Forest and grass resource data fusion method and system

Abstract

The invention relates to the field of data processing, in particular to a forest and grass resource data fusion method and system. The method comprises the following steps: acquiring a data source list of multi-source data, extracting a field mapping table and coordinate reference parameters, executing geometry and attribute standardization, generating a quality label set, and forming an evidence record set; in combination with the target achievement specification and the target version number, a classification mapping rule and topological constraints are extracted, closed constraints and time sequence consistency constraints are registered and counted, and an evidence slot template is generated; a candidate resource unit set and an alignment mapping table are formed through spatial index clustering and evidence association processing, and a fusion judgment result is obtained; and finally, generating a target version and a change difference package according to a fusion judgment result, and completing audit tracing record writing and credibility parameter group updating. According to the method, fusion judgment and versioning management of the multi-source forest and grass resource data under the unified constraint framework are realized, and the consistency, traceability and continuous treatment capability of the fusion process are improved.

Description

Technical Field

[0001] This invention relates to the field of data processing, and in particular to a method and system for fusion of forest and grassland resource data. Background Technology

[0002] In the field of data processing, existing solutions for forestry and grassland resource data fusion typically revolve around the access and organization of data source lists. Data fusion is achieved through the conversion of field and coordinate definitions, the merging of geometry and attributes, the generation of candidate objects, and the formation of output results. However, these solutions suffer from limitations such as difficulty in unifying the field mapping table and coordinate reference parameters, a lack of consistent quality label sets to support geometric and attribute standardization, and incomplete implementation of statistical closure constraints and temporal consistency constraints during the fusion process. Existing methods often rely on spatial index clustering or overlay matching to form candidate sets, and perform alignment and merging in the absence of evidence slot template constraints. Under statistical closure constraints and temporal consistency constraints, this can lead to unstable associations between the evidence record set and the candidate resource unit set, difficulty in unifying the alignment mapping table with classification mapping rules and topological constraints, and challenges in ensuring the stable implementation of fusion adjudication results and change differential packages. For the joint processing of data source lists, evidence record sets, evidence slot templates, alignment mapping tables, and consistency constraints, existing technologies generally suffer from shortcomings such as a lack of structured definition for conflict detection, difficulty in forming verifiable links in constraint consistency decision graphs, and loose connections between target version generation and change extraction and audit traceability record writing. These shortcomings make it difficult to form a continuous and consistent process for data collection and registration, alignment and adjudication, version generation and differential output, audit traceability, and credibility parameter group updates in forestry and grassland resource data fusion application scenarios. As a result, the correspondence between fusion adjudication results and audit traceability records is unclear, and the update of credibility parameter groups lacks a unified data link foundation. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a method for fusing forestry and grassland resource data, comprising:

[0004] S100. Obtain a list of data sources, extract the field mapping table and coordinate reference parameters, perform geometric and attribute standardization and generate a quality label set to obtain a set of evidence records; wherein, the list of data sources includes source identifier, data type, coverage, storage location, collection timestamp and access method marker;

[0005] S200. Obtain the target outcome specification and target version number, extract classification mapping rules and topological constraints, and register statistical closure constraints and temporal consistency constraints for processing, and generate evidence slot templates; wherein, the target outcome specification includes classification system caliber, field caliber and statistical caliber;

[0006] S300. Based on the evidence record set and evidence slot template, perform spatial index clustering and evidence association table generation processing to generate a candidate resource unit set and alignment mapping table, and construct a constraint consistency decision graph to obtain the fusion adjudication result; wherein, the fusion adjudication result includes the final spatial boundary, final classification, final attribute field set, conflict interpretation code and evidence citation list;

[0007] S400. Based on the fusion adjudication result, perform target version generation and change extraction processing to generate a change differential package, and perform audit traceability record writing and credibility parameter group update processing to obtain audit traceability records and credibility parameter groups.

[0008] Furthermore, the process of extracting the field mapping table and coordinate baseline parameters includes:

[0009] The field mapping table includes field name correspondence, unit caliber, encoding dictionary caliber, and null value caliber; the coordinate reference parameters include projection caliber, caliber of the reference plane, and caliber of the coordinate axis order.

[0010] Furthermore, the process of performing geometry and attribute standardization and generating a quality label set includes:

[0011] The geometric standardization process includes geometric type consistency, coordinate benchmark unification, and geometric validity verification. The attribute standardization process includes field extraction, field mapping, encoding dictionary conversion, and null value unification. A quality label set is generated based on a credibility parameter set, which includes timeliness label generation rules, spatial accuracy label generation rules, completeness rate label generation rules, and verification consistency label generation rules. An evidence record set is generated, which includes source identifier, spatial geometry, attribute field set, collection timestamp, processing timestamp, quality label set, credibility score, and evidence digest hash.

[0012] Furthermore, the process of extracting classification mapping rules and topological constraints includes:

[0013] The classification mapping rules include source identifier, source classification field path, target classification field path, mapping conditions, conflict priority marker and missing backfill marker; the topological constraints include boundary closure constraints, adjacent unit non-overlapping constraints, hole handling constraints, cross-administrative boundary trimming constraints and protection red line boundary consistency constraints.

[0014] Furthermore, the process of handling registration and statistical closure constraints and temporal consistency constraints includes:

[0015] The statistical closure constraints include the area summary caliber within the same statistical level, the summary consistency caliber between different statistical levels, and the consistency caliber between the total area and the total area. The temporal consistency constraints include the version inheritance relationship caliber, the change legality caliber, and the consistency caliber with the release timestamp. The slot orchestration process is performed, which includes associating the target result specifications with the data source list and generating evidence slot templates. The evidence slot templates include slot number, slot type, source list, field mapping table, minimum quality threshold, gap filling rules, and conflict interpretation code dictionary. The source list is generated from the set of source identifiers in the data source list that meet the coverage and access status conditions. The minimum quality threshold is bound to the slot type and is consistent with the credibility score caliber of the credibility parameter group. The gap filling rules include trigger conditions, supplementary source suggestions, collection range caliber, field caliber reference, and task write-back caliber.

[0016] Furthermore, the process of performing spatial index clustering and generating evidence association tables includes:

[0017] Spatial index clustering processing includes geometrical validity verification of the spatial geometry in the evidence record set, generating spatial index codes according to a preset spatial index granularity, bucketing the evidence records according to the spatial index codes to generate bucketed evidence clusters, and performing cluster merging within each bucketed evidence cluster according to the spatial geometric overlap relationship to generate a candidate resource unit set; and performing evidence association table generation processing, which includes identifying the evidence digest hash list, source identifier list and collection timestamp list associated with each candidate resource unit within its cluster, and writing the overlap rate, nearest boundary distance and intersection area ratio as association metric fields.

[0018] Furthermore, the process of generating a candidate resource unit set and alignment mapping table, and constructing a constraint-consistent decision graph includes:

[0019] The constraint consistency decision graph construction process includes generating a graph structure with candidate resource unit identifiers. The graph structure includes candidate boundary nodes, candidate classification nodes, and candidate attribute nodes. Node weights are generated based on confidence scores and collection timestamps. Topological constraints, classification mapping rules, statistical closure constraints, and quality gating criteria in the consistency constraint package are mapped as constraint edges. Weighted constraint solving is performed to generate a fusion decision result.

[0020] Furthermore, the process of generating and extracting the target version includes:

[0021] The target version generation process includes retrieving the baseline version number from the version repository, performing version context verification (including baseline version number existence verification, regional range consistency verification, and version number policy consistency verification), and performing versioned assembly processing (including boundary regularization and topology repair by boundary assembly units, code standardization by classification assembly units, unit caliber merging and default value filling by attribute assembly units, and version header information generation by metadata assembly units), thus generating the target version and target version number. The process also includes change extraction processing (including reading the resource unit identifier set, spatial boundary, classification, and attribute field sets from the target version and baseline version, performing resource unit alignment and matching processing, generating a list of additions, deletions, and modifications, and generating boundary change summaries and attribute change summaries to obtain a change differential package). This change differential package includes the baseline version number, target version number, addition list, deletion list, modification list, boundary change summary set, attribute change summary set, alignment matching log reference, and status field.

[0022] Furthermore, the process of writing audit traceability records and updating credibility parameter groups includes:

[0023] The audit traceability record writing process includes, with resource unit identifier as the main thread, aggregating the differential summary in the change differential package and the process log to form an audit link object, which is then formed into an audit traceability record and written to the audit database; and performing a credibility parameter group update process, which includes extracting the differential type distribution, supplementary sampling trigger count summary, conflict interpretation code distribution, evidence source stability summary, and quality gate pass status summary based on the change differential package and the audit traceability record, and generating a credibility parameter group update package.

[0024] Furthermore, a forestry and grassland resource data fusion system includes: a data source access and standardization module, a constraint and slot arrangement module, a candidate unit generation and alignment mapping module, a conflict detection and adjudication module, a version generation and differential output module, and an audit traceability and parameter update module; the modules are connected in sequence to implement the method described in any of the above-mentioned methods.

[0025] The key innovations of this invention include:

[0026] (1) Based on the evidence slot template, the target result specifications and target version number are transformed into an executable evidence requirement skeleton. Through the registration and binding of classification mapping rules, topological constraints, statistical closure constraints and temporal consistency constraints, the unified organization of the data source list, evidence record set and candidate resource unit set is achieved.

[0027] (2) Based on the evidence record set and evidence slot template, the spatial index clustering, evidence association table and alignment mapping table are generated in conjunction, and a constraint consistency decision graph is constructed in the same link. The classification mapping rules, topological constraints, statistical closure constraints and temporal consistency constraints are incorporated into the structured generation process of the fusion adjudication results.

[0028] (3) Based on the integrated adjudication results, the target version generation, change extraction, change differential package generation and audit traceability record writing are integrated, and a versioned traceable link is formed by combining the credibility parameter group update, so that the evidence record set, evidence slot template and alignment mapping table maintain a consistent reference and write-back relationship in the version evolution.

[0029] The following are its main beneficial effects:

[0030] (1) In response to the common problems in the background technology, such as the difficulty in unifying the field mapping table and coordinate reference parameters, and the incomplete carrying of statistical closure constraints and temporal consistency constraints in the fusion process, the evidence slot template is used to structurally carry the target result specifications and target version number. This enables the data source list to form a set of evidence records that can be uniformly called after the geometric and attribute standardization and quality label set are generated. In the subsequent generation of candidate resource unit set and alignment mapping table, the consistency with classification mapping rules, topological constraints, statistical closure constraints and temporal consistency constraints is maintained, reducing the risk of breakage in cross-source evidence organization and constraint citation.

