A geographic information and big data information exchange system and method

Abstract

The application discloses a geographic information and big data information exchange system and method, and relates to the technical field of geographic big data, which comprises the following steps: receiving a geographic data sharing request from an external requester; retrieving four types of source data, i.e., remote sensing images, vector geographic elements, Internet of Things sensing streams and user-generated text data containing geographic tags, which match a target geographic range from a distributed data lake; generating a credibility index of each source data; constructing a dynamic desensitization strategy matrix based on the credibility index and the sensitive level of the use purpose declaration; performing spatial granularity generalization, attribute field masking and timestamp fuzzification processing on the four types of source data according to the dynamic desensitization strategy matrix to generate a minimum necessary geographic data package; and after the minimum necessary geographic data package passes spatial topological consistency verification, encapsulating the minimum necessary geographic data package into a standardized geographic exchange format and pushing the minimum necessary geographic data package to the external requester, so that the accuracy of cross-source exchange of geographic big data is improved.

Description

Technical Field

[0001] This invention relates to the field of geographic big data exchange technology, specifically to an information exchange system and method for geographic information and big data. Background Technology

[0002] With the rapid development of geographic information big data, IoT sensing technology and spatial sharing services, multi-source heterogeneous geographic data such as remote sensing imagery, vector geographic features, IoT sensor streams, and user-generated geographic text have been widely used in many fields such as urban governance, resource exploration, emergency dispatch, and public services. Distributed data lake architecture, with its advantages of massive storage and multi-source aggregation, can realize the centralized collection and unified management of various types of geographic big data, providing a data foundation for cross-departmental and cross-platform geographic data sharing and exchange.

[0003] Current traditional geographic data exchange models mostly adopt a crude approach with fixed permissions and uniform desensitization intensity, which has many technical shortcomings in practical applications. First, existing technologies cannot perform differentiated and quantifiable assessments of the reliability of multi-source geographic data. Geographic data with different collection precision, different source channels, and different update times are shared in a mixed manner. Low-reliability, abnormal, and distorted data can easily flow into the sharing link, resulting in poor accuracy and low reliability of geographic data exchange results. Second, traditional desensitization strategies are mostly configured with fixed rules and cannot be dynamically and adaptively adjusted according to data reliability, data usage purpose, and application scenario sensitivity level. There is a two-way problem: insufficient desensitization intensity in high-sensitivity scenarios may lead to data leakage, while excessive desensitization in low-sensitivity scenarios may lead to loss of data usability value. It is difficult to meet the principle of minimum necessary data security sharing. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a geographic information and big data information exchange system and method, which solves the problems described above.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for exchanging geographic information and big data, comprising:

[0007] Receive geographic data sharing requests from external requesters, the requests including the target geographic area, the required data type, and a statement of purpose for use;

[0008] The system retrieves four types of source data from the distributed data lake that match the target geographic area: remote sensing images, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags.

[0009] Geographic context credibility scoring is performed on the four types of source data respectively to generate a credibility index for each source data.

[0010] Based on the credibility index and the sensitivity level of the stated purpose of use, a dynamic desensitization strategy matrix is ​​constructed;

[0011] Based on the dynamic desensitization strategy matrix, the four types of source data are subjected to spatial granularity generalization, attribute field masking, and timestamp fuzzification to generate the minimum necessary geographic data package;

[0012] After the minimum necessary geographic data packet passes the spatial topology consistency check, it is encapsulated into a standardized geographic exchange format and pushed to the external requester.

[0013] Optionally, the step of performing geographic context credibility scoring on the four types of source data to generate a credibility index for each source data includes:

[0014] For remote sensing images, the first confidence component is calculated based on their cloud cover and geometric positioning deviation;

[0015] For vector geographic features, a second confidence component is assessed based on their most recent quality inspection results and feature integrity.

[0016] For IoT sensor streams, a third confidence component is generated based on sensor calibration status, reporting stability, and position drift.

[0017] For user-generated text data containing geographic tags, a fourth credibility component is generated based on the location accuracy of the geographic tags, the distribution of keywords extracted from the text, sentiment indicators, and the identity authentication status of the publisher.

[0018] The first to fourth confidence components are weighted and fused according to a preset priority rule to obtain the confidence index of the corresponding source data.

[0019] Optionally, the step of constructing a dynamic desensitization strategy matrix based on the credibility index and the sensitivity level of the use purpose declaration includes:

[0020] Parse the stated purpose of use, match it with preset usage classification rules, and determine the sensitivity level identifier;

[0021] Based on the sensitivity level identifier, set the corresponding credibility admission conditions and desensitization strength level;

[0022] If the credibility index of any source data does not meet the admission criteria corresponding to the current sensitivity level, the source data shall be prohibited from participating in the exchange or subjected to corresponding levels of desensitization processing.

