Land resource management data analysis method and system based on cloud computing

Abstract

The invention relates to the technical field of data analysis, in particular to a land resource management data analysis method and system based on cloud computing, and the method comprises the following steps: obtaining a difference time node plot number and a purpose field, recognizing a number evolution chain and establishing mapping, extracting a land class sequence to generate a stability label, analyzing the spatial continuity to construct an attribution path, and analyzing the attribution path. And indexing a utilization rate fluctuation trend paragraph, and identifying a purpose fusion direction to generate a fusion layer set. According to the method, through joint comparison of a serial number structure, coordinate overlapping and a purpose field, a stable land class transition chain is established, it is ensured that land class time sequence features are effectively described, the contiguous direction of a space dot matrix and the consistency of land class affiliation are combined, splicing of the space affiliation relation between land parcels is achieved on the regional level, and the splicing efficiency of the space affiliation relation between land parcels is improved. The identification precision and the spatial expression definition of the land type change are improved, and the spatial analysis capability of the land type change trend and the dynamic perception capability of the land utilization efficiency are effectively enhanced.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a cloud-based method and system for land resource management data analysis. Background Art

[0002] The field of data analytics technology encompasses a system of technical methods for acquiring, organizing, modeling, analyzing, and interpreting various types of structured and unstructured data. Its core components include data acquisition, statistical modeling, multidimensional analysis, visualization, and information mining. This field supports the establishment of scientific evidence in various decision-making and prediction scenarios through the analysis and modeling of data and information flows.

[0003] Among them, the land resource management data analysis method refers to the data processing and analysis method for land resource information such as land use, land ownership, and land classification. The technical matters addressed by this patent cover issues such as land parcel information collection, land type coding identification, integration of time-series use records, standardization of use types, and regional area comparison. Specifically, it is accomplished through methods such as field-level data format conversion, rasterization parsing of land parcel boundaries, regional area hierarchical statistics, and extraction of land type change trends.

[0004] Existing technologies fail to adequately establish number evolution chains when processing land parcel number changes at specific time points, resulting in weak correlations between identical land parcels across multiple time records. When analyzing land use change trends, field value changes lack structural indexing, easily leading to ambiguity in land use change identification. Spatial adjacency relationships are based on static regional divisions, lacking support for identifying the spatial logical continuity under changes in land use classification, causing breaks in classification at the regional level and affecting the integrity of trend judgment and regional aggregation. In urban expansion monitoring, continuous changes in land use and spatial extension cannot form a unified trend expression, limiting decision support efficiency in applications such as dynamic land use assessment and land use clustering feature identification. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a cloud computing-based land resource management data analysis method and system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a cloud computing-based land resource management data analysis method, comprising the following steps: S1: Obtain the sequence of land parcel number field, spatial location matrix and use field recorded under different time nodes in the cloud platform, and perform joint comparison based on the spatial coordinate overlap of land parcel records in different time periods, number structure code and use field continuity to generate a dynamic mapping table of land parcel numbers; S2: Call the land parcel trajectory record identified in the dynamic mapping table of land parcel numbers, extract the value sequence of the land category field corresponding to the time node, identify the differences in the time sequence dimension based on the continuous repetition duration of the field value and the interval of the change node, and generate a land category change stability labeling table. S3: Call the stable segment records of the land parcels in the land category change stability labeling table, make a judgment at the regional level based on the spatial location overlap relationship and the consistency of the land category field, construct a numbered splicing logical path between cloud data nodes, and generate a spatial affiliation splicing sequence structure; S4: Call the spatial attribution splicing sequence structure, arrange the field values ​​in time series, determine whether there is a turning point in the direction continuity, mark the trend node at the position of the turning point relationship, and generate a land use rate change structure segment index table.