[0031] (2) In response to the common problems in the background technology that the alignment mapping table is difficult to unify the classification mapping rules and topological constraints, and the lack of structured standards for conflict detection leads to insufficient stability of the fusion adjudication results, the evidence association table and the alignment mapping table are used to establish a traceable association between the evidence record set and the candidate resource unit set. The constraint consistency decision graph carries the linkage judgment of classification mapping rules, topological constraints, statistical closure constraints and temporal consistency constraints in the same link, so that the fusion adjudication results have a consistent adjudication basis and a verifiable generation path in scenarios with multiple constraints.

[0032] (3) In view of the common problems in the background technology, such as the lack of close connection between target version generation, change extraction and audit traceability record writing, and the lack of a unified data link foundation for the update of the credibility parameter group, by integrating the fusion adjudication result with the target version generation, change differential package generation, audit traceability record writing and credibility parameter group update, the change differential package can maintain a consistent reference relationship with the evidence slot template, alignment mapping table and evidence record set in the version evolution, which is convenient for tracing the source of change and constraint caliber in subsequent version iterations and maintaining the continuous update of the credibility parameter group. Attached Figure Description

[0033] Figure 1A flowchart illustrating a forestry and grassland resource data fusion method provided in this application embodiment;

[0034] Figure 2 This is a structural block diagram of a forestry and grassland resource data fusion system provided in an embodiment of this application. Detailed Implementation

[0035] Example 1: Refer to Figure 1 This is a flowchart illustrating a forestry and grassland resource data fusion method provided in an embodiment of the present invention. The process may include at least steps S100-S400:

[0036] S100. Obtain the data source list, extract the field mapping table and coordinate baseline parameters, perform geometric and attribute standardization and generate a quality label set to obtain the evidence record set.

[0037] S200: Obtain the target result specifications and target version number, extract classification mapping rules and topological constraints, register and statistically process closure constraints and temporal consistency constraints, and generate evidence slot templates;

[0038] S300: Based on the evidence record set and evidence slot template, perform spatial index clustering and evidence association table generation to generate a candidate resource unit set and alignment mapping table, and construct a constraint consistency decision graph to obtain the fusion adjudication result.

[0039] S400. Based on the fusion adjudication result, perform target version generation and change extraction processing to generate a change differential package, and perform audit traceability record writing and credibility parameter group update processing to obtain audit traceability records and credibility parameter groups.

[0040] Step S100 includes at least steps S110-S130:

[0041] S110. Obtain the data source list and register the source identifier and collection timestamp to obtain the data source list;

[0042] Specifically, when the fusion task is initiated, the evidence access module receives the target area boundary, task batch number, and data access credentials issued by the task management module, and arranges the access to multiple forestry and grassland related data sources under the constraints of the target area boundary. The data source list is a structured set of data sources required for this fusion task, including at least source identifier, data type, coverage, storage location, collection timestamp, and access method marker. The source identifier is used to distinguish different data sources such as land survey results, forestry and grassland resource survey results, remote sensing classification products, field verification records, and asset ownership ledgers. The collection timestamp is used to characterize the collection time caliber of the corresponding data source and participates in the timeliness determination of subsequent credibility parameter groups. Further, the evidence access module can consist of a data source discovery submodule, an access adaptation submodule, and a registration and archiving submodule. The data source discovery submodule retrieves available data based on a preset data source directory table and regional index. The access adaptation submodule performs file reading, database query, or interface retrieval for different access methods. The registration and archiving submodule writes the metadata of each retrieved source into the task context, forming a data source list that can be invoked by subsequent steps. Understandably, the data source list does not directly carry the original spatial data, but rather the location and caliber information of the original data, so that subsequent field mapping tables, coordinate reference parameters, and geometric and attribute standardization processing have a unified source entry point.

[0043] In a feasible engineering implementation, the provincial forestry and grassland authorities organize quarterly updates to forestry and grassland resource results. The system automatically triggers a fusion task during the nighttime window. The evidence access module first retrieves the latest land survey results slices, database partitions of the same batch of forestry and grassland resource survey results, remote sensing classification product image rasters and their classification layers, sampling points and verification tables from field verification records, and ownership unit records from asset ownership ledgers within the target area boundary. For each source, the evidence access module writes a source identifier upon access and parses the collection timestamp from the data source's own metadata or interface-returned fields. When the collection timestamp is missing or its format does not conform to the preset time standard, the registration and archiving submodule marks the source as pending completion and archives the reason for the missing information along with the access log. Subsequently, the source entry is still retained in the data source list for subsequent steps to handle missing information and generate quality labels. Furthermore, the evidence access module performs version filtering for scenarios where multiple versions of data exist within the boundary of the same target area from the same source. The filtering criteria include at least the consistency verification of the release timestamp and the version identifier. When version identifiers conflict, the system writes the conflict situation to the task log and generates a manual review prompt. However, the data source list still retains the location information of the candidate versions according to the priority strategy, thereby providing traceable input for the subsequent generation of evidence record sets.

[0044] At the end of S110, the evidence access module records the data source list as the output field name of this step in the task context, and uses the data source list as input for the "data source list" to be called in S120; at the same time, the data source list is associated with S210 in the cross-main step connection so as to establish source constraint relationship when generating target result specification and target version number, and serves as the enumeration basis of source list when slot arrangement is performed in S230.

[0045] S120. Extract the field mapping table and coordinate reference parameters from the data source list and register the encoding dictionary caliber and projection caliber, and generate the field mapping table and coordinate reference parameters.

[0046] Specifically, after receiving the data source list output by S110, the standardization configuration module loads a field mapping table and coordinate reference parameters matching the source identifier for each source entry in the list. The field mapping table is a structured definition of the mapping relationship between the original fields and the target output standardized fields, including at least field name correspondence, unit scope, coding dictionary scope, and null value scope. The coding dictionary scope describes the coding meaning and correspondence of classification fields and attribute fields under different sources, and the null value scope describes the unified expression of missing and invalid fields and is referenced in subsequent geometric and attribute standardization processing. The coordinate reference parameters are a structured definition of the projection scope, datum scope, and coordinate axis order scope used for spatial geometric data. The projection scope is used to unify the projection reference of vector and raster data, the datum scope is used to provide a consistent interpretation of the elevation or geographic coordinate datum, and the coordinate axis order scope is used to constrain the consistency of the axis order of latitude, longitude, or planar coordinates in data exchange.

[0047] Furthermore, the standardized configuration module can be composed of a mapping loading unit, a caliber verification unit, and a registration output unit. The mapping loading unit reads a field mapping table matching the source identifier and data type from the configuration repository or result specification library, and performs consistency verification on the field name matching relationship during the reading process. The verification content includes at least whether the target field exists, whether the field type matches, and whether the unit caliber is an allowed caliber. When the verification fails, the caliber verification unit writes the failed item to the configuration exception log and marks the corresponding field mapping table entry as pending correction. The mapping loading unit also reads the coordinate reference parameters matching the source identifier and performs legality verification on the projection caliber and coordinate axis order caliber. When it is found that the projection caliber declared in the source data is inconsistent with the projection caliber in the configuration repository, the caliber verification unit records the conflict and attaches the conflict source identifier, conflict timestamp, and conflict field path, for S130 to select the declared caliber or the configured caliber during geometric standardization. Understandably, the registration output unit registers the field mapping table and coordinate reference parameters that have been loaded and verified, establishes a binding relationship between them and the source identifier, and writes them into the task context, so that the subsequent processing link can automatically select the corresponding field mapping table and coordinate reference parameters based on the source identifier when reading the original data.

[0048] During the project operation, when the remote sensing classification product source uses raster classification coding, the forestry and grassland resource survey results source uses vector surface features and includes sub-compartment attribute fields, and the land survey results source uses land category coding fields, the field mapping table provides at least a coding dictionary caliber registration for the classification fields, ensuring that the standardized fields of the same target result can be consistently interpreted from different source fields. When the asset ownership ledger source does not include spatial geometry but includes ownership unit numbers, the field mapping table further registers the correspondence between ownership unit numbers and spatial geometric associated fields, for S130 to perform correlation completion during the attribute standardization stage. For coordinate reference parameters, when the sampling points of the field verification record source are expressed in geographic coordinates while the forestry and grassland resource survey results are expressed in planar projection, the coordinate reference parameters provide a unified interpretation entry point in terms of projection caliber and coordinate axis order caliber, and S130 subsequently performs coordinate transformation and axis order correction accordingly.

[0049] At the end of S120, the standardization configuration module writes the "field mapping table" and "coordinate reference parameters" as the output field names of this step into the task context, and uses the field mapping table and the coordinate reference parameters as inputs for the "field mapping table, coordinate reference parameters" call in S130; at the same time, the field mapping table and the coordinate reference parameters serve as the field caliber reference for generating the evidence slot template in S200 during the cross-main step connection, and provide a pre-configuration basis for geometric consistency before the spatial index clustering processing in the subsequent S310.

[0050] S130. Perform geometric and attribute standardization on the data source list according to the field mapping table and coordinate baseline parameters, and generate a quality label set according to the confidence parameter group to obtain the evidence record set.

[0051] Specifically, the evidence standardization module receives the data source list output by S110 and the field mapping table and coordinate reference parameters output by S120. It then retrieves or reads the original data entities one by one according to the data source list and performs geometric standardization and attribute standardization processing on each source data. The geometric standardization processing includes at least geometric type consistency, coordinate reference unification, and geometric validity verification: geometric type consistency is used to convert or derive surface features, line features, and point features according to the geometric expression of the target output specification. For example, when the field verification record is a point feature, it generates a geometric mark associated with the verification range; coordinate reference unification performs projection transformation, reference surface interpretation, and coordinate axis order correction according to the coordinate reference parameters, and records the input projection caliber, output projection caliber, and transformation timestamp in the processing log; geometric validity verification checks for self-intersections, multiple components, void anomalies, non-closed boundaries, etc., and generates geometric quality marks on the check results as part of the subsequent quality label set. The attribute standardization process includes at least field extraction, field mapping, code dictionary conversion, and null value standardization: Field extraction extracts corresponding fields from the original records based on the field mapping table; field mapping writes the corresponding fields into the target field set according to the field name correspondence; code dictionary conversion converts the source code into the target result standard code according to the code dictionary standard; and null value standardization writes missing, invalid, and unknown states into a unified mark according to the null value standard, so that the subsequent evidence slot template has a consistent missing expression when checking slot gaps.

