An adjacent map spot fusion method and system based on FME

Abstract

The application discloses a kind of adjacent graph spot fusion method and system based on FME, it is related to adjacent graph spot fusion technical field, method includes data processing, parameter setting and rule configuration, executes adjacent graph spot fusion, result verification and output and feedback and optimization step, clear fusion standard and rule, unified evaluation index system, optimize spatial relationship processing, the work flow of automation reduces manual intervention, improves processing efficiency;It also provides adjacent graph spot fusion system based on FME, including data preprocessing module, data input module, parameter setting and rule configuration module, comprehensive correlation index calculation module, graph spot fusion module and result verification and output module, executes the adjacent graph spot fusion method based on FME.

Description

Technical Field

[0001] This invention relates to the field of adjacent patch fusion technology, and specifically to an adjacent patch fusion method and system based on FME. Background Technology

[0002] The fusion of adjacent land parcels exhibits significant differences across various application fields, thus posing different demands on the selection of data processing methods and algorithms. For example, in land use research, land parcel fusion primarily focuses on the precise classification and quantitative statistics of land types to ensure the rational allocation and effective utilization of land resources; in environmental monitoring, land parcel fusion emphasizes the spatial correlation and temporal changes between different environmental elements, aiming to improve the accuracy and timeliness of monitoring data to help decision-makers respond to environmental issues in a timely manner; and in urban planning, land parcel fusion focuses on integrating the spatial layout and functional zoning of urban elements to promote sustainable urban development and optimize resource allocation.

[0003] Currently, methods for fusing adjacent polygons mainly rely on traditional spatial analysis and data processing techniques, such as buffer analysis, spatial connectivity, and topology processing. While these methods can achieve the fusion of adjacent polygons in simple scenarios, their processing efficiency and accuracy are significantly insufficient when faced with complex index weights, irregular boundaries, and diverse polygon types. First, existing map patch fusion methods typically require significant manual intervention, resulting in cumbersome and time-consuming processes with low automation. They struggle to adapt flexibly to different application scenarios in dynamically changing environments, limiting data processing efficiency and real-time performance. Second, different application scenarios exhibit significant differences in map patch fusion standards and rules, which traditional methods fail to systematically consider. The lack of clear fusion standards and rules leads to fusion results that do not meet actual application needs. Furthermore, various fields lack unified definitions of the required quality standards, accuracy indicators, and applicable rules for fusion, making effective comparison and integration of results from different projects difficult and increasing the complexity of data use. Third, current map patch fusion methods lack a unified indicator system to evaluate fusion effectiveness. When multiple indicators are involved, the lack of an effective weighting mechanism and a unified indicator system makes it difficult to quantify and analyze the relative importance and influence of each element. Finally, the spatial relationships between adjacent map patches are complex, with irregular boundaries and diverse morphological variations. Existing spatial relationship processing methods cannot effectively capture the spatial interactions and influences between different elements, resulting in insufficient spatial relationship processing capabilities and reducing the reliability and practicality of the fusion results.

[0004] To address the shortcomings of the aforementioned adjacent patch fusion methods in terms of automation, standardization, indicator system, and spatial relationship processing, this patent proposes an adjacent patch fusion method and system based on FME, which improves the efficiency, accuracy, and adaptability of fusion to better meet the diverse needs of various fields. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide a method and system for fusing adjacent patches based on FME, thereby improving the efficiency and accuracy of fusing and providing more reliable data support for geographic information applications.

[0006] On the one hand, this invention provides a method for fusing adjacent patches based on FME.

[0007] S10. Data Preparation: Based on the specific business needs of the industry, collect datasets related to the fusion of adjacent patches, and standardize, clean and correct the collected data to prepare the data for subsequent fusion, that is, obtain the areal data of the patches to be fused.

[0008] S20. Parameter setting and rule configuration: Based on the specific business scenario, analyze the characteristic requirements of different patches, set the fusion parameters, then quantify the key fusion parameters, calculate the comprehensive correlation index and output it.

[0009] The key fusion parameters include the area threshold of the fused patch, the length of the common boundary line between the fused patch and its adjacent patches, whether there are any border lines cutting the patches, and the attribute correlation between the fused patch and its surrounding patches.

[0010] S30. Perform adjacent patch fusion: FME automatically applies the parameters and rules set and configured in S20 to perform fusion operations and automatically identify adjacent patches. At this time, by utilizing FME's powerful streaming processing capabilities, the system can process multiple data sources in parallel, thereby significantly improving overall processing efficiency, quickly responding to large-scale data fusion operations, and ensuring the efficiency and accuracy of the fusion process.

[0011] S40. Result Verification and Output: The fusion results undergo multi-dimensional quality verification to assess their accuracy and consistency, ensuring the final output meets expected quality standards. After verification, the system generates standardized output results, which can be presented as graphical files or database records for easy subsequent analysis and application.

[0012] S50. Feedback and Optimization: Evaluate based on fusion results and expected goals, collect user feedback, analyze feedback information, and adjust parameter settings and rule applications as necessary.

[0013] In step S10, a dataset related to the fusion of adjacent map patches is collected according to the specific business needs of the industry. The specific datasets collected are as follows, taking into account the characteristics of different business scenarios:

[0014] For land acquisition related to reservoir resettlement, relevant land acquisition and land parcel archiving results should be collected, including CAD files and Excel data. The data should include the boundaries, land types, and land ownership information of the resettlement area to ensure that all information related to land acquisition and resettlement is accurately reflected during the map fusion process.

