Project management method based on multi-source monitoring data fusion

Abstract

The invention relates to the technical field of multi-source data fusion, in particular to a project management method based on multi-source monitoring data fusion, and the method improves the data fusion precision, efficiency and query speed, quantifies the reliability difference of monitoring means based on a CRITIC method, introduces an actual availability screening and time index decay model, and improves the data fusion accuracy, efficiency and query speed. Weight space-time dual dynamic optimization is realized, decision deviation is overcome, an objective basis is provided for overlapped pattern spot judgment, and fusion accuracy and credibility are improved. Weight updating is controlled through the QZGXSJ field, total data recalculation is avoided, resource consumption is reduced, the design is compatible with mainstream GIS database standards, seamless connection with an existing system is achieved, and popularization is easy. Meanwhile, global data are stored according to quarterly slices by using a time axis engine, and a project engine is evolved according to a code storage version, so that joint storage is realized, the problem of fragmented management is solved, and the query speed is increased.

Description

Technical Field

[0001] This invention relates to the field of multi-source data fusion technology, and in particular to a project management method based on multi-source monitoring data fusion. Background Technology

[0002] Major projects are construction projects urgently needed by the state and local governments to develop production and improve people's lives. They play a key role in the development of the entire national economy. Monitoring the implementation of major projects is crucial for assessing their progress, optimizing resource allocation, and supporting major decision-making.

[0003] However, for a long time, the monitoring methods for major projects have been singular, mostly in the form of documents and reports. This has led to problems such as the lack of intuitive presentation of project status, inaccurate information scheduling, and lagging human supervision services. By using technologies such as remote sensing satellites, drones, and field surveys to obtain multi-source monitoring data, and regularly integrating the monitoring data, we can obtain the current project construction status and integrate the monitoring data of each period according to the timeline. This not only meets the verification and scheduling needs of regulatory departments, but also facilitates the review of the project's historical progress, providing strong data support for management. Summary of the Invention

[0004] The purpose of this invention is to provide a project management method based on multi-source monitoring data fusion, which aims to solve the problem that the existing multi-source monitoring data fusion methods are relatively limited.

[0005] To achieve the above objectives, this invention provides a project management method based on multi-source monitoring data fusion, comprising the following steps:

[0006] Obtain multi-source monitoring data and project constraints;

[0007] Extract feature data and perform spatial matching and processing;

[0008] Establish core fields, calculate dynamic weights, and merge data;

[0009] The current status of the bidirectional engine storage query project.

[0010] The section on "Obtaining Multi-Source Monitoring Data and Project Constraints" includes the following steps:

[0011] Multi-source monitoring data was obtained through satellites, drones, and field surveys;

[0012] Collect a list of major projects and coordinate files, and obtain the constraint range of major projects through coordinate transformation and attribute organization.

[0013] The "Extracting Feature Data and Performing Spatial Matching and Processing" step includes the following steps:

[0014] By combining artificial intelligence technology, feature data such as bulldozing areas, buildings, and ground greening in multi-source monitoring data can be identified and extracted;

[0015] Parse metadata coordinate parameter information in multi-source monitoring data, convert feature data pixel coordinates to geographic coordinates and perform projection transformation to complete spatial location matching of feature data;

[0016] Vectors are generated using binary masking techniques, and boundary optimization and geometric repair are performed.

[0017] The section on "Establishing core fields, calculating dynamic weights, and fusing data" includes the following steps:

[0018] Establish core fields, check and correct outliers, and set conditions for recalculating dynamic weight values.

[0019] When key fields are cleared, the dynamic weight DTQZ is updated using the CRITIC weighting method;

[0020] Overlapping regions are detected and marked, and the data is merged and updated in order of dynamic weight values.

[0021] The section on "Establishing the Current Status of the Bidirectional Engine Storage Query Project" includes the following steps:

[0022] Store snapshot data of the construction status of each phase and its corresponding original multi-source monitoring data according to the timeline, and support querying historical versions by time engine;

[0023] All related data for the same project are stored together by a unique project code, enabling the dynamic extraction of individual projects from the overall dataset. The project engine can then be used to query the entire lifecycle of a project and trace its evolution.

