Underground municipal pipeline facility general survey system

Abstract

The invention discloses an underground municipal pipeline facility general survey system, and relates to the technical field of urban management data processing, and the system comprises a data collection module which integrates multiple types of detection equipment and cooperatively collects pipeline space, attribute and surrounding environment data; the data preprocessing module unifies coordinates and formats, removes noise and fills missing data; the pipeline feature recognition module automatically recognizes pipeline types and key feature points through an improved algorithm; the pipeline attribute association module associates the space and the static attribute data through the unique code; the multi-source data fusion module integrates multiple types of data and eliminates conflicts to form a unified data set; the data quality verification module evaluates the data quality in multiple dimensions; the general survey result management module generates a standardized result; and the system management module realizes authority control, operation log recording and data security protection. The system improves the accuracy and efficiency of pipeline general survey, realizes efficient fusion of multi-source data, intelligent pipeline identification and controllable quality, and provides reliable support for city management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of urban management data processing, and particularly relates to an underground municipal pipeline facility census system. BACKGROUND

[0002] There are many outstanding problems in the existing underground pipeline census technology. The data collection link relies on single equipment or manual detection, the equipment coordination is poor, the data formats collected by different detection equipment are not unified, the coordinate systems are inconsistent, which leads to great difficulty in data integration; for complex geological conditions or subdivided pipelines, the collection accuracy is insufficient, and data loss or deviation is prone to occur. In the data processing stage, there is a lack of efficient multi-source data fusion mechanism, and the multi-source information such as field detection data, historical archive data and urban GIS data has redundancy and logical conflict, so it is difficult to form a complete and unified data set; pipeline type identification relies on manual judgment, which is low in efficiency and is easily affected by subjective factors, and the identification accuracy of key features such as small diameter pipelines and damaged points is insufficient. In terms of attribute association, the spatial data of the pipeline is disconnected with the static attribute information such as ownership and maintenance records, so that the fine management of "one thing one file" cannot be realized, which brings inconvenience to subsequent maintenance and planning.

[0003] The defects in the data quality and the application link of the results are also remarkable. The existing quality check is limited to a single dimension, and lacks comprehensive evaluation of position accuracy, attribute integrity and logical consistency, so that unqualified data is easy to flow into the subsequent link; the census results are mostly presented in the form of two-dimensional charts, and lack intuitive three-dimensional visualization display, so it is difficult to clearly reflect the spatial distribution relationship of the pipeline and the correlation with the surrounding environment; at the same time, the data updating mechanism is not perfect, and the newly added or changed pipeline data cannot be synchronized in time, which leads to the disconnection between the pipeline data and the actual situation, and increases the safety risk of underground construction. These problems lead to low efficiency, high cost and insufficient data reliability of the census work, which cannot provide accurate and effective data support for urban management, and even may cause safety accidents such as pipeline breakage and pipeline leakage due to data errors, affecting the normal operation of the city and the life of residents. SUMMARY

[0004] The underground municipal pipeline facility census system provided by the present application solves the problems mentioned in the prior art.

[0005] In order to achieve the above purpose, the present application adopts the following technical scheme: an underground municipal pipeline facility census system, comprising the following modules: A data collection module is deployed at the census operation site, integrates multiple detection and aerial survey equipment, and collects core data of underground municipal pipelines and surrounding related data; A data preprocessing module receives raw data, fills in missing data and corrects abnormal data, and provides a basis for subsequent processing; The pipeline feature recognition module is based on a recognition model built with deep learning and trained on a large number of samples. It takes preprocessed data and images as input, extracts pipeline features, and combines them with relevant correlation patterns. The pipeline attribute association module establishes a basic attribute database to store pipeline static attribute information. It uses an 18-bit unique identifier code to associate spatial data, feature data and static attributes, and supports the import, supplementation and query of attribute information. The multi-source data fusion module integrates on-site detection, attributes, historical archives and urban GIS basic data. It adopts a feature-level fusion strategy and eliminates redundancy and conflicts through conflict detection and resolution to form a unified census dataset. The data quality verification module constructs a multi-dimensional evaluation system, verifying data from four aspects: location accuracy, attribute integrity, logical consistency, and data integrity. It marks unqualified data and outputs a verification report. The census results management module organizes qualified data according to industry standards, generates various census results, supports visual display and query operations of results, automatically generates relevant documents, and supports exporting, sharing and printing of results; The system management module sets up a multi-level user permission management mechanism, divides different roles and assigns corresponding permissions, records operation logs and supports related functions, and provides system configuration, data backup and recovery functions.

[0006] Furthermore, it also includes a pipeline risk assessment module. Based on the pipeline spatial data, attribute data, and characteristic data obtained from the survey, combined with surrounding environmental risk factors, geological disaster risk levels, and the frequency of underground construction activities, it constructs a pipeline operation risk assessment system, identifies high-risk pipeline sections and potential safety hazards, generates risk level distribution maps, risk assessment reports, and targeted risk prevention and control suggestions, and supports dynamic adjustment of risk levels and key marking of high-risk areas.