[0015] If the business involves land use, then map data and attribute tables related to land use are collected. This includes spatial distribution information of different land types such as cultivated land, forest land, and grassland, as well as their related attribute data, such as land use type, land ownership, and current use status, so as to reasonably consider the characteristics of various types of land during the integration process.

[0016] For environmental monitoring operations, a dataset of map features related to environmental monitoring should be established, focusing on collecting data from water quality monitoring points, air quality monitoring stations, and their monitoring indicators. Integrating the spatial locations, monitoring parameters, and historical data of this data will provide necessary background information for map feature fusion, thereby improving the scientific validity and practicality of the data.

[0017] For urban planning projects, relevant map data should be collected, including information on urban land use zoning, infrastructure layout, green spaces, and public facilities. Simultaneously, factors such as urban development trends and policies / regulations must be considered to ensure that the integrated data supports urban development decisions.

[0018] Furthermore, in step S10, the collected data is formatted, cleaned, and corrected. Specifically, the data format is standardized, data cleaning includes deduplication, handling missing and outlier values, data correction includes standardizing the coordinate system, ensuring accuracy and consistency, and finally, data quality checks are performed to prepare for subsequent fusion processing.

[0019] In step S20, fusion parameters are set. When setting these parameters, researchers should first consult with experts in the field to ensure that the set spatial and attribute parameters are reasonable and feasible. Expert feedback will provide important support for developing scientific fusion standards, thereby improving the effectiveness and accuracy of the fusion process.

[0020] The following are the specific analysis methods and parameter settings for four main business scenarios:

[0021] Reservoir resettlement land acquisition: In this scenario, researchers need to pay attention to the area threshold of the merged patches, which is recommended to be set to 3.33 square meters, equivalent to 0.01 mu (approximately 0.01 acres). This ensures that patches larger than or equal to this threshold can be labeled and output, requiring users to complete the relevant attribute information; while small, attribute-less patches smaller than this threshold can be effectively merged with adjacent patches. Required attribute parameters include land owner, land type, area, district / county, township, and village.

[0022] Land Use: For land use scenarios, researchers need to analyze the characteristics of different land types and set area thresholds based on specific business scenarios, such as 1 mu (approximately 0.16 acres), to exclude irrelevant small plots. If adjacent plots have the same land type, the correlation calculated using attribute indicators is 1, and subsequent fusion of adjacent plots can proceed. Furthermore, attribute correlation should be weighted based on the similarity of land types, and the calculated comprehensive index will serve as the basis for plot fusion.

[0023] Environmental Monitoring: In environmental monitoring, researchers need to focus on the spatial location and attribute information of monitoring points, setting a low area threshold, such as 0.1 acres, to maintain focus on small monitoring areas. The distance requirements for monitoring points should be based on the actual conditions of the monitoring area to ensure effective integration of monitoring information. Attribute correlation should include environmental data such as water quality and air quality indicators to fully utilize monitoring information during the integration process.

[0024] Urban Planning: For urban planning, a suitable area threshold needs to be set, such as 10 mu (approximately 667 square meters), to consider the main functional zoning of urban land. The setting of the length of public boundaries should take into account the connectivity of urban roads, green spaces, and public facilities to ensure a rational layout of urban space. Attribute relevance should cover urban development indicators, the matching degree between planned uses and the current situation, to improve the scientific nature of the integration process.

[0025] Furthermore, in step S20, the key fusion parameters are quantified one by one, specifically as follows:

[0026] Areas smaller than or equal to an area threshold are merged into their adjacent areas. The selection of merged areas depends on the evaluation of a comprehensive correlation index. The calculation formula is:

[0027] ;

[0028] in, Indicates spatial correlation. Indicates attribute relevance. and Let be the weight coefficient, and satisfy... By appropriately adjusting these two weights, researchers can balance the impact of spatial and attribute correlation in fusion decisions based on specific business needs.

[0029] The spatial correlation is quantified by evaluating the length of the common boundary, the distance between the center points, or whether the patch is cut by the frame line. The length of the common boundary is the length of the common boundary line between the merged patch and the adjacent patches. The distance between the center points is the distance between the center point of the merged patch and the center point of the surrounding adjacent patches.

[0030] Furthermore, in the application scenario of land acquisition for reservoir resettlement, when CAD results are cut by map frame lines during archiving, an indicator of "whether it is a map frame line-cutting patch" is introduced. This indicator marks whether a patch is a map frame line-cutting patch. If two adjacent patches are map frame line-cutting patches, and one of the smaller patches has lost its attributes, the two patches are merged.

[0031] ;

[0032] in, This indicates that the pattern is cut by the pattern line. This indicates that it is not a line cut of the pattern.

[0033] Furthermore, the spatial correlation quantification specifically includes:

[0034] If the length of the common boundary is used as an indicator, then the following formula is used to quantify spatial correlation:

[0035] ;

[0036] in, It is the length of the common boundary line between the merged patch and its adjacent patches. It is the length of the maximum common boundary line of all adjacent polygons;

[0037] If distance is used as an indicator, then the following formula is used to quantify spatial correlation:

[0038] ;

[0039] This formula ensures that the closer the distance, the better. A higher value indicates a stronger spatial correlation between the two. It is the distance between the center point of the merged patch and the center point of an adjacent patch. It is the distance between the center point of the merged patch and the center point of the nearest neighboring patch.

[0040] If a certain weight combination of the two is used, then:

[0041] ;

[0042] in, , This corresponds to the spatial relevance weight value. , The corresponding weight coefficients, and satisfying .