[0024] This invention presents a project management method based on multi-source monitoring data fusion, significantly improving the accuracy and reliability of data fusion. Based on the CRITIC method, it objectively quantifies the inherent reliability differences of different monitoring methods and introduces practical availability screening and a time-exponential decay model, achieving dual dynamic optimization of weights in both space and time. This dynamic weight fusion mechanism effectively overcomes the decision-making bias risks caused by subjective weighting, simple averaging, or static weights, providing an objective and reliable core basis for determining the priority of overlapping patches, thereby significantly improving the accuracy and credibility of multi-source geospatial data fusion results. It also improves computational efficiency and compatibility with engineering projects. The QZGXSJ field controls the on-demand update mechanism of weights, effectively avoiding unnecessary full data recalculation overhead and significantly reducing the computational resource consumption of the system. The design is directly compatible with mainstream GIS database standards, seamlessly connecting to existing geographic information system infrastructure without complex conversions, making it highly implementable and easy to promote and deploy. A bidirectional engine stores and queries the construction status of major projects. The timeline engine stores the full-domain fusion base version by quarterly slices, allowing horizontal switching of time and rapid loading of historical snapshots. The project engine stores version evolution snapshots and version evolution maps by code. By enabling joint storage across the horizontal timeline and vertical projects, the fragmented management model is resolved, and query speed is improved. This addresses the issue of the relatively limited methods for fusion of multi-source monitoring data in existing technologies. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of a project management method based on multi-source monitoring data fusion provided by the present invention.

[0027] Figure 2 This is a flowchart for obtaining multi-source monitoring data and project constraints.

[0028] Figure 3 It is a flowchart for extracting feature data and performing spatial matching and processing.

[0029] Figure 4 It is a flowchart for establishing core fields, calculating dynamic weights, and fusing data.

[0030] Figure 5 This is a flowchart illustrating the current status of the bidirectional engine storage query project.

[0031] Figure 6This is a flowchart of the process that triggers the dynamic weight value recalculation.

[0032] Figure 7 This is a flowchart of the bidirectional engine storage implementation process. Detailed Implementation

[0033] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0034] Please see Figures 1 to 7 This invention provides a project management method based on multi-source monitoring data fusion, comprising the following steps:

[0035] S1 acquires multi-source monitoring data and project constraints;

[0036] S11 acquires multi-source monitoring data through satellites, drones, and field surveys;

[0037] Specifically, basic monitoring data such as remote sensing images, videos, and photos covering the target area are obtained and preprocessed through monitoring methods such as satellites, drones, and field surveys.

[0038] S12 collects a list of major projects and coordinate files, and obtains the constraint range of major projects through coordinate transformation and attribute organization.

[0039] Specifically, compile a list of major projects and coordinate files, and generate constraint ranges for major projects with spatial location and attribute information.

[0040] S2 extracts feature data and performs spatial matching and processing;

[0041] S21 combines artificial intelligence technology to identify and extract characteristic data such as bulldozing areas, buildings, and ground greening from multi-source monitoring data;

[0042] Specifically, by combining artificial intelligence, the construction forms of multi-source data, such as bulldozing areas, buildings, and vegetation, are extracted through surface visual features such as spectrum, texture, and color, as well as geometric and morphological features such as shape, structure, edge, and spatial relationships.

[0043] S22 parses metadata coordinate parameter information in multi-source monitoring data, performs feature data pixel coordinates to geographic coordinates and projection transformation, and completes spatial location matching of feature data.

[0044] Specifically, the metadata information of multi-source data, such as sensor parameters and positioning coordinates, is parsed to obtain accurate geospatial positioning of feature data, and then unified to the same coordinate reference through coordinate transformation.

[0045] S23 generates vectors using binary masking technology and performs boundary optimization and geometric repair processing.

[0046] Specifically, the feature data for determining spatial positioning is vectorized. First, the feature extraction results are converted into a binary mask image. Then, a boundary tracking algorithm is used to extract the mask contour and construct a polygon with a complete topological structure (including hole polygons), clarifying the hierarchical relationship between holes and external polygons. Next, polygon optimization is performed, including geometric simplification, boundary smoothing, and noise removal (removing broken surfaces, abnormal patches, etc.). Finally, geometric repair is performed, executing operations such as self-intersection correction and gap closure to ensure the geometric integrity of the vector data.

[0047] S3 establishes core fields, calculates dynamic weights, and merges data;

[0048] S31 establishes core fields, checks and corrects outliers, and sets the dynamic weight value recalculation trigger condition;

[0049] Specifically, establish core fields and check for outliers. Records with missing information, values ​​outside the specified range, or unreasonable times are marked and corrected. The fields are as follows:

[0050] Monitoring methods (JCSD) record the monitoring methods used in the data source;

[0051] Get Time (HQSJ): Records the time retrieved from the data source;

[0052] Land cover type (DWLX) records the land cover type extracted from features.

[0053] The Dynamic Weights Value (DTQZ) is calculated using the CRITIC weighting method.

[0054] Overlapping marker (CDBJ) is used to mark areas that spatially overlap with other features.