[0007] As a further embodiment of the present invention, the dynamic mapping table for land parcel numbers includes land parcel time series identifiers, number evolution chain structure, spatial location mapping relationship, and land use attribute mapping index; the land use change stability labeling table includes stable segment time range, land use field stability level, change node time interval, and field continuity assessment value; the spatial affiliation splicing sequence structure includes number path chain, spatial continuity judgment result, land use consistency identifier, and regional level affiliation node; and the land use rate change structure segment indexing table includes change trend segment number, trend node location index, fluctuation direction sequence group, and structure division segment index.

[0008] As a further aspect of the present invention, the steps for obtaining the dynamic mapping table of land parcel numbers are as follows: S111: Obtain the land parcel number field, spatial location matrix, and use field sequence recorded under different time nodes in the cloud platform, extract the spatial coordinate group and use classification value corresponding to each land parcel number in the different time period, group them according to the combination characteristics of the number field and coordinate matrix, determine whether there is a continuous structure of use field sequence within the same spatial range, and generate spatial use sequence group value. S112: Call the spatial use sequence group value, perform character-level comparison and semantic encoding judgment on the number field under the time node in the group, identify whether there is a number change trend in the records in the group, and generate a number evolution comparison sequence value by combining the sorting difference of the number under the time node. S113: Based on the numbering evolution comparison sequence value, jointly determine whether the numbering matching chain under the same spatial coordinates and the time period of the different numbering has the characteristics of continuous use field and numbering structure inheritance, calculate the numbering continuity difference value, and combine the matching numbering path within the difference range to obtain the dynamic mapping table of land parcel numbering.

[0009] As a further aspect of the present invention, the steps for obtaining the land use change stability labeling table are as follows: S211: Call the land parcel trajectory record identified in the land parcel number dynamic mapping table, extract the land category field value sequence recorded under continuous time nodes, classify and statistically analyze the same field values ​​based on the distribution of land category field values ​​between continuous time nodes and mark the corresponding time period, and obtain the field value repetition segment information; S212: Based on the duration of each land category field value in the repeated segment information of the field value and the interval time of the subsequent change node, call the time series annotation information to judge the change amplitude between adjacent segments, and identify the node whose change degree exceeds the set land category change judgment threshold, and obtain the land category change interval intensity value. S213: Based on the coverage ratio of the intensity value of the land use change interval and the multi-field value in the whole time series, extract the land use time series stability information of the land parcel, calculate the stability assessment value of the land parcel, and obtain the land use change stability labeling table.

[0010] As a further aspect of the present invention, the step of obtaining the spatial attribution splicing sequence structure specifically includes: S311: Call the stable segment record of the land parcel in the land use change stability labeling table, extract the spatial matrix and adjacent direction identifier fields of multiple segments, and based on the correspondence between the boundary line coordinates, raster code and adjacent fields of the spatial matrix under the same number record, obtain the raster combination matching pair under the same number, construct the corresponding spatial combination node pair, and obtain the number of spatial combination node pairs. S312: Based on the number of spatial combination node pairs, filter node pair combinations that match the spatial location overlap relationship and land use field consistency, extract the grid center coordinates, spatial extension vector and land use field identifier, determine whether there is a continuous path relationship at the regional level, calculate the regional path continuity value, combine the numerical value to determine whether the node connection standard is met, obtain the set of spatial paths that meet the connection conditions, and generate continuous path connection quantity value. S313: Based on the continuous path connection value, select the node sequence with continuous paths in the spatial path set, construct a number splicing structure according to the connection number and path direction order, reorganize the node sequence by number and splicing index classification, and generate a spatial affiliation splicing sequence structure.

[0011] As a further aspect of the present invention, the steps for obtaining the land use rate change structural segment index table are as follows: S411: Call the land utilization rate field data corresponding to the path chain in the spatial attribution splicing sequence structure, establish a one-to-one correspondence between the path chain number and the land identification value, arrange them in time order, and generate a land utilization rate time series value group. S412: Based on the time series value group of land utilization rate, extract the land utilization rate field value under adjacent time nodes, calculate the increase / decrease trend direction mark between adjacent field values, scan the continuous and consistent segments in the trend direction sequence, detect the position of the direction turning point and mark the trend node, and generate the trend node mark coordinate sequence. S413: Call the trend node annotation coordinate sequence and the trend direction sequence of the corresponding path chain, divide the segments according to the trend node position, identify the segments with the same direction as a group of structural segments, and assign start and end position indexes and direction information to each group of structural segments to generate a land use rate change structural segment index table.

