CN-121999346-A - Digital line drawing identification correction method and system

CN121999346ACN 121999346 ACN121999346 ACN 121999346ACN-121999346-A

Abstract

The invention provides a digital line drawing identification correction method and system, the method comprises the steps of constructing a multistage configurable scene rule base, obtaining digital line drawing data to be detected and external reference data, carrying out data preprocessing and space registration, identifying and outputting structural error information based on comparison of the scene rule base and the external reference data, constructing a repair strategy knowledge base, matching or generating a repair strategy from the repair strategy knowledge base according to the structural error information, executing repair operation to obtain corrected digital line drawing data, constructing a correction evaluation model, carrying out rechecking verification and effect evaluation on the corrected digital line drawing data, optimizing the repair strategy knowledge base according to an evaluation result, and outputting final digital line drawing data.

Inventors

HUANG LU
CHEN CHUNHUA
ZHANG LIANG
JIA JIPENG
FAN XUEFENG
DAI LI
HOU AILING
HU HAN

Assignees

湖北省测绘工程院

Dates

Publication Date: 20260508
Application Date: 20260116

Claims (10)

1. The digital line drawing identification correction method is characterized by comprising the following steps of: s1, constructing a multi-level configurable scene rule base; S2, acquiring digital line drawing data to be detected and external reference data, carrying out data preprocessing and space registration, and identifying and outputting structured error information based on comparison of a scene rule base and the external reference data; s3, constructing a repair strategy knowledge base, wherein the repair strategy knowledge base stores a predefined repair function, corresponding applicable conditions and a strategy generation model, and the strategy generation model is used for generating a new repair strategy; S4, matching or generating a repair strategy from a repair strategy knowledge base according to the structured error information, and executing repair operation to obtain corrected digital line drawing data; S5, constructing a correction evaluation model, performing rechecking verification and effect evaluation on the correction digital line drawing data, optimizing a repair strategy knowledge base according to an evaluation result, and outputting final digital line drawing data.
2. The method for digital map recognition correction according to claim 1, wherein said constructing a multi-level configurable scene rule base in step S1 comprises the sub-steps of: s11, defining an atomization rule module by adopting a mode of combining a natural language template and parameterization, wherein structural elements of the rule module comprise an inspection object, an inspection dimension, a spatial relationship, a reference object and a threshold condition; s12, according to different scene types and constraint conditions thereof, instantiating a rule module into an atomic rule, wherein the quality inspection scene types comprise landform inspection, geometric constraint inspection, edge connection inspection, attribute logic inspection, water system traffic inspection and topological relation inspection; S13, combining a plurality of logically related atomic rules into a quality inspection model, wherein the quality inspection model is correspondingly and associatively arranged with the scene type; S14, carrying out association configuration on the quality inspection model and the data item standard and/or the geographic region characteristics to form a quality inspection scheme oriented to specific application; and S15, storing the scene rule base into a database according to a quality inspection scheme, and constructing a multi-stage configurable scene rule base.
3. The method for identifying and correcting a digital map as claimed in claim 2, wherein in step S2, the step of obtaining the digital map data to be inspected and the external reference data, performing data preprocessing and spatial registration, and identifying and outputting the structured error information based on the comparison of the scene rule base and the external reference data includes the following sub-steps: S21, obtaining digital line drawing data to be detected and external reference data; s22, carrying out format standardization and coordinate system unified processing on the digital line drawing data to be detected to obtain standardized digital line drawing data; S23, extracting stable feature points of standardized digital line drawing data and external reference data based on a feature point matching algorithm, calculating space transformation parameters by adopting a least square method, and executing registration operation, and aligning the two to the same space reference to obtain registered digital line drawing data; s24, loading a corresponding quality inspection model and atomic rules from a scene rule base according to a quality inspection scheme, performing topological relation inspection, geometric constraint inspection and attribute logic inspection on each geographic element in the registered digital line drawing data, extracting elements violating the rules and error context information thereof, and generating a primary error layer; S25, carrying out multi-dimensional difference detection on the registered digital line drawing data and external reference data, wherein the difference detection comprises the steps of calculating the deviation of elements in position and shape through a geometric difference detection algorithm, identifying the missing, redundant or partial missing condition of the elements through spatial correlation, and checking the visual consistency of element classification attribute through comparing semantic information extracted from the external reference data; s26, performing spatial association and merging on the primary error layer and the suspicious error area, filtering false detection by combining the geographical semantic context, encoding error information, and generating an error information set comprising unique ID, associated element ID, error type, severity level, spatial geometry, detailed description and repair suggestion fields.
4. The method for identifying and correcting a digital map as claimed in claim 3, wherein in step S3, a repair policy repository is constructed, the repair policy repository storing predefined repair functions and corresponding applicable conditions and a policy generation model for generating a new repair policy, the method comprising the sub-steps of: S31, defining and packaging an atomization repair function for executing basic repair operation according to the error type in the digital line graph, wherein the repair function at least comprises vertex capturing, line segment merging, face gap filling, geometric smoothing, attribute assignment and topology reconstruction, and labeling each repair function with function description, input parameter format and expected output; S32, defining corresponding applicable conditions for each atomization repair function or a composite repair strategy combined by a plurality of atomization repair functions, wherein the applicable conditions at least comprise repairable error types, geometric feature threshold values for triggering repair, element attribute constraints and processing priorities; s33, establishing a strategy generation model based on a machine learning network to generate a new repairing strategy; And S34, carrying out structural storage on the basic repair strategy knowledge base, the model parameters of the strategy generation model and the historical case base to obtain a final repair strategy knowledge base.
