Three-dimensional imaging method, device and computer equipment for metal pipelines

Abstract

The present application relates to a three-dimensional imaging method, device, and computer device for metal pipelines. The method includes: collecting regional magnetic sensing detection data of an overall area through a magnetic gradient detection subsystem to screen target detection data of each target area in the overall area; then generating a regional two-dimensional imaging map of each target area through a decoder of a magnetic method three-dimensional imaging model, and generating a regional three-dimensional imaging map corresponding to each target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model; thereby constructing a pipeline three-dimensional imaging map of the metal pipeline and identifying pipeline position information corresponding to each pipeline feature information in the pipeline three-dimensional imaging map; in the pipeline three-dimensional imaging map, performing feature marking processing on the pipeline three-dimensional imaging map based on the pipeline position information corresponding to each pipeline feature information to obtain a target three-dimensional imaging map of the metal pipeline. Using this method can improve the imaging accuracy of metal pipelines.

Description

Technical Field

[0001] The present application relates to three-dimensional imaging and technical fields, and particularly relates to a three-dimensional imaging method, device, computer device, computer-readable storage medium, and computer program product for a metal pipeline. Background Art

[0002] With the growth of global energy transportation demands, as a core infrastructure, the safety and reliability of oil pipelines face severe challenges. Pipelines have been in service in complex geological and working condition environments for a long time, and are eroded by substances such as humid air and internal electrolytes, resulting in frequent accidents such as energy leakage, oil and gas pipeline explosions, etc. caused by defects such as corrosion, cracks, or deformations, posing a huge threat to people's lives and property safety.

[0003] Traditional detection means such as manual inspections, seismic inversion, or single magnetic flux leakage detection technologies have problems such as low efficiency, insufficient defect quantification accuracy, and difficulty in identifying complex geometric features. For example, although the inversion method based on seismic waves can locate the damaged area, it is limited by the simplified error of the geological model and is difficult to achieve high-resolution imaging of pipeline micro-defects; while the traditional magnetic flux leakage detection signals rely on manual experience for interpretation, and the recognition accuracy for tiny defects or interactive damages is insufficient, thus leading to the imaging accuracy of metal pipelines. Summary of the Invention

[0004] Based on this, it is necessary to provide a three-dimensional imaging method, device, computer device, computer-readable storage medium, and computer program product for a metal pipeline in view of the above technical problems.

[0005] In a first aspect, the present application provides a three-dimensional imaging method for a metal pipeline, including:

[0006] Collect regional magnetic sensing detection data of an overall area through a magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area;

[0007] Based on the target detection data of each target area, generate a regional two-dimensional imaging map of each target area through a decoder of a magnetic method three-dimensional imaging model, and based on the regional two-dimensional imaging map corresponding to each target area, generate a regional three-dimensional imaging map corresponding to each target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model;

[0008] Based on the regional three-dimensional imaging map corresponding to each target area, construct a pipeline three-dimensional imaging map of the metal pipeline, and identify the pipeline position information corresponding to each pipeline feature information in the pipeline three-dimensional imaging map;

[0009] In the three-dimensional imaging map of the pipeline, based on the pipeline position information corresponding to each pipeline feature information, perform feature marking processing on the three-dimensional imaging map of the pipeline to obtain the target three-dimensional imaging map of the metal pipeline.

[0010] Optionally, the screening of the target detection data of each target area in the overall area based on the area magnetic sensing detection data includes:

[0011] Perform interpolation processing on the area magnetic sensing detection data to obtain the magnetic sensing detection distribution matrix of the overall area, and identify each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix;

[0012] Based on each abnormal sub-distribution matrix, in the overall area, screen the area range corresponding to each abnormal sub-distribution matrix as the target area corresponding to each abnormal sub-distribution matrix;

[0013] Use the abnormal sub-distribution matrix corresponding to each target area as the target detection data of each target area.

[0014] Optionally, the generation of the area two-dimensional imaging map of each target area through the decoder of the magnetic method three-dimensional imaging model based on the target detection data of each target area includes:

[0015] For each target area, based on the target detection data of the target area, identify the area edge range of the metal conduit in the target area through a threshold segmentation strategy;

[0016] Based on the magnetic sensing distribution matrix in the area edge range, generate the two-dimensional imaging map of the area edge range through the decoder of the magnetic method three-dimensional imaging model.

[0017] Optionally, the generation of the area three-dimensional imaging map corresponding to each target area through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model based on the area two-dimensional imaging map corresponding to each target area includes:

[0018] For each target area, based on the area two-dimensional imaging map corresponding to the target area, generate the initial area three-dimensional imaging map corresponding to the target area through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model;

[0019] Extract the three-dimensional feature information of the initial area three-dimensional imaging map through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, and based on the three-dimensional feature information, perform image optimization processing on the initial area three-dimensional imaging map through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model to obtain the area three-dimensional imaging map corresponding to the target area.

