Method and system for underground pipeline detection

Abstract

The invention belongs to the technical field of underground pipeline detection, and discloses a method and system for underground pipeline detection, and the method comprises the steps: determining a detection region of an underground pipeline, obtaining a pipeline distribution diagram of the detection region, moving a detection vehicle according to a set driving route, and collecting ground penetrating radar data and a top view of the detection region, obtaining a three-dimensional image and N1 first cross section images of a detection area based on ground penetrating radar data, inputting to-be-analyzed data into a corresponding preset model to obtain a corresponding identification result base, and carrying out calibration processing on the pipeline type of each pipeline in each piece of to-be-analyzed data according to all identification results, and outputting the calibrated identification result of each piece of to-be-analyzed data. According to the technical scheme of the invention, manpower and material resources can be saved, and the precision and accuracy of underground pipeline detection can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of underground pipeline detection, and in particular relates to a method and a system for underground pipeline detection. Background Art

[0002] With the acceleration of urbanization, cities have an increasing demand for underground pipelines. Due to the wide distribution range, diverse types and strong concealment of underground pipeline networks, the construction and development of urban municipal underground pipeline facilities are becoming increasingly prominent. Research on underground pipeline detection technology can quickly and accurately detect the spatial location, attributes, direction and buried depth of underground pipelines without excavation, measure the coordinates and draw a map to achieve visual map management, which is crucial to today's urban pipeline construction and urban modernization.

[0003] Similar prior art includes a Chinese patent application with publication number CN117092644A, which discloses an underground pipeline detection method, which obtains the initial information of the target pipeline from a large database; combines the initial information of the target pipeline and the actual construction situation around the current site to determine the detection plan of the target pipeline; drills a first detection port on the ground at a first detection point, and places a first transmitting device inside the first detection port; drills a second detection port on the ground at a second detection point, and places a second transmitting device inside the second detection port; starts the first transmitting device to transmit a radar signal, and after the signal receiving device receives the reflected radar signal, transmits the data to the data processing device; starts the second transmitting device to transmit a radar signal, and after the signal receiving device receives the reflected radar signal, transmits the data to the data processing device; the data processing device processes the two data to calculate the position and depth information of the target pipeline; summarizes and packages the re-measured target pipeline position and depth information and the surrounding construction information, and uploads them to the large database for storage. When performing underground pipeline detection, this method needs to drill a detection port based on the last detection information, which is highly complex and consumes manpower and material resources. There is also a Chinese patent application with publication number CN111679268A, which discloses a method for detecting underground pipelines, which includes determining a first point on a path perpendicular to the marking tape, sending a radar signal underground through a detector, causing the marking tape to reflect the radar signal to the detector, and calculating the distance between the detector and the marking tape; determining a second point on a path perpendicular to the marking tape, and the second point is determined in a direction away from the marking tape and the first point, sending a radar signal underground through a detector, causing the marking tape to reflect the radar signal to the detector, and calculating the distance between the detector and the marking tape; determining a vertical origin of the marking tape on the vertical path, calculating the distance between the vertical origin and the first point and the distance between the marking tape and the vertical origin; using a direct connection method to detect the vertical distance between one of the pipelines and the ground, placing the detector at the vertical origin, sending a radar signal underground through the detector, calculating the distance between the vertical origin and the pipeline to be tested and the straight-line distance between the pipeline to be tested and the marking tape; determining the positional relationship between the pipeline to be tested and the marking tape, and determining the positional relationship between several other pipelines and the marking tape; and drawing a cross-sectional positional relationship diagram of the marking tape and several pipelines. This method sets multiple detection positions on the ground to send radar signals to the underground and obtain measurement information, and draws a single profile image based on the measurement information, which has low detection efficiency.

[0004] Therefore, it is an urgent problem to provide a method and system for underground pipeline detection to save manpower and material resources and improve the efficiency, precision and accuracy of underground pipeline detection. Summary of the invention

[0005] In view of the above-mentioned technical problems, the present invention provides a method and system for underground pipeline detection.

[0006] In a first aspect, the present invention provides a method for detecting underground pipelines, the method comprising the following steps:

[0007] Step 1: determine the detection area of ​​underground pipelines and obtain the pipeline distribution map of the detection area;

[0008] Step 2: Move the detection vehicle according to the set driving route to collect ground penetrating radar data and a bird's-eye view of the detection area;

[0009] Step 3: Acquire a three-dimensional image and N1 first cross-sectional images of the detection area based on the ground penetrating radar data, wherein the first cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline;

[0010] Step 4: Input the data to be analyzed into the corresponding preset model to obtain the corresponding recognition results, wherein the data to be analyzed includes the pipeline distribution map, the top view, the three-dimensional image and N1 first cross-sectional images, and the recognition results include the pipeline type, pipeline profile and pipeline coordinates;

[0011] Step 5: Calibrate the pipeline type of each pipeline in each data to be analyzed based on all identification results;

[0012] Step 6: Output the calibrated recognition results of each data to be analyzed.

