the RGB space, and a color image is generated by superimposing the two. Examples of images input/output by the image generation unit 110 will be described later with reference to FIG. Alternatively, the image generator 110 may directly input the survey data (B-mode survey data and aperture synthesis survey data) to the embedded object identification unit 120 without synthesizing the color image.

The buried object identification unit 120 uses the buried object learning model 130 to extract the radar response of the buried object from the input color image, and converts the extracted radar reaction into the buried object information (for example, the type, shape, position ( depth, horizontal position), and size), and outputs the derived buried object information to the validity determination unit 140 . The buried object identification unit 120 may derive the buried object information using the buried object learning model 130 based on artificial intelligence, or may derive the buried object information using a rule base or multi-dimensional clustering.

The buried object learning model 130 is an AI (Artificial Intelligence) model configured by a neural network that has learned image data of detection targets such as buried pipes and exclusion targets such as noise. The buried object learning model 130 may be learned by, for example, teacher data in which image data and the type, shape, position, and size of the buried object are associated with each other, as shown in FIG. The teacher data may include the material of the buried object. By including the material of the buried object in the training data, the buried object identification unit 120 can derive the material of the buried object from the exploration data.

The validity determination unit 140 compares the color image generated by the image generation unit 110 with the buried object information output from the buried object identification unit 120, judges the validity of the buried object information, and outputs corrected buried object information. . For example, the types of buried objects (buried pipes, cavities, subgrades, etc.) are corrected, discontinuous sections of buried pipes are interpolated, and material information is added. Details of the processing executed by the validity determination unit 140 will be described later with reference to FIG.

The positioning information integration unit 150 acquires the positioning data acquired at the same time as the exploration data, specifies the position (latitude/longitude information) of the buried object information, associates it with the buried object information, and stores it in the buried object information unit 160 . .

The buried object information unit 160 is a database configured by a storage device for accumulating data, and stores buried object information and position information in association with each other. For example, the buried object information section 160 stores the following data.
When the buried object is a buried pipe Type of buried object: Buried pipe Pipe type: Gas pipe, water pipe, sewer pipe, telecommunication pipe, etc. Size: Diameter [cm]
Materials: Steel pipes, rigid polyethylene, vinyl chloride, etc. Start point coordinates: latitude, longitude, depth End point coordinates: latitude, longitude, depth When buried objects are other than buried pipes Types of buried objects: manholes, cavities, etc. Size: Width [cm], height [cm]
Material: concrete, stone, water, air, etc. Starting point coordinates: latitude, longitude, depth Ending point coordinates: latitude, longitude, depth

The 2D/3D map display unit 170 displays the buried object information acquired from the buried object information unit 160 on the map acquired from the map information unit 300 in a format that meets the user's request based on the position information of the buried object information. Data to be superimposed and displayed is generated and output to the display terminal 400 . The 2D/3D map display unit 170 may display an image of the buried object superimposed on the map in a two-dimensional or three-dimensional format (see FIGS. 6 and 7). The 2D/3D map display unit 170 preferably has a web server function.

The rover 200 is a vehicle that acquires underground exploration data, positioning data, and terrestrial images while traveling on roads. Specifically, the intensity of the reflected wave of the radar wave irradiated to the road surface is measured, and exploration data (for example, a B-mode data image in which the intensity of the reflected wave is visualized) is output. Also, the position of the probe 200 is measured by GNSS or gyro. Furthermore, an omnidirectional camera installed in front of the exploration vehicle 200 captures a moving image of the ground (for example, an all-around image including from the ground surface to the zenith). The transfer of exploration data from the exploration vehicle 200 to the buried object identification system 100 may be via a network or via a storage medium such as a flash memory.

The map information unit 300 is configured by a storage device that accumulates data, and provides map information of the exploration range. The map information unit 300 may be provided outside the buried object identification system 100 as shown in the figure, or may be provided inside the buried object identification system 100 .

The display terminal 400 is configured by a user-operated computer (for example, a smartphone, a tablet terminal, or a portable computer). The display terminal 400 is connected to the buried object identification system 100, for example, through a network, and displays the display data generated by the 2D/3D map display section 170 on the screen. For example, the display terminal 400 may be a dedicated application, but if the buried object identification system 100 has a web server function, the display terminal 400 may run a web browser.