[0020] Optionally, identifying the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline includes:

[0021] Through a three-dimensional feature extraction network, respectively extract the pipeline feature data of the three-dimensional imaging map of the pipeline and the feature types corresponding to the pipeline feature data, and use the pipeline feature data of each feature type as the pipeline feature information of the three-dimensional imaging map of the pipeline;

[0022] In the three-dimensional imaging map of the pipeline, identify the pipeline position information corresponding to each pipeline feature information through a feature positioning strategy.

[0023] In a second aspect, the present application also provides a three-dimensional imaging device for a metal pipeline, including:

[0024] An acquisition module, configured to collect regional magnetic sensing detection data of an overall area through a magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area;

[0025] A generation module, configured to generate a regional two-dimensional imaging map of each target area through a decoder of a magnetic method three-dimensional imaging model based on the target detection data of each target area, and based on the regional two-dimensional imaging map corresponding to each target area, generate a regional three-dimensional imaging map corresponding to each target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model;

[0026] An identification module, configured to construct a three-dimensional imaging map of the metal pipeline based on the regional three-dimensional imaging map corresponding to each target area, and identify the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline;

[0027] A marking module, configured to perform feature marking processing on the three-dimensional imaging map of the pipeline based on the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline to obtain a target three-dimensional imaging map of the metal pipeline.

[0028] Optionally, the acquisition module is specifically configured to:

[0029] Perform interpolation processing on the regional magnetic sensing detection data to obtain a magnetic sensing detection distribution matrix of the overall area, and identify each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix;

[0030] Based on each abnormal sub-distribution matrix, screen the regional range corresponding to each abnormal sub-distribution matrix in the overall area as the target area corresponding to each abnormal sub-distribution matrix;

[0031] Take the abnormal sub - distribution matrix corresponding to each of the target regions as the target detection data for each of the target regions.

[0032] Optionally, the generating module is specifically configured to:

[0033] For each target region, based on the target detection data of the target region, through a threshold segmentation strategy, identify the regional edge range of the metal conduit in the target region;

[0034] Based on the magnetic sensing distribution matrix in the regional edge range, through the decoder of the magnetic three - dimensional imaging model, generate a two - dimensional imaging map of the regional edge range.

[0035] Optionally, the generating module is specifically configured to:

[0036] For each target region, based on the regional two - dimensional imaging map corresponding to the target region, through the three - dimensional reconstructor of the magnetic three - dimensional imaging model, generate an initial regional three - dimensional imaging map corresponding to the target region;

[0037] Through the three - dimensional reconstructor of the magnetic three - dimensional imaging model, extract the three - dimensional feature information of the initial regional three - dimensional imaging map, and based on the three - dimensional feature information, through the three - dimensional reconstructor of the magnetic three - dimensional imaging model, perform image optimization processing on the initial regional three - dimensional imaging map to obtain the regional three - dimensional imaging map corresponding to the target region.

[0038] Optionally, the identifying module is specifically configured to:

[0039] Through a three - dimensional feature extraction network, respectively extract each pipeline feature data of the pipeline three - dimensional imaging map and the corresponding feature type of each pipeline feature data, and take the pipeline feature data of each feature type as the pipeline feature information of the pipeline three - dimensional imaging map;

[0040] In the pipeline three - dimensional imaging map, through a feature localization strategy, identify the pipeline position information corresponding to each pipeline feature information.

[0041] In a third aspect, the present application provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any one of the first aspect are implemented.

[0042] In a fourth aspect, the present application provides a computer - readable storage medium. A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method described in any one of the first aspect are implemented.

[0043] Fifth aspect, the present application provides a computer program product. The computer program product includes a computer program which, when executed by a processor, implements the steps of the method according to any one of the first aspect.

[0044] The above three-dimensional imaging method, device, computer device, computer-readable storage medium and computer program product for metal pipelines collect regional magnetic sensing detection data of the overall area through a magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area; based on the target detection data of each target area, through the decoder of the magnetic method three-dimensional imaging model, generate a regional two-dimensional imaging map of each target area, and based on the regional two-dimensional imaging map corresponding to each target area, through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, generate a regional three-dimensional imaging map corresponding to each target area; based on the regional three-dimensional imaging map corresponding to each target area, construct a pipeline three-dimensional imaging map of the metal pipeline, and identify the pipeline position information corresponding to each pipeline feature information in the pipeline three-dimensional imaging map; in the pipeline three-dimensional imaging map, based on the pipeline position information corresponding to each pipeline feature information, perform feature marking processing on the pipeline three-dimensional imaging map to obtain the target three-dimensional imaging map of the metal pipeline. Through the magnetic gradient detection subsystem constructed by the inventor, this solution performs sensing detection on the water inlet pipeline buried deep underground in the target area, improving the detection accuracy and comprehensiveness of metal pipelines. Then, through the magnetic method three-dimensional imaging model constructed by the inventor, this solution analyzes the detected magnetic gradient data, and finally constructs the target three-dimensional imaging map of the metal pipeline, thereby effectively improving the imaging accuracy of the metal pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can be obtained based on these drawings.