[0043] Furthermore, the weight calculation for the common boundary length specifically involves obtaining the largest value among N values:

[0044] ;

[0045] The corresponding common boundary line length weights are:

[0046] ;

[0047] Where A is the small area of ​​the patch to be merged, N is the number of adjacent patches around it, and K is the neighboring patch of patch A, where K ranges from 1 to N, and the length of the corresponding common boundary line is L. K .

[0048] Furthermore, the weight of the center point distance is calculated by taking the minimum value among N values:

[0049] ;

[0050] The distance weight of the corresponding center point is:

[0051] ;

[0052] Where A is the small area of ​​the image patch to be merged, N is the number of adjacent image patches around it, and K is the neighboring image patch of image patch A, where K ranges from 1 to N, and the distance from the corresponding image patch to image patch A is... .

[0053] Furthermore, attribute relevance encompasses the relationship between the merged map patch and its surrounding map patches in terms of land type, ownership, and economic value. Specifically, in land type-based fusion, if the merged map patch shares the same land type as its adjacent surrounding map patches, its land type relevance is 1; if the land types are similar, the land type relevance can be set to a value close to 1 to reflect their similarity. Similarly, in ownership-based fusion, if the merged map patch shares the same ownership as its adjacent map patches, the ownership relevance between the two map patches is 1 to reflect the consistency of ownership.

[0054] When considering the combined influence of multiple attributes, the attribute correlation is calculated using the following formula. If the attribute correlation is evaluated based on three attributes, the comprehensive attribute correlation evaluation is:

[0055] ;

[0056] in , and These represent the relevance values ​​of the 1st, 2nd, and 3rd attributes, respectively. , , Let be the weight coefficient of the corresponding attribute, and satisfy . .

[0057] On the other hand, the present invention provides a neighboring patch fusion system based on FME, for executing the aforementioned FME-based neighboring patch fusion method, specifically including:

[0058] Data Preprocessing Module: This module receives geospatial data from various sources, supporting formats including Shapefile, GDB, CAD, and Excel. It prepares the data for subsequent fusion of adjacent polygons. Depending on the business requirements, if the provided data is relatively raw, this module may function as a system itself, handling initial data processing and standardization. If there is already isometric data to be fused, this module primarily standardizes the corresponding field attributes to meet system requirements. For example, it can convert labeled CAD sheet data into GDB data containing attributes. In this converted data, some polygons may contain attribute information, while others may not have had their attribute labels manually added to the corresponding cut polygons due to CAD sheet division. Depending on the business requirements, these polygons need to be fused to achieve more accurate spatial data integration.

[0059] Data Input Module: This module is primarily responsible for selecting the areal data of the patches to be fused after processing by the data preprocessing module, providing basic data support for the patch fusion process. Depending on specific business needs, this module can flexibly adapt to different data structures and attribute field naming conventions. Since attribute names in patch data may differ in real-world applications, this module allows users to edit and adjust attribute fields to ensure that the data input meets the requirements of subsequent processing, thus laying a solid foundation for the patch fusion stage.

[0060] Parameter setting and rule configuration module: This module allows users to flexibly set fusion parameters and rules according to specific application needs, including selected fusion standards, weight settings, and spatial relationship processing rules. Users can easily adjust parameters through a graphical interface to adapt to the patch fusion requirements of different scenarios. In addition, this module also allows users to save frequently used configuration templates for quick future use.

[0061] The comprehensive correlation index calculation module is responsible for calculating the comprehensive correlation index between the fused small patch and its surrounding large patches. By analyzing factors such as spatial location, patch attributes, the length of the fusion boundary, and whether the patch is cut by a map frame line, the module generates the corresponding comprehensive correlation index. This calculation process provides a scientific basis for subsequent patch fusion, ensuring that the fusion results better meet actual business needs.

[0062] The patch fusion module is one of the core components of the system. Leveraging the powerful processing capabilities of the FME platform, it automates the fusion of adjacent patches through predefined fusion rules and metrics. This module can handle complex boundaries and diverse patch types, and it fuses patches according to different rules and weights to ensure the accuracy and effectiveness of the fusion results. Simultaneously, the module considers the weighting requirements of complex metrics to ensure optimal fusion results under various conditions.

[0063] Result Verification and Output Module: This module is responsible for verifying the fusion results. By comparing them with the original data, it ensures that the fusion results meet preset quality standards and application requirements. Simultaneously, this module supports outputting the final results in multiple formats for subsequent analysis, decision-making, and visualization.

[0064] The beneficial effects of this invention are:

[0065] 1. This invention formulates specific fusion standards and rules for different business scenarios to ensure the scientific nature and consistency of the fusion process, enabling effective comparison and integration of results from different projects; it establishes a unified indicator system, combined with a weighted mechanism, to quantify the importance of different elements; it optimizes spatial relationship processing, paying particular attention to the spatial relationships between adjacent map features, and employs advanced algorithms to ensure that the fusion results truly reflect actual business characteristics; it fully utilizes the powerful data processing capabilities of the FME platform, designs an automated workflow, reduces manual intervention, improves processing efficiency, and enables the system to quickly adapt to dynamically changing data environments, significantly improving real-time performance.