The buried object identification system 100 may be installed on the ground (for example, a data center) and may analyze in real time exploration data (B-mode exploration data, synthetic aperture exploration data) acquired from the exploration vehicle 200 via a wireless line. Alternatively, the buried object identification system 100 may be mounted on the exploration vehicle 200 to analyze exploration data in real time. In addition, a data collection server that accumulates exploration data acquired from the exploration vehicle 200 is provided between the buried object identification system 100 and the exploration vehicle 200, and the buried object identification system 100 batches the exploration data acquired from the data collection server. May be parsed in processing.

FIG. 3 is a diagram showing the physical configuration of the buried object discrimination system 100 of the embodiment of the present invention.

The buried object identification system 100 of this embodiment is configured by a computer having a processor (CPU) 101 , a memory 102 , an auxiliary storage device 103 and a communication interface 104 .

The processor 101 is an arithmetic unit that executes a program stored in the memory 102 to perform predetermined arithmetic processing. The memory 102 includes ROM, which is a non-volatile storage element, and RAM, which is a volatile storage element. The ROM stores immutable programs (eg, BIOS) and the like. RAM is a high-speed and volatile storage element such as DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by the processor 101 and data used when the programs are executed.

The auxiliary storage device 103 is, for example, a large-capacity, non-volatile storage device such as a magnetic storage device (HDD) or flash memory (SSD), and stores programs executed by the processor 101 and data used when the programs are executed. do. That is, the program is read from the auxiliary storage device 103, loaded into the memory 102, and executed by the processor 101. FIG.

Communication interface 104 is a network interface device that controls communication with other devices (map information unit 300, display terminal 400, etc.) according to a predetermined protocol.

The buried object identification system 100 may have an input interface 105 and an output interface 108 . The input interface 105 is an interface to which a keyboard 106, a mouse 107, etc. are connected and receives input from an operator. The output interface 108 is an interface to which the display device 109, a printer, etc. are connected and which outputs the result of execution of the program in a form that can be visually recognized by the operator.

The program executed by the processor 101 is provided to the embedded object discrimination system 100 via removable media (CD-ROM, flash memory, etc.) or a network, and stored in the non-volatile auxiliary storage device 103, which is a non-temporary storage medium. be. Therefore, the buried object discrimination system 100 preferably has an interface for reading data from removable media.

The buried object discrimination system 100 is a computer system configured on one physical computer or on a plurality of computers configured logically or physically, and operates on the same computer with separate threads. Alternatively, it may operate on a virtual computer built on a plurality of physical computer resources. Also, each functional unit of the buried object identification system 100 may be implemented on a different computer.

FIG. 4 is a flowchart of processing executed by the validity determination unit 140 .

The validity determination unit 140 determines whether the buried object information is appropriate according to the buried object type. For example, the validity determination unit 140 includes a shape identification unit that estimates the validity of the buried object information from the shape of the buried object, a continuity identification unit that identifies continuity with other buried objects, and a continuity identification unit based on the continuity of the buried pipe. A buried pipe interpolator for correcting a plurality of buried pipes into a series of buried pipes, a waveform discriminator for discriminating a material based on an image of exploration data (for example, B-mode exploration data), etc. may be configured by subprograms. In addition to the identification section described above, an identification section that identifies other elements according to the type of the buried object may be provided.

First, when the validity determination unit 140 acquires one piece of buried object information (type, shape, position, size of the buried object) (step 11), the shape identification unit determines whether the type of the buried object is “buried pipe”. is determined (step 12).

If the type of buried object is not "buried pipe" (N at step 12), the shape identification unit determines whether the type of buried object is "roadbed" (step 16). If the type of buried object is not "roadbed" (N at step 16), the buried object information is sent to the processing of step 19 (waveform identification unit). On the other hand, if the type of buried object is "roadbed" (Y in step 16), the shape identification unit determines whether the shape of the buried object is a plate (step 17). Since the roadbed is characterized by its horizontally extending plate-like shape, if the shape is plate-like (Y in step 17), it is determined that the buried object information of the "roadbed" is valid, and the buried object is buried. The object information is sent to the positioning information integration unit 150 . On the other hand, if the shape is not plate-like (N in step 17), the plate-like portion and the rest of the buried object information are separated, and the separated plate-like portion is sent to the positioning information integrating section 150 for separation. The rest of the buried object information is sent to the processing of step 19 (waveform identification section).