[0046] Figure 1 It is a schematic flowchart of the three-dimensional imaging method for a metal pipeline in an embodiment;

[0047] Figure 2 It is a schematic diagram of the detection sequence of the magnetic gradient detection subsystem in an embodiment;

[0048] Figure 3 It is a schematic diagram of the system structure of the magnetic gradient detection subsystem in an embodiment;

[0049] Figure 4Schematic flow chart of a three-dimensional imaging example of a metal pipeline in an embodiment;

[0050] Figure 5 Structural block diagram of a three-dimensional imaging device for a metal pipeline in an embodiment;

[0051] Figure 6 Internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0052] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] The three-dimensional imaging method for a metal pipeline provided by the embodiments of the present application can be applied to the application environment of three-dimensional imaging of a metal pipeline. Among them, this method can be applied to a terminal, can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptop computers, etc. Among them, the terminal uses the magnetic gradient detection subsystem constructed by the inventor to perform sensing detection on the water inlet pipeline buried deep underground in the target area, improving the detection accuracy and comprehensiveness of the metal pipeline. Then, through the magnetic method three-dimensional imaging model constructed by the inventor, the detected magnetic gradient data is analyzed, and finally the target three-dimensional imaging map of the metal pipeline is constructed, thereby effectively improving the imaging accuracy of the metal pipeline.

[0054] In an exemplary embodiment, as Figure 1 shown, a three-dimensional imaging method for a metal pipeline is provided. Taking the application of this method to a terminal as an example, it includes the following steps S101 to S104. Among them:

[0055] Step S101, collect the regional magnetic sensing detection data of the overall area through the magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area.

[0056] In this embodiment, in response to the information upload operation of the staff, the terminal obtains in real time the staff's sequential detection of the overall area through the magnetic gradient detection subsystem according to the detection width of the magnetic gradient detection subsystem as the single detection width, and obtains the regional magnetic sensing detection data of the overall area. Then, the terminal screens the target detection data of each target area in the overall area based on the regional magnetic sensing detection data. As Figure 2As shown, it is a schematic diagram of the detection sequence of the magnetic gradient detection subsystem. The specific detection process will be described in detail later. Among them, the detection width of the magnetic gradient detection subsystem can be 1.2 m. Each target area is the area range where metal substances such as metal pipelines, metal instruments, and metal minerals exist. The specific identification process will be described in detail later.

[0057] Step S102: Based on the target detection data of each target area, through the decoder of the magnetic method three-dimensional imaging model, generate a two-dimensional imaging map of each target area, and based on the two-dimensional imaging map corresponding to each target area, through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, generate a three-dimensional imaging map corresponding to each target area.

[0058] In this embodiment, the terminal, based on the target detection data of each target area, through the decoder of the magnetic method three-dimensional imaging model, generates a two-dimensional imaging map of each target area, and based on the two-dimensional imaging map corresponding to each target area, through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, generates a three-dimensional imaging map corresponding to each target area. Among them, the magnetic method imaging neural network consists of three parts, namely an encoder, a decoder, and a three-dimensional reconstructor. The designs of these three parts are introduced separately below:

[0059] In an embodiment of the present invention, the encoder (Encoder) adopts a network structure design with gradually shrinking levels, which is used to gradually reduce the resolution of the feature map to enhance the perception ability of context information. Specifically, the encoder structure is divided into four stages, and each stage includes a linear neural network and a max pooling layer.

[0060] In each stage, first, the linear neural network extracts the feature information of each measurement point. The max pooling layer performs a downsampling operation on the input feature map to effectively reduce the spatial resolution of the feature map. At the same time, the symmetric feature of the max pooling layer ensures that the entire network does not depend on ordered measurement points.

[0061] The present invention provides a feature decoder, which converts the one-dimensional features extracted by the encoder into two-dimensional features and inputs them into the three-dimensional refinement unit. In this embodiment, the decoder is designed as a one-dimensional to two-dimensional module and a convolutional module.

[0062] The terminal inputs the two-dimensional feature map output by the decoder into the three-dimensional reconstructor module. The three-dimensional reconstructor module is used to convert the two-dimensional feature map into a three-dimensional feature map and perform feature extraction and optimization processing on the three-dimensional feature map. Specifically, the three-dimensional reconstructor module includes multiple three-dimensional residual blocks (abbreviated as 3D RBs) to further extract and refine the converted three-dimensional feature information.

[0063] Preferably, four 3D residual blocks are configured in the 3D reconstructor module, which are connected in series in turn to enhance the expression ability of the 3D feature map layer by layer, so as to obtain a more accurate inversion result of the underground structure. The final output is a 3D density matrix reflecting the distribution of the underground medium, which is used to represent the spatial distribution characteristics of the underground structure.

[0064] Through the above structural design, the accuracy and stability of the inversion result can be effectively improved, especially applicable to application scenarios such as geological exploration, oil and gas reservoir prediction, and other applications requiring high-resolution underground imaging.

[0065] Step S103: Based on the regional 3D imaging maps corresponding to each target area, construct a 3D imaging map of the metal pipeline, and identify the pipeline position information corresponding to each pipeline feature information in the 3D imaging map of the pipeline.

