Land contract management right data format automatic conversion method

Abstract

The invention discloses a land contract management right data format automatic conversion method, and relates to the technical field of geographic information, and the method comprises the following steps: comparing and analyzing a source database standard and a target database standard, and constructing a multi-dimensional mapping relation model; a GIS platform is used as a core, an automatic conversion model is constructed according to the multi-dimensional mapping relation model, and performance optimization is carried out on the model through a distributed computing architecture, a staged Hash association algorithm, a signature space mapping duplicate removal mechanism, an increment migration mechanism, a coordinate batch generation algorithm and a storage optimization strategy; and performing data inspection on the data converted by the automatic conversion model, and if the inspection is not passed, returning an error report. According to the method, efficient and automatic conversion of database formats can be realized, the integrity and availability of data in cross-platform application are guaranteed, and powerful support is provided for smooth development of real estate registration work.

Description

Technical Field

[0001] This invention relates to the field of geographic information technology, specifically to an automated method for converting land contract management right data formats. Background Technology

[0002] With the comprehensive advancement of my country's unified real estate registration system, the data accumulated by the agricultural and rural affairs departments on the confirmation and registration of rural land contractual management rights needs to be migrated to the real estate registration platform to achieve unified management of natural resources. However, the agricultural and rural affairs department's database follows the "Specification for Database of Confirmation and Registration of Rural Land Contractual Management Rights," while the real estate registration database follows the "Standard for Real Estate Registration Database." The two databases differ significantly in terms of field structure, data type, coding rules, and spatial topology.

[0003] Currently, the conversion of agricultural land rights data into real estate standards mainly relies on manual processing or conversion using single software tools. Manual processing is not only extremely inefficient but also prone to human error, making it difficult to meet the needs of large-scale data conversion. Single software tools, limited by their functionality, cannot simultaneously handle the collaborative conversion of spatial and attribute data. Especially when dealing with the massive amounts of data in county-level agricultural land rights databases—often hundreds of thousands of plots and tens of millions of boundary points—the conversion process can take weeks or even months, and the accuracy and completeness of the conversion results are difficult to guarantee. Furthermore, traditional methods lack effective data quality verification mechanisms, leading to converted data frequently being rejected for failing to meet target standards, severely impacting the efficiency of real estate registration work.

[0004] Therefore, there is an urgent need for an efficient, accurate, automated method for converting agricultural economic rights data formats that can handle massive amounts of complex spatial data, in order to solve the above-mentioned technical problems. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an automated conversion method for land contract management right data format, which addresses the shortcomings of the existing technology.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for automatically converting land contractual management right data formats includes the following steps:

[0008] Step S1: Compare and analyze the layer structure, field definition, value domain encoding, and spatial topology constraints in the source database standard and the target database standard. Construct a multidimensional mapping relationship model by establishing a correspondence table between the source layer and the target layer, a mapping table between the source field and the target field, a value domain code conversion table, and a spatial topology rule correspondence table.

[0009] Step S2: Using the GIS platform as the core, the layer correspondence table, field mapping table, code conversion table, and spatial topology rule correspondence table in the multidimensional mapping relationship model are used as input parameters. The data extraction tool, field calculation tool, spatial connection tool, and Python script are connected through the modeling tools provided by the GIS platform to construct an automated conversion model that can automatically perform batch extraction, conversion, and reconstruction of spatial and attribute data based on the above mapping relationship. The automated conversion model is optimized for performance through a distributed computing architecture, a phased hash association algorithm, a signature spatial mapping deduplication mechanism, an incremental migration mechanism, a coordinate batch generation algorithm, and storage optimization strategies.

[0010] Step S3: Perform data integrity, logical consistency, spatial topology correctness, and standard compliance checks on the data transformed by the automated transformation model. If the checks fail, an error report will be returned.

[0011] Furthermore, step S1 specifically includes the following steps:

[0012] Step S1.1: Based on the layers of the target database standard, analyze the source of its attribute information one by one and establish a one-to-many or many-to-one mapping relationship with the source database standard.

[0013] Step S1.2: Based on the layer mapping, according to the differences in field value ranges and the generation logic, establish field mapping rules and value range mapping rules respectively: Field mapping rules are defined as four categories: direct mapping, direct assignment, derived calculation, and irrelevant ignoring; Value range mapping rules are defined as a type of code conversion, which is used to handle the situation where the same semantic field in the source standard and the target standard uses different code values, and form a code conversion table.