[0066] 2. By establishing a standardized fusion process and a flexible parameter configuration mechanism, this invention can adapt to the needs of different fields, ensuring that data processing consistency and efficiency are maintained while meeting the specific objectives of each field. This not only significantly improves the accuracy and precision of data processing but also enhances data interoperability between different fields, providing strong support for the comprehensive application of geographic information. At the same time, this invention solves the problems of fusion between small and narrow patches and adjacent patches caused by partial overlap of boundary lines during the conversion of CAD line data to GDB surface data in the process of land acquisition for reservoir resettlement, as well as attribute loss and spatial inconsistency caused by the splitting of CAD results during archiving, through specific rule definitions. Attached Figure Description

[0067] Figure 1 Flowchart for data preparation in Embodiment 1 of the present invention;

[0068] Figure 2 This is a framework diagram of parameter setting and rule configuration in Embodiment 1 of the present invention;

[0069] Figure 3This is a flowchart of the overall system architecture of Embodiment 2 of the present invention;

[0070] Figure 4 This is a schematic diagram of the data preprocessing interface in Embodiment 2 of the present invention;

[0071] Figure 5 This is a schematic diagram of the interface of the adjacent patch fusion system based on FME in Embodiment 2 of the present invention;

[0072] Figure 6 This is a diagram showing the data preprocessing effect of Embodiment 2 of the present invention;

[0073] Figure 7 This is a schematic diagram of the output of the adjacent patch fusion system based on FME in Embodiment 2 of the present invention after fusion of patches;

[0074] Figure 8 This is a diagram showing the fused pattern after embodiment 2 of the present invention.

[0075] Attached image translation (English):

[0076] Translation Parameter Values: Transform parameter values;

[0077] User Parameters: User parameter values;

[0078] Source DWG / DXF File(s): Original DWG / DXF files;

[0079] Source Microsoft Excel File(s): Original Microsoft Excel files;

[0080] Feature Types to Read: The types of features to be read;

[0081] File Geodatabase: A file-based geodatabase;

[0082] Save As User Parameter Default Values: Saves the default values ​​for user parameters.

[0083] Options; Presets; Run; Cancel;

[0084] spatial_weight: Spatial relevance weight;

[0085] boundary_weight: Weight of the common boundary line length;

[0086] distance_weight: Distance weight from the center point;

[0087] enable_frame_cut: Enables frame cutting;

[0088] Attribute weight: Weight related to the attribute;

[0089] attribute0l_name: The name of attribute 01;

[0090] attribute01_weight: The weight of attribute 01. Detailed Implementation

[0091] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but are not intended to limit the present invention.

[0092] Please see Figures 1-2 A neighboring patch fusion method based on Fiber Optic Mapping (FME) includes the following steps: This invention provides a neighboring patch fusion method based on Fiber Optic Mapping (FME).

[0093] S10. Data Preparation: Based on the specific business needs of the industry, collect datasets related to the fusion of adjacent patches, and standardize, clean and correct the collected data to prepare the data for subsequent fusion, that is, obtain the areal data of the patches to be fused.

[0094] In step S10, a dataset related to the fusion of adjacent map patches is collected according to the specific business needs of the industry. The specific datasets collected are as follows, taking into account the characteristics of different business scenarios:

[0095] For land acquisition related to reservoir resettlement, relevant land acquisition and land parcel archiving results should be collected, including CAD files and Excel data. The data should include the boundaries, land types, and land ownership information of the resettlement area to ensure that all information related to land acquisition and resettlement is accurately reflected during the map fusion process.

[0096] If the business involves land use, then map data and attribute tables related to land use are collected. This includes spatial distribution information of different land types such as cultivated land, forest land, and grassland, as well as their related attribute data, such as land use type, land ownership, and current use status, so as to reasonably consider the characteristics of various types of land during the integration process.

[0097] For environmental monitoring operations, a dataset of map features related to environmental monitoring should be established, focusing on collecting data from water quality monitoring points, air quality monitoring stations, and their monitoring indicators. Integrating the spatial locations, monitoring parameters, and historical data of this data will provide necessary background information for map feature fusion, thereby improving the scientific validity and practicality of the data.

[0098] For urban planning projects, relevant map data should be collected, including information on urban land use zoning, infrastructure layout, green spaces, and public facilities. Simultaneously, factors such as urban development trends and policies / regulations must be considered to ensure that the integrated data supports urban development decisions.

[0099] In step S10, the collected data is formatted, cleaned, and corrected. Specifically, the data format is standardized, data cleaning includes deduplication, handling missing and outlier values, data correction includes standardizing the coordinate system, ensuring accuracy and consistency, and finally, data quality is checked to prepare for subsequent fusion processing.

[0100] S20. Parameter Setting and Rule Configuration: Based on the specific business scenario, analyze the characteristic requirements of different patches, set fusion parameters, then quantify the key fusion parameters, calculate the comprehensive correlation index, and output it. The key fusion parameters include the area threshold of the fused patch, the length of the common boundary line between the fused patch and adjacent patches, whether there are any border lines cutting the patches, and the attribute correlation between the fused patch and its surrounding patches.

[0101] In step S20, fusion parameters are set. When setting these parameters, researchers should first consult with experts in the field to ensure that the set spatial and attribute parameters are reasonable and feasible. Expert feedback will provide important support for developing scientific fusion standards, thereby improving the effectiveness and accuracy of the fusion process.

[0102] The following are the specific analysis methods and parameter settings for four main business scenarios:

[0103] Reservoir resettlement land acquisition: In this scenario, researchers need to pay attention to the area threshold of the merged patches, which is recommended to be set to 3.33 square meters, equivalent to 0.01 mu (approximately 0.01 acres). This ensures that patches larger than or equal to this threshold can be labeled and output, requiring users to complete the relevant attribute information; while small, attribute-less patches smaller than this threshold can be effectively merged with adjacent patches. Required attribute parameters include land owner, land type, area, district / county, township, and village.