[0013] Specifically, step 5 includes:

[0014] Step 51: extract any pipeline in the three-dimensional image, and obtain the first type and first center coordinates of any pipeline;

[0015] Step 52: Based on the first center coordinate, determine whether there is a second pipeline of the first type within the preset range of the second data to be analyzed; if yes, generate a first comparison set based on any pipeline and the second pipeline; if no, proceed to step 55, wherein the second data to be analyzed includes a pipeline distribution map, a top view and a first cross-sectional image, and the center point of the preset range is the first center coordinate;

[0016] Step 53, calculating the first score values ​​of all pipelines in the first comparison set, and determining whether the first score value is greater than or equal to a first preset value, if so, proceeding to step 54, if not, proceeding to step 55;

[0017] Step 54: determine that the pipeline types of all pipelines in the first comparison set are the first type, and mark the first type on the corresponding pipeline of each data to be analyzed;

[0018] Step 55: determine whether there is a third pipeline whose pipeline type is different from the first type within the preset range of the second data to be analyzed; if yes, generate a second comparison set based on any pipeline and the third pipeline; if no, mark the first type on any pipeline of the three-dimensional image;

[0019] Step 56, calculating the second score values ​​of all pipelines in the second comparison set, and determining whether the second score value is greater than or equal to a second preset value, if so, proceeding to step 57, if not, proceeding to step 58;

[0020] Step 57: Determine the pipeline type corresponding to the second comparison set based on the specific identification parameter values ​​of any pipeline and the third pipeline, and define it as the third type. At the same time, set the pipeline types of all pipelines in the second comparison set to the third type, and mark the third type on the corresponding pipeline of each data to be analyzed;

[0021] Step 58: Mark the pipeline type corresponding to each pipeline in the second comparison set on the corresponding pipeline of the corresponding data to be analyzed.

[0022] Specifically, when there are N2 second pipelines within the preset range of any second data to be analyzed, N3 first comparison sets are generated, the first set score value of each first comparison set is calculated respectively, and the maximum value of the first set score value is taken as the first score value, where N2 and N3 are positive integers greater than 1.

[0023] Specifically, when there are N4 third pipelines within the preset range of any second data to be analyzed, N5 second comparison sets are generated, the second set score value of each second comparison set is calculated respectively, and the maximum value of the second set score value is taken as the second score value, where N4 and N5 are positive integers greater than 1.

[0024] Specifically, the identification result includes a specific identification parameter value corresponding to the pipeline type of each pipeline, and step 57 includes:

[0025] Step 571, obtaining a specific identification parameter value corresponding to each pipeline in the second comparison set;

[0026] Step 572, sorting the specific identification parameter values ​​in ascending order;

[0027] Step 573: define the pipeline type corresponding to the maximum value of the specific identification parameter as the third type.

[0028] Specifically, the scoring value is calculated by the three-dimensional intersection ratio of all pipelines in the comparison set or the average value of the distance between the center point coordinates of all pipelines, wherein the scoring value includes a first scoring value and a second average value, and the comparison set includes a first comparison set and a second comparison set.

[0029] Specifically, the detection vehicle travels back and forth along the pipeline arrangement direction, and step 3 includes:

[0030] Step 31, based on the acquisition position of the ground penetrating radar data, obtaining a second cross-sectional image at the acquisition position, wherein the second cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline;

[0031] Step 32: stitch all the second cross-sectional images located on the first horizontal plane corresponding to any pipeline cross section to generate an initial cross-sectional image corresponding to the first horizontal plane, and after traversing all the horizontal planes corresponding to the pipeline cross sections, generate N1 initial cross-sectional images;

[0032] Step 33, extract any initial cross-sectional image, analyze any initial cross-sectional image, and obtain all reflection wave curves in any initial cross-sectional image;

[0033] Step 34, extract any reflection wave curve, determine whether any reflection wave curve is a curve corresponding to the pipeline, if so, retain it, if not, delete it, after traversing all reflection wave curves, obtain a first cross-sectional image corresponding to any initial cross-sectional image;

[0034] Step 35, after traversing all the initial cross-sectional images, generate N1 first cross-sectional images;

[0035] Step 36: Generate a three-dimensional image based on all the first cross-sectional images.