[0007] Furthermore, it also includes a dynamic update module, which establishes a dynamic update mechanism for pipeline data. It supports the on-site collection of new pipeline data, pipeline change data, and pipeline maintenance data through mobile terminal collection devices. The collection of new pipeline data must be associated with the original pipeline network topology, and the change data must be confirmed by the auditor before it takes effect. The system automatically compares the differences between the data before and after the update.

[0008] Furthermore, the multi-source data fusion module introduces a multi-source data credibility weighted fusion model, which completes the weight allocation and fusion calculation of different data sources through the following formula: in This represents the integrated pipeline data value after merging. Indicates the first The credibility coefficient of each data source Indicates the first The original data values ​​of each data source, data redundancy of the first logical conflict degree of the first total number of data sources participating in fusion, dynamically adjusting weights through credibility, redundancy and conflict degree of each data source, data sources with high credibility, low redundancy and low conflict occupy a higher proportion in fusion results.

[0009] Further, in the pipeline feature recognition module, an improved U-Net image segmentation algorithm is adopted, attention mechanisms of channel attention and spatial attention are added in parallel to strengthen the extraction of pipeline edge features and key feature points, and Dice loss and cross-entropy loss are combined to improve the recognition accuracy of small-diameter pipelines and fine branch pipelines, and the contour coordinates, key feature point coordinates and type recognition results of the pipeline are output synchronously.

[0010] Further, in the pipeline risk assessment module, a pipeline risk level quantification calculation model is introduced, and the risk level is accurately quantified through the following formula: wherein represents the risk level quantification value of the pipeline segment, represents the pipeline basic risk coefficient, represents the influence weight of the first class risk factor, represents the actual influence degree of the first class risk factor, represents the effectiveness of the first class risk factor, represents the occurrence probability of the first class risk factor, represents the total number of risk factor categories.

[0011] Further, in the data quality checking module, a quality threshold dynamic adjustment mechanism is set, and the position accuracy deviation threshold, attribute filling rate threshold and logical consistency judgment threshold are automatically adjusted according to different pipeline types, different survey areas and different geological conditions. The original high-standard quality threshold is strictly implemented in the core area of the city and the newly built pipeline area, and the user can manually adjust the quality threshold according to the actual survey demand.

[0012] Further, in the survey result management module, a three-dimensional visualization engine is integrated, pipeline data is superimposed with urban topographic and geomorphic data and building data, a three-dimensional visualization model of underground municipal pipelines is constructed, three-dimensional roaming, three-dimensional sectioning and three-dimensional measurement operations of the pipeline are supported, the spatial distribution relationship, burial depth and positional relationship with the surrounding environment of the pipeline are intuitively displayed, and a three-dimensional result report and a three-dimensional schematic diagram are synchronously generated.

[0013] ​​Further, a mobile terminal cooperative work module is further included, a mobile terminal application program suitable for smart phones and tablet computers is developed, field position data of the pipeline is collected by using a mobile terminal GPS positioning, the appearance features and auxiliary facilities of the pipeline are recorded by using a photographing and video recording function, and pipeline attribute information is manually input, thereby supporting an offline collection mode.

[0014] Further, in the system management module, a data security protection mechanism is added, an AES-256 encryption algorithm is used to encrypt and store and transmit sensitive pipeline data, a data access log audit function is used to record all access operations of the sensitive data, and a data leakage early warning mechanism is set.

[0015] Compared with the prior art, the present application has the following beneficial effects: At the data collection and preprocessing level, the system integrates multiple types of detection equipment, supports multi-device cooperative work and real-time data transmission, adapts to different underground medium environments and pipeline types, and greatly improves the comprehensiveness and efficiency of data collection. The preprocessing module ensures the consistency and accuracy of the data through a series of accurate processing such as coordinate unification, format standardization, noise removal and missing value filling, thereby laying a solid foundation for the subsequent links. Compared with the traditional single device collection and manual processing mode, the data collection accuracy and processing efficiency are significantly improved, and the data deviation and redundancy are effectively reduced.

[0016] At the pipeline identification and data fusion level, the pipeline feature identification model based on deep learning combines the attention mechanism and the optimized loss function to realize the accurate automatic identification of different types of pipelines and key features such as damaged points and leakage points, thereby greatly reducing the cost of manual intervention and improving the accuracy and integrity of the identification. The multi-source data fusion model scientifically eliminates data conflicts through credibility weighted calculation, integrates multi-source data such as field detection, historical archives and city GIS, forms a complete and unified census data set, and at the same time, the pipeline attribute association module realizes the accurate connection of spatial data and static attribute information through unique identification coding, realizes the fine management of "one object one file", and solves the problem of traditional data fragmentation and loose association.

[0017] At the quality control and result application level, the multi-dimensional data quality checking system evaluates the data quality from the aspects of position accuracy, attribute integrity, logical consistency and data integrity, marks unqualified data and outputs a report, thereby ensuring the high reliability of the census data. The census result management module supports dual display of two-dimensional charts and three-dimensional visualization models, intuitively presents the spatial distribution of the pipeline and the relationship with the surrounding environment, and automatically generates various work reports and statistical reports to meet the application requirements of different scenes. The dynamic update module realizes real-time update and historical tracking of the pipeline data, and the mobile terminal cooperative work module supports real-time cooperation on site, thereby greatly improving the flexibility and timeliness of the census work.