[0014] Step S1.3: Define the spatial data topology relationship transformation rules from the source database standard to the target database standard.

[0015] Furthermore, in step S2, the phased hash association algorithm specifically includes the following steps:

[0016] Build a hash index for the smaller tables involved in the join based on the join key;

[0017] Broadcast the hash index to all compute nodes;

[0018] On each computing node, local hash matching is performed on the data.

[0019] Furthermore, in step S2, the signature space mapping deduplication mechanism specifically includes the following steps:

[0020] The fields of the records to be deduplicated are combined and mapped into an m-dimensional signature vector through multiple independent hash functions;

[0021] A random hyperplane is used to partition the m-dimensional signature vector space so that similar records fall into the same partition or adjacent partitions;

[0022] Calculate the signature vector for a new record and locate its partition. Calculate the similarity of the vector with the historical records in that partition and adjacent partitions. Determine whether it is a duplicate based on a preset threshold.

[0023] Furthermore, in step S2, the coordinate batch generation algorithm specifically includes the following steps:

[0024] Morton codes are used to sort parcel data, and spatial partitioning boundaries are adaptively determined by minimizing an objective function that includes load balancing and spatial compactness terms.

[0025] Within each segment, the vertex coordinates of the land parcels are extracted in batches using the geometric object iterator of the GIS platform;

[0026] Establish a topology reuse cache for boundary lines shared by adjacent parcels to avoid duplicate generation.

[0027] Furthermore, in step S2, the distributed computing architecture consists of a master node and computing nodes. The master node is responsible for task sharding and scheduling, the computing nodes are responsible for executing data processing tasks within the shards, the master node monitors the load of each node and dynamically allocates tasks, and finally the master node merges the results of each node and rebuilds the index.

[0028] Furthermore, step S3 specifically includes the following steps:

[0029] Step S3.1: Use data quality inspection software to check the field integrity, value range compliance, and logical consistency of the converted data;

[0030] Step S3.2: Use the topology inspection tool of the GIS platform to verify the geometric and topological relationships of the converted spatial data;

[0031] Step S3.3: Based on the errors found by the data quality inspection software and topology inspection, locate the root cause of the problem and generate an error report.

[0032] Furthermore, in step S2, the storage optimization strategy includes: storing all intermediate data and result data uniformly in a file geodatabase format, utilizing binary storage, spatial and attribute indexes to reduce disk input / output overhead.

[0033] Furthermore, in step S2, the incremental migration mechanism includes: marking data change timestamps in the source database, reading the configured time points through a Python script, extracting only the changed data with timestamps greater than those timestamps for conversion, and merging it with existing data in the target database.

[0034] Furthermore, the distributed computing architecture has a fault tolerance mechanism, in which the master node reassigns its unfinished tasks to other computing nodes when any computing node fails.

[0035] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0036] 1. This invention reduces the time complexity of multi-table association from product level to linear level through a phased hash association algorithm, and compresses the processing time from several hours to minutes; through a coordinate batch generation algorithm and GPU hardware acceleration, the generation time of boundary points is shortened, and the overall conversion efficiency is improved.

[0037] 2. The signature space mapping deduplication mechanism of this invention not only supports accurate deduplication, but also identifies similar duplicates caused by input errors through similarity thresholds, avoiding the misjudgment problem of traditional Bloom filters; the multidimensional mapping relationship model covers layers, fields, value ranges and spatial topology rules, ensuring the integrity and compliance of the converted data.

[0038] 3. This invention is based on an automated conversion model with a distributed computing architecture, which supports full-process automation from source data reading, conversion, reconstruction to target data entry, without manual intervention; it has a built-in log recording module and incremental migration mechanism to support automatic synchronization of subsequent daily changes, thereby improving the efficiency of daily change processing.

[0039] 4. The distributed cluster of this invention supports dynamic task allocation and node fault tolerance. When a computing node fails, the master node automatically reassigns tasks to ensure the stable execution of large-scale data conversion tasks. The file geodatabase storage format supports data capacity of more than 1TB, meeting the storage needs of tens of millions of boundary points.