[0036] Specifically, the process before step 34 includes:

[0037] Step 331, obtaining a specific component image of any initial cross-sectional image;

[0038] Step 332: remove the specific component image from any initial cross-sectional image, and obtain a third cross-sectional image corresponding to any initial cross-sectional image;

[0039] Step 333: perform Fourier transform on the specific component image and the third cross-sectional image respectively to generate a first spectrum map and a second spectrum map, and perform smoothing processing on the first spectrum map and the second spectrum map to generate a third spectrum map and a fourth spectrum map;

[0040] Step 334: Obtain a first frequency value corresponding to the maximum amplitude of the third spectrum graph and a second frequency value corresponding to the maximum amplitude of the fourth spectrum graph, and use an average value of the first frequency value and the second frequency value as a demarcation value.

[0041] Specifically, step 34 includes:

[0042] Step 341, performing Fourier transform on any reflected wave curve to generate a fifth spectrum diagram;

[0043] Step 342: selecting a first frequency range before the demarcation value and a second frequency range after the demarcation value on the fifth frequency spectrum based on the demarcation value;

[0044] Step 343, calculating a first accumulated value of all amplitudes within the first frequency range and a second accumulated value of all amplitudes within the second frequency range, and calculating a ratio of the second accumulated value to the first accumulated value;

[0045] Step 344: when the ratio is greater than or equal to the third preset value, it is determined that any reflected wave curve is a curve corresponding to the pipeline; when the ratio is less than the third preset value, it is determined that any reflected wave curve is not a curve corresponding to the pipeline.

[0046] In a second aspect, the present invention also provides a system for underground pipeline detection, the system comprising: a detection vehicle and a detection and analysis device, the detection and analysis device comprising a first data acquisition module, a second data acquisition module, an image generation module, an analysis and recognition module, a type calibration module and a result output module;

[0047] The detection vehicle is equipped with ground penetrating radar equipment and image acquisition equipment;

[0048] A first data acquisition module is used to determine the detection area of ​​underground pipelines and obtain a pipeline distribution map of the detection area;

[0049] The second data acquisition module is used to move the detection vehicle according to the set driving route to collect ground penetrating radar data and a bird's-eye view of the detection area;

[0050] An image generation module, used to obtain a three-dimensional image and N1 first cross-sectional images of the detection area according to the ground penetrating radar data, wherein the first cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline;

[0051] An analysis and recognition module is used to input the data to be analyzed into a corresponding preset model to obtain corresponding recognition results, wherein the data to be analyzed includes a pipeline distribution map, a top view, a three-dimensional image and N1 first cross-sectional images, and the recognition results include pipeline type, pipeline profile, and pipeline coordinates;

[0052] A type calibration module is used to calibrate the pipeline type of each pipeline in each data to be analyzed according to all identification results;

[0053] The result output module is used to output the calibrated recognition results of each data to be analyzed.

[0054] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0055] 1. Install ground-penetrating radar equipment and image acquisition equipment on the detection vehicle, obtain ground-penetrating radar data based on the ground-penetrating radar equipment, and obtain ground images of the detection area through the image acquisition equipment, saving manpower and material resources;

[0056] 2. Based on the ground penetrating radar data, a three-dimensional image and multiple cross-sectional data of the detection area are obtained. Then, the three-dimensional image, cross-sectional data, ground image and pipeline distribution map are analyzed respectively, and the pipeline type, pipeline profile and pipeline coordinates of each pipeline in the multiple data are identified respectively. Then, the identification results are calibrated to obtain the final identification results, which improves the precision and accuracy of underground pipeline detection;

[0057] 3. The recognition results of three-dimensional images, cross-sectional data, ground images and pipeline distribution maps are output respectively, and the underground pipeline detection efficiency is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0059] Figure 1 A flow chart of a method for underground pipeline detection according to the present invention;

[0060] Figure 2 It is a modular schematic diagram of the system for underground pipeline detection of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Obviously, the specific embodiments described herein are only used to explain the present invention and are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0062] It should be noted that if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0063] Figure 1 FIG. 1 is a flow chart of an embodiment of a method for underground pipeline detection provided by the present invention, and the flow chart specifically includes the following steps:

[0064] Step 1: Determine the detection area of ​​underground pipelines and obtain the pipeline distribution map of the detection area.

[0065] The pipeline distribution map displays the pipeline identification, pipeline type, pipeline size, pipeline shape and designed location of all pipelines designed and installed in the detection area.

[0066] The pipelines in the detection area may be pipelines of the same type or pipelines of different types. For example, the pipeline types include water pipes, gas pipes, oil pipes, power lines, communication lines, and the like.

[0067] Step 2: Move the detection vehicle along the set driving route to collect ground penetrating radar data and a bird's-eye view of the detection area.

[0068] The detection vehicle is provided with a ground penetrating radar device and an image acquisition device. The ground penetrating radar device is used to collect ground penetrating radar data of the detection area, and the image acquisition device is used to collect a top view of the detection area. There can be multiple ground penetrating radar devices, which are arranged in sequence and installed in the width direction of the detection vehicle, and there can also be multiple image acquisition devices.

[0069] The above-mentioned top view is a ground image of the entire detection area generated by stitching multiple local images.