[0018] In addition, the pipeline risk assessment module accurately identifies high-risk pipeline segments and potential hazards by comprehensive quantification of multi-dimensional risk factors, providing scientific basis for risk prevention and control and operation and maintenance decision-making; the data security protection mechanism ensures the storage and transmission safety of sensitive data, and the multi-level permission division and operation log recording of the system management module ensure the stable operation of the system and the compliant use of data. Overall, the present application significantly improves the accuracy, efficiency and intelligent level of the underground municipal pipeline survey, provides comprehensive and reliable data support for urban planning and construction, construction and operation and maintenance, emergency rescue, effectively reduces the safety risk, saves the management cost, ensures the safe and stable operation of the urban underground pipeline network, and has important practical value and social significance. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A schematic block diagram of a kind of underground municipal pipeline facility census system is provided in the present application; Figure 2 A bar chart for data acquisition accuracy comparison of different census methods; Figure 3 A broken line graph for pipeline identification accuracy rate change with pipeline diameter of different models; Figure 4 Core area pipeline damage point density distribution thermal diagram. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the present application.

[0021] In the description of the present application, it should be understood that the orientations or positional relationships indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0023] Reference Figures 1 to 4 A survey system for underground municipal pipeline facilities, comprising the following modules: The data acquisition module, deployed at the survey site, integrates pipeline detectors, high-precision GPS positioning equipment, ground-penetrating radar, and UAV aerial survey equipment to collect spatial location data, burial depth data, material information, pipe diameter, flow direction information, valve location, and distribution of ancillary facilities of underground municipal pipelines. It also collects surrounding geological conditions, topographic data, and historical pipeline file numbers. Data synchronization between multiple devices is achieved through wireless communication protocols, and the acquisition frequency is dynamically adjusted according to the density of pipelines to adapt to the acquisition needs of various underground media environments and pipeline types. The data preprocessing module receives the raw data transmitted by the data acquisition module, uses an adaptive filtering algorithm to remove environmental interference noise from the probe data, maps the spatial data collected by different devices to the WGS84 unified geographic coordinate system through coordinate system transformation, uses data format standardization processing to convert various heterogeneous data into the GIS industry common format, uses interpolation algorithm to fill in local missing data, and corrects abnormal fluctuation data through data smoothing processing, providing an accurate foundation for subsequent data processing. The pipeline feature recognition module is based on deep learning to build a pipeline type recognition model. The model is trained with massive pipeline detection sample data, covering pipeline features under different geological conditions. The preprocessed detection data and ground-penetrating radar image data are input, and the spatial morphological features, gray-scale distribution features, and texture features of the pipeline are extracted through convolutional neural networks. Combined with the correlation between pipeline material, pipe diameter and underground medium, the module can automatically identify water supply pipelines, drainage pipelines, gas pipelines, power pipelines and communication pipelines, and simultaneously identify key feature points such as pipeline damage points, leakage points, and interface locations. The pipeline attribute association module establishes a basic pipeline attribute database, storing static attribute information such as pipeline ownership, construction year, design standards, maintenance records, and service life. It adopts a unique 18-bit digital structure for identification, including area code, pipeline type code, and construction year code. Through this code, the collected spatial data, feature data, and static attribute information are associated and matched to achieve a one-to-one correspondence between spatial location and attribute information. It supports batch import, manual supplementation, and quick query of attribute information. The multi-source data fusion module integrates on-site detection data acquired by the data acquisition module, attribute data from the pipeline attribute association module, historical pipeline survey archive data, and urban GIS basic data. It uses a feature-level fusion strategy to extract the core information from each data source, identifies data contradictions through a logical conflict detection algorithm, resolves conflicts based on the credibility priority of the data source, eliminates redundant information and logical conflicts between multi-source data, and forms a complete and unified pipeline survey dataset. The data quality verification module constructs a multi-dimensional data quality assessment system, verifying the fused pipeline data from four dimensions: location accuracy, attribute integrity, logical consistency, and data integrity. Location accuracy is assessed by the coordinate deviation from known control points, with a deviation threshold set at ±5cm. Attribute integrity is assessed by the fill rate of required attribute fields, which must reach over 95%. Logical consistency is assessed by the matching relationship between pipeline material and diameter, and the compatibility of pipeline burial depth with geological conditions. Data integrity is assessed by the coverage ratio between the actual survey area and the planned survey area. Unqualified data is marked and a quality verification report is output. The census results management module organizes the pipeline data that has passed quality verification according to industry standard formats such as the "Technical Specification for Urban Underground Pipeline Detection" and the "Construction Specification for Underground Pipeline Database", generating census results such as pipeline spatial database, attribute database, pipeline distribution map, and profile map. It supports the visualization, layered browsing, zooming in and out, and roaming query operations of the census results, and automatically generates documents such as census work report, data statistical report, and problem rectification suggestions. It also supports the export, sharing, and printing of the results data. The system management module features a multi-level user access control mechanism, defining different roles such as administrators, census workers, and data auditors. Administrators have full access, census workers only have data collection and upload permissions, and data auditors are responsible for data quality verification. The module records all user operation logs, including operation time, content, and results. It supports querying and exporting operation logs, as well as abnormal operation alerts. The module also provides system parameter configuration, data backup, and data recovery functions to ensure stable system operation and data security.