[0012] As a further aspect of the present invention, the method further includes step S5: S5: Call the trend change paragraph in the land use rate change structure paragraph indexing table, load the adjacency direction identifier and use field value of the plot in the cloud platform layer service, identify the consistency of the use field change cluster direction and the spatial arrangement concentration, filter the regional paragraphs with change trend clustering characteristics in the adjacency direction relationship, and generate the adjacent regional land type clustering structure layer set. The adjacent area land use clustering structure layer set includes trend clustering segments, spatial distribution concentration index, distribution of usage field change direction, and adjacent direction association information.

[0013] As a further aspect of the present invention, the steps for obtaining the adjacent region land class fusion structure layer set are specifically as follows: S511: Call the trend change paragraph in the land use rate change structure paragraph indexing table, load the adjacent direction identifier and use field value of the plot in the cloud platform layer service, match the use field difference value of the adjacent plot according to the plot number corresponding to the paragraph number, and record the number of adjacent field directions that are consistent with the use change direction of the paragraph, and generate the adjacent use direction matching number. S512: Based on the number of adjacent use direction matching, extract the spatial position combination of adjacent plots in the coordinate system for each plot, compare the consistency direction of use field changes within the direction combination, and generate a list of continuous use change combinations. S513: Based on the list of continuous combinations of land use changes, analyze the combinations of land use changes that simultaneously possess directional consistency and concentrated continuity within the spatial range, extract the land parcel segment numbers with combination characteristics, and generate a set of adjacent area land use clustering structure layers.