5. The method as claimed in claim 4, wherein the creating a policy generation model based on the machine learning network in the step S33 for generating a repair new policy comprises the following sub-steps: S331, extracting error characteristics and corresponding successful repair strategy sequences from the historical correction cases, and constructing a training sample set, wherein the error characteristics comprise error types, geometric contexts and element attributes; S332, constructing a strategy generation model based on a machine learning network, and inputting a training sample set into the strategy generation model for training so as to learn the association relationship between the error characteristics and the effective restoration strategy; s333, for undefined error types in the basic repair strategy knowledge base, inputting the corresponding error characteristics into a trained strategy generation model, generating a plurality of repair strategy sequences, and giving confidence scores to each strategy.
6. The method for identifying and correcting a digital map as set forth in claim 5, wherein in step S4, the steps of matching or generating a repair policy from a repair policy knowledge base based on the structured error information and performing a repair operation to obtain corrected digital map data include the following sub-steps: S41, screening a plurality of candidate repair strategies from a repair strategy knowledge base according to the error type and the context characteristics of the structured error information, sorting according to the confidence scores of the strategies, and selecting the strategy with the highest confidence if the strategy with the direct match exists; S42, according to the selected repair strategy, performing topology error, geometric deformation and attribute repair operation on the corresponding error elements, wherein, The topology error repairing operation is used for capturing and merging suspension points based on neighborhood searching, splitting and reconstructing self-intersecting line segments and closing the gaps based on boundary inference; performing correction based on space transformation parameters on the integral position deviation of the element by the geometric deformation restoration operation, and performing smoothing based on spline interpolation on the local geometric distortion; Performing attribute deduction and assignment according to the attribute of the adjacent element or semantic information of the reference data by the repair operation of the attribute deficiency or error; s43, writing the repaired result back into the digital line drawing data to generate corrected digital line drawing data.
7. The method for identifying and correcting digital map as set forth in claim 6, wherein the step S5 of constructing a correction evaluation model, performing rechecking verification and effect evaluation on the corrected digital map data, optimizing a repair strategy knowledge base according to the evaluation result, and outputting final digital map data comprises the sub-steps of: s51, establishing a strategy effect evaluation model, and setting evaluation indexes, wherein the evaluation indexes comprise a repair success rate, a geometric precision improvement degree, an attribute correction accuracy rate and an execution efficiency; S52, in the process of executing the repair, recording a repair operation log, wherein the log comprises error information before the repair, a repair strategy, repair operation parameters, element states after the repair and an execution time stamp; s53, after the repair operation is completed, re-executing the corresponding atomic rule check for the repaired element, and calculating each evaluation index value; s54, carrying out weighted comprehensive calculation on the evaluation index value based on a preset weight coefficient to obtain a comprehensive evaluation result of the repair operation; s55, presetting a first evaluation threshold and a second evaluation threshold, wherein the first evaluation threshold is larger than the second evaluation threshold; If the comprehensive evaluation result is larger than the first evaluation threshold, judging that the repair strategy is good, increasing the confidence score of the current repair strategy according to a preset increase proportion, and expanding the application condition range of the current repair strategy through a threshold expansion factor; If the comprehensive evaluation result is greater than or equal to the second evaluation threshold and the comprehensive evaluation result is less than or equal to the first evaluation threshold, judging that the general repair strategy is a general repair strategy, reducing the confidence score of the current repair strategy according to a preset reduction ratio, and reducing the application condition range of the current repair strategy by a threshold reduction factor; if the comprehensive evaluation result is smaller than or equal to the second evaluation threshold value, judging that the strategy is invalid, removing the strategy from the recommendation list, and initiating a manual repair flow for the corresponding error; S56, taking the error characteristics, the repair strategies and the corresponding comprehensive evaluation results after each repair operation as new training samples to be stored in a historical case library for optimizing and training a strategy generation model; S57, integrating all the verified and optimized repair results, and outputting final digital line drawing data.
8. A digital map recognition correction system implemented by the digital map recognition correction method according to any one of claims 1 to 7, comprising: the rule base construction module is used for constructing a multi-level configurable scene rule base; The error recognition module is used for acquiring the digital line drawing data to be detected and external reference data, carrying out data preprocessing and space registration, and recognizing and outputting structured error information based on comparison of a scene rule base and the external reference data; The system comprises a knowledge base construction module, a restoration strategy knowledge base, a restoration strategy generation module and a restoration strategy generation module, wherein the knowledge base construction module is used for constructing a restoration strategy knowledge base, the restoration strategy knowledge base stores a predefined restoration function, corresponding applicable conditions and a strategy generation model, and the strategy generation model is used for generating a new restoration strategy; The matching execution module is used for matching or generating a repair strategy from a repair strategy knowledge base according to the structured error information and executing repair operation to obtain corrected digital line drawing data; And the evaluation output module is used for constructing a correction evaluation model, executing rechecking verification and effect evaluation on the correction digital line drawing data, optimizing a repair strategy knowledge base according to an evaluation result and outputting final digital line drawing data.
9. An electronic device comprising at least one processor, at least one memory, a communication interface and a bus, wherein the processor, the memory, the communication interface complete communication with each other via the bus, wherein the memory stores a digital map identification correction method program executable by the processor, wherein a digital map identification correction method program is configured to implement a digital map identification correction method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the storage medium has stored thereon a digital map recognition correction method program which, when executed, implements the digital map recognition correction method according to any one of claims 1 to 7.