[0024] This invention also includes a pipeline risk assessment module. Based on the pipeline spatial data, attribute data, and characteristic data obtained from the survey, combined with surrounding environmental risk factors, geological disaster risk levels, and the frequency of underground construction activities, the risk factors include the degree of pipeline corrosion, the intensity of surrounding construction disturbance, the rate of geological subsidence, and changes in groundwater level, a pipeline operation risk assessment system is constructed to identify high-risk pipeline sections and potential safety hazards, generate risk level distribution maps, risk assessment reports, and targeted risk prevention and control suggestions, and support dynamic adjustment of risk levels and key marking of high-risk areas.

[0025] This invention also includes a dynamic update module, which establishes a dynamic update mechanism for pipeline data. It supports the on-site collection of new pipeline data, pipeline change data, and pipeline maintenance data through mobile terminal collection devices. The collection of new pipeline data must be associated with the original pipeline network topology, and the change data must be confirmed by the auditor before it takes effect. The system automatically compares the differences between the data before and after the update, marks the change content, change time, and change reason, and realizes real-time updates and historical version traceability of pipeline survey data, so that the pipeline data is consistent with the actual situation.

[0026] In this invention, the multi-source data fusion module introduces a multi-source data credibility weighted fusion model, which uses the following formula to achieve weight allocation and fusion calculation for different data sources: in This represents the integrated pipeline data value after merging. Indicates the first The reliability coefficient of each data source is calculated by comprehensively considering the accuracy of the data acquisition equipment and the accuracy of historical data. Indicates the first The original data values ​​of each data source, Indicates the first The data redundancy of a data source is calculated by the proportion of duplicate information between that data source and other data sources. Indicates the first The logical conflict level of a data source is calculated by the degree of deviation between that data source and the standard data. This represents the total number of data sources participating in the fusion. The weights of each data source are dynamically adjusted based on its credibility, redundancy, and conflict level. Data sources with high credibility, low redundancy, and low conflict level will have a higher weight in the fusion result.

[0027] In this invention, the pipeline feature recognition module employs an improved U-Net image segmentation algorithm. By adding a parallel attention mechanism of channel attention and spatial attention, it enhances the extraction of pipeline edge features and key feature points, optimizes the network loss function, and uses a weighted combination of Dice loss and cross-entropy loss to improve the recognition accuracy of small-diameter pipelines and fine-branched pipelines. It simultaneously outputs the pipeline contour coordinates, key feature point coordinates, and type recognition results, supports manual correction and secondary confirmation of recognition results, and improves the accuracy and completeness of pipeline recognition.

[0028] In this invention, the pipeline risk assessment module introduces a pipeline risk level quantification calculation model, which achieves accurate quantification of risk level through the following formula: in This represents a quantitative value indicating the risk level of a pipeline segment, ranging from 0 to 100. A higher value indicates a higher risk level. The pipeline foundation risk coefficient is determined comprehensively based on the pipeline material, service life, and design standards. Indicates the first The impact weights of risk factors were determined using the analytic hierarchy process (AHP) to ascertain the importance of each risk factor. Indicates the first The actual impact of risk factors is determined based on on-site survey data and historical statistical data. Indicates the first The effectiveness of prevention and control measures for risk factors is assessed through an evaluation of the implementation results of the implemented prevention and control measures. Indicates the first The probability of occurrence of risk factors is calculated based on historical risk event statistics for the region. It represents the total number of risk factors, and achieves a scientific classification of risk levels through the comprehensive quantification of multi-dimensional risk factors.

[0029] In this invention, the data quality verification module is equipped with a dynamic adjustment mechanism for quality thresholds. Based on different pipeline types, different survey areas, and different geological conditions, the location accuracy deviation threshold, attribute fill rate threshold, and logical consistency judgment threshold are automatically adjusted. In geologically complex areas and old pipeline areas, the location accuracy deviation threshold can be relaxed to ±10cm, and the attribute fill rate threshold is adjusted to over 90%. In urban core areas and newly built pipeline areas, the original high-standard quality thresholds are strictly enforced. Users can manually adjust the quality thresholds according to actual survey needs to generate personalized quality verification schemes.

[0030] In this invention, the survey results management module integrates a 3D visualization engine, supports the import of .obj / .fbx format models, overlays pipeline data with urban topography and building data to construct a 3D visualization model of underground municipal pipelines, supports 3D roaming, 3D sectioning, and 3D measurement operations of pipelines, and achieves a 3D measurement accuracy of ±3cm. It intuitively displays the spatial distribution relationship, burial depth, and positional relationship of pipelines with the surrounding environment, and simultaneously generates 3D results reports and 3D schematic diagrams, providing intuitive data support for pipeline planning, design, construction, and maintenance.

[0031] This invention also includes a mobile collaborative operation module, which develops a mobile application adapted to smartphones and tablets. This application allows field surveyors to receive survey tasks, view the survey scope, and retrieve historical data via mobile devices. It uses GPS positioning on the mobile device to collect pipeline location data, records pipeline appearance features and ancillary facilities through photo and video recording functions, and allows manual input of pipeline attribute information. It supports offline data collection mode with an offline storage capacity of no less than 10GB. Data is stored in environments without network access. After network recovery, incremental transmission is used to reduce network bandwidth usage, and the data is automatically synchronized to the system server, enabling real-time collaboration between field operations and back-end management.