[0014] The cloud-based land resource management data analysis system is used to execute the aforementioned cloud-based land resource management data analysis method. The system includes: The spatial trajectory extraction module obtains the sequence of land parcel number field, spatial location matrix and use field recorded under different time nodes in the cloud platform, constructs the number evolution chain of the same land parcel number field in the time series, and generates a dynamic mapping table of land parcel numbers. The land use change identification module calls the dynamic mapping table of the land parcel number, extracts the land category field value of the corresponding time node of the number record, divides the continuous repetition of field value into segments, and obtains the land category change stability label table. The attribution path construction module calls the land use change stability labeling table, collects the spatial location matrix field and adjacent direction field of the land parcels corresponding to the stable segment, compares the spatial overlap between adjacent land parcels with the consistency of the land use field, and obtains the spatial attribution splicing sequence structure when the path is continuous and the attributes are consistent. The utilization rate trend segmentation module calls the spatial attribution splicing sequence structure, extracts the land utilization rate field value of the plots in the path, analyzes the direction of change by time, divides the trend segment at the position where the direction changes, and generates a land utilization rate change structure segment index table. The fusion region filtering module uses the land use rate change structure segment index table to load the adjacent direction field and use field of the corresponding land parcels in the trend segment, analyzes the frequency of change and spatial concentration distribution of the use field in the same direction, and obtains the land type fusion structure layer set of adjacent areas.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by analyzing the evolution of land parcel numbers at different time points, the numbering structure, coordinate overlap, and land use field are jointly compared to establish a stable land use change chain, ensuring that the temporal characteristics of land use are effectively characterized. By combining the adjacency direction of the spatial matrix with the consistency of land use classification, the spatial classification relationship between land parcels is spliced ​​at the regional level, making the continuity and change path between land parcels have a clear logic. The directional fluctuation of the utilization rate value sequence is structurally analyzed and trend turning points are extracted, realizing the systematic indexing of trend changes. Through spatial clustering analysis of the direction of land use change and adjacency relationship, regional segments with trend clustering characteristics are identified, improving the identification accuracy and spatial expression clarity of land use change, and effectively enhancing the spatial analysis capability of land use change trends and the dynamic perception capability of land use efficiency. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the workflow of the present invention; Figure 2 This is a flowchart illustrating the process of obtaining the dynamic mapping table for land parcel numbers in this invention. Figure 3 This is a flowchart illustrating the process of obtaining the land cover change stability labeling table in this invention. Figure 4This is a flowchart illustrating the process of obtaining the spatial attribution splicing sequence structure in this invention. Figure 5 This is a flowchart illustrating the process of obtaining the index table for the structural segments of land use rate changes in this invention. Figure 6 This is a flowchart illustrating the process of obtaining the adjacent region land class fusion structure layer set in this invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0019] Please see Figure 1 This invention provides a technical solution: a cloud-based land resource management data analysis method, comprising the following steps: S1: Obtain the sequence of land parcel number field, spatial location matrix and use field recorded under different time nodes in the cloud platform. Perform joint comparison based on the spatial coordinate overlap of land parcel records in different time periods, number structure encoding and use field continuity to identify the number evolution record chain of the same spatial land parcel under different time periods and generate a dynamic mapping table of land parcel numbers. S2: Call the land parcel trajectory record identified in the dynamic mapping table of land parcel numbers, extract the value sequence of the land category field corresponding to the time node, identify the differences in the time sequence dimension based on the continuous repetition duration of the field value and the interval of the change node, and generate a land category change stability labeling table. S3: Call the stable segment records of the land parcels in the land use change stability labeling table, extract the spatial matrix and adjacent direction identifier fields of adjacent records through the cloud platform analysis interface, and make a judgment at the regional level based on the spatial location overlap relationship and the consistency of the land use field. When the spatial overlap association and land use classification in the regional path have continuity, construct a numbered splicing logical path between cloud data nodes to generate a spatial classification splicing sequence structure. S4: Call the land use rate field data corresponding to the path chain in the spatial affiliation splicing sequence structure, arrange the field values ​​in time series, construct the change sequence group by the fluctuation direction relationship of adjacent utilization rate values ​​in the time dimension, determine whether there is a turning point in the direction continuity, mark the trend node at the position of the turning point relationship, and structurally divide the continuous fluctuation path according to the trend direction to generate a land use rate change structure segment index table. S5: Call the trend change paragraphs in the land use rate change structure paragraph indexing table, load the adjacency direction identifiers and use field values ​​of the plots in the cloud platform layer service, identify the consistency of the use field change clustering direction and the spatial arrangement concentration, filter the regional paragraphs with change trend clustering characteristics in the adjacency direction relationship, and generate the adjacent regional land type clustering structure layer set. The dynamic mapping table for land parcel numbers includes land parcel time series identifiers, number evolution chain structure, spatial location mapping relationship, and land use attribute mapping index. The land use change stability labeling table includes the time range of stable sections, stability level of land use field, time interval of change nodes, and field continuity assessment value. The spatial attribution splicing sequence structure includes number path chain, spatial continuity judgment result, land use consistency identifier, and regional hierarchical attribution node. The land use rate change structure segment indexing table includes change trend segment number, trend node location index, fluctuation direction sequence group, and structural division segment index. The adjacent area land use clustering structure layer set includes change trend clustering segments, spatial arrangement concentration index, distribution of change direction of land use field, and adjacent direction association information.