[0070] Step 3: Acquire a three-dimensional image and N1 first cross-sectional images of the detection area based on the ground penetrating radar data, wherein the first cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline.

[0071] The first cross-sectional image is a cross-sectional image of the detection area perpendicular to the ground and perpendicular to the length direction of the pipeline. N1 is related to the number of times the ground-penetrating radar equipment sends radar signals underground. For example, if the detection vehicle travels along the driving route once and sends radar signals underground 10 times, then N1 is 10; if the detection vehicle travels along the driving route twice and sends a total of 20 radar signals underground, 10 radar signals are sent underground when driving forward and 10 radar signals are sent underground when driving in the reverse direction, then N1 is 10.

[0072] Specifically, the detection vehicle travels back and forth along the pipeline arrangement direction, and step 3 includes:

[0073] Step 31: Based on the acquisition position of the ground penetrating radar data, a second cross-sectional image at the acquisition position is acquired, wherein the second cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline.

[0074] Step 32: stitch all second cross-sectional images located on the first horizontal plane corresponding to any pipeline cross section to generate an initial cross-sectional image corresponding to the first horizontal plane. After traversing all horizontal planes corresponding to the pipeline cross sections, N1 initial cross-sectional images are generated.

[0075] Step 33: extract any initial cross-sectional image, analyze any initial cross-sectional image, and obtain all reflection wave curves in any initial cross-sectional image.

[0076] Step 34, extract any reflection wave curve, determine whether any reflection wave curve is a curve corresponding to the pipeline, if so, retain it, if not, delete it, and after traversing all reflection wave curves, obtain a first cross-sectional image corresponding to any initial cross-sectional image.

[0077] Step 35: After traversing all the initial cross-sectional images, N1 first cross-sectional images are generated.

[0078] Step 36: Generate a three-dimensional image based on all the first cross-sectional images.

[0079] The edges of the detection area perpendicular to the pipeline arrangement direction are divided into M segments, and M driving routes are generated. The detection vehicle travels back and forth along the M driving routes. The value of M is set according to the coverage range of the ground penetrating radar equipment on the detection vehicle.

[0080] The first horizontal plane is a plane where any pipeline cross section is located, and all second cross-sectional images located on the first horizontal plane are spliced ​​to generate an initial cross-sectional image corresponding to the first horizontal plane.

[0081] When collecting radar detection data, a three-dimensional coordinate system is established, and the axis parallel to the driving route is defined as the x-axis, the axis perpendicular to the driving route is defined as the y-axis, and the axis vertically downward is defined as the z-axis. After the driving route is generated, the x-coordinate value is set each time the radar signal is emitted underground. Based on this, the detection vehicle can emit radar signals at a fixed x-coordinate value when traveling back and forth along the driving route. Based on this, the generated initial cross-sectional image is a continuous cross-sectional image covering all underground pipeline features on the first horizontal plane.

[0082] Since the initial cross-sectional image is generated by stitching together multiple second cross-sectional images arranged in parallel, if there is a "V"-shaped waveform on the second cross-sectional image, two adjacent second cross-sectional images may generate a "W"-shaped waveform after stitching together, and there is an "Λ"-shaped interference waveform. In order to improve the reliability and accuracy of underground pipeline detection, it is necessary to identify and remove the interference waveform.

[0083] Specifically, the process before step 34 includes:

[0084] Step 331: Acquire a specific component image of any initial cross-sectional image.

[0085] Step 332: remove the specific component image from any initial cross-sectional image, and obtain a third cross-sectional image corresponding to any initial cross-sectional image.

[0086] Step 333: Perform Fourier transform on the specific component image and the third cross-sectional image respectively to generate a first spectrum map and a second spectrum map, and perform smoothing processing on the first spectrum map and the second spectrum map to generate a third spectrum map and a fourth spectrum map.

[0087] Step 334: Obtain a first frequency value corresponding to the maximum amplitude of the third spectrum graph and a second frequency value corresponding to the maximum amplitude of the fourth spectrum graph, and use an average value of the first frequency value and the second frequency value as a demarcation value.

[0088] Specific components are signals that appear continuously in the initial cross-sectional images. They usually represent general characteristics of the underground environment (external environmental characteristics of the pipeline) rather than specific targets or anomalies.

[0089] Exemplarily, a specific component image is acquired through median filtering, Gaussian filtering, bandpass filtering, principal component analysis, etc.

[0090] Smoothing the first spectrum of the specific component image and the second spectrum of the third cross-sectional image can effectively remove random noise in the radar detection data while retaining the basic trends and characteristics of the signal for subsequent analysis.

[0091] The first frequency value of the third spectrum graph is related to the specific component image (i.e., background noise), and the second frequency value of the fourth spectrum graph is related to the signal after removing the specific component. By using the average of the first frequency value and the second frequency value, the middle point of these characteristics can be captured, thereby better performing effective waveform analysis.