[0023] Optionally, the step of performing spatial granularity generalization, attribute field masking, and timestamp blurring on the four types of source data based on the dynamic desensitization strategy matrix to generate a minimum necessary geographic data package includes:

[0024] For point features in vector geographic features, their credibility index is mapped to the corresponding hierarchical units in a preset-level geospatial grid system to achieve spatial generalization.

[0025] For pixels in high-density areas of remote sensing images, pixel-level masking is performed on the matched feature outlines based on a pre-defined list of sensitive feature types.

[0026] The timestamps of IoT sensor streams are superimposed with random noise that meets preset privacy budget constraints to achieve time fuzzification;

[0027] The identifiable identity information and precise coordinate information in the user-generated text data containing geographic tags are replaced with irreversible regional aggregation codes.

[0028] Optionally, the step of encapsulating the minimum necessary geographic data packet into a standardized geographic exchange format and pushing it to the external requester after the minimum necessary geographic data packet passes the spatial topology consistency check includes:

[0029] Verify whether the generalized vector geographic features meet the preset spatial topology rules;

[0030] Verify whether the spatial alignment error between the anonymized remote sensing image and the vector boundary is within the allowable range;

[0031] After passing all consistency checks, the minimum necessary geographic data packets are converted into an exchange format that conforms to the Open Geospatial Consortium standard;

[0032] An invisible digital watermark and structured usage permission control metadata are added, the package is completed, and the data is pushed to the external requester.

[0033] Optionally, before generating the minimum necessary geographic data packet, the following steps are also included:

[0034] Detect whether the target geographic area has spatial overlap with the preset geographic information security control area;

[0035] When there is spatial overlap, the sensitivity level identifier of the stated purpose of use is increased, and the dynamic desensitization strategy matrix is ​​reconstructed based on the increased sensitivity level.

[0036] Optionally, the mapping of its credibility index to the corresponding hierarchical unit in a preset-level geospatial grid system includes:

[0037] Establish a mapping relationship between credibility index intervals and geospatial grid levels;

[0038] When the credibility index of a point element is in the high credibility range, its original spatial coordinates are retained.

[0039] When it is in a medium confidence interval, it is mapped to the center location of a medium-granularity geographic grid;

[0040] When in a low confidence range, it is mapped to the center location of a coarse-grained geographic grid.

[0041] Optionally, the method further includes:

[0042] Record the requester's identity information, target geographical range, de-identification strategy used, and unique identifier of the output data packet in each data exchange process;

[0043] The records are written into a distributed ledger system with tamper-proof features to form a traceable audit chain for geographic data exchange.

[0044] Optionally, generating the fourth confidence component includes:

[0045] The basic confidence level of a geographic tag is determined based on its positioning method. Tags based on satellite positioning with a positioning error less than a preset threshold are given a higher confidence level, while tags based on base station positioning or obtained by manually inputting place names by the user are given a lower confidence level.

[0046] For tags obtained based on place name resolution, their confidence level is further adjusted according to the number of candidate matches returned by the geocoding service; the fewer the number of candidates, the higher the confidence level.

[0047] A geographic information and big data information exchange system, the system comprising: The geographic sharing request parsing and receiving module receives geographic data sharing requests from external requesters. The requests include the target geographic area, the required data type, and a declaration of the purpose of use. The multi-source geographic data distributed retrieval module retrieves four types of source data from the distributed data lake that match the target geographic area: remote sensing images, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags. The geographic context credibility scoring module performs geographic context credibility scoring on the four types of source data respectively, and generates a credibility index for each source data. The dynamic desensitization strategy matrix construction module constructs a dynamic desensitization strategy matrix based on the credibility index and the sensitivity level of the use purpose declaration. The minimum necessary geographic data desensitization processing module performs spatial granularity generalization, attribute field masking, and timestamp fuzzification processing on the four types of source data according to the dynamic desensitization strategy matrix to generate a minimum necessary geographic data package. The compliance verification and standardized push output module encapsulates the minimum necessary geographic data packet into a standardized geographic exchange format and pushes it to the external requester after the spatial topology consistency verification is passed.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention constructs a differentiated credibility quantification scoring system for multi-source geographic data. For four types of source data—remote sensing imagery, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags—it matches exclusive credibility assessment dimensions to each. Combining multiple dimensions such as image quality, feature completeness, equipment calibration status, tag positioning accuracy, and publishing qualifications, it generates credibility components and weights them together to achieve accurate quantification of the credibility of various heterogeneous geographic data. This effectively identifies low-credibility, distorted, and abnormal data, thereby improving the overall reliability and accuracy of geographic data exchange from the source.