[0066] In this embodiment, the terminal constructs a 3D imaging map of the metal pipeline based on the regional 3D imaging maps corresponding to each target area, and identifies the pipeline position information corresponding to each pipeline feature information in the 3D imaging map of the pipeline. Specifically, the terminal performs 3D stitching processing on the regional 3D imaging maps corresponding to each target area according to the spatial connection relationship between the target areas to obtain the 3D imaging map of the metal pipeline. Then, the terminal uses an image feature recognition network to identify the pipeline position information corresponding to each pipeline feature information in the 3D imaging map of the pipeline. Among them, each pipeline feature information includes, but is not limited to, pipeline abnormal feature information such as pipeline damage, pipeline deformation, pipeline fracture, pipeline collapse, and pipeline blockage. Among them, the pipeline feature recognition network can be a linear neural network.

[0067] Step S104: In the 3D imaging map of the pipeline, based on the pipeline position information corresponding to each pipeline feature information, perform feature marking processing on the 3D imaging map of the pipeline to obtain the target 3D imaging map of the metal pipeline.

[0068] In this embodiment, the terminal performs feature marking processing on the 3D imaging map of the pipeline based on the pipeline position information corresponding to each pipeline feature information to obtain the target 3D imaging map of the metal pipeline. Among them, the feature marking is used to mark the abnormal information and abnormal positions of the metal pipeline in the 3D imaging map of the pipeline, which is convenient for the staff to locate, identify, and warn the relevant information of the metal pipeline.

[0069] Based on the above solution, through the magnetic gradient detection subsystem constructed by the inventor, the buried water inlet pipeline in the target area underground is sensed and detected, which improves the detection accuracy and comprehensiveness of the metal pipeline. Then, through the magnetic method 3D imaging model constructed by the inventor, the detected magnetic gradient data is analyzed, and finally the target 3D imaging map of the metal pipeline is constructed, thus effectively improving the imaging accuracy of the metal pipeline.

[0070] Optionally, based on the regional magnetic sensing detection data, the target detection data of each target area in the overall area is screened, including: performing interpolation processing on the regional magnetic sensing detection data to obtain the magnetic sensing detection distribution matrix of the overall area, and identifying each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix; based on each abnormal sub-distribution matrix, in the overall area, screening the area range corresponding to each abnormal sub-distribution matrix as the target area corresponding to each abnormal sub-distribution matrix; and taking the abnormal sub-distribution matrix corresponding to each target area as the target detection data of each target area.

[0071] In this embodiment, the terminal performs interpolation processing on the regional magnetic sensing detection data to obtain the magnetic sensing detection distribution matrix of the overall area, and identifies each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix. Specifically, determining the reasonable segmentation of the key area according to the detected magnetic anomaly data helps to reduce the computational amount of the inversion imaging. Since the magnetic field vector roughly decays according to the cube of the distance, and the magnetic gradient tensor roughly decays according to the fourth power of the distance, at a certain distance from the metal pipeline, if the magnetic field vector no longer plays a role, the magnetic gradient tensor will also no longer play a role. This article selects to perform regional segmentation based on the magnetic field vector data or lower-order data (such as the total magnetic anomaly). Since the original data is measured during movement, the TMA data is first interpolated to form the data matrix D on the grid.

[0072] Then, the terminal, based on each abnormal sub-distribution matrix, screens the area range corresponding to each abnormal sub-distribution matrix in the overall area as the target area corresponding to each abnormal sub-distribution matrix. Finally, the terminal takes the abnormal sub-distribution matrix corresponding to each target area as the target detection data of each target area. Among them, the role of the convolution kernel K is to extract the steeper part in the data matrix D. If the data at a certain place is less affected by the target, the data at that place should be close to the background field and relatively smooth as a whole, so the absolute value of Dconv is smaller. If the data at a certain place is greatly affected by the target, the data at that place should change violently, so the absolute value of Dconv is also relatively large. Therefore, threshold processing can be performed on Dconv. Among them, the terminal defines the points where the absolute value of the convolution result is less than the threshold as "boundary points". In the case where the target distribution is relatively sparse, these "boundary points" will divide the entire area into multiple sub-areas. Under different thresholds, the area segmentation results will also be different. If the threshold is selected too large, multiple sub-areas will be divided into the same area, losing the meaning of area division; if the threshold is selected too small, the amount of data in each area is too small to achieve the imaging of the target.

[0073] Based on the above scheme, by first performing interpolation processing on the regional magnetic sensing detection data, the recognition accuracy and comprehensiveness of the distribution information of the regional magnetic sensing detection data in the target area are improved.

[0074] Optionally, based on the target detection data of each target area, a two-dimensional imaging map of each target area is generated through the decoder of the magnetic method three-dimensional imaging model, including: for each target area, based on the target detection data of the target area, the regional edge range of the metal conduit in the target area is identified through a threshold segmentation strategy; based on the magnetic sensing distribution matrix in the regional edge range, a two-dimensional imaging map of the regional edge range is generated through the decoder of the magnetic method three-dimensional imaging model.