[0092] Specifically, step 34 includes:

[0093] Step 341: Perform Fourier transform on any reflected wave curve to generate a fifth spectrum diagram.

[0094] Step 342: Select a first frequency range before the delimiting value and a second frequency range after the delimiting value on the fifth frequency spectrum based on the delimiting value.

[0095] Step 343: Calculate a first accumulated value of all amplitudes within the first frequency range and a second accumulated value of all amplitudes within the second frequency range, and calculate a ratio of the second accumulated value to the first accumulated value.

[0096] Step 344: when the ratio is greater than or equal to the third preset value, it is determined that any reflected wave curve is a curve corresponding to the pipeline; when the ratio is less than the third preset value, it is determined that any reflected wave curve is not a curve corresponding to the pipeline.

[0097] The spans of the first frequency range and the second frequency range are equal, that is, the lengths of the first frequency range and the second frequency range are equal.

[0098] The third preset value is set according to the experience of those skilled in the art or according to actual application scenarios, and the present application embodiment does not limit this.

[0099] When a radar signal is transmitted underground, the reflected waves generated by real underground structures (such as pipelines, rock layers, etc.) usually show a certain degree of continuity and stability in the frequency domain. If the reflected waves still have significant power at higher frequencies, this indicates that the signal has a strong energy distribution.

[0100] If the ratio is greater than or equal to the third preset value, it means that the reflected wave still maintains a high energy level after attenuation, and the reflected wave curve is determined to be the curve corresponding to the pipeline. If the ratio is less than the third preset value, it means that the reflected wave decays rapidly in the high-frequency region and the attenuation amplitude is large, and the reflected wave curve is determined to be not the curve corresponding to the pipeline, that is, the interference curve.

[0101] Step 4: Input the data to be analyzed into the corresponding preset model to obtain the corresponding recognition results, wherein the data to be analyzed includes a pipeline distribution map, a top view, a three-dimensional image and N1 first cross-sectional images, and the recognition results include pipeline type, pipeline outline, and pipeline coordinates.

[0102] The above preset model is a neural network model.

[0103] Exemplarily, the model corresponding to the pipeline distribution map is the first preset model, and the pipeline distribution map is input into the first preset model to obtain a first recognition result corresponding to the pipeline distribution map; the model corresponding to the top view is the second preset model, and the top view is input into the second preset model to obtain a second recognition result corresponding to the top view; the model corresponding to the three-dimensional image is the third preset model, and the three-dimensional image and generation parameters are input into the third preset model to obtain a third recognition result corresponding to the three-dimensional image; the model corresponding to the cross-sectional image is the fourth preset model, and the first cross-sectional image and the above-mentioned generation parameters are input into the fourth preset model to obtain a fourth recognition result corresponding to the first cross-sectional image, and the above-mentioned generation parameters are tinting intensity value and / or dielectric constant, etc.

[0104] Before performing step 5, the sizes of the pipeline distribution map, the top view and the first cross-sectional image are adjusted based on the three-dimensional image, so that the adjusted pipeline distribution map, the top view and the first cross-sectional image are aligned with the three-dimensional image in the three-dimensional space, thereby ensuring the accuracy and consistency of the data to be analyzed in the three-dimensional space, reducing the errors generated in the data acquisition process, and improving the calibration accuracy of the pipeline type.

[0105] Step 5: Calibrate the pipeline type of each pipeline in each data to be analyzed based on all identification results.

[0106] Specifically, step 5 includes:

[0107] Step 51: extract any pipeline in the three-dimensional image, and obtain the first type and first center coordinates of any pipeline.

[0108] Step 52: Based on the first center coordinate, determine whether there is a second pipeline of the first type within the preset range of the second data to be analyzed; if so, generate a first comparison set based on any one pipeline and the second pipeline; if not, proceed to step 55, wherein the second data to be analyzed includes a pipeline distribution map, a top view and a first cross-sectional image, and the center point of the preset range is the first center coordinate.

[0109] Step 53, calculate the first score value of all pipelines in the first comparison set, and determine whether the first score value is greater than or equal to a first preset value. If so, proceed to step 54, if not, proceed to step 55.

[0110] Step 54: Determine that the pipeline types of all pipelines in the first comparison set are the first type, and mark the first type on the corresponding pipeline of each data to be analyzed.

[0111] Step 55, determine whether there is a third pipeline whose pipeline type is different from the first type within the preset range of the second data to be analyzed. If so, generate a second comparison set based on any pipeline and the third pipeline. If not, mark the first type on any pipeline in the three-dimensional image.

[0112] Step 56, calculate the second score values ​​of all pipelines in the second comparison set, and determine whether the second score value is greater than or equal to a second preset value. If so, proceed to step 57, if not, proceed to step 58.