[0040] 5. The similarity measurement formula proposed in this invention not only considers the geometric relationship between signature vectors, but also introduces global statistical information, making the similarity judgment more accurate. It is especially suitable for similar and duplicate recognition scenarios caused by input errors in agricultural economic rights data. Attached Figure Description

[0041] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0042] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0043] Figure 2 This is a flowchart of the phased hash association algorithm according to an embodiment of the present invention;

[0044] Figure 3 This is a flowchart of the signature space mapping deduplication mechanism according to an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] like Figure 1 As shown, an automated conversion method for land contractual management right data format includes the following steps:

[0047] Step S1: Compare and analyze the layer structure, field definition, value domain encoding, and spatial topology constraints in the source database standard and the target database standard. Construct a multidimensional mapping relationship model by establishing a correspondence table between the source layer and the target layer, a mapping table between the source field and the target field, a value domain code conversion table, and a spatial topology rule correspondence table.

[0048] Step S2: Using the GIS platform as the core, the layer correspondence table, field mapping table, code conversion table, and spatial topology rule correspondence table in the multidimensional mapping relationship model are used as input parameters. The data extraction tool, field calculation tool, spatial connection tool, and Python script are connected through the modeling tools provided by the GIS platform to construct an automated conversion model that can automatically perform batch extraction, conversion, and reconstruction of spatial and attribute data based on the above mapping relationship. The automated conversion model is optimized for performance through a distributed computing architecture, a phased hash association algorithm, a signature spatial mapping deduplication mechanism, an incremental migration mechanism, a coordinate batch generation algorithm, and storage optimization strategies.

[0049] Step S3: Perform data integrity, logical consistency, spatial topology correctness, and standard compliance checks on the data transformed by the automated transformation model. If the checks fail, an error report will be returned.

[0050] The source standard is the "Specification for Database of Confirmation and Registration of Rural Land Contractual Management Rights"; the target standard is the "Standard for Real Estate Registration Database"; and the agricultural economic rights data is the data of the results of confirmation and registration of rural land contractual management rights.

[0051] Step S1 specifically includes the following steps:

[0052] Step S1.1: Based on the layers of the target database standard, analyze the source of its attribute information one by one and establish a one-to-many or many-to-one mapping relationship with the source database standard.

[0053] Step S1.2: Based on the layer mapping, according to the differences in field value ranges and the generation logic, establish field mapping rules and value range mapping rules respectively: Field mapping rules are defined as four categories: direct mapping, direct assignment, derived calculation, and irrelevant ignoring; Value range mapping rules are defined as a type of code conversion, which is used to handle the situation where the same semantic field in the source standard and the target standard uses different code values, and form a code conversion table.

[0054] Step S1.3: Define the spatial data topology relationship transformation rules from the source database standard to the target database standard.

[0055] Based on the 25 relevant layers in the target standard, we analyze their information requirements one by one. For example, the basic information layer of the land parcel in the target standard has geometric information and some attributes derived from the land parcel layer of the source standard, and the ownership information comes from the contractor table and the contract management right registration book table. We establish a one-to-many or many-to-one layer correspondence relationship.

[0056] Based on layer mapping, and refined to the field level, the field mapping relationship is defined into five rules according to the differences in field value ranges and generation logic: Rule A (direct mapping): field name and value range are completely consistent; Rule B (direct assignment): field value is fixed; Rule C (code conversion): value ranges are different but can be mapped; Rule D (derived calculation): needs to be generated through an algorithm; Rule E (irrelevant ignore): fields that exist in the target standard but have no corresponding information in the source standard.

[0057] The first type is fields related to land contract management rights that can be directly loaded. The value range of these fields in the land contract management rights database is exactly the same as that in the real estate database, and no processing is required. Fields such as identification code, land parcel boundaries, and land plot code can be completed directly through ArcGIS add tools.

[0058] The second type is fields that can be directly filled in according to the "Real Estate Registration Database Standard". The content of these fields can be clearly defined according to the "Real Estate Registration Database Standard", and the content of these fields is uniform for each county. For example, the district / county code field can be directly assigned a value using a field calculation tool or a Python script.