[0075] In this embodiment, for each target area, the terminal identifies the regional edge range of the metal conduit in the target area based on the target detection data of the target area through a threshold segmentation strategy. Then, the terminal generates a two-dimensional imaging map of the regional edge range based on the magnetic sensing distribution matrix in the regional edge range through the decoder of the magnetic method three-dimensional imaging model. Specifically, the terminal uses the feature decoder provided by the present invention (abbreviated as the decoder), which converts the one-dimensional features extracted by the encoder into two-dimensional features and inputs them into the three-dimensional refiner. In this embodiment, the decoder is designed as a one-dimensional to two-dimensional module and a convolutional module.

[0076] Based on the above solution, after first identifying the regional edge range of the metal conduit, and then through the feature decoder designed by the present invention, the one-dimensional features are converted into two-dimensional features, thereby generating a two-dimensional imaging map, which can effectively improve the accuracy and stability of the inversion result, and is especially suitable for application scenarios that require high-resolution underground imaging.

[0077] Optionally, based on the two-dimensional imaging map of each target area, a three-dimensional imaging map of each target area is generated through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, including: for each target area, based on the two-dimensional imaging map of the target area, an initial three-dimensional imaging map of the target area is generated through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model; through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, the three-dimensional feature information of the initial three-dimensional imaging map is extracted, and based on the three-dimensional feature information, the initial three-dimensional imaging map is subjected to image optimization processing through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model to obtain the three-dimensional imaging map of the target area.

[0078] In this embodiment, for each target area, the terminal generates an initial three-dimensional imaging map of the target area based on the two-dimensional imaging map of the area corresponding to the target area through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model. Then, the terminal extracts the three-dimensional feature information of the initial three-dimensional imaging map through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, and based on the three-dimensional feature information, performs image optimization processing on the initial three-dimensional imaging map through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model to obtain the three-dimensional imaging map of the target area. Specifically, the terminal inputs the two-dimensional feature map output by the decoder into the three-dimensional reconstructor module, and the three-dimensional reconstructor module is used to convert the two-dimensional feature map into a three-dimensional feature map and perform feature extraction and optimization processing on the three-dimensional feature map. Specifically, the three-dimensional reconstructor module includes a plurality of three-dimensional residual blocks (3D Residual Blocks, abbreviated as 3D RBs) to further extract and refine the converted three-dimensional feature information.

[0079] Preferably, four three-dimensional residual blocks are configured in the three-dimensional reconstructor module, which are connected in series in sequence to enhance the expression ability of the three-dimensional feature map layer by layer, so as to obtain a more accurate underground structure inversion result. The final output is a three-dimensional density matrix reflecting the underground medium distribution, which is used to represent the spatial distribution characteristics of the underground structure.

[0080] Based on the above solution, a three-dimensional imaging map of the metal pipeline is constructed through the three-dimensional reconstruction module, effectively improving the accuracy of the constructed three-dimensional imaging map.

[0081] Optionally, identifying the pipeline position information corresponding to each pipeline feature information in the pipeline three-dimensional imaging map includes: respectively extracting the pipeline feature data of the pipeline three-dimensional imaging map and the feature type corresponding to each pipeline feature data through a three-dimensional feature extraction network, and using the pipeline feature data of each feature type as the pipeline feature information of the pipeline three-dimensional imaging map; in the pipeline three-dimensional imaging map, identifying the pipeline position information corresponding to each pipeline feature information through a feature positioning strategy.

[0082] In this embodiment, the terminal respectively extracts the pipeline feature data of the pipeline three-dimensional imaging map and the feature type corresponding to each pipeline feature data through a three-dimensional feature extraction network, and uses the pipeline feature data of each feature type as the pipeline feature information of the pipeline three-dimensional imaging map. Among them, the three-dimensional feature extraction network is a linear neural network. The feature types corresponding to the pipeline feature data include, but are not limited to, the abnormal information of each pipeline anomaly of the metal pipeline. For example, the pipeline break type, the pipeline deformation type, the pipeline structure anomaly type, the pipeline.

[0083] In the three-dimensional imaging diagram of the pipeline, the terminal identifies the pipeline position information corresponding to each pipeline feature information through a feature localization strategy. Among them, the feature localization strategy is that the terminal identifies the three-dimensional image data corresponding to the pipeline feature information based on each pipeline feature information, and based on the three-dimensional image data, performs image data adaptation processing in the three-dimensional imaging diagram of the pipeline to obtain the pipeline position information corresponding to each pipeline feature information. Among them, the magnetic method imaging neural network protected by this solution is an inversion algorithm combined with a deep neural network. It uses the magnetic field data collected from a large number of simulation experiments to organize and build a data set for training, and combines the measured data to correct the network parameters to achieve three-dimensional imaging of the defects on the inner and outer walls of the metal pipeline and the distribution trend of the pipeline. Among them, the neural network can learn the highly non-linear mapping relationship between the magnetic field perturbation and the defect morphology, and has stronger generalization ability and robustness than the traditional analytical inversion algorithm. The terminal can realize functions such as defect morphology recognition, depth estimation, and size determination, significantly improving the detection accuracy and efficiency.