[0049] 2. This invention establishes a dynamic de-identification strategy matrix based on data credibility and sensitivity level of usage scenarios. Unlike fixed de-identification rules, it can adaptively adjust the de-identification intensity and processing logic according to the data usage purpose, regional security level, and data credibility. It implements differentiated protection for geographic data with different credibility levels and different sensitivity scenarios, avoiding the risk of data leakage in highly sensitive scenarios and preventing the loss of data value caused by excessive de-identification in low-sensitivity scenarios. This invention designs exclusive and refined de-identification processing logic for four types of heterogeneous geographic data, realizing comprehensive de-identification processing of spatial granularity generalization, attribute field masking, timestamp blurring, and privacy information encoding replacement. At the same time, it combines the credibility level to adaptively match the spatial grid generalization granularity, taking into account both data de-identification security and geospatial feature integrity. By superimposing noise processing with privacy budget constraints, pixel-level sensitive masking, and irreversible encoding replacement, it avoids the risk of leakage of accurate geographic information and user privacy information, and adapts to the needs of secure sharing of multiple types of geographic data. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating the steps of the present invention. Figure 2 This is a system architecture diagram of the present invention. Detailed Implementation

[0051] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0052] Example 1: Refer to Figure 1 As shown, a method for exchanging geographic information and big data includes:

[0053] Step 101: Receive a geographic data sharing request from an external requester, the request including the target geographic area, the required data type, and a statement of purpose for use;

[0054] Step 102: Retrieve four types of source data from the distributed data lake that match the target geographic area: remote sensing images, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags.

[0055] Step 103: Perform geographic context credibility scoring on the four types of source data respectively to generate a credibility index for each source data.

[0056] Step 104: Based on the credibility index and the sensitivity level of the use purpose declaration, construct a dynamic desensitization strategy matrix;

[0057] Step 105: Based on the dynamic desensitization strategy matrix, perform spatial granularity generalization, attribute field masking, and timestamp fuzzification on the four types of source data to generate the minimum necessary geographic data package;

[0058] Step 106: After the minimum necessary geographic data packet passes the spatial topology consistency check, it is encapsulated into a standardized geographic exchange format and pushed to the external requester.

[0059] Through the above technical solutions, this invention addresses the technical shortcomings of traditional geographic data sharing and exchange, such as fixed desensitization rules, inability to quantify data credibility, insufficient differentiation in processing multi-source heterogeneous data, inability to balance sharing security and availability, and lack of traceability in data exchange. It achieves differentiated credibility assessment of multi-source geographic data, adaptive matching of scene sensitivity levels, dynamic layered desensitization, topology compliance verification, and full-process audit traceability, effectively improving the security, accuracy, compliance, and traceability of geographic big data sharing and exchange, and realizing secure and controllable data sharing under the principle of minimum necessity.

[0060] In some embodiments, performing geographic context credibility scoring on the four types of source data to generate a credibility index for each source data includes: for remote sensing imagery, calculating a first credibility component based on its cloud coverage and geometric positioning deviation; for vector geographic features, evaluating a second credibility component based on its most recent quality inspection results and feature integrity; for IoT sensor streams, generating a third credibility component based on sensor calibration status, reporting stability, and location drift; for user-generated text data containing geographic tags, generating a fourth credibility component based on the positioning accuracy of the geographic tags, the distribution of keywords extracted from the text, sentiment indicators, and the identity authentication status of the publisher; and weighting and fusing the first to fourth credibility components according to a preset priority rule to obtain the credibility index of the corresponding source data.

[0061] Specifically, in this invention, all credibility components adopt a unified calculation logic of index normalization + reverse mapping + weighted summation to uniformly map all original physical indicators to the credibility range of 0 to 1. The larger the value, the higher the credibility of the data.

[0062] The calculation process for the first confidence component of remote sensing imagery is as follows: First, the original cloud cover ratio and original geometric positioning deviation index of the remote sensing imagery are collected. The cloud cover ratio is the ratio of the total number of cloud pixels to the total number of pixels in the image, and the geometric positioning deviation is the average offset distance between the image feature pixels and the real geographic coordinates. The two original indices are compressed to the standard range of 0-1 using the maximum-minimum normalization algorithm to obtain the normalized cloud cover coefficient and the normalized deviation coefficient. Since the larger the cloud cover and the larger the positioning deviation, the lower the data confidence, a reverse mapping calculation method is adopted. The two reverse coefficients are weighted and summed by preset cloud cover weight and deviation weight to finally obtain the first confidence component in the range of 0-1. The smaller the cloud cover and the more accurate the geometric positioning, the higher the value of the first confidence component.