[0052] Furthermore, after completing geometric and attribute standardization, the evidence standardization module calls the credibility calculation unit to load the credibility parameter set and generates a quality label set accordingly. The credibility parameter set is a structured configuration of the quality label item list and weighted caliber, including at least timeliness label generation rules, spatial accuracy label generation rules, completeness rate label generation rules, and verification consistency label generation rules. Specifically, the timeliness label generation rules generate grading marks based on the time difference between the collection timestamp and the task timestamp; the spatial accuracy label generation rules generate grading marks based on the resolution, measurement accuracy, or geometric transformation residual declared by the data source; the completeness rate label generation rules generate grading marks based on the missing key fields and the geometric legality verification results; and the verification consistency label generation rules generate grading marks based on the consistency comparison results between field verification records and historical records from the same source. Understandably, the credibility parameter group is one of the essential minimum sets for realizing the core improvement of this invention. Together with the field mapping table and the coordinate reference parameters, it constitutes the three key configurations for generating the evidence record set. On this basis, optional extended functions include using different quality label granularities for different data source types, or introducing multiple rounds of sampling review records for verifying consistency labels. However, these extended functions do not change the field composition of the evidence record set.

[0053] During operation, the evidence standardization module generates an evidence record for each standardized record and writes it into the evidence record set. The evidence record set is a structured set output from this step, containing at least a source identifier, spatial geometry, attribute field set, acquisition timestamp, processing timestamp, quality tag set, confidence score, and evidence digest hash. The processing timestamp is automatically generated and written by the system upon completion of standardization, distinguishing between the acquisition and processing timestamps. The evidence digest hash is obtained by the evidence digest generation unit through digest calculation of the source identifier, spatial geometry digest, and attribute field set digest, and is used together with the source identifier for constructing the subsequent evidence citation list. For the confidence score, the confidence calculation unit calculates the quality tag set based on the weighted caliber of the confidence parameter group and writes it into the evidence record. When a quality tag set of an evidence record contains missing or conflicting items, the confidence calculation unit marks the evidence record as low-confidence and writes it into the anomaly log, while still including the evidence record in the evidence record set. This ensures that the subsequent conflict detection processing in S330 can cover low-confidence inputs and generate a conflict event set. Furthermore, when writing the evidence record set, the evidence standardization module performs deduplication registration on duplicate records with the same source identifier and the same spatial index unit. The deduplication registration retains the evidence digest hash and deduplication reason code of the deduplicated record. The deduplication reason code can be referenced in the subsequent conflict interpretation code generation.

[0054] In engineering implementation examples, when there are subtle differences in the boundary representation of the same spatial area between land survey results and forestry and grassland resource survey results, geometric standardization processing uniformly projects the data based on coordinate reference parameters and records the transformation residuals, and the quality label set generates spatial accuracy labels accordingly. When remote sensing classification products provide raster codes for classification fields and there are empty areas, attribute standardization processing converts the classification codes according to the coding dictionary caliber and marks the missing status corresponding to the empty values ​​according to the null value caliber, and the quality label set generates completeness labels accordingly. When field verification records provide verification points and can be compared with the small plot attributes of forestry and grassland resource survey results for consistency, the verification consistency label generation rules write the comparison conclusions into the quality label set and participate in the writing of the credibility score. After the evidence record set is generated, the system uses it as the factual basis for generating the evidence slot template in subsequent step S200. During the cross-main step transition, the evidence record set is used by the "evidence record set" in step S310 for spatial index clustering. Simultaneously, the data source list continues to be used as a source constraint for step S210 association, forming a configuration link from S110 to S210 and then to S230, providing traceable input for conflict resolution and version differentiation in steps S300 and S400. Furthermore, the credibility parameter set may be updated in step S430 and then flow back to step S130 as input, forming an iterative link for credibility configuration. However, in this step, the credibility parameter set loaded at task startup is still used as the minimum required set for generating the quality tag set.

[0055] In summary, the technical effects of this step are as follows: By unifying multi-source forestry and grassland related data sources into a set of evidence records containing quality label sets, credibility scores, and evidence summary hashes, the conflict detection and adjudication steps obtain traceable evidence input and consistent field definitions; at the same time, geometric and attribute standardization processing solidifies coordinate benchmark parameters and field mapping tables into the evidence generation chain, so that the subsequent operation of evidence slot templates and constraint consistency decision diagrams has a stable source of pre-constraints.

[0056] Step S200 includes at least steps S210-S230:

[0057] S210. Obtain the target deliverable specification and target version number and associate them with the data source list to obtain the target deliverable specification and target version number;

[0058] Specifically, after receiving the data source list output by S110, the slot template module first reads the target result specification that matches the batch number of the fusion task from the result specification library, and then reads the target version number bound to the target result specification from the version governance library. The target result specification is a set of structured constraints followed by this forestry and grassland resource result release, which includes at least three types of scope items: classification system scope, field scope, and statistical scope. The classification system scope describes the classification level and coding scope of forest land, grassland and related land types in the result. The field scope describes the field name, field type, allowed value range, unit scope and null value scope of each field in the result library. The statistical scope describes the summary scope and rounding scope of area summary, hierarchical summary and cross-level verification. Furthermore, the slot template module associates the target output specification with the data source list through the source association unit. This association registration includes at least the mapping relationship between the source identifier and the fields that can be covered in the target output specification, the coverage relationship between the source identifier and the classification levels that can be covered in the classification system, and the participation relationship between the source identifier and the statistical levels that can participate in the aggregation in the statistical system. During the registration process, the source association unit performs availability verification on each source identifier. The verification includes at least whether the coverage area intersects with the target area boundary, whether the collection timestamp is resolvable, and whether the storage location is accessible. When a source identifier is found to have access failure or a missing collection timestamp, the source identifier is recorded in the task exception log and maintained in the data source list. Subsequent steps, during the consistency constraint package generation and slot orchestration stages, set differentiated gating rules for this abnormal source identifier. Understandably, the target version number is used to identify the output version marker and release timestamp corresponding to this output release. After reading the target version number, the slot template module binds it to the fusion task batch number and writes it into the task context, so that when S410 obtains the baseline version number and generates the target version, it has a version alignment entry point. After the slot template module completes the acquisition of the target result specification and the target version number and completes the association registration with the data source list, it writes the "target result specification" and "target version number" as output field names into the task context, and uses the target result specification as input for the "target result specification" of S220 to call. At the same time, it uses the target version number as a cross-main step connection field for the subsequent version generation link of S400 to reference.

[0059] In a feasible engineering embodiment, the system is deployed on the provincial forestry and grassland "one map" results update platform. The task management module triggers fusion tasks at fixed quarterly windows, and the slot template module reads the currently released specifications. Its classification system is aligned with the provincial forestry and grassland classification code and is compatible with the land survey land category code mapping. Its field scope covers the unified naming and allowed value range of fields such as map patch identifier, ownership, forest and grass species, age group, canopy density, and management measures. Its statistical scope covers the area summary and cross-level verification scope at the county, township, and forest compartment levels. The target version number is read in the same batch and bound to the task batch number. Then, the target result specification is associated with the data source list and registered, so that subsequent constraint extraction and slot arrangement can explicitly determine the data coverage capability of different sources and form auditable records.

[0060] S220. Extract classification mapping rules and topological constraints from the target outcome specifications, register and statistically analyze closure constraints and temporal consistency constraints, and generate a consistency constraint package;

[0061] Specifically, the consistency constraint package generation unit receives the target result specification output by S210, first performs rule expansion on the classification system caliber to obtain classification mapping rules; the classification mapping rules are a set of structured rules for converting classification codes from different sources to target classification codes, including at least the mapping rules from land survey result land type codes to target classification codes, the mapping rules from remote sensing classification product classification codes to target classification codes, and the mapping rules from forest and grassland resource survey result forest and grassland species codes to target classification codes; wherein, each mapping rule includes at least the source identifier, source classification field path, target classification field path, mapping conditions, conflict priority marker, and missing backfill marker. The conflict priority marker is used to form the constraint weight caliber in the subsequent constraint consistency decision map construction stage, and the missing backfill marker is used to form the supplementary collection trigger caliber in the slot gap supplementary collection rule generation stage. Furthermore, the consistency constraint package generation unit extracts and registers topological constraints from the target outcome specification. The topological constraints include at least boundary closure constraints, adjacent unit non-overlapping constraints, hole handling constraints, cross-administrative boundary trimming constraints, and protection red line boundary consistency constraints. When registering topological constraints, the consistency constraint package generation unit simultaneously registers the scope of application of the constraints and the constraint conflict handling criteria. The scope of application of the constraints includes at least the target area boundary and spatial index granularity, and the constraint conflict handling criteria includes at least the conflict event category and the conflict interpretation code generation criteria, so that when generating the conflict event set, S330 can classify topological conflicts and generate verifiable interpretation information.

[0062] Furthermore, the consistency constraint package generation unit registers closed-loop constraints for statistical definitions. These closed-loop constraints are a structured set of constraints for area aggregation and cross-level verification across multiple statistical levels in the results database. They include at least the area aggregation definition within the same statistical level, the aggregation consistency definition between different statistical levels, and the consistency definition between the total area of ​​different categories and the total area. When a source data lacks a key field for statistical analysis or exhibits spatial geometric anomalies, these closed-loop constraints can trigger the generation of statistical closure conflicts during subsequent conflict detection and processing. Additionally, the consistency constraint package generation unit registers temporal consistency constraints. These temporal consistency constraints include at least version inheritance relationship definitions, change legality definitions, and publication timestamp consistency definitions. Version inheritance relationship definitions describe the inheritance marker generation rules between candidate resource units and resource units in historical results versions. Change legality definitions describe the acceptable range of boundary changes and attribute changes. Publication timestamp consistency definitions describe the verification criteria between the collection timestamp, processing timestamp, and target version number publication timestamp. Understandably, the consistency constraint package generation unit verifies the completeness of the rules before generating the consistency constraint package. When there are gaps in the classification mapping rules or unregistered items in the topology constraints, the gap items are written into the rule gap log and the gap mark is retained and entered into the consistency constraint package. Subsequently, in the slot orchestration stage, S230 generates differentiated gap supplementation rules or manual review triggering conditions based on the gap mark.

[0063] After the consistency constraint package is generated, the "consistency constraint package" is written as an output field name into the task context, and the consistency constraint package is used as input for S230 to call the "consistency constraint package". At the same time, the consistency constraint package is used by S330 to build the constraint consistency decision graph in the cross-main step connection, so that conflict detection, constraint registration and adjudication solution share the same constraint caliber and have audit consistency.

[0064] S230. Arrange slots for the data source list and consistency constraint package, register the minimum quality threshold and gap sampling rules, and generate evidence slot templates.