[0084] Based on the above solution, by extracting the pipeline features and then identifying the pipeline position information corresponding to each pipeline feature information, the accuracy of pipeline anomaly recognition for metal pipelines is improved.

[0085] In an exemplary embodiment, a three-dimensional imaging system for a metal pipeline is provided. The system includes a magnetic gradient detection subsystem and an imaging subsystem, where:

[0086] The imaging subsystem is connected to the magnetic gradient detection subsystem. Among them, the connection method can be by wired connection or wireless connection.

[0087] The magnetic gradient detection subsystem includes a regular tetrahedron-shaped housing and fluxgate sensors. The fluxgate sensors are fixedly connected to the four corners of the regular tetrahedron-shaped housing. Specifically, as Figure 3 shown, the magnetic gradient detection subsystem is a regular tetrahedron-shaped orthogonal fluxgate magnetic gradient detection system, which is composed of four orthogonal fluxgate sensors and fixed by a regular tetrahedron-shaped housing. The directions of these four fluxgate sensors are the same, and the distance between each two is kept at 20 cm, forming a magnetic gradient detection system. The magnetic gradient calculation formula at the measurement point is:

[0088]

[0089] Among them, the magnetic gradient detection subsystem is arranged in a regular tetrahedron structure, and four fluxgate sensors are arranged in the vertices of the regular tetrahedron. The spatial symmetry of the regular tetrahedron enables the system to have an isotropic magnetic field sensing ability. Compared with traditional two-dimensional arrays or linear arrangements, it can collect magnetic field disturbance information from different directions more comprehensively and uniformly. This structure is particularly suitable for detecting complex defects on the inner and outer walls of cylindrical metal pipelines (such as oil pipelines, gas pipelines, etc.), improving the resolution and accuracy of three-dimensional imaging.

[0090] The fluxgate sensor has excellent low-frequency magnetic field sensitivity and can effectively detect the magnetic field disturbance caused by metal defects. Among them, compared with traditional magnetic sensors such as Hall sensors, the fluxgate performs better in low-frequency magnetic detection and is suitable for non-contact detection of deep defects.

[0091] The imaging subsystem is used to receive the regional magnetic sensing detection data of the overall area collected by the magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area; based on the target detection data of each target area, through the decoder of the magnetic method three-dimensional imaging model, generate the regional two-dimensional imaging map of each target area, and based on the regional two-dimensional imaging map corresponding to each target area, through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model, generate the regional three-dimensional imaging map corresponding to each target area; based on the regional three-dimensional imaging map corresponding to each target area, construct the three-dimensional imaging map of the metal pipeline, and identify the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline; in the three-dimensional imaging map of the pipeline, based on the pipeline position information corresponding to each pipeline feature information, perform feature marking processing on the three-dimensional imaging map of the pipeline to obtain the target three-dimensional imaging map of the metal pipeline.

[0092] This application also provides a three-dimensional imaging example of a metal pipeline. As shown in Figure 4, the specific processing process includes the following steps:

[0093] Step S401, collect the regional magnetic sensing detection data of the overall area through the magnetic gradient detection subsystem.

[0094] Step S402, perform interpolation processing on the regional magnetic sensing detection data to obtain the magnetic sensing detection distribution matrix of the overall area, and identify each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix.

[0095] Step S403, based on each abnormal sub-distribution matrix, in the overall area, screen the regional range corresponding to each abnormal sub-distribution matrix as the target area corresponding to each abnormal sub-distribution matrix.

[0096] Step S404, use the abnormal sub-distribution matrix corresponding to each target area as the target detection data of each target area.

[0097] Step S405: For each target area, based on the target detection data of the target area, identify the range of the edge of the metal conduit in the target area through a threshold segmentation strategy.

[0098] Step S406: Based on the magnetic sensing distribution matrix in the range of the edge of the area, generate a two-dimensional imaging map of the range of the edge of the area through the decoder of the magnetic three-dimensional imaging model.

[0099] Step S407: For each target area, based on the two-dimensional imaging map of the area corresponding to the target area, generate an initial three-dimensional imaging map of the area corresponding to the target area through the three-dimensional reconstructor of the magnetic three-dimensional imaging model.

[0100] Step S408: Through the three-dimensional reconstructor of the magnetic three-dimensional imaging model, extract the three-dimensional feature information of the initial three-dimensional imaging map of the area, and based on the three-dimensional feature information, through the three-dimensional reconstructor of the magnetic three-dimensional imaging model, perform image optimization processing on the initial three-dimensional imaging map of the area to obtain the three-dimensional imaging map of the area corresponding to the target area.

[0101] Step S409: Based on the three-dimensional imaging maps of the areas corresponding to the respective target areas, construct a three-dimensional imaging map of the metal pipeline.

[0102] Step S410: Through the three-dimensional feature extraction network, extract the respective pipeline feature data of the three-dimensional imaging map of the pipeline and the feature types corresponding to the respective pipeline feature data, and use the pipeline feature data of each feature type as the pipeline feature information of the three-dimensional imaging map of the pipeline.