[0104] Land Use: For land use scenarios, researchers need to analyze the characteristics of different land types and set area thresholds based on specific business scenarios, such as 1 mu (approximately 0.16 acres), to exclude irrelevant small plots. If adjacent plots have the same land type, the correlation calculated using attribute indicators is 1, and subsequent fusion of adjacent plots can proceed. Furthermore, attribute correlation should be weighted based on the similarity of land types, and the calculated comprehensive index will serve as the basis for plot fusion.

[0105] Environmental Monitoring: In environmental monitoring, researchers need to focus on the spatial location and attribute information of monitoring points, setting a low area threshold, such as 0.1 acres, to maintain focus on small monitoring areas. The distance requirements for monitoring points should be based on the actual conditions of the monitoring area to ensure effective integration of monitoring information. Attribute correlation should include environmental data such as water quality and air quality indicators to fully utilize monitoring information during the integration process.

[0106] Urban Planning: For urban planning, a suitable area threshold needs to be set, such as 10 mu (approximately 667 square meters), to consider the main functional zoning of urban land. The setting of the length of public boundaries should take into account the connectivity of urban roads, green spaces, and public facilities to ensure a rational layout of urban space. Attribute relevance should cover urban development indicators, the matching degree between planned uses and the current situation, to improve the scientific nature of the integration process.

[0107] In step S20, the key fusion parameters are quantified one by one, specifically as follows:

[0108] Areas smaller than or equal to an area threshold are merged into their neighboring areas. The selection of merged areas depends on an evaluation of a comprehensive correlation index. The calculation formula is:

[0109] ;

[0110] in, Indicates spatial correlation. Indicates attribute relevance. and Let be the weight coefficient, and satisfy... By appropriately adjusting these two weights, researchers can balance the impact of spatial and attribute correlation in fusion decisions based on specific business needs.

[0111] The spatial correlation is quantified by evaluating the length of the common boundary, the distance between the center points, or whether the patch is cut by the frame line. The length of the common boundary is the length of the common boundary line between the merged patch and the adjacent patches. The distance between the center points is the distance between the center point of the merged patch and the center point of the surrounding adjacent patches.

[0112] In the application scenario of land acquisition for reservoir resettlement, CAD software introduces an indicator of "whether a piece of map is cut by a map frame line" when the archived results are segmented by map frame lines. This indicator marks whether a piece of map is cut by a map frame line. If two adjacent pieces of map are cut by a map frame line, and one of the smaller pieces has lost its attributes, the two pieces of map are merged.

[0113] ;

[0114] in, This indicates that the pattern is cut by the pattern line. This indicates that it is not a line cut of the pattern.

[0115] The spatial correlation quantification is specifically as follows:

[0116] If the length of the common boundary is used as an indicator, then the following formula is used to quantify spatial correlation:

[0117] ;

[0118] in, It is the length of the common boundary line between the merged patch and its adjacent patches. It is the length of the maximum common boundary line of all adjacent polygons;

[0119] If distance is used as an indicator, then the following formula is used to quantify spatial correlation:

[0120] ;

[0121] This formula ensures that the closer the distance, the better. A higher value indicates a stronger spatial correlation between the two. It is the distance between the center point of the merged patch and the center point of an adjacent patch. It is the distance between the center point of the merged patch and the center point of the nearest neighboring patch.

[0122] If a certain weight combination of the two is used, then:

[0123] ;

[0124] in, , This corresponds to the spatial relevance weight value. , The corresponding weight coefficients, and satisfying .

[0125] The weight calculation for the length of the common boundary is specifically as follows: obtain the largest value among N values.

[0126] ;

[0127] The corresponding common boundary line length weights are:

[0128] ;

[0129] Where A is the small area of ​​the patch to be merged, N is the number of adjacent patches around it, and K is the neighboring patch of patch A, where K ranges from 1 to N, and the length of the corresponding common boundary line is L. K .

[0130] The weight of the center point distance is calculated by taking the minimum value among N values:

[0131] ;

[0132] The distance weight of the corresponding center point is:

[0133] ;

[0134] Where A is the small area of ​​the image patch to be merged, N is the number of adjacent image patches around it, and K is the neighboring image patch of image patch A, where K ranges from 1 to N, and the distance from the corresponding image patch to image patch A is... .

[0135] Attribute relevance encompasses the relationship between the merged map patch and its surrounding map patches in terms of land use type, ownership, and economic value. Specifically, in land use-based fusion, if the merged map patch shares the same land use type as its adjacent surrounding map patches, its land use relevance is 1; if the land use types are similar, the land use relevance can be set to a value close to 1 to reflect their similarity. Similarly, in ownership-based fusion, if the merged map patch shares the same ownership as its adjacent map patches, the ownership relevance between the two map patches is 1 to reflect the consistency of ownership.

[0136] This embodiment demonstrates the structure of attribute correlation calculation, including the correlation assessment method for land type, utilization rate, location value, and comprehensive attributes.

[0137] The land use relevance weight calculation is as follows: A is a small area of ​​map patch that needs to be merged; N is the number of surrounding adjacent map patches; the land use type of map patch A is land use type X; and map patch K is the adjacent map patch of map patch A, where the value of K ranges from 1 to N, and the corresponding land use type of the map patch is Y. k Then, the similarity of land types is:

[0138] ;

[0139] The land category similarity function can be used to construct a land category similarity matrix, from which the similarity coefficients can be read.