If the type of buried object is "buried pipe" (Y in step 12), the shape identification unit determines whether the aspect ratio is within the normal range (step 13). If the aspect ratio is not within the normal range (N at step 13), the buried object information is sent to the processing of step 19 (waveform identification section). On the other hand, if the aspect ratio is within the normal range (Y in step 13), the continuity identifying unit determines continuity with the nearby buried pipe (step 14). If there is no continuous buried pipe nearby (N at step 14), the buried object information is sent to the processing of step 19 (waveform identification section). On the other hand, if there is a continuous buried pipe nearby (Y in step 14), it is determined that the two buried pipes are detected as being separated, and the buried object information of the discontinuous section where the buried pipe is separated is displayed. It is generated by the buried pipe interpolation unit (step 15).

For example, as shown in FIG. 6, since the buried pipe 512 is present in the extension direction of the buried pipe 511, it is assumed that both are continuous, and there is a discontinuous section to be interpolated by the interpolated buried pipe between them. 521 and interpolated with buried pipe 531 . Similarly, between the buried pipes 512 and 513, between the buried pipes 514 and 515, and between the buried pipes 516 and 517 are discontinuous sections 522, 523 to be interpolated with the interpolated buried pipes. 524 and interpolated with buried pipes 532 , 533 , 534 .

After that, the waveform identification unit determines whether the type of the buried object is "buried pipe" (step 19). If the type of buried object is not "buried pipe" (N at step 19), go to step 22; If the type of buried object is "buried pipe" (Y in step 19), the waveform identification unit determines whether the length of the pipe is equal to or greater than a predetermined value (step 20). If the length of the pipe is equal to or greater than the predetermined value (Y in step 20), the waveform identifying section determines the material based on the characteristics of the waveform data, and adds the material information to the buried object information (step 21).

If the length of the pipe is shorter than a predetermined value (N in step 20), the waveform identification unit compares the survey data image with cavity features, and if the survey data image has cavity features, the buried object type is set to "cavity". (step 23), and if the image of the exploration data does not have cavity features, the type of buried object is changed to "other buried object" (step 24).

In addition to the determination processing shown in the flowchart, the following determination processing may be performed depending on the position, size, and material of the pipe.
・A buried object with unidirectional continuity in radar response and equal horizontal and depth dimensions is a buried pipe.
・A buried object with independent radar response and cavity-specific characteristics in response amplitude is a cavity.
・A subgrade is a buried object that has a constant depth in which radar response can be seen and that has continuity in the horizontal direction.
・A buried object whose type cannot be determined from its material and shape, although there is a radar response, is another buried object.
・The reaction of buried objects in places shallower than the subgrade is hollow (there are no buried objects in places shallower than the subgrade).
・A buried pipe with a diameter of 100 cm or more at a position shallower than 150 cm may be misidentified.
・A buried pipe that is not connected to other buried pipes and is independent is a residual pipe.
・The sloping buried pipe connected to the manhole is a sewage pipe.
・A buried pipe with a diameter of 50 cm or more at a position deeper than 150 cm is a sewage pipe.
- Buried pipes made of concrete are sewage pipes.
・The material of gas pipes deeper than 100 cm is metal.
- Buried pipes identified in the asphalt may be misidentified.
- Buried pipes that are clogged with soil, concrete, etc. are left pipes.

5A and 5B are diagrams showing examples of images input and output by the image generation unit 110. FIG.

FIG. 5A shows B-mode search data input to the image generator 110. FIG. The B-mode exploration data is image data in which the amplitude information of the radar reflected wave received by the underground exploration radar is displayed as brightness. A plurality of ground penetrating radar antennas are arranged in the exploration vehicle 200 so as to be orthogonal to the vehicle traveling direction (lateral line direction), and a plurality of B-mode exploration data are generated in one scan.

FIG. 5B shows aperture synthesis exploration data input to the image generator 110 . Aperture synthesis survey data is image data obtained by subjecting the reflection of the underground exploration radar to aperture synthesis and converting the reflection of the radar into 3D.