[0113] Step 57: Determine the pipeline type corresponding to the second comparison set based on the specific identification parameter values ​​of any pipeline and the third pipeline, and define it as the third type. At the same time, set the pipeline type of all pipelines in the second comparison set to the third type, and mark the third type on the corresponding pipeline of each data to be analyzed.

[0114] Step 58: Mark the pipeline type corresponding to each pipeline in the second comparison set on the corresponding pipeline of the corresponding data to be analyzed.

[0115] The first preset value, the preset range, and the second preset value are set according to the experience of those skilled in the art or according to the actual application scenario, and the embodiments of the present application do not limit this. For example, the first preset value is 0.9, and the first preset value and the second preset value may be equal or unequal.

[0116] If the first score value is greater than or equal to the first preset value, it means that the pipelines identified as the same type in the first comparison set are the same pipeline. If the first score value is less than the first preset value, it means that the pipelines identified as the same type in the first comparison set are not the same pipeline, and a second comparison set is further generated, and the pipelines in the second comparison set are identified as pipelines of different types.

[0117] If the second score value is greater than or equal to the second preset value, it means that the pipelines identified as different types in the second comparison set are the same pipeline. Due to the recognition error, the same pipeline is identified as different pipeline types, and the pipeline type corresponding to the second comparison set is determined based on the specific recognition parameter value. If the second score value is less than the second preset value, it means that the pipelines identified as different types in the second comparison set are not the same pipeline.

[0118] Preferably, after completing the pipeline type calibration processing of all pipelines in the three-dimensional image, determine whether there are any uncalibrated pipelines in any second data to be analyzed. If so, calibrate the uncalibrated pipelines in any second data to be analyzed according to the method of steps 51 to 58.

[0119] Specifically, when there are N2 second pipelines within the preset range of any second data to be analyzed, N3 first comparison sets are generated, the first set score value of each first comparison set is calculated respectively, and the maximum value of the first set score value is taken as the first score value, where N2 and N3 are positive integers greater than 1.

[0120] For example, if the pipeline type of any pipeline A11 in the three-dimensional image is a water pipe, if there are two second pipelines A21 and A22 of the pipeline type being water pipes within the preset range of the first cross-sectional image, and there is one second pipeline of the pipeline type being water pipe within the preset range of the pipeline distribution map and the top view, which are A31 and A41 respectively, then two first comparison sets (A11, A21, A31, A41) and (A11, A22, A31, A41) can be generated; if the preset range of the first cross-sectional image contains There are two second pipelines A21 and A22 whose pipeline type is water pipes, there are two second pipelines A31 and A32 whose pipeline type is water pipes within the preset range of the pipeline distribution map, and there is one second pipeline A41 whose pipeline type is water pipe within the preset range of the top view, then four first comparison sets (A11, A21, A31, A41), (A11, A21, A32, A41), (A11, A22, A31, A41) and (A11, A22, A32, A41) can be generated.

[0121] Specifically, when there are N4 third pipelines within the preset range of any second data to be analyzed, N5 second comparison sets are generated, the second set score value of each second comparison set is calculated respectively, and the maximum value of the second set score value is taken as the second score value, where N4 and N5 are positive integers greater than 1.

[0122] Specifically, the identification result includes a specific identification parameter value corresponding to the pipeline type of each pipeline, and step 57 includes:

[0123] Step 571: Obtain the specific identification parameter value corresponding to each pipeline in the second comparison set.

[0124] Step 572: Sort the specific identification parameter values ​​in ascending order.

[0125] Step 573: define the pipeline type corresponding to the maximum value of the specific identification parameter as the third type.

[0126] Exemplarily, the above-mentioned specific identification parameter is the recognition accuracy of the pipeline type.

[0127] The pipeline type recognition result with a low specific recognition parameter value is regarded as a misplaced recognition result, and the pipeline type recognition result with a high specific recognition parameter value is regarded as a correct recognition result.

[0128] Specifically, the scoring value is calculated by the three-dimensional intersection ratio of all pipelines in the comparison set or the average value of the distance between the center point coordinates of all pipelines, wherein the scoring value includes a first scoring value and a second average value, and the comparison set includes a first comparison set and a second comparison set.

[0129] The calculation methods of the first score value and the second average value may be the same or different.

[0130] According to the technical solution of the present invention, the comprehensiveness and accuracy of underground pipeline detection are improved by integrating the pipeline distribution map, the top view, the three-dimensional image and the first cross-sectional image.

[0131] Step 6: Output the calibrated recognition results of each data to be analyzed.

[0132] Assign a unique identifier to each pipeline in each data to be analyzed, and mark the pipeline identifier, pipeline type, pipeline shape, coordinates of each vertex of the pipeline, pipeline size, etc. of each pipeline at the corresponding position.