[0065] Specifically, the slot orchestration engine receives the data source list output by S110 and the consistency constraint package output by S220, and combines it with the target result specification output by S210 to orchestrate the evidence slots required for resource unit generation, forming an evidence slot template. The evidence slot template is a structured definition of the set of evidence slots that each candidate resource unit needs to fill before fusion adjudication. It includes at least slot number, slot type, source list, field mapping table, minimum quality threshold, gap acquisition rules, and conflict interpretation code dictionary. Among them, the slot number is used to uniquely identify the slot and support subsequent acquisition task packages to be located by slot. The slot type includes at least boundary evidence slots, classification evidence slots, attribute evidence slots, and quality threshold slots. The source list is generated from the set of source identifiers in the data source list that meet the coverage and access status conditions. The field mapping table references the field paths that match the slot type in the field mapping table generated and registered by S120. The minimum quality threshold is the gate control caliber of the confidence score and is bound to the slot type. The gap acquisition rules describe the acquisition source suggestions, acquisition range caliber, and trigger condition caliber when the slot is missing, the slot has low confidence, or the slot conflicts. The conflict interpretation code dictionary describes the mapping caliber between the conflict event category and the interpretation code.

[0066] Furthermore, the slot orchestration engine first filters source coverage capabilities during the orchestration process, and then registers the minimum quality threshold for the slot. Specifically, the slot orchestration engine establishes a source candidate set for each slot type. The source candidate set is generated based on at least the collection timestamp, data type, and field coverage capability of the source identifier. When a source identifier has a rule gap log marker or an access anomaly log marker, the slot orchestration engine still includes the source identifier in the source candidate set, but registers a stricter credibility score gate for it in the minimum quality threshold registration, and registers the manual review trigger condition in the gap supplementation rule. The minimum quality threshold, as one of the minimum set parameters of the core improvement of this invention, should at least be bound to the slot type and consistent with the credibility score of the credibility parameter group. On this basis, preferably, the minimum quality threshold can be bound to the source identifier subdivision to form a source-differentiated gate, but this preferred extension does not change the basic field composition of the evidence slot template.

[0067] Furthermore, the slot orchestration engine generates and registers gap acquisition rules. These gap acquisition rules include at least triggering conditions, suggested acquisition sources, acquisition scope definitions, field references, and task write-back definitions. The triggering conditions include at least a slot missing marker, a confidence score below the minimum quality threshold marker, a classification mapping rule gap marker, and a topological constraint conflict marker. Suggested acquisition sources are generated from the data source list according to priority and may include field verification records or remote sensing classification products as acquisition sources. The acquisition scope definition includes at least the target area boundary, the candidate resource unit boundary involved in the conflict event, and the buffer range definition. The field references point to the field definitions of the target outcome specification and reference the field paths in the field mapping table. The task write-back definition describes the source identifier and acquisition timestamp registration method for the evidence record set written back after the acquisition task is completed. Understandably, these gap acquisition rules form the pre-control information for the subsequent S300 stage evidence slot template-driven candidate resource unit generation and conflict adjudication, and form an auditable rule basis link when the confidence parameter group update package is triggered for write-back in S430.

[0068] In the engineering implementation, the system generates evidence slot templates for county-level forestry and grassland resource integration tasks. The source list for boundary evidence slots includes land survey results and forestry and grassland resource survey results; the source list for classification evidence slots includes remote sensing classification products and forestry and grassland resource survey results; the source list for attribute evidence slots includes forestry and grassland resource survey results and asset ownership ledgers; and the source list for quality threshold slots includes field verification records. When field verification records are missing or collection timestamps are missing, the slot arrangement engine registers the field verification supplementary collection trigger conditions and the scope of supplementary collection in the gap supplementary collection rules. The subsequent task management module generates supplementary collection task packages based on this and writes back the evidence record set after the supplementary collection is completed, forming an automated iteration link within the same task batch. Furthermore, the conflict interpretation code dictionary of the evidence slot template maintains consistency with the conflict event categories of the consistency constraint package, enabling S330 to output conflict interpretation codes and associate them with the evidence citation list when generating the fusion adjudication results.

[0069] After the slot arrangement is completed, the "evidence slot template" is written into the task context as an output field name, and the evidence slot template is used as input for S310 to call the "evidence slot template". At the same time, the evidence slot template is used by S330 for conflict detection processing and constraint consistency decision graph construction in the cross-main step connection, and provides slot positioning caliber for evidence reference tracing when S400 generates change differential packages.

[0070] In summary, the technical effects of this step are as follows: By extracting the target outcome specifications into a consistency constraint package and jointly compiling it with the data source list to generate evidence slot templates, the subsequent fusion link obtains auditable slot definitions, minimum quality thresholds, and gap filling rules; at the same time, classification mapping rules, topological constraints, statistical closure constraints, and temporal consistency constraints form a unified standard in the consistency constraint package, enabling the conflict detection and adjudication stage to have a stable constraint entry point and supporting the traceability record of the version governance link.

[0071] Step S300 includes at least steps S310-S330:

[0072] S310. Obtain the evidence record set and evidence slot template, perform spatial index clustering processing and generate an evidence association table to obtain the candidate resource unit set;

[0073] Specifically, the candidate unit module reads the evidence records to be processed from the evidence record set output and registered in the preceding S130, and simultaneously reads the slot number, slot type, source list, field mapping table, minimum quality threshold, and gap collection rules from the evidence slot template output and registered in the preceding S230, as upstream inputs for this step; wherein, each evidence record in the evidence record set contains at least a source identifier, spatial geometry, attribute field set, collection timestamp, quality tag set, credibility score, and evidence digest hash, and the evidence slot template is used to specify the evidence slot set and gating criteria required by the candidate resource unit in the subsequent conflict detection and adjudication stage. Furthermore, the candidate unit module performs spatial index clustering processing on the evidence record set. This spatial index clustering processing is completed collaboratively by the spatial index generation unit and the clustering merging unit. The spatial index generation unit first reads the spatial geometry of the evidence records and performs geometric legality verification. The geometric legality verification includes at least loop closure verification, self-intersection verification, empty geometry verification, and coordinate range verification. For evidence records with abnormalities, the abnormality code is written into the task abnormality log and retained for subsequent processing. After the verification is completed, the spatial index generation unit generates spatial index encoding for the spatial geometry according to a preset spatial index granularity and binds and registers the spatial index encoding with the evidence digest hash. The preset spatial index granularity is one of the smallest sets of core parameters in this step. The preset spatial index granularity is registered in the same source as the target region boundary granularity in the task context, so that different batches of fusion tasks have reproducible slicing calibers in the candidate resource unit set generation stage. Subsequently, the clustering and merging unit performs initial bucketing of the evidence records according to the spatial index encoding, generating bucketed evidence clusters, and performs clustering and merging within each bucketed evidence cluster according to the spatial geometric overlap relationship; the overlap relationship is calculated by the overlap calculation unit, which calculates the overlap rate, intersection area ratio and nearest distance to the boundary for any two evidence records in spatial geometry. The calculation caliber of the overlap rate, intersection area ratio and nearest distance to the boundary is consistent with the coordinate reference parameters, and the calculation results are written into the clustering intermediate log for reuse in subsequent alignment mapping registration processing. Understandably, the clustering and merging unit introduces slot constraint gating when performing clustering and merging. The slot constraint gating reads the slot type and source list in the evidence slot template and adopts a differentiated merging strategy for evidence records with different source identifiers. When the source identifier of an evidence record belongs to the source list of the boundary evidence slot, the clustering and merging unit prioritizes keeping the spatial geometry of the evidence record as a candidate boundary candidate and marks the spatial geometry of other source identifiers in the same cluster as candidate alignment objects. When the source identifier of an evidence record only appears in the source list of the attribute evidence slot, the clustering and merging unit allows it to enter the same candidate resource unit when the spatial geometric overlap rate is low but the nearest distance of the boundary meets the gating criteria, so as to avoid the loss of attribute evidence due to geometric differences.After each clustering is completed, a candidate resource unit identifier is generated and a candidate resource unit set is formed. The candidate resource unit identifier contains a spatial index encoding fragment and a task batch fragment and is uniquely registered in the candidate resource unit set.

[0074] While generating the candidate resource unit set, the candidate unit module generates an evidence association table, which is generated by the association construction unit. The association construction unit identifies the evidence digest hash list, source identifier list, and collection timestamp list associated with each candidate resource unit within its cluster, and writes the overlap rate, nearest boundary distance, and intersection area ratio as association metric fields into the evidence association table. The evidence association table also records the slot attribution mark of the evidence record in the evidence slot template. The slot attribution mark is determined by the slot matching unit based on the source identifier and field mapping table. In the determination process, multiple field paths with the same source identifier are selected using the field caliber priority caliber and the selection code is recorded. After completing spatial index clustering and association construction, the candidate unit module writes the "candidate resource unit set" and "evidence association table" into the task context as output field names, and uses the candidate resource unit set and the evidence association table as inputs for S320 to call. Simultaneously, the candidate resource unit set serves as the resource unit alignment entry point for S410 to generate the target version during cross-main step connections, and the evidence association table provides evidence reference location information for S430 audit traceability record writing during cross-main step connections. In one engineering embodiment, the system triggers this step in the provincial fusion task using nighttime batch processing. After detecting that the evidence record set has been entered into the database and the evidence slot template has been published, the task management module automatically triggers the candidate unit module to perform spatial index clustering. When the size of a bucketed evidence cluster exceeds the task limit, the candidate unit module triggers a task slicing strategy, splitting the excess buckets into spatial index encoded sub-fragments and serializing the clustering and merging. Simultaneously, the splitting record is written to the task exception log, enabling subsequent alignment mapping registration and adjudication processing to be replayed using the same slicing caliber.

[0075] S320. Extract geometric alignment indicators and semantic alignment indicators from the candidate resource unit set and the evidence association table, perform alignment mapping registration processing, and generate an alignment mapping table.

[0076] Specifically, the alignment mapping registration unit reads the candidate resource unit identifiers and candidate spatial geometry sets from the candidate resource unit set output by S310. Simultaneously, it reads the evidence digest hash list, source identifier list, association metric field, and slot attribution flag associated with each candidate resource unit identifier from the evidence association table output by S310, serving as the input source for this step. Further, the alignment mapping registration unit extracts geometric alignment indicators for each candidate resource unit identifier. These geometric alignment indicators are generated by the geometric alignment calculation subunit. Within the same candidate resource unit, the geometric alignment calculation subunit reads back the evidence record set according to the evidence digest hash in the evidence association table to obtain the spatial geometry of the corresponding evidence record, and calculates the geometric alignment indicator for any two spatial geometries. The geometric alignment indicator includes at least the overlap rate, the closest boundary distance, the boundary offset metric, and the topological relationship flag. The boundary offset metric is calculated by the boundary sampling subunit after equidistant sampling of the two boundaries, calculating the point-to-line distance distribution and registering the quantile. The topological relationship flag is determined by the topological discrimination subunit, which determines inclusion, intersection, disjointness, or adjacent relationships and writes them into the alignment mapping table. Understandably, the geometric alignment computation subunit downgrades evidence records with geometrically valid abnormal codes. The downgrade process marks the spatial geometry of these records as unsuitable for candidate boundary generation, but still allows them to participate in the extraction of semantic alignment indicators for attribute fields. The downgraded marker is then written into the quality gating field of the alignment mapping table for use in subsequent conflict detection processing.