[0032] In this invention, the system management module includes a data security protection mechanism. It employs the AES-256 encryption algorithm to encrypt and store sensitive pipeline data during transmission. It also uses a data access log auditing function to record all access operations for sensitive data, with the data access logs being retained for at least one year. A data leakage early warning mechanism is also set up, which automatically triggers an alert when abnormal access or batch export risks are detected. The system regularly generates data security reports and supports integration with the city's public safety management platform to achieve secure sharing and compliant use of sensitive pipeline data.

[0033] The following two examples further illustrate the specific implementation of this system: Example 1: Application of Comprehensive Underground Municipal Pipeline Survey in Urban Core Areas This embodiment targets the underground municipal pipeline survey scenario covering three streets in the core urban area with a total area of ​​8 square kilometers. It covers five types of pipelines, including water supply pipelines, drainage pipelines, gas pipelines, power pipelines, and communication pipelines. The underground municipal pipeline facility survey system of this invention is applied to achieve multi-device collaborative data collection, multi-source data fusion, and three-dimensional visualization output, and fully implements all functional modules.

[0034] 1. Multi-module collaborative operation The data acquisition module is deployed at the survey site, integrating pipeline detectors, high-precision GPS positioning equipment, ground-penetrating radar, and UAV aerial surveying equipment. UAV aerial surveying covers a large area, acquiring topographic data; ground-penetrating radar detects pipelines at a depth of 3-5 meters, outputting radar images; the pipeline detector collects parameters such as pipeline material and diameter; the GPS positioning equipment records the spatial coordinates of the pipelines, with the acquisition frequency adjusted according to pipeline density: 1 meter per acquisition point in core sections and 3 meters per acquisition point in ordinary sections. Simultaneously, it collects soil type, groundwater level, and historical pipeline file numbers from the surrounding geological conditions, achieving real-time data synchronization across multiple devices via wireless communication protocols.

[0035] After receiving the raw data, the data preprocessing module uses an adaptive filtering algorithm to remove environmental interference noise, with the filter window size dynamically adjusted according to the degree of data fluctuation. Through coordinate system transformation, GPS positioning data and ground-penetrating radar data are uniformly mapped to the WGS84 geographic coordinate system, with the transformation error controlled within ±3cm. Data in txt, csv, and img formats output from different devices are standardized and converted to the shp format commonly used in the GIS industry. A linear interpolation algorithm is used to fill in locally missing data, with the missing duration not exceeding 5 minutes. A moving average method is used for data smoothing to correct abnormal fluctuations.

[0036] The pipeline feature recognition module employs an improved U-Net image segmentation algorithm, adding a parallel mechanism of channel attention and spatial attention. Channel attention enhances the extraction of pipeline material-related features, while spatial attention highlights the pipeline edge contours. Inputting preprocessed ground-penetrating radar images and detection data, the module extracts spatial morphological features, grayscale distribution features, and texture features through three convolutional layers. Combining this with the correlation between pipeline material and diameter (e.g., gas pipelines are mostly metal with diameters between 100-300mm), the module automatically identifies five pipeline types. Simultaneously, it identifies key feature points such as pipeline damage points (areas with abrupt grayscale changes in radar images) and interface locations (points where pipe diameter changes). The recognition results support manual correction before being submitted to the next module.

[0037] The pipeline attribute association module establishes a basic pipeline attribute database, storing static information such as ownership unit, construction year, and design standards. It uses an 18-digit unique identifier code, with the first 6 digits representing the region, the middle 6 digits representing the pipeline type, and the last 6 digits representing the construction year. This code associates collected spatial coordinates and feature data with static attribute information. For example, code 110101-GS-201005 corresponds to a water supply pipeline in Dongcheng District, Beijing, constructed in May 2010. This achieves a one-to-one correspondence between spatial data and attribute information, supporting batch import and manual supplementation of attribute information.

[0038] The multi-source data fusion module integrates field survey data, attribute data, 2015 historical census archive data, and urban GIS basic data, and introduces a multi-source data credibility weighted fusion model: set up =4 refers to four types of data sources, including on-site detection data. =0.9, calculated based on a combined accuracy of 98% for the equipment and 95% for historical accuracy. =100 means the original pipe diameter is 100mm. =0.1, which means a redundancy of 10%. =0.05, which means a conflict level of 5%; historical archive data =0.7、 =98、 =0.3、 =0.1; GIS data =0.8、 =102、 =0.2、 =0.08; Attribute data =0.95、 =100、 =0.05、 =0.03. The calculation process is as follows: Field detection item = 0.9 × 100 × exp(-0.1 × 0.05) = 90 × 0.995 ≈ 89.55; Historical archive item = 0.7 × 98 × exp(-0.3 × 0.1) = 68.6 × 0.97 ≈ 66.54; GIS data item = 0.8 × 102 × exp(-0.2 × 0.08) = 81.6 × 0.984 ≈ 80.3; Attribute data item = 0.95 × 100 × exp(-0.05 × 0.03) = 95 × 0.9985 ≈ 94.86; =89.55+66.54+80.3+94.86≈331.25, and the pipe diameter data after fusion is taken as 331.25 / 4≈82.8mm, which is the standardized result, eliminating data redundancy and conflicts.