[0133] Figure 2 FIG. 2 is a schematic diagram of a structure of an embodiment of a system for underground pipeline detection provided by the present invention. Figure 2 As shown, the system includes: a detection vehicle 10 and a detection and analysis device 20, the detection and analysis device 20 includes a first data acquisition module 201, a second data acquisition module 202, an image generation module 203, an analysis and recognition module 204, a type calibration module 205 and a result output module 206.

[0134] The detection vehicle 10 is provided with a ground penetrating radar device 101 and an image acquisition device 102 .

[0135] The first data acquisition module 201 is used to determine the detection area of ​​underground pipelines and obtain the pipeline distribution map of the detection area.

[0136] The second data acquisition module 202 is used to move the detection vehicle according to the set driving route to collect ground penetrating radar data and a bird's-eye view of the detection area.

[0137] The image generation module 203 is used to obtain a three-dimensional image and N1 first cross-sectional images of the detection area according to the ground penetrating radar data, wherein the first cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline.

[0138] The analysis and identification module 204 is used to input the data to be analyzed into the corresponding preset model to obtain the corresponding identification results, wherein the data to be analyzed includes a pipeline distribution map, a top view, a three-dimensional image and N1 first cross-sectional images, and the identification results include pipeline type, pipeline contour, and pipeline coordinates.

[0139] The type calibration module 205 is used to calibrate the pipeline type of each pipeline in each data to be analyzed according to all the recognition results.

[0140] The result output module 206 is used to output the calibrated recognition result of each data to be analyzed.

[0141] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0142] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0143] The above-mentioned embodiments only express the preferred implementation mode of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for detecting underground pipelines, characterized in that: The steps include: Step 1: determine the detection area of ​​underground pipelines and obtain a pipeline distribution map of the detection area; Step 2: moving the detection vehicle according to the set driving route to collect ground penetrating radar data and a bird's-eye view of the detection area; Step 3: acquiring a three-dimensional image and N1 first cross-sectional images of the detection area based on the ground penetrating radar data, wherein the first cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline; Step 4: input the data to be analyzed into the corresponding preset model to obtain the corresponding recognition result, wherein the data to be analyzed includes the pipeline distribution map, the top view, the three-dimensional image and N1 first cross-sectional images, and the recognition result includes the pipeline type, pipeline profile, and pipeline coordinates; Step 5: Calibrate the pipeline type of each pipeline in each of the data to be analyzed based on all the identification results; Step 6: Output the calibrated recognition results of each of the data to be analyzed respectively.

2. The method according to claim 1, characterized in that The step 5 comprises: Step 51: extract any pipeline in the three-dimensional image, and obtain a first type and a first center coordinate of any pipeline; Step 52: Based on the first center coordinate, determine whether there is a second pipeline of the first type within the preset range of the second data to be analyzed; if yes, generate a first comparison set based on any of the pipelines and the second pipeline; if no, proceed to step 55, wherein the second data to be analyzed includes the pipeline distribution map, the top view and the first cross-sectional image, and the center point of the preset range is the first center coordinate; Step 53, calculating the first score values ​​of all pipelines in the first comparison set, and determining whether the first score value is greater than or equal to a first preset value, if so, proceeding to step 54, if not, proceeding to step 55; Step 54: determine that the pipeline types of all pipelines in the first comparison set are the first type, and mark the first type on the corresponding pipeline of each data to be analyzed; Step 55, determining whether there is a third pipeline whose pipeline type is different from the first type within the preset range of the second data to be analyzed, if yes, generating a second comparison set based on any of the pipelines and the third pipeline, if no, marking the first type on any of the pipelines in the three-dimensional image; Step 56, calculating the second score values ​​of all pipelines in the second comparison set, and determining whether the second score value is greater than or equal to a second preset value, if so, proceeding to step 57, if not, proceeding to step 58; Step 57: Determine the pipeline type corresponding to the second comparison set based on the specific identification parameter value of any of the pipelines and the third pipeline, and define it as the third type; at the same time, set the pipeline type of all pipelines in the second comparison set to the third type, and mark the third type on the corresponding pipeline of each of the data to be analyzed; Step 58: Mark the pipeline type corresponding to each pipeline in the second comparison set on the corresponding pipeline of the corresponding data to be analyzed.

3. The method according to claim 2, characterized in that When there are N2 second pipelines within the preset range of any second data to be analyzed, N3 first comparison sets are generated, and the first set score value of each first comparison set is calculated respectively, and the maximum value of the first set score value is taken as the first score value, wherein N2 and N3 are positive integers greater than 1.

4. The method according to claim 2, characterized in that: When there are N4 third pipelines within the preset range of any second data to be analyzed, N5 second comparison sets are generated, and the second set score value of each second comparison set is calculated respectively, and the maximum value of the second set score value is taken as the second score value, where N4 and N5 are positive integers greater than 1.