FIG. 5C shows a color image output by the image generator 110. FIG. As described above, the color image output by the image generation unit 110 is obtained by placing the B-mode exploration data in red, the aperture synthesis exploration data in blue, and the intensity of each exploration data (image brightness) in the RGB space. is a superimposed image. The color image output by the image generation unit 110 includes the features of both the B-mode exploration data and the aperture synthesis exploration data, and the features of both exploration data are mapped on independent axes. By allowing the learning model 130 to learn and inputting the one type of image to the buried object identification unit 120, the buried object can be appropriately determined.

In particular, B-mode survey data is useful for judging the depth and material of buried objects, but it has the property of being difficult to judge the horizontal position and size. On the other hand, synthetic aperture survey data is useful for determining the horizontal position and size of buried objects, but it is difficult to determine the material. Therefore, by using an image obtained by synthesizing the B-mode search data and the synthetic aperture search data, the processing cost can be suppressed and highly accurate identification can be realized.

FIG. 6 is a diagram showing an example of the buried object display screen 500 displayed on the display terminal 400. As shown in FIG.

The display data generated by the 2D/3D map display unit 170 is displayed on the display device of the display terminal 400 .

As described above, there are discontinuous sections 521 to 524 of the buried pipe in the displayed exploration range. In general, buried pipes are laid long along the direction of travel of the road, or laid in a direction crossing the road, and are not interrupted in the middle. Therefore, the continuity identification unit of the validity determination unit 140 described above interpolates the discontinuous sections 521 to 524 with the buried pipes 531 to 534 and identifies them as continuous buried pipes. The buried pipe identified by the buried object identification unit 120 and the interpolated buried pipe obtained by interpolating the discontinuous section may be displayed in different manners (for example, different colors) so as to be distinguished from each other.

In addition, buried pipes 541 to 543 oblique to the direction of travel of the road are detected. As described above, since buried pipes are usually laid in the longitudinal or transverse direction of the road, the oblique buried pipes 541-543 may be erroneously detected. The buried object learning model 130 is made to learn the information of the erroneously detected oblique buried pipe, and the accuracy of identifying the buried object can be improved.

Also, multiple cavities are identified. For example, as described above, a buried pipe identified at a shallow position is assumed to be erroneously detected and determined to be hollow.

FIG. 7 is a diagram showing an example of another buried object display screen 600 displayed on the display terminal 400. As shown in FIG.

On the buried object display screen 600, the recognized buried object is displayed on the map according to the latitude and longitude information. When the user designates the buried object, the information (for example, accuracy score) of the buried object identified by the buried object identification unit 120 is displayed.

As described above, in the buried object discrimination system of the embodiment of the present invention, the buried object learning model 130 is used to derive buried object information including the position and size of the buried object from the underground exploration image. Since it has an identification unit 120, a buried object information unit 160 that stores derived buried object information, and an output unit (2D/3D map display unit 170) that outputs the derived buried object information, a skilled worker can easily perform the work. It is possible to determine the status of buried objects from the results of radar exploration without having to In addition, it is possible to grasp the positions of buried pipes and cavities to be constructed without conducting trial excavation. In addition, it is possible to grasp the existence and position of buried objects that will serve as construction barriers without conducting test excavation. In addition, it is possible to design a new pipeline without conducting a preliminary survey using drawings or trial drilling.

In addition, since the buried object identification unit 120 derives buried object information including the type and shape of the buried object from the underground exploration image, the type and shape of the buried object, which conventionally required a high degree of expertise, can be easily determined. I can judge.

In addition, since the image generation unit 110 is provided for synthesizing the B-mode exploration data and aperture synthesis exploration data to generate an underground exploration image to be input to the buried object identification unit 120, the processing cost can be suppressed and highly accurate identification can be achieved. realizable.

In addition, since the buried object learning model 130 is composed of a neural network learned from buried object information associated with an underground exploration image, the knowledge of an expert who identifies radar responses can be replaced by AI technology based on Deep Learning.

Moreover, since it has a validity determination unit 140 that determines the validity of the type and shape of the derived buried object, as a post-processing of identification by the buried object identification unit 120, the type of the buried object is closely examined, and insufficient information is obtained. can be interpolated.