[0063] The calculation process for the second credibility component of vector geographic features is as follows: A quality inspection timeliness scoring interval and a feature integrity scoring interval are pre-set. The most recent quality inspection time of the vector data is quantified for timeliness; the closer the quality inspection time is to the current time, the higher the timeliness quantification score. Integrity checks are performed on the vector features, counting the number of voids, broken lines, and missing attribute fields, and calculating the feature integrity rate, which is the ratio of the number of valid complete features to the total number of features. The quality inspection timeliness score and feature integrity rate are normalized to the 0-1 range, and then linearly weighted and fused according to preset quality inspection weights and integrity weights to obtain the second credibility component. The higher the quality inspection level and the higher the feature integrity, the higher the second credibility component.

[0064] The calculation process for the third credibility component of IoT sensor streams is as follows: Three indicators are quantified sequentially: sensor calibration status, data reporting stability, and location drift. Calibration status is quantified by calibration cycle deviation, with high scores assigned to on-time calibration and low scores to overdue calibration. Reporting stability is quantified by data packet loss rate per unit time, with lower packet loss rates resulting in higher stability scores. Location drift is quantified by the spatial distance deviation between the real-time sensor location and the reference deployment location, with smaller drift distances resulting in higher scores. These three indicators are normalized to the 0-1 range and weighted and summed according to preset device weights, stability weights, and drift weights to obtain the third credibility component, which fully characterizes the real-time credibility of the sensor data.

[0065] The calculation process for the fourth credibility component of user-generated text data containing geographic tags is as follows: Four indicators are quantified respectively: geographic tag positioning accuracy, text keyword matching degree, text sentiment tendency, and publisher authentication level. Positioning accuracy is quantified by coordinate error distance; keyword distribution is quantified by the proportion of effective geographic semantic keyword matching; sentiment tendency is quantified by the proportion of positive objective text and negative subjective interference text; and authentication status is quantified by real-name / anonymous level. The four indicators are normalized to the standard range of 0-1, and then weighted and fused according to preset positioning weight, text quality weight, and user qualification weight to obtain the initial fourth credibility component.

[0066] The final credibility index fusion calculation process is as follows: preset the component priority weight coefficients corresponding to the four types of data sources, and different types of data sources correspond to exclusive weight ratios to ensure that the core evaluation indicators account for a higher proportion; multiply the normalized first, second, third, and fourth credibility components by their corresponding priority weights and sum them up to obtain the initial fusion score; then perform secondary extreme value constraint and normalization processing on the initial fusion score to eliminate abnormal extreme value interference, and finally generate a unified credibility index that is strictly in the range of 0 to 1. The higher the value, the stronger the credibility, authenticity, and usability of the corresponding source data in the geographic context.

[0067] Furthermore, the generation of the fourth confidence component also includes: determining the basic confidence level of the geographic tag based on its positioning method, wherein tags based on satellite positioning with a positioning error less than a preset threshold are given a higher confidence level, and tags based on base station positioning or obtained by manually inputting place names are given a lower confidence level; for tags obtained based on place name resolution, the confidence level is further adjusted according to the number of candidate matches returned by the geocoding service, the fewer the number of candidates, the higher the confidence level.

[0068] Specifically, the basic confidence level grading calculation process is as follows: a positioning error threshold is preset, and the original positioning error of the satellite positioning tag is compared with the difference. When the error is less than the threshold, a preset high basic confidence score is assigned; when the error is greater than the threshold, the score is assigned linearly decreasing according to the error gradient. For base station positioning and manual place name resolution tags, a benchmark low confidence score is uniformly assigned. The confidence level fine-tuning calculation process is as follows: the total number of place name candidate matches returned by the geocoding interface is counted, and a preset candidate number range is used as the judgment standard. The fine-tuning coefficient is calculated through an inverse proportional mapping formula. The fewer the number of candidate matches, the larger the fine-tuning coefficient. Finally, the fine-tuning coefficient is superimposed on the basic confidence score to obtain the corrected fourth confidence component, thereby realizing the refined grading and quantitative calculation of the confidence of crowdsourced text geographic data.

[0069] In some embodiments, constructing a dynamic desensitization strategy matrix based on the credibility index and the sensitivity level of the use purpose declaration includes: parsing the use purpose declaration, matching it with preset usage classification rules, and determining a sensitivity level identifier; setting corresponding credibility access conditions and desensitization intensity levels according to the sensitivity level identifier; and prohibiting the source data from participating in the exchange or applying desensitization processing of corresponding intensity when the credibility index of any source data does not meet the access conditions corresponding to the current sensitivity level.