[0020] Please see Figure 2 The specific steps for obtaining the dynamic mapping table of land parcel numbers are as follows: S111: Obtain the land parcel number field, spatial location matrix, and use field sequence recorded under different time nodes in the cloud platform, extract the spatial coordinate group and use classification value corresponding to each land parcel number in the different time period, group them according to the combination characteristics of the number field and coordinate matrix, determine whether there is a continuous structure of use field sequence within the same spatial range, and generate spatial use sequence group value. To obtain the sequence of land parcel number fields, spatial location points, and usage fields recorded under different time nodes, it is necessary to first perform structured extraction on data from any cloud platform within the target area. Taking 2022 and 2023 as two different time nodes as examples, obtain the list of land parcel number fields under the 2022 time node, such as {G101, G102, G103}, and simultaneously extract the spatial coordinate points {(120.001, 30.001), (120.002, 30.002), (120.003, 30.003)} under the corresponding numbers. The sequence of land use fields {residential, commercial, industrial} is also extracted when obtaining the 2023 time node. The number field {G201, G202, G103}, the coordinate matrix {(120.001, 30.001), (120.002, 30.003), (120.003, 30.003)}, and the use field {residential, office, industrial} are extracted. By mapping the number field to the coordinate matrix one by one, it can be determined that the spatial location of land plot G103 is consistent with the number in the two time nodes, and the spatial use sequence group value is generated.

[0021] S112: Call the spatial purpose sequence group value, perform character-level comparison and semantic encoding judgment on the number field under the time node in the group, identify whether there is a number change trend in the records in the group, and generate number evolution comparison sequence value by combining the sorting difference of the number under the time node. At each time point within the group, the numbering field was structurally decomposed at the bit level. Taking the numbers "G101" and "G201" as examples, they were decomposed into the prefix "G" and the serial number fields "101" and "201". Since their prefixes are the same, they were inferred to be consecutive numbers within the same numbering system. By comparing the difference in the serial number field (201-101=100) with the change in sorting position, the numbers were sorted within the group. The differences in the usage field and spatial matrix structure between the previous and subsequent numbers were calculated. The absolute value of the serial number difference was used as one of the trend indicators. If the serial numbers within the same group showed a continuous increasing trend... Furthermore, if the fluctuation of the usage field is less than 0.1 (judged by the difference in quantitative coding), then the numbering evolution path is considered continuous. At the same time, it is necessary to exclude numbering changes caused by non-structural changes and to set up the migration of numbering segments due to management adjustments. False judgments can be eliminated by judging whether the minimum numbering interval within the numbering segment changes significantly. If the interval within the numbering segment changes from 5 to 100, it can be judged as a migration. After the above analysis, the numbering evolution trend path within the group is formed. The numbering path structure of each group is sorted in chronological order to obtain the numbering evolution sequence. Each group is assigned a unique numbering path ID, and the numbering evolution comparison sequence value is obtained.

[0022] S113: Based on the numbering evolution comparison sequence value, jointly determine whether the numbering matching chain in different time periods under the same spatial coordinates has continuous use field and numbering structure inheritance characteristics, calculate the numbering continuity difference value, and combine the matching numbering path within the difference range to obtain the dynamic mapping table of land parcel numbers; The serial number continuity difference value is a numerical indicator that measures the consistency between the serial number and use changes of land parcels within the same spatial range at two time points. It integrates serial number structure, use field and spatial location changes, and after normalization, reflects the similarity or deviation of their evolutionary relationship. The smaller the value, the more continuous the serial number and use correspondence between the two time points. The larger the value, the more significant the difference in the serial number and use correspondence between the two time points, making it difficult to regard them as the same evolutionary chain. A numbering index table is established based on spatial coordinates (such as latitude and longitude coordinates). The numbering evolution sequence of different years or time periods is extracted. Changes in the numbering of the same plot need to be categorized and integrated. By comparing the changing trends of the numbers in different time periods, it is determined whether there is a continuous inheritance characteristic in the numbering structure. If a plot is numbered A123 in 2020, A124 in 2021, and A125 in 2022, then it is inferred that its numbering exhibits a structurally increasing pattern. During this process, the numbering information corresponding to the spatial coordinates can be extracted using GIS software. A numbering time series table is established using database management tools such as PostgreSQL. SQL queries are executed to compare the numbering sequences. It is necessary to jointly determine whether the matching chain of numbers has consistency in the continuous use field. This requires comparing the use field, such as the land use code (e.g., R for residential, C for commercial, M for mixed), with the use field in the same numbering sequence. Whether it changes or not, if we set the use field of a certain land parcel number A123 to R, A124 to R, and A125 to R, then we can determine that the use field is continuous. By executing a Python script to read the number and use field in batches and perform field consistency judgment, a binary judgment result is generated for "whether consistent" (1 indicates consistency, 0 indicates inconsistency). The continuity difference value of the number is calculated. The difference value = the proportion of inconsistent fields in the length of the number change sequence. If the length of the continuous sequence is set to 3 and the number of inconsistencies is 0, then the difference value = 0 / 3 = 0. If there is a different use field in the sequence, then the difference value = 1 / 3 = 0.33. A difference threshold such as 0.2 needs to be set as the benchmark value. If the difference value > 0.2, then the continuity of the number is determined to be insufficient. By comparing the difference of the number matching chain of different land parcels and time periods multiple times, a set of number paths within the difference range is formed, and a dynamic mapping table of land parcel numbers is obtained.