[0059] The third type is fields filled in through conversion of land contract management right data. The "Standard Value Range of Land Contract Management Right Database" and the "Standard Value Range of Real Estate Database" are not completely the same for this type of field, but they can be converted to each other through certain algorithms. Taking the "yes" field as an example, the code for "no" for land contract management right is 2, while the code for "no" for real estate is 0. The conversion of this field can be completed by using a field calculation tool to calculate 2 as 0.

[0060] The fourth type is fields filled in according to real estate-related rules. Such fields can be filled in according to real estate-related rules. For example, the land parcel code field can be obtained from the cadastral sub-zone layer by using a spatial connection tool (6-digit county-level administrative division code, 3-digit cadastral zone code, and 3-digit cadastral sub-zone code), and then filled in with a 2-digit land parcel feature code and a 5-digit land parcel sequence number by using a Python script.

[0061] The fifth type consists of fields that are unrelated to land contract management rights and do not need to be filled in, such as plot ratio, building density, and building height limit;

[0062] Define spatial data transformation rules, such as converting the topological relationships of the source standard plot layer into rules that meet the target standard requirements for non-overlapping surfaces and seamless splicing.

[0063] In step S2, the phased hash association algorithm specifically includes the following steps:

[0064] Build a hash index for the smaller tables involved in the join based on the join key;

[0065] Broadcast the hash index to all compute nodes;

[0066] On each computing node, local hash matching is performed on the data.

[0067] In step S2, the signature space mapping deduplication mechanism specifically includes the following steps:

[0068] The fields of the records to be deduplicated are combined and mapped into an m-dimensional signature vector through multiple independent hash functions;

[0069] A random hyperplane is used to partition the m-dimensional signature vector space so that similar records fall into the same partition or adjacent partitions;

[0070] Calculate the signature vector for a new record and locate its partition. Calculate the similarity of the vector with the historical records in that partition and adjacent partitions. Determine whether it is a duplicate based on a preset threshold.

[0071] In step S2, the coordinate batch generation algorithm specifically includes the following steps:

[0072] Morton codes are used to sort parcel data, and spatial partitioning boundaries are adaptively determined by minimizing an objective function that includes load balancing and spatial compactness terms.

[0073] Within each segment, the vertex coordinates of the land parcels are extracted in batches using the geometric object iterator of the GIS platform;

[0074] Establish a topology reuse cache for boundary lines shared by adjacent parcels to avoid duplicate generation.

[0075] During the generation of the family member table, a many-to-many relationship needs to be established between the contracted land information table and the family member table. Business rules require that the final number of records in the family member table equals the number of land parcels owned by the household multiplied by the number of family members (excluding the head of household). The time complexity of a traditional nested loop join is... ,in and The number of records in the two tables is separate, which is completely infeasible when the data volume reaches millions;

[0076] like Figure 2As shown, this invention employs a phased hash association algorithm to reduce complexity to linear time. Specifically, it includes the following steps: traversing the family member table, using the contractor's code as the key, aggregating all member records under the same contractor (excluding the head of household) into a list, and constructing a hash table. The time complexity is [not specified]. In a distributed environment, Spark's broadcast variable or MPI's broadcast mechanism is used to broadcast the hash table to all compute nodes. Each node holds a read-only copy in its local memory. On each compute node, the local shards of the contracted land parcel information table are traversed, and the contractor's code is extracted from each record. This code is then processed using the hash table. Get the corresponding member list within the time limit and generate Result records, among which, This indicates the number of plots of land owned by that household. Indicates the number of members, with a total time complexity of O(n). ,in It is the theoretical lower bound for generating the result set.

[0077] If the number of records in a table is less than 1 / 10 of the memory capacity, it is considered a small table and broadcasting is used; otherwise, hash partitioning is used.

[0078] like Figure 3 As shown, in scenarios such as generating land contract management right registration information, it is necessary to avoid duplicate records of combinations such as real estate unit number + member certificate number. Traditional Bloom filters have a high false positive rate and cannot handle similar duplicates. This invention proposes a signature space mapping algorithm:

[0079] For records to be deduplicated It contains Key fields Define the signature vector for ,in, It is the k-th independent hash function applied to the field concatenation result. The value of k ranges from 1 to m, where m is the dimension of the signature space. In this embodiment, it is set to 128. The hash function used is MurmurHash3, which generates 128 independent hash values ​​through different seeds.