[0070] Specifically, the sensitivity level determination calculation process is as follows: A multi-dimensional scenario semantic lexicon and level mapping rules are pre-constructed; the user-input usage purpose declaration is segmented, keywords are extracted, and semantic vectors are calculated; the text semantic vector is compared with the vectors of each preset scenario template; the scenario category corresponding to the maximum similarity is taken as the final usage classification; and the corresponding four sensitivity level identifiers (ordinary, intermediate, advanced, and top secret) are matched. The admission criteria and desensitization strength matching calculation process is as follows: A corresponding credibility admission threshold range and desensitization strength parameter range are preset for each sensitivity level. The higher the confidence index, the higher the access confidence threshold and the greater the desensitization strength coefficient. The confidence index of each of the four types of source data is compared with the current level access threshold. If the confidence index is less than the access threshold, the data is determined not to meet the exchange conditions and the exchange is directly prohibited. If the access conditions are met, the accurate desensitization strength coefficient is calculated by linear interpolation based on the ratio of the difference between the confidence index and the threshold. Finally, a complete dynamic desensitization strategy matrix is ​​constructed with four-dimensional parameters: data type, confidence index range, scene sensitivity level, and desensitization strength coefficient, so as to realize the quantification, adaptability, and computable matching of desensitization rules.

[0071] In some embodiments, before generating the minimum necessary geographic data packet, the method further includes: detecting whether the target geographic range has spatial intersection with a preset geographic information security control area; when spatial intersection exists, increasing the sensitivity level identifier of the purpose declaration, and reconstructing the dynamic desensitization strategy matrix based on the increased sensitivity level. Specifically, the spatial intersection determination calculation process is as follows: The set of vector boundary coordinates of the user's requested target geographic area and the set of vector boundary coordinates of the preset security control area are obtained respectively. A surface-to-surface overlay topology calculation algorithm is used to traverse the intersection and containment relationships of the two sets of boundary coordinates, calculating the overlap ratio of the overlapping area to the total area of ​​the requested area. When the overlap ratio is greater than a preset overlap threshold, a valid spatial intersection is determined, triggering a sensitivity level upgrade mechanism. The level upgrade and strategy reconstruction process is as follows: Multi-level upgrade rules are set according to the overlap ratio gradient; the higher the overlap ratio, the higher the upgrade level. After the upgrade is completed, the current scene sensitivity level identifier is updated, the corresponding admission threshold and desensitization strength parameters are replaced, the credibility comparison and desensitization coefficient calculation are re-executed, and an updated dynamic desensitization strategy matrix is ​​iteratively generated to achieve adaptive enhanced protection of sensitive areas.

[0072] In some embodiments, the step of performing spatial granularity generalization, attribute field masking, and timestamp blurring on the four types of source data according to the dynamic desensitization strategy matrix to generate a minimum necessary geographic data package includes: mapping point features in vector geographic features to corresponding hierarchical units in a preset-level geospatial grid system based on their credibility index to achieve spatial generalization; performing pixel-level masking on the outlines of matched features in high-density areas of remote sensing images based on a preset list of sensitive feature types; superimposing random noise that meets preset privacy budget constraints on the timestamps of IoT sensor streams to achieve temporal blurring; and replacing identifiable identity information and precise coordinate information in user-generated text data containing geographic tags with irreversible regional aggregation codes.

[0073] Specifically, the complete calculation process for spatial generalization of vector point features is as follows: A one-to-one mapping data table between confidence index intervals and grid granularity levels is pre-established, dividing confidence intervals into high, medium, and low categories, corresponding to a 10m fine grid, a 50m medium grid, and a 100m coarse-grained grid, respectively; the confidence index of the vector point feature is obtained, and its confidence level is determined through interval comparison, matching the corresponding spatial grid level; wherein, mapping the confidence index to the corresponding hierarchical unit in the preset-level geospatial grid system includes: establishing a mapping relationship between confidence index intervals and geospatial grid levels; when the confidence index of a point feature is in the high confidence interval, its original spatial coordinates are retained; when it is in the medium confidence interval, it is mapped to the center position of the medium-grained geographic grid; when it is in the low confidence interval, it is mapped to the center position of the coarse-grained geographic grid.

[0074] The grid coordinate mapping calculation process is as follows: high-confidence interval data directly reuses the original latitude and longitude plane coordinates; medium and low-confidence interval data calculate the row and column numbers of the corresponding grid cells according to the matched grid granularity, and then use the grid row and column numbers to solve the plane coordinates of the grid center point in reverse, replacing the original precise coordinates with the grid center coordinates to complete spatial generalization; in this embodiment, a multi-level grid system is preset, including 10m, 50m, and 100m granularity levels. High-confidence interval data has high authenticity and no risk of distortion, and retains the original precise coordinates; medium-confidence data has slight deviations, and is mapped to the center of the 50m grid to achieve moderate generalization; low-confidence data has large deviations and high security risks, and is mapped to the 100m coarse-grained grid to completely obscure the precise point information, balancing data availability and security.