[0023] Please see Figure 3 The specific steps for obtaining the land use change stability labeling table are as follows: S211: Call the land parcel trajectory record identified in the land parcel number dynamic mapping table, extract the land category field value sequence recorded by the land parcel at continuous time nodes, classify and statistically analyze the same field values ​​based on the distribution of land category field values ​​between continuous time nodes, mark the corresponding time period, and obtain the information of repeated field value segments. First, the mapping table structure is parsed. Each plot number corresponds to a set of time series trajectories. Based on the unique plot number field, the associated time nodes are read sequentially. At each time node, the land type field value is extracted. For example, plot number A101 has the following records from January 2022 to December 2023: the land type field value is "farmland" in January 2022, "farmland" from February 2022 to October 2022, and "forestland" from November 2022 onwards. This forms two consecutive field segments. After extracting the field values, the land type is then... The classification is based on whether the fields remain unchanged, that is, whether the field values ​​in consecutive nodes are consistent. If they are consistent, they are classified into the same segment. If the land type of plot B207 is "grassland" from January to June 2023, then the segment is classified as a duplicate segment. Then, its start and end time and field value are recorded. During the execution process, the field values ​​need to be standardized and matched. The code 01 should be uniformly set for farmland, 02 for forest land, and 03 for grassland to avoid interference from the naming conflicts of different data sources on duplicate identification. After the field value segmentation processing of each plot within the time period is completed in sequence, the information of duplicate field value segments is obtained.

[0024] S212: Based on the duration of each land category field value in the repeated segment information and the interval between subsequent change nodes, call the time series annotation information to judge the change magnitude between adjacent segments, identify nodes whose change degree exceeds the set land category change judgment threshold, and obtain the land category change interval intensity value. The duration of each land category field value segment can be calculated by subtracting the start time from the end time. For example, if a segment starts in March 2023 and ends in November 2023, its duration is 9 months. Simultaneously, it checks if subsequent segments have different land category field values. If a land category change occurs, the time interval between this change node and the end node of the previous segment is recorded. During execution, it starts from the end node of the current segment, reads the start time of the next segment, and compares whether the land category code has changed. If it changes from "01" farmland to "02" forest land, it is considered a land category change node. The magnitude of the change can be determined by... The land use category change level table is calculated by setting the change from farmland to forest land as a level 2 change, with a change magnitude level of 2. If the change level is greater than the set threshold, it is judged as a structural land use category change. The set threshold is 1.5, and changes greater than this value are marked as significant changes. The change level needs to be set in conjunction with the actual land classification of the application area. For the segment identified as a land use category change node, a change magnitude value is assigned. Combined with its segment span, the change intensity is calculated. The calculation formula is the change level divided by the segment duration. If the level is set to 3 and the duration is 6 months, the change intensity is 0.5, forming the land use category change interval intensity value.