[0080] Divide the m-dimensional signature space into T disjoint regions. The system employs a randomized hyperplane partition: each region is composed of a set of randomly generated m-dimensional unit vectors. and dynamically adjusted thresholds Definition, that is This partitioning ensures that similar points in the signature space fall into the same or adjacent partitions with a high probability.

[0081] in, The signature vector is an m-dimensional integer vector consisting of m independent hash values ​​of the record to be deduplicated. Let represent the m-dimensional integer vector space, and let represent the set of all possible signature vectors. Let t represent the t-th randomly generated m-dimensional unit vector. This represents the threshold value for the t-th partition. Let represent the set of all signature vectors that satisfy the inequality;

[0082] For new records Calculate its signature vector Locate the partition This invention retrieves the signature vector sets of existing records from the partition and adjacent partitions, calculates the similarity, and proposes a similarity measurement method that integrates vector distance and partition sparseness to replace the traditional cosine similarity. The specific formula is as follows:

[0083] ;

[0084] in, and Represents the m-dimensional signature vector of the new record and the existing record. This represents the distance decay factor, which controls the strength of the influence of distance on similarity. This indicates the total number of existing records. Represents the signature vector Partition The number of records in the record, Indicates the similarity between the new record and the existing record;

[0085] The distance decay factor is set to 0.1 to 0.5. If the data noise is large and a certain degree of similarity and repetition is allowed, the lower limit of 0.1 can be used; if strict and accurate deduplication is required, the upper limit of 0.5 or higher can be used.

[0086] like This represents the preset threshold, which is usually set to 0.98. This threshold can be dynamically adjusted based on the historical data repetition rate. When the repetition rate is high, the threshold can be appropriately reduced to increase sensitivity. This algorithm supports the identification of similar duplicates caused by data entry errors and does not have the misjudgment problem of Bloom filters.

[0087] Let the total number of existing records be N, and the number of partitions be T. Then, each deduplication process only requires approximately [number missing] records. Compare records, i.e. Compared to full comparison Efficiency has been significantly improved.

[0088] When generating boundary points and boundary lines from parcel layers, traditional parcel-by-parcel processing is extremely inefficient. This invention proposes a batch coordinate generation algorithm based on spatial partitioning and topological reuse.

[0089] Morton codes are used to maintain spatial proximity for any parcel center point. The floating-point coordinates are converted to 32-bit integers, and then the Morton code is calculated using bit interleaving. It is obtained by interleaving the binary bits of x and y;

[0090] After sorting all land parcels according to Morton code, they need to be divided into K partitions, and the set of partition boundaries needs to be defined. The j-th partition contains the index. The objective function for sharding is to minimize the weighted sum of the load balancing and spatial compactness terms for each shard, and the specific formula is as follows:

[0091] ;

[0092] in, This represents the number of land parcels in the j-th segment. This represents the average number of parcels per plot, which is the ratio of the total number of parcels to the target number of plots. This represents the expansion degree of the j-th tract, which is the ratio of the area of ​​the smallest enclosing rectangle of all parcels in the tract to the total area of ​​the parcels within that tract, measuring spatial compactness. This represents the balance factor, used to balance load balancing and spatial compactness. This represents the expansion of all fragments. The maximum value;

[0093] A greedy algorithm is used to find the optimal partition boundary. Starting from the first parcel, parcels are added to the current partition sequentially. When the number of parcels in a partition exceeds... At that time, divide into new segments. This represents the relaxation factor, with a value of 0.2. Its function is to allow the fragment size to fluctuate around the average value, avoiding the destruction of spatial locality in pursuit of strict load balancing. Ultimately, this ensures that each fragment satisfies both computational load balancing and spatial locality, reducing the computational overhead of sharing across fragment boundary lines in the future.

[0094] Within each slice, ArcPy's arcpy.da.SearchCursor is used to read all parcel geometry objects at once, and explode_to_points=True is set to extract all vertex coordinates in batches and store them in a NumPy array to reduce API call overhead;

[0095] Establish a cache for boundary lines shared by adjacent parcels: Use the dictionary edge_cache, with the string of the coordinates of the two endpoints of the boundary line as the key and the FID of the generated boundary line as the value. When generating the boundary line of a new parcel, first check if the key corresponding to the coordinates of its two endpoints exists in edge_cache. If it exists, the FID is directly referenced. If it does not exist, it is created and added to the cache.