[0075] The pixel-level masking calculation process for remote sensing images is as follows: read the preset list of sensitive land cover types, use a semantic segmentation algorithm to classify the features of each pixel in the remote sensing image, calculate the similarity between each pixel and the feature template of the sensitive land cover, and filter out sensitive pixel areas with similarity higher than the preset matching threshold; count the contour boundary coordinates of the sensitive areas, perform full-coverage assignment masking processing on all pixels inside the contour, uniformly replace them with general surface pixel values, completely eliminate classified visual information, and retain only non-sensitive general surface textures.

[0076] The process of fuzzifying the timestamp of IoT sensor streams is as follows: A differential privacy strict privacy budget constraint mechanism is adopted, and the maximum privacy budget threshold for a single data exchange is preset; based on the privacy budget, the amplitude range of random noise is calculated in reverse, and small-amplitude random noise conforming to a Gaussian distribution is generated within the range. The noise values ​​are linearly superimposed on the original precise timestamp to complete the timestamp perturbation calculation; at the same time, the slope deviation of the time series curve after noise superposition is constrained to be within a preset allowable range to ensure that the overall time series change trend remains unchanged. Under the premise of satisfying differential privacy security constraints, time information fuzzification is achieved, and precise time series data leakage is prevented.

[0077] The user-generated text data encoding replacement calculation process is as follows: Precise latitude and longitude coordinates, user identification, device number, and other privacy fields are extracted from the text. Based on the desensitization level corresponding to the desensitization strategy matrix, the corresponding level of regional aggregation encoding rules is matched. According to the administrative grid division standard, precise single-point coordinates are mapped to corresponding block codes. A fixed aggregation code is generated through an irreversible hash algorithm. The code retains only macro-level regional information and cannot reverse-parse the original precise location and user identity, completely eliminating the risk of privacy leakage. Through the above multi-dimensional, quantifiable, and differentiated desensitization processing, the minimum necessary geographic data package is finally generated, retaining only the necessary business information and without sensitive redundant content.

[0078] In some embodiments, the step of encapsulating the minimum necessary geographic data packet into a standardized geographic exchange format and pushing it to the external requester after the minimum necessary geographic data packet passes the spatial topology consistency verification includes: verifying whether the generalized vector geographic features meet the preset spatial topology rules; verifying whether the spatial alignment error between the anonymized remote sensing image and the vector boundary is within the allowable range; after passing all consistency verifications, converting the minimum necessary geographic data packet into an exchange format that conforms to the Open Geospatial Consortium standard; attaching an invisible digital watermark and structured access control metadata, completing the encapsulation and pushing it to the external requester.

[0079] Specifically, the vector topology consistency verification calculation process is as follows: a preset vector topology compliance threshold system is established, including a threshold for overlapping area of ​​surfaces, a threshold for area of ​​voids in surfaces, a threshold for the number of line segment intersections, and a threshold for boundary break distance. Topology traversal calculations are performed on each desensitized vector element, and the quantitative index values ​​of various topology errors are statistically analyzed. The measured values ​​are compared with the corresponding preset thresholds. When all measured values ​​are less than the threshold, topology compliance is determined. If the values ​​exceed the threshold, topology anomalies are determined, and topology errors are automatically corrected or abnormal elements are removed to ensure the integrity and compliance of the vector data structure. The image and vector alignment error calculation process is as follows: multiple sets of image and vector corresponding feature points are uniformly selected, and the planar Euclidean distance between the image pixel coordinates and the vector coordinates is calculated point by point. The average alignment error of multiple sets of distances is obtained, and the average alignment error is compared with a preset pixel error threshold. If the error is within the allowable range, spatial alignment compliance is determined, avoiding raster vector misalignment and data distortion after desensitization.

[0080] After all verifications are passed, heterogeneous multi-source geographic data is uniformly converted into a universal exchange format that conforms to the OGC Open Geospatial Consortium standard, adapting to the access and parsing requirements of various geographic information platforms and business systems, and improving cross-platform exchange compatibility. At the same time, an invisible anti-counterfeiting digital watermark is embedded in the data packet, binding the unique data identifier, sharing subject, and usage scenario information to prevent data tampering and illegal transfer. Structured access control metadata is added to limit the scope of use, forwarding permissions, and validity period of the data packet, so as to achieve controllable permissions after data sharing. Finally, standardized encapsulation is completed and pushed to external requesters.

[0081] In some embodiments, the method further includes: recording the requester's identity information, target geographical range, desensitization strategy adopted, and unique identifier of the output data packet in each data exchange process; and writing the records into a distributed ledger system with tamper-proof characteristics to form a traceable audit chain for geographic data exchange. Specifically, the system records every geographic data sharing and exchange activity in a structured, full-element log, retaining core information such as the requesting entity's identity, requested geographic area, data type, anonymization strategy parameters, unique data packet ID, exchange time, and target recipient. It then performs hash digest calculations on the log data to generate a unique data fingerprint. The plaintext logs and hash fingerprints are stored together on the blockchain in a distributed ledger. Leveraging the immutability and end-to-end traceability of blockchain, the system solidifies the record of each exchange activity, forming a continuous, complete, and verifiable audit chain. This enables the entire geographic data sharing process to be monitored, traceable, and accountable, completely resolving the problems of no data retention, difficulty in traceability, and ambiguous responsibility in traditional data exchange.