[0140] The utilization rate is calculated based on relevance weights. A represents a small area of ​​map patch that needs to be merged, N is the number of adjacent map patches around it, the utilization rate of map patch A is X, and map patch K represents adjacent map patches of map patch A, where K ranges from 1 to N, and the corresponding utilization rate of the adjacent map patch is Y. k Then the utilization rate is similar:

[0141] ;

[0142] The utilization rate function can be set based on relevant zoning data to determine the correlation between utilization rates. The closer the utilization rates of two plots of land are (the smaller the difference), the higher the correlation. If the utilization rates are the same, the correlation is 1.

[0143] The location value relevance weight calculation is as follows: A represents a small area of ​​map patch that needs to be merged, N represents the number of surrounding adjacent map patches, the location value coefficient of map patch A is X, and map patch K represents adjacent map patches of map patch A, where the value of K ranges from 1 to N, and the corresponding map patch value coefficient is Y. k Then, the similarity of locational value:

[0144] ;

[0145] The location value coefficient function can be calculated based on the distance to important facilities (such as transportation hubs). If the distance is close, the value is close to 1. If the distance is far, the location value correlation gradually decreases. The closer the location value coefficients are, the higher the correlation. If the neighborhood committee value coefficients are the same, the correlation is 1.

[0146] The overall attribute correlation is:

[0147] .

[0148] in, For land type similarity, the values ​​are the same. ; For the value of utilization correlation, the same ; For the location value relevance value, the same ; , , These are the weight coefficients of the three components in the overall attribute correlation, satisfying... .

[0149] S30. Perform adjacent patch fusion: FME automatically applies the parameters and rules set and configured in S20 to perform fusion operations and automatically identify adjacent patches. At this time, by utilizing FME's powerful streaming processing capabilities, the system can process multiple data sources in parallel, thereby significantly improving overall processing efficiency, quickly responding to large-scale data fusion operations, and ensuring the efficiency and accuracy of the fusion process.

[0150] S40. Result Verification and Output: The fusion results undergo multi-dimensional quality verification to assess their accuracy and consistency, ensuring the final output meets expected quality standards. After verification, the system generates standardized output results, which can be presented as graphical files or database records for easy subsequent analysis and application.

[0151] S50. Feedback and Optimization: Evaluate based on fusion results and expected goals, collect user feedback, analyze feedback information, and adjust parameter settings and rule applications as necessary.

[0152] In the reservoir resettlement and land acquisition project, early CAD map data has been archived. To meet the needs of the information system construction, this data needs to be converted to GDB spatial database format. During the early measurement process, the area calculation based on the CAD map was set with an area threshold of 0.01 mu (equivalent to 3.34 square meters). However, during the conversion of CAD line data to GDB polygon data, it was discovered that due to repeated measurements and overlapping of CAD boundary lines, small, narrow polygons appeared in the converted polygon data. These small, narrow polygons need to be merged into adjacent polygons.

[0153] To address this issue, the research team set boundary line length as a comprehensive relevance index for fusion, with a weight of 1, to ensure that boundary line length is given priority as a comprehensive index for fusion of adjacent patches during the fusion process. Meanwhile, the project often suffers from the loss of attribute markers for smaller plots during archiving due to map sheet segmentation. To address this, if attribute loss occurs in patches converted to polygons, their surrounding patches can be directly integrated based on the criterion of "whether the patch is cut by the map frame line."

[0154] After optimization, for map features larger than 3.34 square meters, if attributes are missing, the relevant attribute information needs to be supplemented in the CAD drawing to ensure data integrity. For map features smaller than 3.34 square meters, fusion can be performed based on the criterion of whether they are map features cut by a map frame line, without needing to supplement attribute information. This improves the efficiency and accuracy of fusion between adjacent map features and ensures that the final data output meets the user's actual needs and expected standards, providing reliable data support for subsequent information management of reservoir resettlement and land acquisition.

[0155] In water resource monitoring projects, the attributes of some water body patches fail to fully reflect their relevance to water quality and quantity monitoring. The attribute data for some small water bodies lack key water quality parameters and quantity information, such as chemical oxygen demand (COD), turbidity, flow rate, and water volume, resulting in these patches not being effectively utilized during data fusion.

[0156] To address this issue, the research team updated the attribute correlation calculation. The specific steps are as follows:

[0157] (1) Data collection and analysis: Researchers first collected relevant water quality and quantity monitoring data and assessed the importance of different water body patches in the monitoring project. Through analysis of historical data, the correlation between water quality and quantity and each patch was determined.

[0158] (2) Design of Attribute Correlation Indicators: Based on the analysis results, the research team designed new attribute correlation indicators, focusing on the weights of water quality parameters and water quantity parameters. When synthesizing the attribute correlation indicators, water quality and water quantity parameters were assigned different weight coefficients to highlight their influence on monitoring. For example, the weight of water quality parameters was set to 0.7, and the weight of water quantity parameters was set to 0.3.

[0159] (3) Calculation formula adjustment: The updated formula for calculating attribute correlation is:

[0160] Attribute correlation = 0.7 × water quality index + 0.3 × water quantity index

[0161] This formula is used to evaluate the correlation of comprehensive attributes of each water body patch, so that the monitoring needs can be more accurately reflected during the patch fusion process.