[0077] Further, the alignment mapping registration unit extracts semantic alignment indicators for each candidate resource unit identifier. These indicators are generated by the semantic alignment calculation subunit. The semantic alignment calculation subunit reads back the attribute field sets of the evidence records from the evidence record set and merges them into slot-based categories according to the slot attribution markers in the evidence association table, resulting in a slot candidate attribute set. This slot candidate attribute set includes at least a classification candidate attribute set and a non-classification candidate attribute set. The classification candidate attribute set is the set of field paths matching the classification evidence slots, and the non-classification candidate attribute set is the set of field paths matching the attribute evidence slots. Subsequently, the semantic alignment calculation subunit extracts semantic alignment indicators for candidate attribute values ​​from different source identifiers within the same slot. These indicators include at least a coding consistency marker, a field caliber consistency marker, and a conflict item count. The coding consistency marker registers whether the classification codes of different source identifiers are the same. The field caliber consistency marker registers whether the field unit, allowed value range, and null value caliber are consistent. The conflict item count records the number of fields with differing candidate attribute values ​​within the same slot. The semantic alignment calculation subunit, while extracting semantic alignment indicators, simultaneously registers the collection timestamp sequence and credibility score sequence of evidence records and writes them into the time and credibility fields of the alignment mapping table. This allows the subsequent constraint consistency decision graph construction stage to reference the same alignment mapping registration caliber to complete node weight generation. After extracting geometric alignment indicators and semantic alignment indicators, the alignment mapping registration unit generates an alignment mapping table. The alignment mapping table includes at least candidate resource unit identifiers, evidence digest hash pairs, source identifier pairs, geometric alignment indicators, semantic alignment indicators, quality gating fields, collection timestamp sequences, and credibility score sequences. The alignment mapping table is written into the task context as an output field name and is used as input for S330's "alignment mapping table" call. Simultaneously, the alignment mapping table is referenced by S420 when generating boundary change summaries and attribute change summaries during cross-main step connections. It is used to perform calibrated calculations of geometric and attribute differences before and after changes and write them into the change difference package.

[0078] In one engineering embodiment, the system spatially calculates the overlap rate and boundary offset metric of land survey results and forestry and grassland resource survey results within the same candidate resource unit, and calculates the coding consistency mark for the classification fields of remote sensing classification products and forestry and grassland resource survey results. When the geometric alignment index shows that the boundary offset metric exceeds the gate control caliber and the semantic alignment index shows that there is a conflict in the classification coding, the alignment mapping registration unit marks the candidate resource unit as a high-conflict candidate and writes it into the task anomaly log. The subsequent S330 conflict detection processing reads the mark and enters the priority adjudication path. When the semantic alignment index shows that there is an inconsistency in the field caliber, the alignment mapping registration unit writes the field caliber inconsistency code into the alignment mapping table and records the trigger source identifier at the same time, so that the field mapping table registered in the slot arrangement stage has a correction entry point in the subsequent adjudication stage.

[0079] S330. Perform conflict detection and processing on the evidence slot template, alignment mapping table and consistency constraint package, construct the constraint consistency decision graph, and generate the fusion adjudication result;

[0080] Specifically, the decision graph adjudication module receives the evidence slot template output by S230, the alignment mapping table output by S320, and the consistency constraint package output by S220, and reads back the candidate resource unit set and the evidence association table from the task context as input sources for conflict detection processing and constraint consistency decision graph construction. Further, the conflict detection unit traverses the alignment mapping table records according to the candidate resource unit identifier and performs slot-level conflict detection in conjunction with the slot number, slot type, and minimum quality threshold of the evidence slot template; slot-level conflict detection includes at least the registration processing of geometric conflict detection, classification conflict detection, attribute conflict detection, quality conflict detection, and statistical closure conflict detection. Geometric conflict detection reads the overlap rate, nearest boundary distance, boundary offset metric, and topological relationship marker from the alignment mapping table, and combines this with the topological constraints in the consistency constraint package to determine whether there are any issues such as unclosed candidate boundaries, overlapping candidate units, crossing administrative boundaries, or violations of hole handling standards. When a geometric conflict is detected, the conflict detection unit writes the conflict type code, candidate resource unit identifier, conflict-related evidence digest hash pair, and triggered topological constraint marker into the conflict registration log, and maps this record to the conflict factor of the constraint consistency decision graph in the subsequent mapping stage. Classification conflict detection reads the coding consistency marker and conflict item count from the alignment mapping table, and combines this with the classification mapping rules in the consistency constraint package to determine whether there are discrepancies after mapping classification codes from different source identifiers to the target classification code. When discrepancies exist, the conflict detection unit simultaneously reads the credibility score sequence and the collection timestamp sequence, calculates the conflict priority caliber, and writes it into the conflict registration log, enabling the mapping stage to assign weights according to evidence credibility scores and timeliness calibers. Attribute conflict detection reads the field consistency marker and the set of non-classified candidate attributes, and combines it with the field mapping table in the evidence slot template to determine whether there are inconsistencies in unit scope, allowed value range, or null value scope for attribute fields in the same slot. When an attribute conflict occurs, the conflict detection unit writes the trigger field path, trigger source identifier, and slot number into the conflict registration log, and associates it with the field reference in the gap acquisition rule for subsequent generation of acquisition trigger conditions. Quality conflict detection reads the quality gate field and confidence score sequence in the alignment mapping table, and compares the confidence score with the minimum quality threshold in the evidence slot template. When it is lower than the gate scope, a quality conflict is registered, and the low-confidence evidence digest hash is written to the low-confidence queue. Statistical closure conflict detection reads the statistical closure constraints in the consistency constraint package, and reads back the summary scope of the candidate space geometry area and the classification slot candidate results from the candidate resource unit set, performs statistical summarization and cross-layer verification. When the verification does not meet the statistical closure constraints, a statistical closure conflict is registered and written to the statistical verification log.Understandably, the conflict detection unit adopts a hierarchical triggering strategy for different conflict types. When a quality conflict occurs and there is a suggestion for a source of supplementary collection in the gap collection rule, the supplementary collection triggering condition is written into the task triggering log. Subsequently, the task management module reads the triggering log to generate a supplementary collection task package and waits for it to be written back. When a geometric conflict occurs and the topology constraint marking is triggered, the candidate resource unit identifier is written into the priority adjudication queue, so that the constraint consistency decision graph construction stage is solved in the order of the queue, maintaining a controllable operation process under large-scale tasks.

[0081] Furthermore, after completing conflict detection and processing, the decision graph adjudication module constructs a constraint consistency decision graph. The constraint consistency decision graph construction unit generates a graph structure according to the candidate resource unit identifier. The candidate nodes of the graph structure are jointly determined by the evidence association table and the alignment mapping table. Each candidate node includes at least candidate boundary nodes, candidate classification nodes, and candidate attribute nodes. Candidate boundary nodes come from the spatial geometric candidates in the source list of boundary evidence slots. Candidate classification nodes come from the classification candidate attribute set after mapping the classification evidence slots. Candidate attribute nodes come from the non-classification candidate attribute set after mapping the attribute evidence slots. The node weight is generated by the weight calculation unit. The weight calculation unit reads the credibility score and collection timestamp of the evidence record and combines it with the temporal consistency constraints registered in the consistency constraint package to generate a time-effect decay caliber. Then, the time-effect decay caliber is combined with the credibility score to calculate the node weight. The minimum set parameters required for the node weight calculation include the credibility score, collection timestamp, and temporal consistency constraint marker. Other statistical verification logs and conflict priority calibers are written into the graph structure as preferred extended fields but do not change the basic composition of the node weight. The constraint edges of the graph structure are generated by the constraint mapping unit. This unit maps topological constraints, classification mapping rules, statistical closure constraints, and quality gating criteria to constraint edges and binds them to candidate node pairs. Each constraint edge is also bound to a conflict interpretation code dictionary entry, enabling the generation of conflict interpretation codes and evidence citation lists when outputting the fusion decision result. Subsequently, the solution execution unit performs weighted constraint solving on the constraint consistency decision graph. Under the premise of satisfying topological constraints, classification mapping rules, and quality gating criteria, the weighted constraint solving process selects combinations of candidate boundaries, candidate classifications, and candidate attributes from the candidate nodes, and performs statistical closure constraints as a verification step. When multiple feasible combinations exist, the solution execution unit sorts them according to the node weight combination score and selects the optimal combination. The sorting process is written to the solution audit log, which records the candidate node set, constraint edge set, triggered conflict registration log entries, and the final selected evidence digest hash set. If the solution execution unit cannot obtain a feasible combination solution under the quality gating criteria and statistical closure constraints, the gap filling rule and manual review trigger condition are triggered. The candidate resource unit identifier and slot number are written to the trigger log, and the temporary decision result is output. The temporary decision result is marked as pending review in the task context. After the subsequent field verification record is written back or the manual review is written back to the evidence record set, the task management module triggers recalculation. The recalculation calls the same evidence slot template, the same consistency constraint package, and reuses the alignment mapping registration criteria, so that the fusion link forms a sustainable automated iterative process with consistent audit traceability records.

[0082] After the execution unit completes the solution process, a fusion adjudication result is generated. This result includes at least the final spatial boundary, final classification, and final attribute field set, and simultaneously generates conflict interpretation codes and an evidence reference list. Conflict interpretation codes are generated from a conflict interpretation code dictionary based on the triggered conflict type code and the binding relationship between constraint edges. The evidence reference list is generated by combining evidence digest hashes, source identifiers, and collection timestamps, and is bound and registered with candidate resource unit identifiers. The decision graph adjudication module writes the "fusion adjudication result" into the task context as an output field name and uses it as the input source for the subsequent "fusion adjudication result" in S410. Simultaneously, it writes the graph structure summary of the constraint consistency decision graph into the intermediate cache of the audit traceability record for use during the S430 audit traceability record writing process. In the engineering implementation, when the county-level integration task encounters candidate resource units with large boundary offsets and classification coding discrepancies between the land survey results and forestry and grassland resource survey results during the solution phase, the system generates node weights by combining the credibility score and the collection timestamp, forms constraint edges according to topological constraints and classification mapping rules, and performs cross-layer verification on the county-level stratified area summary during the statistical closure constraint verification phase. When the verification is not satisfied, the gap supplementation rule is triggered, the missing slot number and supplementation source suggestion are written into the trigger log and the adjudication result is temporarily stored. After the field verification record is written back, the recalculation is triggered and the final integration adjudication result is output. The entire process retains a verifiable record in the solution audit log.