[0082] Example 2: Refer to Figure 2 As shown, the present invention also provides an information exchange system for geographic information and big data, comprising: The geographic data sharing request parsing and receiving module is used to receive geographic data sharing requests from external requesters. The requests include the target geographic range, the required data type, and a declaration of the purpose of use. It is responsible for the access verification, parameter parsing, and structured storage of external requests, providing input basis for subsequent data retrieval and strategy generation. The multi-source geographic data distributed retrieval module is used to retrieve four types of source data from the distributed data lake that match the target geographic area: remote sensing images, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags; based on spatial indexing and type filtering mechanisms, it realizes accurate collection of multi-source heterogeneous geographic data and removal of redundant data. The geographic context credibility scoring module is used to perform geographic context credibility scoring on the four types of source data respectively, and generate a credibility index for each source data. It configures exclusive evaluation dimensions for different data types, and achieves accurate quantitative classification of data credibility through a complete quantitative calculation process of multi-indicator normalization, weighted fusion, and extreme value correction. The dynamic desensitization strategy matrix construction module is used to construct a dynamic desensitization strategy matrix based on the credibility index and the sensitivity level of the use purpose declaration; it realizes semantic quantification of scene sensitivity level, comparison of credibility access threshold, and interpolation calculation of desensitization intensity coefficient to generate a quantified and executable differentiated desensitization rule matrix. The minimum necessary geographic data desensitization module is used to perform spatial granularity generalization, attribute field masking, and timestamp blurring on the four types of source data according to the dynamic desensitization strategy matrix to generate minimum necessary geographic data packages; and to perform fine-grained quantitative desensitization calculations such as grid mapping calculation, pixel masking determination, differential privacy noise superposition, and irreversible encoding replacement on the four types of data to balance data security and usability. The compliance verification and standardized push output module is used to encapsulate the minimum necessary geographic data packet into a standardized geographic exchange format and push it to the external requester after the spatial topology consistency verification is passed; it completes topology error quantification verification, spatial alignment error calculation, standardized format conversion, watermark hash calculation, permission metadata encapsulation and secure push, and at the same time, it links with the audit module to retain exchange records.

[0083] In summary, this invention addresses the shortcomings of traditional geographic big data sharing, such as rigid desensitization, unquantifiable data quality, insufficient security protection, and untraceable exchange behavior, through an integrated technical solution that combines multi-source geographic data credibility quantification assessment, scenario-adaptive dynamic desensitization, differentiated and refined data quantification processing, spatial topology error compliance verification, standardized secure output, and distributed ledger traceability. It significantly improves the security, accuracy, compliance, and versatility of geographic information big data exchange, and is suitable for the secure sharing needs of geographic data in various government, scientific research, and public service scenarios.

[0084] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for exchanging geographic information and big data, characterized in that, include: Receive geographic data sharing requests from external requesters, the requests including the target geographic area, the required data type, and a statement of purpose for use; The system retrieves four types of source data from the distributed data lake that match the target geographic area: remote sensing images, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags. Geographic context credibility scoring is performed on the four types of source data respectively to generate a credibility index for each source data. Based on the credibility index and the sensitivity level of the stated purpose of use, a dynamic desensitization strategy matrix is ​​constructed; Based on the dynamic desensitization strategy matrix, the four types of source data are subjected to spatial granularity generalization, attribute field masking, and timestamp fuzzification to generate the minimum necessary geographic data package; After the minimum necessary geographic data packet passes the spatial topology consistency check, it is encapsulated into a standardized geographic exchange format and pushed to the external requester.

2. The method for exchanging geographic information and big data according to claim 1, characterized in that, The process of performing geographic context credibility scoring on the four types of source data to generate a credibility index for each source data includes: For remote sensing images, the first confidence component is calculated based on their cloud cover and geometric positioning deviation; For vector geographic features, a second confidence component is assessed based on their most recent quality inspection results and feature integrity. For IoT sensor streams, a third confidence component is generated based on sensor calibration status, reporting stability, and position drift. For user-generated text data containing geographic tags, a fourth credibility component is generated based on the location accuracy of the geographic tags, the distribution of keywords extracted from the text, sentiment indicators, and the identity authentication status of the publisher. The first to fourth confidence components are weighted and fused according to a preset priority rule to obtain the confidence index of the corresponding source data.