[0055] Weight update time (QZGXSJ) records the time of dynamic weight calculation or update.

[0056] Construct an R-tree spatial index. Insert the unique identifier (ID) of the geometric object along with the MBR into the R-tree, and associate vector polygons with monitoring methods (JCSD), acquisition time (HQSJ), and land cover type (DWLX) for easy and fast retrieval and querying.

[0057] Establish a spatial indexing function to divide the data into spatial grids. :

[0058]

[0059] in For grid cell boundaries, It is a geometric polygon in geographic space.

[0060] When S32 clears the key fields, it uses the CRITIC weighting method to update the dynamic weight DTQZ;

[0061] Specifically, the QZGXSJ field stores the last calculation timestamp of the dynamic weight value. Its update logic is to clear the field when the current monitoring data source changes or when the time difference between the current time and the field exceeds the threshold T_max. For the part of QZGXSJ that is empty, the dynamic weight DTQZ is automatically recalculated. After the weight calculation is completed, the field is updated with the current system time, as shown in Figure x.

[0062] The CRITIC weighting method is used to assign weights to the reliability of data sources obtained by different monitoring methods, such as satellite, UAV, and field survey, to obtain the inherent weight values ​​for the three monitoring methods. Then, the actual weight values ​​are adjusted according to the actual monitoring methods within the same geographical area. The specific steps are as follows:

[0063] The first step is to select sample data (n) from existing data on monitoring data at different times and locations. Each sample contains reliability indices for three monitoring methods, namely geometric and attribute accuracy of the data, forming a matrix with columns corresponding to monitoring methods and rows corresponding to samples:

[0064]

[0065] The second step is to calculate the contrast intensity (standard deviation). The following is the standard deviation of monitoring method j:

[0066]

[0067] The third step is to calculate the conflict (correlation coefficient):

[0068] First, calculate the Pearson correlation coefficient matrix R among the monitoring methods:

[0069]

[0070] Next, calculate the conflict of the j-th method:

[0071]

[0072] Fourth step, calculate the information carrying capacity of the j-th method (Cj combines volatility and independence; the larger the value, the higher the weight):

[0073]

[0074] Fifth step: Calculate the normalized weight values ​​for different monitoring methods:

[0075]

[0076] Step 6: Review the monitoring methods used within the same geographical area and set... Simultaneously, a time-exponential decay model is used for weight reduction, and settings are made. To obtain the adjusted weight values:

[0077]

[0078] The adjusted weight values ​​are stored in the dynamic weight DTQZ field to determine the priority order of overlapping patches.

[0079] S33 detects and marks overlapping areas, and then merges and updates the data in order of dynamic weight values.

[0080] Specifically, when spatial overlap occurs in the polygons, an overlap marker (CDBJ) field is used to record the overlap markers. For the overlapping markers, the following rules apply: If a high-weight polygon completely covers a low-weight polygon, the complete geometry and attributes of the high-weight polygon are retained, and the low-weight polygon is deleted. If a high-weight polygon is within the range of a low-weight polygon, or if the high-weight and low-weight polygons intersect, the overlapping portion of the low-weight polygon is deleted, and the original data of the non-overlapping area is retained. Gwh represents a polygon with higher weight, and Gwl represents a polygon with lower weight, as detailed below.

[0081]

[0082] S4 establishes a bidirectional engine to store and query the current status of the project.

[0083] S41 stores snapshot data of the construction status of each phase and its corresponding original multi-source monitoring data according to the time axis, and supports querying historical versions by time engine;

[0084] Specifically, a timeline base table is established to store snapshots of all projects across the entire domain at specific points in time. Core fields include timestamp (unique identifier of the data point in time), global snapshot, and original data identifier (metadata associated with the original data used to generate this snapshot). A project lifecycle table is established to store snapshots of individual projects at different points in time. Core fields include project unique number, project timestamp associated with the timeline base table, project snapshot (data snapshot of the corresponding range of the project at that point in time extracted from the global slice), and project original data identifier (metadata associated with the original data used to generate the snapshot for that project). When the fusion of a new phase of multi-source monitoring data is completed, and the constraints of major projects are applied, a fused data snapshot F is generated across the entire domain project scope. F is stored as a new record in the timeline base table, with the timestamp representing the current processing period, and the corresponding original data identifier is recorded. From the current major project's full-domain snapshot F, a subset of the project's data at this point in time is cropped according to the project's defined scope to form a project-level snapshot S. Snapshot S is written as a new record into the project's full lifecycle table, associated with the corresponding project unique number, project timestamp, and original data identifier. By continuously writing to S at different points in time, a full lifecycle data view of the project is naturally accumulated.