[0096] Coordinate transformation, distance calculation and other intensive geometric operations are offloaded to the GPU. The coordinate array is processed in batches through parallel threads. CUDA or CuPy is used to transfer the coordinate array to the GPU memory and start the kernel function to process each coordinate point in parallel.

[0097] In step S2, the distributed computing architecture consists of a master node and computing nodes. The master node is responsible for task sharding and scheduling, and the computing nodes are responsible for executing data processing tasks within the shards. The master node monitors the load of each node and dynamically allocates tasks. Finally, the master node merges the results of each node and rebuilds the index.

[0098] Deploy a distributed cluster consisting of one master node and 10-20 compute nodes: The master node encapsulates each shard as an independent task based on the spatial sharding results of S2.3.1. The task metadata includes the FID list of parcels within the shard and the data paths of the attribute tables to be associated. The master node receives heartbeats from each slave node every 10 seconds, and the heartbeats include CPU utilization, memory utilization, and current task progress. When a node is idle (CPU < 30% and no tasks), a new task is allocated from the task queue; when a node is overloaded (CPU > 80%), the allocation of new tasks is paused. RabbitMQ or Redis is used as the task queue, with the master node as the producer and slave nodes as consumers to ensure balanced task distribution. A task timeout is set (e.g., 30 minutes), and tasks that are not completed within the timeout period are re-added to the queue. When a node fails, its unfinished tasks are taken by other nodes. After each node completes its processing, the results are written to an independent temporary feature class. The master node finally merges the results using the Merge tool and calls AddSpatialIndex to rebuild the global spatial index.

[0099] Let the fragmentation and distribution time be... The maximum execution time of a node is The merger time is The total time consumed is Through load balancing Approximately linear speedup is achieved by approaching the theoretical minimum.

[0100] Step S3 specifically includes the following steps:

[0101] Step S3.1: Use data quality inspection software to check the field integrity, value range compliance, and logical consistency of the converted data;

[0102] Step S3.2: Use the topology inspection tool of the GIS platform to verify the geometric and topological relationships of the converted spatial data;

[0103] Step S3.3: Based on the errors found by the data quality inspection software and topology inspection, locate the root cause of the problem and generate an error report.

[0104] Import the converted .gdb database into the Ministry of Natural Resources' Real Estate Registration Stock Data Quality Inspection Software to check field integrity, value range compliance, logical consistency, etc.

[0105] Using the topology toolset of the GIS platform, we verified topological rules such as no overlap of parcels and that boundary lines must be covered by parcel boundaries.

[0106] Based on the error report, locate the root cause of the problem. If it is a problem with the quality of the original data, perform data cleaning. If it is a problem with the transformation algorithm, return to step S2 to optimize the model parameters or script until the data completely passes the quality inspection.

[0107] In step S2, the storage optimization strategy includes: storing all intermediate and result data in a unified file geodatabase format, and using binary storage, spatial and attribute indexes to reduce disk input / output overhead.

[0108] In step S2, the incremental migration mechanism includes: marking data change timestamps in the source database, reading the configured time points through a Python script, extracting only the changed data with timestamps greater than those timestamps for conversion, and merging it with existing data in the target database.

[0109] A last update timestamp field was added to each layer in the source database. During subsequent routine changes, a Python script was used to extract only the changed data whose timestamps were greater than the last migration time. This data was then transformed using the same algorithm as the full dataset and merged with existing data in the target database. Since the amount of routine change data is typically less than 1% of the full dataset, this resulted in an efficiency improvement of over 90%.

[0110] The distributed computing architecture has a fault tolerance mechanism. When any computing node fails, the master node reassigns its unfinished tasks to other computing nodes.