[0083] In summary, the technical effects of this step are as follows: Driven by the evidence slot template, alignment mapping table, and consistency constraint package, the conflict detection and processing completes conflict registration at the slot level and maps the constraint scope to the constraint consistency decision graph. The adjudication output carries both the conflict interpretation code and the evidence citation list, so that the fused adjudication result has a traceable evidence chain and a recalculated solution record.

[0084] Step S400 includes at least steps S410-S430:

[0085] S410. Obtain the fusion decision result and the baseline version number, perform target version generation processing, and obtain the target version;

[0086] Specifically, the version generation module reads the resource unit identifier, final spatial boundary, final classification, and final attribute field set from the fusion adjudication result output and registered in the database by the preceding S330. Simultaneously, it reads the evidence citation list and conflict interpretation code bound to the fusion adjudication result and uses them as the main input for this step. The version generation module also obtains the baseline version number from the version repository. This baseline version number corresponds to the target version released in the previous round or a specified historical release version, and together with the task batch identifier, regional scope identifier, and result version number strategy of this round of fusion tasks, constitutes the version context for this step. Further, when performing target version generation, the version generation module first performs version context verification, which includes baseline version number existence verification, regional scope consistency verification, and version number strategy consistency verification. The regional scope consistency verification is completed by determining the inclusion and adjacency relationships between the final spatial boundary in the fusion adjudication result and the regional scope boundary corresponding to the baseline version number. If the determination fails, an exception code is written to the version exception log and entered into the manual review queue, while the temporary state of the current round of fusion adjudication result is retained. Furthermore, the version generation module performs versioned assembly processing on the fusion decision result. This versioned assembly processing is collaboratively completed by the boundary assembly unit, classification assembly unit, attribute assembly unit, and metadata assembly unit. The boundary assembly unit performs boundary regularization and topology repair on the final spatial boundary according to a unified coordinate reference. Topology repair includes at least boundary closure checks, duplicate point cleanup, merging of narrow gaps, and unification of hole diameters. The repair actions during the repair process are recorded in the boundary repair log. The classification assembly unit standardizes the final classification according to the target classification reference. This standardization includes classification code validity verification and mapping. The subsequent encoding domain verification and null value caliber verification are performed, and the verification results are written to the classification verification log. The attribute assembly unit merges the final attribute field set according to the field caliber, verifies the allowed value range, and fills in the default value. Default value filling is only performed on field paths marked as defaultable in the evidence slot template. For field paths marked as required, a gap mark is written and the field paths are added to the supplementary collection trigger queue. The metadata assembly unit generates the version header information of the target version. The version header information includes at least the target version number, baseline version number, task batch identifier, generation timestamp, evidence citation summary, conflict explanation summary, and confidence parameter group snapshot reference. Understandably, the version generation module introduces an atomic commit strategy during the assembly process, generating middleware for boundaries, categories, attributes, and metadata respectively and performing consistency checks. The consistency checks include resource unit identifier consistency checks, field caliber consistency checks, and statistical closure caliber checks. When the consistency checks pass, the version generation module merges the middleware and writes it into the target version repository to obtain the target version. At the same time, it generates a target version index record and updates the "target version - baseline version number" association table.The version generation module writes the "target version" and its corresponding "target version number" into the task context as output field names, and uses the target version as the input source for the subsequent S420 extraction of the resource unit addition, deletion, and modification list. Simultaneously, the target version number is registered as a candidate baseline version number that can be called when the next round of similar fusion tasks is executed (S410). Boundary repair logs, classification verification logs, and attribute verification logs during the version generation process are temporarily stored in the audit cache for subsequent S430 audit traceability record writing and processing. In one engineering embodiment, in the quarterly forestry and grassland resource update task, the system triggers the version generation module to automatically execute this step using the county as the processing unit. When the task management module detects that the fusion adjudication result has been entered into the database and the baseline version number is in a readable state, it automatically triggers version context verification and version assembly processing. When a required field gap marker or boundary closure anomaly code appears during the assembly process, the system writes the corresponding resource unit identifier into the supplementary collection trigger queue and temporarily stores the target version as pending supplementary collection. After the field verification record is written back or the asset ownership ledger is written back, the task management module triggers incremental recalculation under the same target version number and completes the final write.

[0087] S420. Extract the list of added, deleted, and modified resource units from the target version and baseline version number, perform boundary change summary and attribute change summary generation processing, and generate a change differential package;

[0088] Specifically, the differential generation module reads the resource unit identifier set, final spatial boundary, final classification, and final attribute field set from the target version output by S410. Simultaneously, it reads the resource unit identifier set, baseline spatial boundary, baseline classification, and baseline attribute field set corresponding to the baseline version from the version repository according to the baseline version number, and uses both as input sources for differential calculation. Further, the differential generation module first performs resource unit alignment and matching processing, which is completed collaboratively by the identifier matching unit and the spatial matching unit. The identifier matching unit performs direct matching using the resource unit identifier as the primary key, obtaining directly matched pairs and an unmatched set. The unmatched set includes newly added candidates for the target version and deleted candidates for the baseline version. The spatial matching unit performs spatial proximity matching on the unmatched set, determining the nearest distance, overlap rate, and topological relationship between the unmatched units in the target version and the unmatched units in the baseline version to identify possible identifier changes or merging / splitting scenarios. The determination results are written to the alignment and matching log, enabling subsequent auditing and tracing to recalculate differential attribution. The differential generation module extracts a list of newly added, deleted, and modified resource units based on the alignment and matching results. The newly added list includes resource unit identifiers that exist in the target version but not in the baseline version; the deleted list includes resource unit identifiers that exist in the baseline version but not in the target version; and the modified list includes resource unit identifiers that exist in both versions but whose boundaries or attributes have changed. The determination of the modified list is completed by the change determination unit, which calculates change markers for boundaries and attributes respectively. Further, the differential generation module performs boundary change summary generation processing, which is generated by the boundary differential unit. The boundary differential unit extracts the target spatial boundary and baseline spatial boundary for each resource unit identifier in the modified list, calculates the boundary offset summary, area difference summary, and topology change summary, and writes the summary results into the boundary change summary field. The boundary offset summary obtains the offset distribution through boundary sampling comparison and registers a caliber description; the area difference summary obtains the difference value through boundary area calculation and registers a directional marker; and the topology change summary generates caliber entries such as "separated to intersecting, intersecting to separated, and containing to adjacent" based on changes in topology relationship markers.Furthermore, the differential generation module performs attribute change summary generation processing, which is generated by the attribute differential unit. The attribute differential unit extracts the target attribute field set and the baseline attribute field set for each resource unit identifier in the modification list, aligns them according to field scope, and generates field-level differential records. The field-level differential records contain at least the field path, baseline value summary, target value summary, change type code, and scope description code. The change type code is used to register additions, deletions, or replacements, and the scope description code is used to register changes in unit scope, classification scope, or null value scope. For differentials involving classification fields, the attribute differential unit simultaneously generates a classification change summary and establishes a correlation with the boundary change summary, enabling subsequent audit traceability to verify the linkage relationship of "boundary change - classification change - attribute change" at the same resource unit dimension. Understandably, the differential generation module introduces the reference caliber of the credibility parameter group when generating boundary change summaries and attribute change summaries. If the evidence reference summary of the corresponding resource unit identifier in the target version shows a pending supplementary collection mark, the differential generation module adds a "pending supplementary collection differential mark" to the summary of the resource unit identifier and writes it into the status field of the change differential package, so that the subsequent audit traceability record writing process can distinguish between "final differential" and "temporary differential". After completing the extraction of the new, delete, and modify lists and the generation of boundary change summaries and attribute change summaries, the differential generation module generates the change differential package. The change differential package includes at least the baseline version number, target version number, new list, delete list, modify list, boundary change summary set, attribute change summary set, alignment matching log reference, and status field. The "change differential package" is written into the task context as the output field name, and the change differential package is used as the input source for the subsequent S430 audit traceability record writing process. At the same time, the new, delete, and modify lists in the change differential package can be used by the results release module for incremental release orchestration, and the boundary change summary and attribute change summary can be used by the business verification module for sampling task orchestration. In one engineering embodiment, when the annual asset ownership ledger update and the forestry and grassland resource survey results update are superimposed, the system performs differential generation on the baseline version and the target version of the same area. When the spatial matching unit identifies a resource unit as being merged or split, the differential generation module marks the resource unit as a structural change and generates a structural boundary change summary. At the same time, it records the master-slave association mark in the attribute differential. Subsequent audit traceability record writing processing can use this to restore the link source and timestamp of the merged or split.

[0089] S430. Perform audit traceability record writing processing on the change differential package and generate a credibility parameter group update package, generating audit traceability records and credibility parameter groups;

[0090] Specifically, the audit traceability module reads the baseline version number, target version number, list of additions, deletions and modifications, set of boundary change summaries, set of attribute change summaries, alignment matching log references and status fields from the change differential package output by S420. At the same time, it reads back the boundary repair logs, classification verification logs and attribute verification logs temporarily stored in S410 from the audit cache, and reads back the summary information of the evidence reference list from the evidence repository according to the evidence reference summary, as the input source for audit writing. Furthermore, the audit traceability module performs audit traceability record writing processing, which is completed collaboratively by the link aggregation unit, the record shaping unit, and the write submission unit. The link aggregation unit uses the resource unit identifier as the main line to aggregate the differential summary and the list of additions, deletions, and modifications in the change differential package, and connects the alignment matching log reference, boundary repair log reference, classification verification log reference, and attribute verification log reference to form a resource unit-level audit link object. The audit link object includes at least a version link segment, an evidence link segment, a differential link segment, and a process link segment. The version link segment records the association between the baseline version number and the target version number, the evidence link segment records the binding relationship between the evidence reference summary and the source identifier and the collection timestamp, the differential link segment records the boundary change summary and the attribute change summary, and the process link segment records the key log references generated during the assembly and verification process. The record shaping unit shapes the audit link object into a writable audit trace record. The audit trace record includes at least the audit record number, resource unit identifier, audit timestamp, version link summary, evidence citation summary, differential summary, process summary, status field, and recalculation entry marker. The recalculation entry marker is used to register the input index positions required for replaying the current round of differential and version generation. These input index positions include the target version number, baseline version number, alignment matching log reference, and evidence citation summary index. The write submission unit writes the audit trace record to the audit database according to an atomic commit strategy and generates a write receipt. The write receipt includes a write batch identifier and a failure retry flag. When the write failure retry flag is triggered, the write submission unit writes the failure reason code to the audit exception log and enters the rewrite queue. The rewrite queue is triggered by the task management module according to the retry window to prevent audit link breaks.