[0162] (4) Optimization of the fusion process: During the fusion of patches, researchers evaluated them based on the updated attribute correlation index. Patches with high attribute correlation were prioritized for fusion to ensure that key water quality and quantity information was not lost. This improvement enabled relevant water body patches to better retain important monitoring data during the fusion process.

[0163] (5) Validation and Feedback: After optimization, the research team validated the updated attribute correlation calculation results and found that the quality of the fused patches was significantly improved. Meanwhile, user feedback showed that the fused data better met the actual needs of water resource monitoring, improving the reliability and effectiveness of the monitoring results.

[0164] This method of optimizing attribute correlation calculation makes the fusion process of water body patches more scientific and reasonable, ensuring that the fusion results can better support water resource management and decision analysis.

[0165] Through these specific implementations, researchers can effectively optimize parameter settings and rule adjustments, thereby improving the efficiency and accuracy of adjacent patch fusion and ensuring that the final output meets the user's actual needs and expected standards.

[0166] Example 2

[0167] See Figures 3-8 , Figure 3The overall architecture of a neighboring patch fusion system based on FME is demonstrated, including the various functional modules and their interrelationships. Specifically, an FME-based neighboring patch fusion system includes:

[0168] Data Preprocessing Module: This module receives geospatial data from various sources, supporting formats including Shapefile, GDB, CAD, and Excel. It prepares the data for subsequent fusion of adjacent polygons. Depending on the business requirements, if the provided data is relatively raw, this module may function as a system itself, handling initial data processing and standardization. If there is already isometric data to be fused, this module primarily standardizes the corresponding field attributes to meet system requirements. For example, it can convert labeled CAD sheet data into GDB data containing attributes. In this converted data, some polygons may contain attribute information, while others may not have had their attribute labels manually added to the corresponding cut polygons due to CAD sheet division. Depending on the business requirements, these polygons need to be fused to achieve more accurate spatial data integration. Figure 4 This is a schematic diagram of the data preprocessing interface, which shows the layout of the system's user interface and intuitively presents the interactive effects when the user performs operations.

[0169] Data Input Module: This module is primarily responsible for selecting the areal data of the patches to be fused after processing by the data preprocessing module, providing basic data support for the patch fusion process. Depending on specific business needs, this module can flexibly adapt to different data structures and attribute field naming conventions. Since attribute names in patch data may differ in real-world applications, this module allows users to edit and adjust attribute fields to ensure that the data input meets the requirements of subsequent processing, thus laying a solid foundation for the patch fusion stage.

[0170] Parameter setting and rule configuration module: This module allows users to flexibly set fusion parameters and rules according to specific application needs, including selected fusion standards, weight settings, and spatial relationship processing rules. Users can easily adjust parameters through a graphical interface to adapt to the patch fusion requirements of different scenarios. In addition, this module also allows users to save frequently used configuration templates for quick future use.

[0171] The comprehensive correlation index calculation module is responsible for calculating the comprehensive correlation index between the fused small patch and its surrounding large patches. By analyzing factors such as spatial location, patch attributes, the length of the fusion boundary, and whether the patch is cut by a map frame line, the module generates the corresponding comprehensive correlation index. This calculation process provides a scientific basis for subsequent patch fusion, ensuring that the fusion results better meet actual business needs.

[0172] The patch fusion module is one of the core components of the system. Leveraging the powerful processing capabilities of the FME platform, it automates the fusion of adjacent patches through predefined fusion rules and metrics. This module can handle complex boundaries and diverse patch types, and it fuses patches according to different rules and weights to ensure the accuracy and effectiveness of the fusion results. Simultaneously, the module considers the weighting requirements of complex metrics to ensure optimal fusion results under various conditions. Figure 5 This is a schematic diagram of the interface of the FME-based adjacent patch fusion system.

[0173] Result Verification and Output Module: This module is responsible for verifying the fusion results. By comparing them with the original data, it ensures that the fusion results meet preset quality standards and application requirements. Simultaneously, this module supports outputting the final results in multiple formats for subsequent analysis, decision-making, and visualization. Figure 6 The image shows the results of data preprocessing, illustrating data examples before and after the fusion of adjacent patches, reflecting the differences in patch data before and after fusion. Figure 7 This is a schematic diagram of the output after fusing patches in an FME-based neighbor patch fusion system. Figure 8 This is a display image of the merged image patches.