Description

Digital line drawing identification correction method and system Technical Field The invention relates to the technical field of geographic information data processing, in particular to a digital line drawing identification correction method and a digital line drawing identification correction system. Background The digital line drawing is used as a space data carrier in the core of the geographic information field and is widely applied to the key fields of national soil mapping, urban planning, traffic construction, natural resource management, emergency disaster relief, intelligent government affairs and the like, the data quality of the digital line drawing directly determines the reliability of space information analysis, decision support and downstream application systems, for example, in urban rail transit planning, the road coordinate precision and pipeline attribute integrity in DLG directly influence engineering design accuracy, and in natural resource validation, the topological consistency of the ground boundary of the DLG is the core basis of property definition, so that the quality inspection and correction link of DLG data is a key control point in the space data production flow. The DLG data production process is complex, the element types are numerous, the space relation is interweaved, and the complexity determines the difficulty of DLG data quality control. In the current digital line drawing processing field, a complete and efficient repair and evaluation system is generally lacking, so that the error repair efficiency in the digital line drawing is low and the accuracy is not enough, a system-integrated repair strategy knowledge base is not available for providing a targeted repair scheme, an effective correction evaluation model is also lacking for comprehensively and objectively evaluating the repair effect, the repair strategy is difficult to dynamically optimize according to the evaluation result, the quality of finally output digital line drawing data is uneven, reliable data support cannot be provided for geographic information application, and the accuracy and reliability of related application are seriously affected. Disclosure of Invention In view of the above, the invention provides a digital line drawing identification correction method and a system, which realize efficient and accurate repair of digital line drawing errors, dynamic optimization of strategies and improvement of final data quality by constructing a repair strategy knowledge base and a correction evaluation model, and provide high-quality data support for geographic information application. The technical scheme of the invention is realized in that the digital line drawing identification correction method comprises the following steps: s1, constructing a multi-level configurable scene rule base; S2, acquiring digital line drawing data to be detected and external reference data, carrying out data preprocessing and space registration, and identifying and outputting structured error information based on comparison of a scene rule base and the external reference data; s3, constructing a repair strategy knowledge base, wherein the repair strategy knowledge base stores a predefined repair function, corresponding applicable conditions and a strategy generation model, and the strategy generation model is used for generating a new repair strategy; S4, matching or generating a repair strategy from a repair strategy knowledge base according to the structured error information, and executing repair operation to obtain corrected digital line drawing data; S5, constructing a correction evaluation model, performing rechecking verification and effect evaluation on the correction digital line drawing data, optimizing a repair strategy knowledge base according to an evaluation result, and outputting final digital line drawing data. On the basis of the above technical solution, preferably, the constructing a multi-level configurable scene rule base in step S1 includes the following sub-steps: s11, defining an atomization rule module by adopting a mode of combining a natural language template and parameterization, wherein structural elements of the rule module comprise an inspection object, an inspection dimension, a spatial relationship, a reference object and a threshold condition; s12, according to different scene types and constraint conditions thereof, instantiating a rule module into an atomic rule, wherein the quality inspection scene types comprise landform inspection, geometric constraint inspection, edge connection inspection, attribute logic inspection, water system traffic inspection and topological relation inspection; S13, combining a plurality of logically related atomic rules into a quality inspection model, wherein the quality inspection model is correspondingly and associatively arranged with the scene type; S14, carrying out association configuration on the quality inspection model and the data item standard and/o