[0111] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0112] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for automatically converting land contractual management right data formats, characterized in that, Includes the following steps: Step S1: Compare and analyze the layer structure, field definition, value domain encoding, and spatial topology constraints in the source database standard and the target database standard. Construct a multidimensional mapping relationship model by establishing a correspondence table between the source layer and the target layer, a mapping table between the source field and the target field, a value domain code conversion table, and a spatial topology rule correspondence table. Step S2: Using the GIS platform as the core, the layer correspondence table, field mapping table, code conversion table, and spatial topology rule correspondence table in the multidimensional mapping relationship model are used as input parameters. The data extraction tool, field calculation tool, spatial connection tool, and Python script are connected through the modeling tools provided by the GIS platform to construct an automated conversion model that can automatically perform batch extraction, conversion, and reconstruction of spatial and attribute data based on the above mapping relationship. The automated conversion model is optimized for performance through a distributed computing architecture, a phased hash association algorithm, a signature spatial mapping deduplication mechanism, an incremental migration mechanism, a coordinate batch generation algorithm, and storage optimization strategies. Step S3: Perform data integrity, logical consistency, spatial topology correctness, and standard compliance checks on the data transformed by the automated transformation model. If the checks fail, an error report will be returned.

2. The method according to claim 1, characterized in that, Step S1 specifically includes the following steps: Step S1.1: Based on the layers of the target database standard, analyze the source of its attribute information one by one and establish a one-to-many or many-to-one mapping relationship with the source database standard. Step S1.2: Based on the layer mapping, according to the differences in field value ranges and the generation logic, establish field mapping rules and value range mapping rules respectively: Field mapping rules are defined as four categories: direct mapping, direct assignment, derived calculation, and irrelevant ignoring; Value range mapping rules are defined as a type of code conversion, which is used to handle the situation where the same semantic field in the source standard and the target standard uses different code values, and form a code conversion table. Step S1.3: Define the spatial data topology relationship transformation rules from the source database standard to the target database standard.

3. The method according to claim 2, characterized in that, In step S2, the phased hash association algorithm specifically includes the following steps: Build a hash index for the smaller tables involved in the join based on the join key; Broadcast the hash index to all compute nodes; On each computing node, local hash matching is performed on the data.

4. The method according to claim 3, characterized in that, In step S2, the signature space mapping deduplication mechanism specifically includes the following steps: The fields of the records to be deduplicated are combined and mapped into an m-dimensional signature vector through multiple independent hash functions; A random hyperplane is used to partition the m-dimensional signature vector space so that similar records fall into the same partition or adjacent partitions; Calculate the signature vector for a new record and locate its partition. Calculate the similarity of the vector with the historical records in that partition and adjacent partitions. Determine whether it is a duplicate based on a preset threshold.

5. The method according to claim 4, characterized in that, In step S2, the coordinate batch generation algorithm specifically includes the following steps: Morton codes are used to sort parcel data, and spatial partitioning boundaries are adaptively determined by minimizing an objective function that includes load balancing and spatial compactness terms. Within each segment, the vertex coordinates of the land parcels are extracted in batches using the geometric object iterator of the GIS platform; Establish a topology reuse cache for boundary lines shared by adjacent parcels to avoid duplicate generation.

6. The method according to claim 5, characterized in that, In step S2, the distributed computing architecture consists of a master node and computing nodes. The master node is responsible for task sharding and scheduling, and the computing nodes are responsible for executing data processing tasks within the shards. The master node monitors the load of each node and dynamically allocates tasks. Finally, the master node merges the results of each node and rebuilds the index.

7. The method according to claim 6, characterized in that, Step S3 specifically includes the following steps: Step S3.1: Use data quality inspection software to check the field integrity, value range compliance, and logical consistency of the converted data; Step S3.2: Use the topology inspection tool of the GIS platform to verify the geometric and topological relationships of the converted spatial data; Step S3.3: Based on the errors found by the data quality inspection software and topology inspection, locate the root cause of the problem and generate an error report.

8. The method according to claim 7, characterized in that, In step S2, the storage optimization strategy includes: storing all intermediate and result data in a unified file geodatabase format, and using binary storage, spatial and attribute indexes to reduce disk input / output overhead.

9. The method according to claim 8, characterized in that, In step S2, the incremental migration mechanism includes: marking data change timestamps in the source database, reading the configured time points through a Python script, extracting only the changed data with timestamps greater than those timestamps for conversion, and merging it with existing data in the target database.

10. The method according to claim 9, characterized in that, The distributed computing architecture has a fault tolerance mechanism. When any computing node fails, the master node reassigns its unfinished tasks to other computing nodes.

Images