[0085] S42 stores all related data of the same project by unique project code, enabling the dynamic extraction of individual projects from the overall dataset and the use of the project engine to query the entire lifecycle of the project and trace its changes.

[0086] Specifically, the system provides a query interface, with an underlying bidirectional query architecture built on a timeline engine and a project engine, and relies on distributed storage to achieve high-performance data access. During queries, it efficiently integrates information from three core data storage areas: a time-domain snapshot storage area, a raw data storage area (unprocessed raw data), and a project lifecycle storage area. The timeline engine handles time-based queries, i.e., queries for the overall state at a specific point in time; the project engine handles related queries for different project stages, i.e., queries for the complete lifecycle trajectory of a specific project.

[0087] Beneficial effects

[0088] This invention discloses a multi-source monitoring data fusion technology method for major projects. By establishing a trigger weight value update field and a dynamic weight value update algorithm, the data can be automatically updated on schedule. At the same time, a bidirectional engine storage is established to realize project storage and query from both the time axis engine and the project engine.

[0089] 1. Significantly improves the accuracy and reliability of data fusion. Based on the CRITIC method, the inherent reliability differences of different monitoring methods are objectively quantified, and a practical availability screening and time exponential decay model are introduced to achieve dual dynamic optimization of weights in both space and time. This dynamic weight fusion mechanism effectively overcomes the decision-making bias risks caused by subjective weighting, simple averaging, or static weights, providing an objective and reliable core basis for determining the priority of overlapping patches, thereby significantly improving the accuracy and credibility of multi-source geospatial data fusion results.

[0090] 2. Improved computational efficiency and project compatibility. The on-demand update mechanism for weights controlled by the QZGXSJ field effectively avoids unnecessary full data recalculation overhead, significantly reducing the system's computational resource consumption. The design is directly compatible with mainstream GIS database standards, allowing seamless integration with existing geographic information system infrastructure without complex conversions. It is highly implementable and easy to deploy and promote.

[0091] 3. Two-way engine storage for querying the construction status of major projects. The timeline engine stores a unified global database by quarterly slices, allowing for horizontal switching between time periods and rapid loading of historical snapshots. The project engine stores version evolution snapshots and version evolution graphs by code. This achieves joint storage across the horizontal timeline and vertical projects, resolving the fragmented management problem and improving query speed.

[0092] The above-disclosed embodiments are merely preferred embodiments of a project management method based on multi-source monitoring data fusion according to the present invention. They should not be construed as limiting the scope of the present invention. Those skilled in the art can understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A project management method based on multi-source monitoring data fusion, characterized in that, Includes the following steps: Obtain multi-source monitoring data and project constraints; Extract feature data and perform spatial matching and processing; Establish core fields, calculate dynamic weights, and merge data; The current status of the bidirectional engine storage query project.

2. The project management method based on multi-source monitoring data fusion as described in claim 1, characterized in that, The "Acquiring Multi-Source Monitoring Data and Project Constraints" section includes the following steps: Multi-source monitoring data was obtained through satellites, drones, and field surveys; Collect a list of major projects and coordinate files, and obtain the constraint range of major projects through coordinate transformation and attribute organization.

3. The project management method based on multi-source monitoring data fusion as described in claim 1, characterized in that, The "Extracting Feature Data and Performing Spatial Matching and Processing" section includes the following steps: By combining artificial intelligence technology, feature data such as bulldozing areas, buildings, and ground greening in multi-source monitoring data can be identified and extracted; Parse metadata coordinate parameter information in multi-source monitoring data, convert feature data pixel coordinates to geographic coordinates and perform projection transformation to complete spatial location matching of feature data; Vectors are generated using binary masking techniques, and boundary optimization and geometric repair are performed.

4. The project management method based on multi-source monitoring data fusion as described in claim 1, characterized in that, The "Establishing core fields, calculating dynamic weights, and fusing data" section includes the following steps: Establish core fields, check and correct outliers, and set conditions for recalculating dynamic weight values. When key fields are cleared, the dynamic weight DTQZ is updated using the CRITIC weighting method; Overlapping regions are detected and marked, and the data is merged and updated in order of dynamic weight values.

5. The project management method based on multi-source monitoring data fusion as described in claim 1, characterized in that, The "Current Status of Establishing a Two-Way Engine Storage Query Project" includes the following steps: Store snapshot data of the construction status of each phase and its corresponding original multi-source monitoring data according to the timeline, and support querying historical versions by time engine; All related data for the same project are stored together by a unique project code, enabling the dynamic extraction of individual projects from the overall dataset. The project engine can then be used to query the entire lifecycle of a project and trace its evolution.

Images