[0091] Furthermore, the audit traceability module generates a credibility parameter group update package, which is generated by the credibility update unit. The credibility update unit takes the change differential package and audit traceability records as input, extracts credibility-related update elements, and forms an update package structure. These update elements include at least the differential type distribution, a summary of supplementary sampling trigger times, a conflict interpretation code distribution, an evidence source stability summary, and a quality gate pass status summary. Specifically, the credibility update unit calculates the differential type distribution for the added, deleted, and modified lists; it statistically analyzes the supplementary sampling trigger times for resource unit identifiers marked as pending supplementary sampling differential tags in the status field; and it statistically analyzes the conflict interpretation code distribution for the conflict interpretation summaries in the audit traceability records. The credibility update unit also performs stability analysis on the source identifier and collection timestamp in the evidence citation summary, generates an evidence source stability summary, and binds it and the quality gate pass status summary to the same update batch identifier of the credibility parameter group update package. Understandably, when generating the credibility parameter group update package, the credibility update unit retains the minimum set parameter scope. The minimum set parameter scope includes at least the source identifier, collection timestamp, quality label summary, conflict interpretation code summary, and supplementary collection trigger summary. The remaining statistical fields are written into the update package as optional extensions without changing the availability of the update package. When the system is in lightweight operation mode, the credibility update unit only writes the minimum set parameter scope, so that subsequent similar fusion tasks can still complete quality gating and evidence priority calculation based on the credibility parameter group.

[0092] After completing the above processing, the audit traceability module generates the audit traceability record and the credibility parameter group, and writes the "audit traceability record" and "credibility parameter group" into the task context as output field names. The audit traceability record provides a location entry point for subsequent version queries and differential verification. The credibility parameter group serves as a reference input for the next round of fusion tasks, executing S130 to generate the quality label set and executing S230 to register the minimum quality threshold. It can also be used as the basis for updating node weights when constructing the constraint consistency decision graph in subsequent S330. In one engineering embodiment, in a scenario of multi-source data overlay updates across years, the system aggregates the differential summary, evidence citation summary, and assembly verification log of each county resource unit into an audit link object and writes it into the audit database. When a regulatory inspection triggers a verification of a resource unit identifier, the system locates the target version number, baseline version number, and evidence citation summary index through the recalculation entry mark in the audit traceability record, matches the log references according to the same alignment, replays the differential generation link, and thus realizes the recalculated call of the audit traceability record.

[0093] In summary, the technical effects of this step are as follows: By aggregating and atomically writing the change differential package, the audit traceability record connects the version link, evidence link, differential link, and process link at the resource unit level, and synchronously forms a credibility parameter group update package. This enables subsequent fusion tasks to reuse credibility and audit indexes for quality gating, priority calculation, and differential verification under the same caliber.

[0094] Example 2: Figure 2 A structural block diagram of a forestry and grassland resource data fusion system according to an embodiment of the present invention is shown. Figure 2 As shown, the structure may include:

[0095] The data source access and standardization module 01 is used to obtain a list of data sources and register source identifiers and collection timestamps, extract field mapping tables and coordinate reference parameters and register coding dictionary calibers and projection calibers, perform geometric and attribute standardization processing on the data source list according to the field mapping table and coordinate reference parameters, and generate a quality label set according to the confidence parameter group to obtain an evidence record set. Specifically, the data source access and standardization module receives land survey results, forest and grassland resource survey results, remote sensing classification products, field verification records, and asset ownership ledgers from external data entry operations, and registers the data entry from each source as an entry in the data source list. The entry contains at least two fields: source identifier and collection timestamp. When the entry is generated, the collection batch identifier and entry verification status record bits are written synchronously to represent the consistent reference relationship of the data source list in the same fusion task. After forming the data source list, the data source access and standardization module reads field definitions and spatial reference information from each entry in the data source list, generates the field mapping table and coordinate reference parameters, and writes the encoding dictionary caliber and projection caliber into the corresponding record fields of the field mapping table and coordinate reference parameters during the generation process, so that subsequent geometric and attribute standardization processing has unified field names, field types, enumeration value calibers, and spatial coordinate expression calibers. The geometric and attribute standardization processing includes performing coordinate reference unification, geometric legality verification, overlap self-intersection and hole checks, and boundary closure checks on geometric objects, and performing field missing checks, type conversion, enumeration mapping, unit caliber normalization, outlier marking, and source field retention mapping on attribute fields; when a geometric object is detected to be missing or an attribute field is missing beyond a preset range, the abnormal status is written into the data source list's entry verification status record position, and the corresponding record is included in the abnormal mark field of the evidence record set. When the data source access and standardization module outputs the standardization results, it generates the quality label set for each standardization result according to the credibility parameter group, and encapsulates the quality label set together with the corresponding geometry, attributes, source identifier, and collection timestamp into the evidence record set. The evidence record set, together with the data source list, the field mapping table, and the coordinate reference parameters, is provided to the constraint and slot arrangement module, and the evidence record set is provided to the candidate unit generation and alignment mapping module as the input object for spatial index clustering processing.

[0096] The constraint and slot orchestration module 02 is used to obtain the target output specification and target version number and associate them with the data source list. It extracts classification mapping rules and topological constraints from the target output specification and registers statistical closure constraints and temporal consistency constraints to generate a consistency constraint package. It then performs slot orchestration on the data source list and consistency constraint package and registers the minimum quality threshold and gap sampling rules to generate evidence slot templates. Specifically, the constraint and slot orchestration module receives the data source list output from the data source access and standardization module, and receives the target output specification and target version number input by the fusion task configuration. The target output specification includes three types of specification items: classification system caliber, field caliber, and statistical caliber. The constraint and slot orchestration module establishes an association registration relationship between the target version number and the data source list to form a data source reference set under the same target version number, so that subsequent slot orchestration is performed on a unified source set. The constraint and slot orchestration module extracts the classification mapping rules and the topological constraints from the target deliverable specification. The classification mapping rules describe the mapping relationship between classification fields from different data sources to classification fields in the target deliverable specification. The topological constraints describe the intersection, containment, disjointness, boundary connection, and closure validity constraints between geometric objects. Simultaneously, the constraint and slot orchestration module registers statistical closure constraints and temporal consistency constraints into the constraint entries of the consistency constraint package. The statistical closure constraints include closure verification rules for resource unit area and category counts after statistical aggregation. The temporal consistency constraints include change inheritance rules and timestamp consistency verification rules between the target version number and the baseline version number. After the consistency constraint package is generated, the constraint and slot orchestration module performs slot orchestration on the data source list and the consistency constraint package. Slot orchestration includes establishing evidence slots for each field specification of the target outcome, specifying the source list reference range for each evidence slot, binding the field mapping item corresponding to the field mapping table to each evidence slot, and writing the minimum quality threshold into the threshold field of the corresponding evidence slot. When slot orchestration detects missing fields or quality labels that do not meet the minimum quality threshold in the source list, it writes the gap filling rule into the gap handling field of the corresponding evidence slot and registers the gap status as a record to be filled. The consistency constraint package and the evidence slot template output by the constraint and slot orchestration module are provided to the candidate unit generation and alignment mapping module, and the evidence slot template and the consistency constraint package are provided to the conflict detection and adjudication module as input objects for conflict detection processing and constraint consistency decision graph construction.

[0097] The candidate unit generation and alignment mapping module 03 is used to obtain the evidence record set and evidence slot template, perform spatial index clustering processing, generate an evidence association table and obtain a candidate resource unit set, extract geometric alignment indicators and semantic alignment indicators from the candidate resource unit set and the evidence association table, perform alignment mapping registration processing, and generate an alignment mapping table. Specifically, the candidate unit generation and alignment mapping module receives the evidence record set output from the data source access and standardization module, and receives the evidence slot template output from the constraint and slot arrangement module. The spatial index clustering processing establishes spatial index keys on the geometric objects of the evidence record set, and merges evidence records in the same spatial neighborhood into the same clustering unit according to the spatial index keys. During the clustering process, the candidate unit generation and alignment mapping module extracts the coverage, boundary set, and attribute candidate set for each clustering unit, and generates a unit record for the candidate resource unit set. The unit record contains a unit identifier, boundary candidate reference, and attribute candidate reference for interface with subsequent conflict detection processing. The candidate unit generation and alignment mapping module simultaneously generates the evidence association table when forming the candidate resource unit set. The evidence association table records the association between the unit identifiers of the candidate resource unit set and the evidence records of the evidence record set. This association includes source identifier references, collection timestamp references, quality tag set references, and field mapping item references, thus providing a traceable evidence citation path during subsequent alignment mapping registration processing. The candidate unit generation and alignment mapping module extracts geometric alignment indicators and semantic alignment indicators from the candidate resource unit set and the evidence association table. The geometric alignment indicators include descriptions of overlap and distance relationships between candidate boundaries, while the semantic alignment indicators include descriptions of classification consistency and field caliber consistency after mapping according to the classification mapping rules. The alignment mapping registration process writes the registration fields of the geometric alignment indicators and semantic alignment indicators into the unit record of each candidate resource unit, generating the alignment mapping table. This alignment mapping table, along with the candidate resource unit set, is provided to the conflict detection and adjudication module as input for conflict detection processing. Simultaneously, the evidence association table is provided to the conflict detection and adjudication module for generating evidence citation associations that fuse adjudication results.

[0098] The conflict detection and adjudication module 04 is used to perform conflict detection processing on the evidence slot template, alignment mapping table, and consistency constraint package, and construct a constraint consistency decision graph to generate a fusion adjudication result. Specifically, the conflict detection and adjudication module receives the evidence slot template and the consistency constraint package output from the constraint and slot arrangement module, and receives the alignment mapping table output from the candidate unit generation and alignment mapping module. The conflict detection processing scans the registration fields of the alignment mapping table slot by slot according to the slot arrangement result of the evidence slot template, generates conflict records for slots that do not meet the minimum quality threshold, slots with gap filling rule triggering conditions, and slots with geometric alignment index contradictions or semantic alignment index contradictions, and establishes an association between the conflict records and the unit identifiers of the corresponding candidate resource units. After a conflict record is generated, the conflict detection and adjudication module performs constraint verification on the conflict record according to the classification mapping rules, topological constraints, statistical closure constraints, and temporal consistency constraints of the consistency constraint package, forming a constraint item reference set, and constructing a constraint consistency decision graph based on this set. The constraint consistency decision graph includes candidate boundary reference nodes, candidate classification reference nodes, candidate attribute reference nodes, and the constraint relationships between them. Furthermore, the registration results of the quality label set and credibility parameter group in the evidence record set are mapped to the node weight registration field for weight reference during adjudication. The conflict detection and adjudication module executes an adjudication process on the constraint consistency decision graph. The adjudication process includes combining and filtering candidate boundaries, candidate categories, and candidate attributes according to constraint consistency rules; eliminating combinations that violate topological constraints or statistical closure constraints; inheriting combinations that satisfy temporal consistency constraints; and writing the finally selected boundary references, category references, and attribute references into the fusion adjudication result. The fusion adjudication result provides the version generation and differential output module as input for the target version generation process. At the same time, the evidence reference association of the fusion adjudication result is written into the reference backfill field of the evidence association table for the audit traceability and parameter update module to read during the audit traceability record writing process.