5. The method according to claim 2, characterized in that: The identification result includes the specific identification parameter value corresponding to the pipeline type of each pipeline, and the step 57 includes: Step 571, obtaining the specific identification parameter value corresponding to each pipeline in the second comparison set; Step 572: sort the specific identification parameter values ​​in ascending order; Step 573: Define the pipeline type corresponding to the maximum value of the specific identification parameter as the third type.

6. The method according to claim 2, characterized in that The scoring value is calculated by the three-dimensional intersection ratio of all pipelines in the comparison set or the average value of the distances between the center point coordinates of all the pipelines, wherein the scoring value includes the first scoring value and the second average value, and the comparison set includes the first comparison set and the second comparison set.

7. The method according to claim 1, characterized in that The detection vehicle travels back and forth along the pipeline arrangement direction, and the step 3 comprises: Step 31, based on the acquisition position of the ground penetrating radar data, acquiring a second cross-sectional image at the acquisition position, wherein the second cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline; Step 32: stitching all the second cross-sectional images located on the first horizontal plane corresponding to any pipeline cross section to generate an initial cross-sectional image corresponding to the first horizontal plane, and after traversing all the horizontal planes corresponding to the pipeline cross sections, generating N1 initial cross-sectional images; Step 33, extracting any initial cross-sectional image, analyzing any of the initial cross-sectional images, and obtaining all reflected wave curves in any of the initial cross-sectional images; Step 34, extract any reflection wave curve, determine whether any of the reflection wave curves is a curve corresponding to the pipeline, if so, retain it, if not, delete it, and after traversing all the reflection wave curves, obtain the first cross-sectional image corresponding to any of the initial cross-sectional images; Step 35, after traversing all the initial cross-sectional images, generating N1 first cross-sectional images; Step 36: Generate the three-dimensional image based on all the first cross-sectional images.

8. The method according to claim 7, characterized in that The step 34 includes: Step 331, obtaining a specific component image of any of the initial cross-sectional images; Step 332: remove the specific component image from any of the initial cross-sectional images, and obtain a third cross-sectional image corresponding to any of the initial cross-sectional images; Step 333: perform Fourier transform on the specific component image and the third cross-sectional image respectively to generate a first spectrum map and a second spectrum map, and perform smoothing on the first spectrum map and the second spectrum map to generate a third spectrum map and a fourth spectrum map; Step 334: Obtain a first frequency value corresponding to the maximum amplitude of the third spectrum graph and a second frequency value corresponding to the maximum amplitude of the fourth spectrum graph, and use an average value of the first frequency value and the second frequency value as a demarcation value.

9. The method according to claim 8, characterized in that The step 34 comprises: Step 341, performing Fourier transform on any of the reflected wave curves to generate a fifth spectrum diagram; Step 342: Selecting a first frequency range before the demarcation value and a second frequency range after the demarcation value on the fifth frequency spectrum based on the demarcation value; Step 343: Calculate a first accumulated value of all amplitudes within the first frequency range and a second accumulated value of all amplitudes within the second frequency range, and calculate a ratio of the second accumulated value to the first accumulated value; Step 344: when the ratio is greater than or equal to a third preset value, it is determined that any of the reflected wave curves is a curve corresponding to the pipeline; when the ratio is less than the third preset value, it is determined that any of the reflected wave curves is not a curve corresponding to the pipeline.

10. A system for underground pipeline detection, used to implement the method according to any one of claims 1 to 9, characterized in that: include: A detection vehicle and a detection and analysis device, wherein the detection and analysis device comprises a first data acquisition module, a second data acquisition module, an image generation module, an analysis and recognition module, a type calibration module and a result output module; The detection vehicle is provided with a ground penetrating radar device and an image acquisition device; The first data acquisition module is used to determine the detection area of ​​underground pipelines and obtain the pipeline distribution map of the detection area; The second data acquisition module is used to move the detection vehicle according to the set driving route to collect the ground penetrating radar data and the top view of the detection area; The image generation module is used to obtain a three-dimensional image and N1 first cross-sectional images of the detection area according to the ground penetrating radar data, wherein the first cross-sectional image is an image perpendicular to the ground and perpendicular to the length direction of the pipeline; The analysis and recognition module is used to input the data to be analyzed into the corresponding preset model to obtain the corresponding recognition result, wherein the data to be analyzed includes the pipeline distribution map, the top view, the three-dimensional image and N1 first cross-sectional images, and the recognition result includes the pipeline type, pipeline profile, and pipeline coordinates; The type calibration module is used to calibrate the pipeline type of each pipeline in each of the data to be analyzed according to all the identification results; The result output module is used to respectively output the calibrated recognition results of each of the data to be analyzed.

Images