Based on the derived type and size of the buried object, if the type is a buried pipe and the length of the buried pipe is shorter than a predetermined value, the validity determination unit 140 determines the type of the buried object as a hollow pipe. Or change to another buried object, so that the type of buried object identified from the radar response can be corrected to the appropriate type.

Further, when it is determined that the type of the derived buried object is a buried pipe and the position of the buried pipe has continuity with other buried pipes, the validity determination unit 140 determines that the buried object and the other buried pipe Since the buried object information for interpolating the buried pipe in the discontinuous portion between the pipes is generated, the discontinuous portion can be interpolated even if the radar response is intermittent.

In addition, since the validity determination unit 140 determines the material of the buried object from the waveform of the B-mode exploration data, the material of the buried object can be determined using the B-mode exploration data useful for determining the material.

It should be noted that the present invention is not limited to the embodiments described above, but includes various modifications and equivalent configurations within the scope of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not necessarily limited to those having all the described configurations. Also, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Moreover, the configuration of another embodiment may be added to the configuration of one embodiment. Further, additions, deletions, and replacements of other configurations may be made for a part of the configuration of each embodiment.

In addition, each configuration, function, processing unit, processing means, etc. described above may be realized by hardware, for example, by designing a part or all of them with an integrated circuit, and the processor realizes each function. It may be realized by software by interpreting and executing a program to execute.

Information such as programs, tables, and files that implement each function can be stored in storage devices such as memories, hard disks, SSDs (Solid State Drives), or recording media such as IC cards, SD cards, and DVDs.

In addition, the control lines and information lines indicate those considered necessary for explanation, and do not necessarily indicate all the control lines and information lines necessary for mounting. In practice, it can be considered that almost all configurations are interconnected.

100 buried object discrimination system 101 processor 102 memory 103 auxiliary storage device 104 communication interface 110 image generation unit 120 buried object identification unit 130 buried object learning model 140 validity judgment unit 150 positioning information integration unit 160 buried object information unit 170 2D/3D map Display unit 200 Survey vehicle 300 Map information unit 400 Display terminal

Claims (8)

A buried object identification system that generates information on a buried object using an underground exploration image,
a calculation unit that executes predetermined calculation processing;
A storage unit that stores data referred to in the arithmetic processing,
an image generation unit that generates an underground exploration image by synthesizing the B-mode exploration data and the synthetic aperture exploration data;
a buried object identification unit that uses the buried object learning model to derive buried object information including the position and size of the buried object from the underground exploration image generated by the image generation unit ;
an embedded object information unit that stores the derived embedded object information;
and an output unit for outputting the derived buried object information. The buried object discrimination system according to claim 1,
The buried object identifying unit derives buried object information including the type and shape of the buried object from the underground exploration image. The buried object discrimination system according to claim 1 or 2,
A buried object discrimination system, wherein the buried object learning model is composed of a neural network learned from buried object information associated with an underground exploration image. The buried object discrimination system according to claim 2,
An embedded object determination system, comprising a validity determination unit that determines validity of the derived type and shape of the embedded object. The buried object discrimination system according to claim 4,
If the type is a buried pipe and the length of the buried pipe is shorter than a predetermined value based on the derived type and size of the buried object, the validity determination unit determines the type of the buried object as a hollow pipe. or another buried object. The buried object discrimination system according to claim 4 or 5,
When it is determined that the derived buried object is a buried pipe and the position of the buried pipe is continuous with another buried pipe, the validity determination unit determines whether the buried object and the other buried pipe A buried object identification system characterized by generating buried object information that interpolates an buried pipe at a discontinuous portion between pipes. The buried object discrimination system according to any one of claims 4 to 6,
The buried object determination system, wherein the validity determination unit determines the material of the buried object from the waveform of the B-mode exploration data. A buried object identification method in which a buried object identification system generates information on a buried object using an underground exploration image,
The buried object identification system is composed of a computer having a calculation unit that executes predetermined calculation processing and a storage unit that stores data referred to in the calculation processing,
The buried object discrimination method includes:
an image generation procedure for synthesizing the B-mode exploration data and the synthetic aperture exploration data to generate an underground exploration image;
a buried object identification procedure for deriving buried object information including the position and size of the buried object from the underground exploration image generated in the image generation procedure using the buried object learning model;
a buried object information storing procedure for storing the derived buried object information;
and an output procedure for outputting the derived buried object information.

Images