[0099] The version generation and differential output module 05 is used to obtain the fusion adjudication result and the baseline version number, and perform target version generation processing to obtain the target version. It extracts a list of added, deleted, and modified resource units from the target version and the baseline version number, and generates a boundary change summary and an attribute change summary, generating a change differential package. Specifically, the version generation and differential output module receives the fusion adjudication result output by the conflict detection and adjudication module, and reads the baseline version record corresponding to the baseline version number from the version repository. The target version generation processing includes writing the final boundary reference, final classification reference, and final attribute reference from the fusion adjudication result into the resource unit record of the target version, registering the association between the target version number and the baseline version number as a version inheritance record, and writing the collection timestamp reference and processing timestamp reference related to the timing consistency constraint into the version metadata field of the target version. After obtaining the target version, the version generation and differential output module performs differential comparison processing on the target version and the baseline version record corresponding to the baseline version number. This differential comparison processing performs set comparison on the resource unit identifier set and generates a list of added, deleted, and modified resource units. It also extracts boundary reference differences and attribute field differences for each resource unit in the modification list, forming boundary change summaries and attribute change summaries. When the differential comparison processing detects missing geometric objects or attribute fields in a resource unit, preventing the formation of a summary, it writes the abnormal entry into the abnormal marker field of the resource unit addition, deletion, and modification list and associates the abnormal entry with the conflict record reference field of the fusion adjudication result. The version generation and differential output module encapsulates the resource unit addition, deletion, and modification list, boundary change summary, attribute change summary, target version number, baseline version number, and fusion task identifier into a change differential package. This change differential package is provided to the audit traceability and parameter update module as an input object for audit traceability record writing processing. Simultaneously, it registers the target version as a searchable version record in the version reading channel for external results services, for subsequent version reading and backtracking processing.

[0100] The audit traceability and parameter update module 06 is used to perform audit traceability record writing processing on the change differential package and generate a credibility parameter group update package, generating audit traceability records and credibility parameter groups. Specifically, the audit traceability and parameter update module receives the change differential package output by the version generation and differential output module, and reads the evidence reference association backfill field and conflict record reference field of the fusion adjudication result. It writes the resource unit addition, deletion and modification list, boundary change summary, attribute change summary and evidence reference association in the change differential package into the record entries of the audit traceability record. The record entries include the baseline version number, target version number, fusion task identifier, source identifier reference, collection timestamp reference, field mapping table reference, coordinate benchmark parameter reference and consistency constraint package reference, thereby maintaining the link correspondence from input object to output product in the same record entry. The audit traceability record writing process performs consistency verification on the record entries during the writing process. The consistency verification includes record entry field missing verification, version number association verification, and evidence reference association existence verification. When the consistency verification is not satisfied, the record entry is marked as abnormal and the original change differential package reference is retained for subsequent manual review channel reading. After the audit traceability record is generated, the audit traceability and parameter update module generates a credibility parameter group update package based on the distribution of quality tag sets and the statistical results of conflict records registered in the audit traceability record. The credibility parameter group update package is then applied to the credibility parameter group to generate a new credibility parameter group registration record. The credibility parameter group is sent back to the data source access and standardization module as an input object for the subsequent generation of quality tag sets. The audit traceability record is also provided to the constraint and slot orchestration module as a version traceability input object when registering the association between the target result specification and the target version number, thus forming a closed-loop record and parameter update link during system operation.

Claims

1. A method for fusing forest and grassland resource data, characterized in that, include: S100. Obtain a list of data sources, extract the field mapping table and coordinate reference parameters, perform geometric and attribute standardization and generate a quality label set to obtain a set of evidence records; wherein, the list of data sources includes source identifier, data type, coverage, storage location, collection timestamp and access method marker; S200. Obtain the target outcome specification and target version number, extract classification mapping rules and topological constraints, and register statistical closure constraints and temporal consistency constraints for processing, and generate evidence slot templates; wherein, the target outcome specification includes classification system caliber, field caliber and statistical caliber; S300. Based on the evidence record set and evidence slot template, perform spatial index clustering and evidence association table generation processing to generate a candidate resource unit set and alignment mapping table, and construct a constraint consistency decision graph to obtain the fusion adjudication result; wherein, the fusion adjudication result includes the final spatial boundary, final classification, final attribute field set, conflict interpretation code and evidence citation list; S400. Based on the fusion adjudication result, perform target version generation and change extraction processing to generate a change differential package, and perform audit traceability record writing and credibility parameter group update processing to obtain audit traceability records and credibility parameter groups.

2. The method according to claim 1, characterized in that, The process of extracting the field mapping table and coordinate baseline parameters includes: The field mapping table includes field name correspondence, unit caliber, encoding dictionary caliber, and null value caliber; the coordinate reference parameters include projection caliber, caliber of the reference plane, and caliber of the coordinate axis order.

3. The method according to claim 1, characterized in that, The process of performing geometry and attribute standardization and generating a quality label set includes: The geometric standardization process includes geometric type consistency, coordinate benchmark unification, and geometric validity verification. The attribute standardization process includes field extraction, field mapping, encoding dictionary conversion, and null value unification. A quality label set is generated based on a credibility parameter set, which includes timeliness label generation rules, spatial accuracy label generation rules, completeness rate label generation rules, and verification consistency label generation rules. An evidence record set is generated, which includes source identifier, spatial geometry, attribute field set, collection timestamp, processing timestamp, quality label set, credibility score, and evidence digest hash.

4. The method according to claim 1, characterized in that, The process of extracting classification mapping rules and topological constraints includes: The classification mapping rules include source identifier, source classification field path, target classification field path, mapping conditions, conflict priority marker and missing backfill marker; the topological constraints include boundary closure constraints, adjacent unit non-overlapping constraints, hole handling constraints, cross-administrative boundary trimming constraints and protection red line boundary consistency constraints.

5. The method according to claim 1, characterized in that, The process of handling registration and statistical closure constraints and time-series consistency constraints includes: The statistical closure constraints include the area summary caliber within the same statistical level, the summary consistency caliber between different statistical levels, and the consistency caliber between the total area and the total area. The temporal consistency constraints include the version inheritance relationship caliber, the change legality caliber, and the consistency caliber with the release timestamp. The slot orchestration process is performed, which includes associating the target result specifications with the data source list and generating evidence slot templates. The evidence slot templates include slot number, slot type, source list, field mapping table, minimum quality threshold, gap filling rules, and conflict interpretation code dictionary. The source list is generated from the set of source identifiers in the data source list that meet the coverage and access status conditions. The minimum quality threshold is bound to the slot type and is consistent with the credibility score caliber of the credibility parameter group. The gap filling rules include trigger conditions, supplementary source suggestions, collection range caliber, field caliber reference, and task write-back caliber.

6. The method according to claim 1, characterized in that, The process of performing spatial index clustering and generating evidence association tables includes: Spatial index clustering processing includes geometrical validity verification of the spatial geometry in the evidence record set, generating spatial index codes according to a preset spatial index granularity, bucketing the evidence records according to the spatial index codes to generate bucketed evidence clusters, and performing cluster merging within each bucketed evidence cluster according to the spatial geometric overlap relationship to generate a candidate resource unit set; and performing evidence association table generation processing, which includes identifying the evidence digest hash list, source identifier list and collection timestamp list associated with each candidate resource unit within its cluster, and writing the overlap rate, nearest boundary distance and intersection area ratio as association metric fields.

7. The method according to claim 1, characterized in that, The process of generating a set of candidate resource units and an alignment mapping table, and constructing a constraint-consistent decision graph includes: The constraint consistency decision graph construction process includes generating a graph structure with candidate resource unit identifiers. The graph structure includes candidate boundary nodes, candidate classification nodes, and candidate attribute nodes. Node weights are generated based on confidence scores and collection timestamps. Topological constraints, classification mapping rules, statistical closure constraints, and quality gating criteria in the consistency constraint package are mapped as constraint edges. Weighted constraint solving is performed to generate a fusion decision result.

8. The method according to claim 1, characterized in that, The process of generating and extracting changes to the target version includes: The target version generation process includes retrieving the baseline version number from the version repository, performing version context verification (including baseline version number existence verification, regional range consistency verification, and version number policy consistency verification), and performing versioned assembly processing (including boundary regularization and topology repair by boundary assembly units, code standardization by classification assembly units, unit caliber merging and default value filling by attribute assembly units, and version header information generation by metadata assembly units), thus generating the target version and target version number. The process also includes change extraction processing (including reading the resource unit identifier set, spatial boundary, classification, and attribute field sets from the target version and baseline version, performing resource unit alignment and matching processing, generating a list of additions, deletions, and modifications, and generating boundary change summaries and attribute change summaries to obtain a change differential package). This change differential package includes the baseline version number, target version number, addition list, deletion list, modification list, boundary change summary set, attribute change summary set, alignment matching log reference, and status field.

9. The method according to claim 1, characterized in that, The process of writing audit trace records and updating credibility parameter groups includes: The audit traceability record writing process includes, with resource unit identifier as the main thread, aggregating the differential summary in the change differential package and the process log to form an audit link object, which is then formed into an audit traceability record and written to the audit database; and performing a credibility parameter group update process, which includes extracting the differential type distribution, supplementary sampling trigger count summary, conflict interpretation code distribution, evidence source stability summary, and quality gate pass status summary based on the change differential package and the audit traceability record, and generating a credibility parameter group update package.

10. A forestry and grassland resource data fusion system, characterized in that, include: The module comprises a data source access and standardization module, a constraint and slot orchestration module, a candidate unit generation and alignment mapping module, a conflict detection and adjudication module, a version generation and differential output module, and an audit traceability and parameter update module; the modules are connected in sequence to implement the method described in any one of claims 1-9.

Images