3. The method for exchanging geographic information and big data according to claim 1, characterized in that, The construction of a dynamic desensitization strategy matrix based on the credibility index and the sensitivity level of the stated purpose of use includes: Parse the stated purpose of use, match it with preset usage classification rules, and determine the sensitivity level identifier; Based on the sensitivity level identifier, set the corresponding credibility admission conditions and desensitization strength level; If the credibility index of any source data does not meet the admission criteria corresponding to the current sensitivity level, the source data shall be prohibited from participating in the exchange or subjected to corresponding levels of desensitization processing.

4. The method for exchanging geographic information and big data according to claim 1, characterized in that, Based on the dynamic desensitization strategy matrix, the four types of source data are subjected to spatial granularity generalization, attribute field masking, and timestamp fuzzification to generate a minimum necessary geographic data package, including: For point features in vector geographic features, their credibility index is mapped to the corresponding hierarchical units in a preset-level geospatial grid system to achieve spatial generalization. For pixels in high-density areas of remote sensing images, pixel-level masking is performed on the matched feature outlines based on a pre-defined list of sensitive feature types. The timestamps of IoT sensor streams are superimposed with random noise that meets preset privacy budget constraints to achieve time fuzzification; The identifiable identity information and precise coordinate information in the user-generated text data containing geographic tags are replaced with irreversible regional aggregation codes.

5. The method for exchanging geographic information and big data according to claim 1, characterized in that, The step of encapsulating the minimum necessary geographic data packet into a standardized geographic exchange format and pushing it to the external requester after the minimum necessary geographic data packet passes the spatial topology consistency check includes: Verify whether the generalized vector geographic features meet the preset spatial topology rules; Verify whether the spatial alignment error between the anonymized remote sensing image and the vector boundary is within the allowable range; After passing all consistency checks, the minimum necessary geographic data packets are converted into an exchange format that conforms to the Open Geospatial Consortium standard; An invisible digital watermark and structured usage permission control metadata are added, the package is completed, and the data is pushed to the external requester.

6. The method for exchanging geographic information and big data according to claim 1, characterized in that, Before generating the minimum necessary geographic data packet, the following is also included: Detect whether the target geographic area has spatial overlap with the preset geographic information security control area; When there is spatial overlap, the sensitivity level identifier of the stated purpose of use is increased, and the dynamic desensitization strategy matrix is ​​reconstructed based on the increased sensitivity level.

7. The method for exchanging geographic information and big data according to claim 4, characterized in that, The mapping of corresponding hierarchical units in a preset-level geospatial grid system based on their credibility index includes: Establish a mapping relationship between credibility index intervals and geospatial grid levels; When the credibility index of a point element is in the high credibility range, its original spatial coordinates are retained. When it is in a medium confidence interval, it is mapped to the center location of a medium-granularity geographic grid; When in a low confidence range, it is mapped to the center location of a coarse-grained geographic grid.

8. The method for exchanging geographic information and big data according to claim 1, characterized in that, The method further includes: Record the requester's identity information, target geographical range, de-identification strategy used, and unique identifier of the output data packet in each data exchange process; The records are written into a distributed ledger system with tamper-proof features to form a traceable audit chain for geographic data exchange.

9. A method for exchanging geographic information and big data according to claim 2, characterized in that, The generation of the fourth confidence component includes: The basic confidence level of a geographic tag is determined based on its positioning method. Tags based on satellite positioning with a positioning error less than a preset threshold are given a higher confidence level, while tags based on base station positioning or obtained by manually inputting place names by the user are given a lower confidence level. For tags obtained based on place name resolution, their confidence level is further adjusted according to the number of candidate matches returned by the geocoding service; the fewer the number of candidates, the higher the confidence level.

10. A geographic information and big data information exchange system, used to implement the geographic information and big data information exchange method according to any one of claims 1 to 9, characterized in that, The system includes: The geographic sharing request parsing and receiving module receives geographic data sharing requests from external requesters. The requests include the target geographic area, the required data type, and a declaration of the purpose of use. The multi-source geographic data distributed retrieval module retrieves four types of source data from the distributed data lake that match the target geographic area: remote sensing images, vector geographic features, IoT sensor streams, and user-generated text data with geographic tags. The geographic context credibility scoring module performs geographic context credibility scoring on the four types of source data respectively, and generates a credibility index for each source data. The dynamic desensitization strategy matrix construction module constructs a dynamic desensitization strategy matrix based on the credibility index and the sensitivity level of the use purpose declaration. The minimum necessary geographic data desensitization processing module performs spatial granularity generalization, attribute field masking, and timestamp fuzzification processing on the four types of source data according to the dynamic desensitization strategy matrix to generate a minimum necessary geographic data package. The compliance verification and standardized push output module encapsulates the minimum necessary geographic data packet into a standardized geographic exchange format and pushes it to the external requester after the spatial topology consistency verification is passed.

Images