[0039] The data quality verification module evaluates data from multiple dimensions: location accuracy is compared with known control points, and the deviation is ≤ ±5cm; the required fields for attribute integrity have a fill rate of 96%; the logical consistency verification shows that the pipeline material and diameter match, and the diameter of the metal gas pipeline is between 100-300mm, which meets the design standards; the actual survey coverage of data integrity is 98% of the planned scope, 3 missing attribute data are marked, and a quality verification report is output.

[0040] The survey results management module organizes data according to industry standards, generates pipeline space database and attribute database, constructs a 3D visualization model, supports importing .obj format models, realizes 3D roaming, sectioning operation with a sectioning depth between 0-5 meters, measurement operation with an accuracy of ±3cm, intuitively displays the relationship between pipeline burial depth and surrounding buildings; automatically generates survey work reports and data statistics reports, including statistics on the length of 5 types of pipelines and the number of damage points.

[0041] The system management module has three levels of permissions: administrators configure system parameters and back up data, census workers collect and upload data, and data auditors verify data quality; all operation logs are recorded and stored for one year, and can be queried and exported; sensitive data is encrypted with AES-256, and an alert is triggered when abnormal batch export behavior is detected.

[0042] 2. Application effect data Table 1 is a comparison table of the application performance of the urban core area pipeline survey system: Table 1 shows that existing census systems suffer from low data collection accuracy and insufficient identification accuracy. Multi-source data fusion relies on manual integration, resulting in low efficiency. Results are only displayed in two dimensions with loose attribute relationships. This invention's system improves collection accuracy to ±5cm through multi-device collaborative collection and precise preprocessing; it improves the U-Net algorithm and multi-source data fusion model, achieving a pipeline identification accuracy of 95%; the automated fusion process increases efficiency by 4 times; and three-dimensional visualization intuitively presents the spatial relationships of pipelines, with attribute association integrity reaching 96%, enabling refined management of "one item, one file," fully meeting the high-precision and high-efficiency requirements of pipeline census in urban core areas.

[0043] Example 2: Application of Underground Pipeline Survey and Risk Assessment in Old Urban Areas This embodiment focuses on the underground pipeline survey scenario of four old streets covering a total area of ​​6 square kilometers. It aims to solve problems such as missing data, untimely updates, and unclear risks and hazards of old pipelines. The system of this invention is applied to realize the integration of survey, update, and risk assessment, and fully implements all functional modules.

[0044] 1. Multi-module collaborative operation The data acquisition module integrates a pipeline detector, GPS positioning device, and ground-penetrating radar. Addressing the characteristics of old urban areas with numerous narrow alleys, dense and severely aged pipelines, it employs a handheld pipeline detector for close-range data collection, GPS positioning devices at 1 meter per collection point, and ground-penetrating radar to detect pipelines 2-4 meters underground. Simultaneously, it collects data on the distribution of miscellaneous fill soil and records of underground construction activities. The mobile collaborative operation module deploys a smartphone-compatible application. Survey personnel receive survey tasks, view historical pipeline data, record pipeline damage by taking photos, and manually enter ownership information. It supports offline data collection with a storage capacity of 15GB, storing data in environments without network access and incrementally transferring it to the server after network recovery.

[0045] The data preprocessing module removes environmental noise from the raw data, converts the collected data to the WGS84 coordinate system, standardizes it to shp format, uses interpolation algorithms to fill in missing data from old pipelines with a missing duration of no more than 10 minutes, and corrects fluctuating data using the moving average method.

[0046] The pipeline feature recognition module adopts an improved U-Net algorithm, which focuses on improving the recognition accuracy of small-diameter pipelines with a diameter of less than 50mm and old and damaged pipelines. It automatically identifies 5 types of pipelines, as well as damage points and leakage points. The recognition results are submitted after being manually confirmed a second time.

[0047] The pipeline attribute association module uses an 18-bit unique identifier to associate collected data with static attribute information. For older pipelines, it supplements information such as construction year and maintenance records, realizing the correspondence between spatial data and attribute information, and supports manual supplementation of attribute information.

[0048] The multi-source data fusion module integrates field detection data, attribute data, 2010 historical census data, and urban GIS data. Through a multi-source data credibility weighted fusion model, it eliminates data conflicts and forms a unified dataset.

[0049] The dynamic update module establishes a data update mechanism. When new pipeline data is collected, it is associated with the original pipeline network topology. For example, when a new water supply pipeline is added, it connects to the original main pipeline. Changes to data, such as pipeline material changes, take effect after confirmation by the data auditor. The module automatically compares the differences before and after the update, marks the content, time and reason of the change, and enables historical version tracing and retains update records for the past 5 years.