[0103] Step S411: In the three-dimensional imaging map of the pipeline, identify the pipeline position information corresponding to each pipeline feature information through a feature positioning strategy.

[0104] Step S412: In the three-dimensional imaging map of the pipeline, based on the pipeline position information corresponding to the respective pipeline feature information, perform feature marking processing on the three-dimensional imaging map of the pipeline to obtain the target three-dimensional imaging map of the metal pipeline.

[0105] It should be understood that although the various steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the indication of the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limitation, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same moment, but can be executed at different moments, and the execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0106] Based on the same inventive concept, an embodiment of the present application further provides a three-dimensional imaging device for a metal pipeline for implementing the three-dimensional imaging method of the metal pipeline involved above. The implementation solution for solving the problem provided by this device is similar to the implementation solution described in the above method. Therefore, the specific limitations in one or more embodiments of the three-dimensional imaging device for a metal pipeline provided below can refer to the limitations on the three-dimensional imaging method of the metal pipeline in the above text, and will not be repeated here.

[0107] In an exemplary embodiment, as Figure 5 shown, a three-dimensional imaging device for a metal pipeline is provided, including: a collection module 510, a generation module 520, an identification module 530, and a marking module 540, where:

[0108] The collection module 510 is configured to collect regional magnetic sensing detection data of the overall area through a magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area;

[0109] The generation module 520 is configured to, based on the target detection data of each target area, generate a regional two-dimensional imaging map of each target area through a decoder of a magnetic method three-dimensional imaging model, and based on the regional two-dimensional imaging maps corresponding to each target area, generate a regional three-dimensional imaging map corresponding to each target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model;

[0110] The identification module 530 is configured to, based on the regional three-dimensional imaging maps corresponding to each target area, construct a three-dimensional imaging map of the metal pipeline and identify the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline;

[0111] The marking module 540 is configured to perform feature marking processing on the three-dimensional imaging map of the pipeline based on the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline to obtain a target three-dimensional imaging map of the metal pipeline.

[0112] Optionally, the collection module 510 is specifically configured to:

[0113] Perform interpolation processing on the regional magnetic sensing detection data to obtain a magnetic sensing detection distribution matrix of the overall area, and identify each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix;

[0114] Based on each abnormal sub-distribution matrix, in the overall area, screen the regional range corresponding to each abnormal sub-distribution matrix as the target area corresponding to each abnormal sub-distribution matrix;

[0115] Use the abnormal sub-distribution matrix corresponding to each target area as the target detection data for each target area.

[0116] Optionally, the generating module 520 is specifically configured to:

[0117] For each target area, based on the target detection data of the target area, identify the regional edge range of the metal conduit in the target area through a threshold segmentation strategy;

[0118] Based on the magnetic sensing distribution matrix in the regional edge range, generate a two-dimensional imaging map of the regional edge range through the decoder of the magnetic three-dimensional imaging model.

[0119] Optionally, the generating module 520 is specifically configured to:

[0120] For each target area, based on the two-dimensional imaging map of the corresponding area of the target area, generate an initial three-dimensional imaging map of the corresponding area of the target area through the three-dimensional reconstructor of the magnetic three-dimensional imaging model;

[0121] Extract the three-dimensional feature information of the initial three-dimensional imaging map through the three-dimensional reconstructor of the magnetic three-dimensional imaging model, and based on the three-dimensional feature information, perform image optimization processing on the initial three-dimensional imaging map through the three-dimensional reconstructor of the magnetic three-dimensional imaging model to obtain the three-dimensional imaging map of the corresponding area of the target area.

[0122] Optionally, the identifying module 530 is specifically configured to:

[0123] Extract the pipeline feature data of each pipeline in the pipeline three-dimensional imaging map and the corresponding feature type through a three-dimensional feature extraction network, and use the pipeline feature data of each feature type as the pipeline feature information of the pipeline three-dimensional imaging map;

[0124] In the pipeline three-dimensional imaging map, identify the pipeline position information corresponding to each pipeline feature information through a feature positioning strategy.

[0125] Each module in the above three-dimensional imaging device for metal pipelines can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0126] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 6As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a three-dimensional imaging method for metal pipes. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0127] Those skilled in the art can understand that Figure 6 the structure shown in the figure is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0128] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, it implements the steps corresponding to the three-dimensional imaging method for metal pipes.

[0129] In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, it implements the steps corresponding to the three-dimensional imaging method for metal pipes.

[0130] In an embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, it implements the steps corresponding to the three-dimensional imaging method for metal pipes.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0132] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0133] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this application.

[0134] The above-described embodiments merely represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several variations and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application shall be subject to the appended claims.

Claims

1. A three-dimensional imaging method for a metal pipeline, characterized in that, The method includes: Collecting regional magnetic sensing detection data of the overall area through a magnetic gradient detection subsystem, and screening target detection data of each target area in the overall area based on the regional magnetic sensing detection data; Based on the target detection data of each target area, generating a regional two-dimensional imaging map of each target area through a decoder of a magnetic method three-dimensional imaging model, and based on the regional two-dimensional imaging map corresponding to each target area, generating a regional three-dimensional imaging map corresponding to each target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model; Based on the regional three-dimensional imaging map corresponding to each target area, constructing a three-dimensional imaging map of the metal pipeline, and identifying pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline; In the three-dimensional imaging map of the pipeline, based on the pipeline position information corresponding to each pipeline feature information, performing feature marking processing on the three-dimensional imaging map of the pipeline to obtain a target three-dimensional imaging map of the metal pipeline.