[0174] Although the present invention has been described in detail above with general descriptions and specific embodiments, the scope of protection of the present invention is not limited thereto. Modifications or improvements can be made to the present invention, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A neighboring patch fusion method based on FME, characterized in that, Includes the following steps: S10. Data preparation: Based on the specific business needs of the industry, collect datasets related to the fusion of adjacent patches, and unify, clean and correct the format of the collected data to prepare the data for subsequent fusion, that is, to obtain the areal data of the patches to be fused. S20. Parameter setting and rule configuration: Analyze the characteristic requirements of different patches based on specific business scenarios, set fusion parameters, then quantify the key fusion parameters, calculate the comprehensive correlation index and output it. The key fusion parameters include the area threshold of the fused patch, the length of the common boundary line between the fused patch and its adjacent patches, whether there are any patches cut by the frame line, and the attribute correlation between the fused patch and its surrounding patches. The quantification of key fusion parameters specifically involves... Areas smaller than or equal to an area threshold are merged into their adjacent areas. The selection of merged areas depends on the evaluation of a comprehensive correlation index. The calculation formula is: ； in, Indicates spatial correlation. Indicates attribute relevance. and Let be the weight coefficient, and satisfy... ; The spatial correlation is quantified by evaluating the length of the common boundary, the distance between the center points, or whether the patch is cut by the frame line. The length of the common boundary is the length of the common boundary line between the merged patch and the adjacent patches. The distance between the center points is the distance between the center point of the merged patch and the center points of the surrounding adjacent patches. The spatial correlation is quantified, specifically as follows: If the length of the common boundary is used as an indicator, then the following formula is used to quantify spatial correlation: ； in, It is the length of the common boundary line between the merged patch and its adjacent patches. It is the length of the maximum common boundary line of all adjacent polygons; If distance is used as an indicator, then the following formula is used to quantify spatial correlation: ； in, It is the distance between the center point of the merged patch and the center point of an adjacent patch. It is the distance between the center point of the merged patch and the center point of its nearest neighboring patch; If a certain weight combination of the two is used, then: ； in, , This corresponds to the spatial relevance weight value. , The corresponding weight coefficients, and satisfying ; S30. Perform adjacent patch fusion: FME automatically applies the parameters and rules set and configured in S20 to perform fusion operations and automatically identifies adjacent patches; S40. Result Verification and Output: Perform multi-dimensional quality verification on the fusion results, evaluate the accuracy and consistency of the fusion results, and obtain standardized output results; S50. Feedback and Optimization: Evaluate based on fusion results and expected goals, collect user feedback, analyze feedback information, and adjust parameter settings and rule applications.

2. The adjacent patch fusion method based on FME according to claim 1, characterized in that, In step S10, a dataset related to the fusion of adjacent map patches is collected according to the specific business needs of the industry. Specifically: If it is a land acquisition project for reservoir resettlement, then collect the relevant land acquisition and resettlement map archive results, including CAD files and Excel data results; the data includes the boundaries, land types, and land ownership information of the resettlement area; If it is a land use business, collect land use-related map data and attribute tables, including spatial distribution information of different land types and their attribute data. The data includes land use type, land ownership, and current use status. If it is an environmental monitoring operation, a dataset of map features related to environmental monitoring will be established to collect data on water quality monitoring points, air quality monitoring stations and their monitoring indicators. If it is urban planning business, then collect map data related to urban planning, including urban land use zoning, infrastructure layout, green space and public facilities.

3. The adjacent patch fusion method based on FME according to claim 1, characterized in that, In step S10, the collected data is formatted, cleaned, and corrected. Specifically, the data format is standardized, data cleaning includes deduplication, handling missing and outlier values, data correction includes standardizing the coordinate system, ensuring accuracy and consistency, and finally, data quality is checked.

4. The adjacent patch fusion method based on FME according to claim 1, characterized in that, The determination of whether a feature is a frame-cutting feature is used to indicate whether a feature is a frame-cutting feature. If two adjacent features are frame-cutting features, then if one of the features has lost its attributes, the two features are merged. ； in, This indicates that the pattern is cut by the pattern line. This indicates that it is not a line cut of the pattern.

5. The adjacent patch fusion method based on FME according to claim 1, characterized in that, The weight calculation for the length of the common boundary is specifically as follows: obtain the largest value among N values. ； The corresponding common boundary line length weights are: ； Where A is the small area of ​​the patch to be merged, N is the number of adjacent patches around it, and K is the neighboring patch of patch A, where K ranges from 1 to N, and the length of the corresponding common boundary line is L. K .

6. The adjacent patch fusion method based on FME according to claim 1, characterized in that, The weight of the center point distance is calculated by taking the minimum value among N values: ； The distance weight of the corresponding center point is: ； Where A is the small area of ​​the image patch to be merged, N is the number of adjacent image patches around it, and K is the neighboring image patch of image patch A, where K ranges from 1 to N, and the distance from the corresponding image patch to image patch A is... .

7. The adjacent patch fusion method based on FME according to claim 1, characterized in that, When considering the combined influence of multiple attributes, the attribute correlation is calculated using the following formula. If the attribute correlation is evaluated based on three attributes, the comprehensive attribute correlation evaluation is: ； in , and These represent the relevance values ​​of the 1st, 2nd, and 3rd attributes, respectively. , , Let be the weight coefficient of the corresponding attribute, and satisfy . .

8. A neighboring patch fusion system based on FME, characterized in that, The method for performing the FME-based neighbor patch fusion method according to any one of claims 1-7 specifically includes: Data preprocessing module: This module receives geospatial data from various sources, supporting formats including Shapefile, GDB, CAD, and Excel. It performs format unification, cleaning, and correction on the collected data to prepare it for subsequent fusion. Data Input Module: This module selects the areal data of the patches to be merged after being processed by the data preprocessing module, providing basic data support for the patch fusion process. Depending on specific business needs, this module adapts to different data structures and naming conventions for attribute fields, allowing users to edit and adjust attribute fields. Parameter setting and rule configuration module: This module sets fusion parameters and rules according to specific application requirements. Users can adjust the parameters through a graphical interface to adapt to the patch fusion requirements in different scenarios. This module also has the function of saving commonly used configuration templates for quick access. Comprehensive correlation index calculation module: This module calculates the comprehensive correlation index between the merged small patch and its surrounding large patches. It generates the corresponding comprehensive correlation index by analyzing factors such as spatial location, patch attributes, length of the fusion boundary, and whether the patch is cut by the map frame line. Map patch fusion module: This module is based on the FME platform and realizes the automatic fusion of adjacent map patches through the set fusion rules and indicators; Result Verification and Output Module: This module verifies the fusion result by comparing it with the original data, and outputs the final result in multiple formats, including graphic files and database records.

Images