[0050] The pipeline risk assessment module, based on survey data and combined with risk factors including pipeline corrosion level, intensity of surrounding construction disturbance, and geological settlement rate, introduces a quantitative calculation model for risk level: set up =3 means 3 types of risk factors =0.8, determined comprehensively based on cast iron material, 30-year service life, and low design standards. =0.4, which represents the corrosion degree weight. =0.9 indicates severe corrosion. A value of 0.3 indicates low effectiveness of prevention and control measures. =0.7 means the probability of occurrence is high; =0.3, which is the construction disturbance weight. =0.8 means frequent construction. A value of 0.4 indicates that the effectiveness of prevention and control measures is moderate. =0.6, meaning the probability of occurrence is moderate; =0.3, which is the geological subsidence weight. =0.7 indicates moderate settlement. =0.2 means that the effectiveness of prevention and control measures is low. =0.5, which is the probability of occurrence. The calculation process is as follows: Summation term = 0.4×0.9×exp(-0.3×0.7)+0.3×0.8×exp(-0.4×0.6)+0.3×0.7×exp(-0.2×0.5)=0.36×0.81+0.24×0.79+0.21×0.91≈0.29+0.19+0.19≈0.67; =0.8×0.67≈0.54, which is a standardized score of 54. The risk level is determined to be medium risk, and risk prevention and control recommendations are generated, including regular testing and partial replacement.

[0051] Data quality verification module evaluation: Position accuracy deviation ≤ ±8cm, i.e., the threshold is relaxed for old areas; attribute fill rate 92%; logical consistency verification shows that the diameter of old cast iron pipelines is between 50-200mm, which is consistent with the actual situation; data integrity 97%; 2 logical conflict data points are marked.

[0052] The census results management module generates two-dimensional distribution maps, three-dimensional models, and risk level distribution maps, and automatically generates problem rectification suggestions, including detection of medium-risk pipelines within 3 months; the system management module encrypts sensitive data, records operation logs, and supports data backup and recovery.

[0053] 2. Application effect data Table 2 is a comparison table of the application performance of the pipeline survey and risk assessment system in old urban areas: Table 2 shows that existing survey systems cannot collect data offline, rely on manual input for data updates, resulting in low efficiency, lack of risk assessment capabilities, and inability to trace historical versions, leading to incomplete hazard identification. The system of this invention supports offline collection and incremental transmission, improving data update efficiency by 7 times; through a multi-dimensional risk factor quantification model, the risk assessment accuracy reaches 92%; it supports 5-year historical version tracing, achieving 94% completeness in hazard identification, accurately locating risky sections in aging pipelines, providing a scientific basis for renovation and maintenance, effectively reducing construction safety risks and the probability of pipeline leakage accidents, and meeting the core needs of pipeline survey and risk prevention in old urban areas.

[0054] Reference Figure 2 This diagram visually illustrates the differences in data collection accuracy among different survey methods. Manual detection relies entirely on the experience of operators, is highly susceptible to subjective factors, and has the worst accuracy. Single-device detection is limited by underground media and environmental interference, with an accuracy of only ±10cm. Traditional survey systems, lacking optimized multi-device collaboration mechanisms, maintain an accuracy of ±8cm. This invention addresses the high-precision requirements of pipeline management in urban core areas, controlling the data collection accuracy to ±5cm. For the complex geological conditions of older urban areas, the threshold is relaxed to ±8cm, still outperforming traditional methods. The high-precision collected data provides a reliable foundation for subsequent feature identification and data fusion, balancing the needs of refined management in core areas with the practical application requirements of older districts.

[0055] Reference Figure 3 This figure clearly illustrates the impact of pipe diameter on recognition accuracy and the technical advantages of the model in this invention. Traditional CNN models are insufficient for feature extraction from small-diameter pipes, achieving only 65% ​​accuracy for 50mm pipes. Even with increased pipe diameter, the overall accuracy remains low. This invention improves the U-Net model by incorporating a dual-attention mechanism of channel and spatial elements, enhancing feature extraction for small-diameter pipes. The accuracy is increased to 88% for 50mm pipes, and maintains a high accuracy of over 95% for large-diameter scenarios. This addresses the industry pain point of low accuracy in identifying numerous small-diameter, old pipes in aging urban areas, ensuring accurate identification of pipes across the entire diameter range.

[0056] Reference Figure 4 This map visually illustrates the differences in the density of damage points among different sub-regions and types of pipelines within the core area. Area A2 has the highest density of damage points for drainage pipelines, reaching 6 per square kilometer, while area A5 has the lowest density for communication pipelines, at only 0.2 per square kilometer. Compared to traditional two-dimensional tables, the heat map uses color gradients to quickly locate areas and pipeline types with high damage densities, facilitating focused inspection and maintenance of drainage pipelines in area A2 by survey personnel. This visualization method improves the efficiency of survey results application and provides an intuitive spatial distribution basis for pipeline maintenance planning.