2. The method according to claim 1, characterized in that The screening of the target detection data of each target area in the overall area based on the regional magnetic sensing detection data includes: Performing interpolation processing on the regional magnetic sensing detection data to obtain a magnetic sensing detection distribution matrix of the overall area, and identifying each abnormal sub-distribution matrix in the magnetic sensing detection distribution matrix; Based on each abnormal sub-distribution matrix, screening the regional range corresponding to each abnormal sub-distribution matrix in the overall area as the target area corresponding to each abnormal sub-distribution matrix; Taking the abnormal sub-distribution matrix corresponding to each target area as the target detection data of each target area.

3. The method according to claim 2, wherein The generating of the regional two-dimensional imaging map of each target area through a decoder of a magnetic method three-dimensional imaging model based on the target detection data of each target area includes: For each target area, based on the target detection data of the target area, identifying the regional edge range of the metal conduit in the target area through a threshold segmentation strategy; Based on the magnetic sensing distribution matrix in the regional edge range, generating a two-dimensional imaging map of the regional edge range through a decoder of a magnetic method three-dimensional imaging model.

4. The method according to claim 1, wherein The generating of the regional three-dimensional imaging map corresponding to each target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model based on the regional two-dimensional imaging map corresponding to each target area includes: For each target area, based on the regional two-dimensional imaging map corresponding to the target area, generating an initial regional three-dimensional imaging map corresponding to the target area through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model; Extracting three-dimensional feature information of the initial regional three-dimensional imaging map through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model, and based on the three-dimensional feature information, performing image optimization processing on the initial regional three-dimensional imaging map through a three-dimensional reconstructor of the magnetic method three-dimensional imaging model to obtain the regional three-dimensional imaging map corresponding to the target area.

5. The method according to claim 1, characterized in that, The identifying of the pipeline position information corresponding to each pipeline feature information in the three-dimensional imaging map of the pipeline includes: Through a three-dimensional feature extraction network, extract the pipeline feature data of the pipeline three-dimensional imaging map and the corresponding feature types of each pipeline feature data respectively, and use the pipeline feature data of each feature type as the pipeline feature information of the pipeline three-dimensional imaging map; In the pipeline three-dimensional imaging map, identify the pipeline position information corresponding to each pipeline feature information through a feature positioning strategy.

6. A three-dimensional imaging system for a metal pipeline, characterized in that, The system includes a magnetic gradient detection subsystem and an imaging subsystem, where: The imaging subsystem is connected to the magnetic gradient detection subsystem; The magnetic gradient detection subsystem includes a regular tetrahedron-shaped housing and fluxgate sensors, and the fluxgate sensors are fixedly connected to the four corners of the regular tetrahedron-shaped housing; The imaging subsystem is configured to receive the regional magnetic sensing detection data of the overall area collected by the magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area; based on the target detection data of each target area, generate the regional two-dimensional imaging maps of each target area through the decoder of the magnetic method three-dimensional imaging model, and based on the regional two-dimensional imaging maps corresponding to each target area, generate the regional three-dimensional imaging maps corresponding to each target area through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model; based on the regional three-dimensional imaging maps corresponding to each target area, construct the pipeline three-dimensional imaging map of the metal pipeline, and identify the pipeline position information corresponding to each pipeline feature information in the pipeline three-dimensional imaging map; in the pipeline three-dimensional imaging map, perform feature marking processing on the pipeline three-dimensional imaging map based on the pipeline position information corresponding to each pipeline feature information to obtain the target three-dimensional imaging map of the metal pipeline.

7. A three-dimensional imaging device for a metal pipeline, characterized in that, The device includes: An acquisition module, configured to collect the regional magnetic sensing detection data of the overall area through the magnetic gradient detection subsystem, and based on the regional magnetic sensing detection data, screen the target detection data of each target area in the overall area; A generation module, configured to generate the regional two-dimensional imaging maps of each target area through the decoder of the magnetic method three-dimensional imaging model based on the target detection data of each target area, and generate the regional three-dimensional imaging maps corresponding to each target area through the three-dimensional reconstructor of the magnetic method three-dimensional imaging model based on the regional two-dimensional imaging maps corresponding to each target area; An identification module, configured to construct the pipeline three-dimensional imaging map of the metal pipeline based on the regional three-dimensional imaging maps corresponding to each target area, and identify the pipeline position information corresponding to each pipeline feature information in the pipeline three-dimensional imaging map; A marking module, configured to perform feature marking processing on the pipeline three-dimensional imaging map in the pipeline three-dimensional imaging map based on the pipeline position information corresponding to each pipeline feature information to obtain the target three-dimensional imaging map of the metal pipeline.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images