[0057] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A system for surveying underground municipal pipeline facilities, characterized in that, Includes the following modules: The data acquisition module, deployed at the survey site, integrates various detection and aerial survey equipment to collect core data on underground municipal pipelines and related data from the surrounding area. The data preprocessing module receives raw data, fills in missing data, corrects abnormal data, and provides a foundation for subsequent processing. The pipeline feature recognition module is based on a recognition model built with deep learning and trained on a large number of samples. It takes preprocessed data and images as input, extracts pipeline features, and combines them with relevant correlation patterns. The pipeline attribute association module establishes a basic attribute database to store pipeline static attribute information. It uses an 18-bit unique identifier code to associate spatial data, feature data and static attributes, and supports the import, supplementation and query of attribute information. The multi-source data fusion module integrates on-site detection, attributes, historical archives and urban GIS basic data. It adopts a feature-level fusion strategy and eliminates redundancy and conflicts through conflict detection and resolution to form a unified census dataset. The data quality verification module constructs a multi-dimensional evaluation system, verifying data from four aspects: location accuracy, attribute integrity, logical consistency, and data integrity. It marks unqualified data and outputs a verification report. The census results management module organizes qualified data according to industry standards, generates various census results, supports visual display and query operations of results, automatically generates relevant documents, and supports exporting, sharing and printing of results; The system management module sets up a multi-level user permission management mechanism, divides different roles and assigns corresponding permissions, records operation logs and supports related functions, and provides system configuration, data backup and recovery functions.

2. The underground municipal pipeline facility survey system according to claim 1, characterized in that, It also includes a pipeline risk assessment module, which, based on the pipeline spatial data, attribute data and characteristic data obtained from the survey, combined with the surrounding environmental risk factors, geological disaster risk levels and the frequency of underground construction activities, constructs a pipeline operation risk assessment system, identifies high-risk pipeline sections and potential safety hazards, generates risk level distribution maps, risk assessment reports and targeted risk prevention and control suggestions, and supports dynamic adjustment of risk levels and key marking of high-risk areas.

3. The underground municipal pipeline facility survey system according to claim 1, characterized in that, It also includes a dynamic update module, which establishes a dynamic update mechanism for pipeline data. It supports the on-site collection of new pipeline data, pipeline change data, and pipeline maintenance data through mobile terminal collection devices. The collection of new pipeline data must be associated with the original pipeline network topology, and the change data must be confirmed by the auditor before it takes effect. It automatically compares the differences between the data before and after the update.

4. The underground municipal pipeline facility survey system according to claim 1, characterized in that, The multi-source data fusion module introduces a multi-source data credibility weighted fusion model, which completes the weight allocation and fusion calculation of different data sources through the following formula: in This represents the integrated pipeline data value after merging. Indicates the first The credibility coefficient of each data source Indicates the first The original data values ​​of each data source, Indicates the first Data redundancy of each data source Indicates the first Logical conflict level of each data source This represents the total number of data sources participating in the fusion. The weights of each data source are dynamically adjusted based on its credibility, redundancy, and conflict level. Data sources with high credibility, low redundancy, and low conflict level will have a higher weight in the fusion result.

5. The underground municipal pipeline facility survey system according to claim 1, characterized in that, The pipeline feature recognition module employs an improved U-Net image segmentation algorithm. By adding a parallel attention mechanism of channel attention and spatial attention, it enhances the extraction of pipeline edge features and key feature points. It also uses a weighted combination of Dice loss and cross-entropy loss to improve the recognition accuracy of small-diameter pipelines and fine-branched pipelines. The module simultaneously outputs the pipeline contour coordinates, key feature point coordinates, and type recognition results.

6. The underground municipal pipeline facility survey system according to claim 2, characterized in that, The pipeline risk assessment module introduces a pipeline risk level quantification calculation model, which uses the following formula to accurately quantify the risk level: in This represents a quantitative value indicating the risk level of a pipeline segment. This indicates the risk coefficient of the pipeline foundation. Indicates the first The weight of the impact of risk factors Indicates the first The actual impact of risk factors Indicates the first The effectiveness of prevention and control measures for risk factors. Indicates the first The probability of occurrence of risk factors. This indicates the total number of risk factor categories.

7. The underground municipal pipeline facility survey system according to claim 1, characterized in that, The data quality verification module includes a dynamic adjustment mechanism for quality thresholds. It automatically adjusts the location accuracy deviation threshold, attribute fill rate threshold, and logical consistency judgment threshold based on different pipeline types, survey areas, and geological conditions. The original high-standard quality thresholds are strictly enforced in urban core areas and newly constructed pipeline areas. Users can also manually adjust the quality thresholds according to their actual survey needs.

8. The underground municipal pipeline facility survey system according to claim 1, characterized in that, The survey results management module integrates a 3D visualization engine, which overlays pipeline data with urban topography and building data to construct a 3D visualization model of underground municipal pipelines. It supports 3D roaming, 3D sectioning, and 3D measurement operations of pipelines, intuitively displaying the spatial distribution relationship, burial depth, and positional relationship of pipelines with the surrounding environment, and simultaneously generating 3D results reports and 3D schematic diagrams.

9. The underground municipal pipeline facility survey system according to claim 1, characterized in that, It also includes a mobile collaborative operation module, which develops a mobile application adapted to smartphones and tablets, uses mobile GPS positioning to collect pipeline site location data, records pipeline appearance features and ancillary facilities through photo and video recording functions, allows manual input of pipeline attribute information, and supports offline data collection mode.

10. The underground municipal pipeline facility survey system according to claim 1, characterized in that, The system management module includes a data security protection mechanism. It uses the AES-256 encryption algorithm to encrypt and store sensitive pipeline data during transmission, records all access operations of sensitive data using a data access log auditing function, and sets up a data leakage early warning mechanism.

Images