[0025] S213: Based on the coverage ratio of the intensity value of land use change interval and the multi-field value in the whole time series, extract the land use time series stability information of the land parcel, calculate the stability assessment value of the land parcel, and obtain the land use change stability labeling table. The stability assessment value represents the balance and continuity of land use changes for a particular plot throughout the entire time series. It comprehensively considers the degree of fluctuation in the intensity of change, the amplitude of fluctuation in repeated sections, and the frequency distribution of land use changes. The smaller the value, the more stable the land use and the stronger the regularity of change; the larger the value, the more frequent the changes in the plot and the stronger the structural fluctuation. When extracting land use type temporal stability information for a plot, it is necessary to collect land use type change data for the plot throughout the entire time series (set from 2000 to 2025). For the same plot, land use type codes (e.g., agricultural land A, construction land B, unused land C, etc.) at different time periods should be recorded year by year. Using GIS tools such as ArcGIS combined with SQL scripts, the land use type should be compared to see if it has changed each year, and the intensity value of the land use type change interval should be calculated. The interval intensity value can be defined as the ratio of the number of land use type changes per unit time to the total duration. If a plot experiences 5 land use type changes between 2000 and 2025, then the interval intensity value = 5 / 25 = 0.2. Simultaneously, multiple field values ​​such as land use, land use rate, and land cover type need to be extracted, and their coverage ratio within the entire time series should be calculated. The coverage ratio = duration of a certain field in a consistent state / total duration. If the land use is continuously agricultural land for 20 years, then the coverage ratio = 20. / 25=0.8. A Python script is needed to batch calculate the interval intensity values ​​and multi-field coverage ratios of the land parcels, storing the results in an intermediate result table. When calculating the land parcel stability assessment value, a weighted average method can be used: Stability assessment value = S - W1 × Interval intensity value + W2 × Coverage ratio. S is set as the baseline value of 1, and W1 and W2 are weight coefficients. For example, W1=0.5 and W2=0.5. If the interval intensity value is 0.2 and the coverage ratio is 0.8, then the stability assessment = 1 - 0.5 × 0.2 + 0.5 × 0.8 = 1 - 0.1 + 0.4 = 1.3. The weight coefficients need to be set in conjunction with the data fluctuation range, using the average of multiple land parcel data as the baseline. For example, if the average interval intensity value is 0.3 and the average coverage ratio is 0.7 obtained from 20 sample land parcels, weights can be set to maintain a balanced contribution to different indicators. Multiple rounds of simulation verification are required to obtain a land use change stability labeling table.

[0026] Please see Figure 4 The specific steps for obtaining the spatial attribution splicing sequence structure are as follows: S311: Call the stable segment records of the land parcels in the land use change stability labeling table, extract the spatial matrix and adjacent direction identification fields of multiple segments, and based on the correspondence between the boundary line coordinates, raster codes and adjacent fields of the spatial matrix under the same number record, obtain the raster combination matching pairs under the same number, construct the corresponding spatial combination node pairs, and obtain the number of spatial combination node pairs. The spatial point matrix data field and the adjacent direction identifier field are extracted. The spatial point matrix data field consists of the raster code of each stable segment, and the adjacent direction identifier field records the direction information of adjacent cells in that segment. In the actual implementation, the record item numbered "ZB001" is obtained. This item's spatial point matrix contains 9 grid points, and its coding range is G1001 to G1009. The boundary line coordinates are quantized using a Cartesian coordinate system, with the unit being meters. At the same time, its adjacent direction identifier is matched and read. The "East" and "South" directions are set to "1", and the other directions are set to "0". A raster combination pair structure based on ZB001 is constructed, with raster G1001 and G1002 forming a pair. The center point coordinates of (500, 700) and (510, 700) are extracted respectively. The adjacent direction is determined to be "East" based on the difference between the two points. On this basis, the nine grid nodes are combined and matched to obtain a total of 12 pairs of adjacent node combinations. The coding pair and the adjacent direction are recorded in the form of a two-dimensional array. For example, the direction of the combination pair (G1001, G1002) is "East", and the direction of the combination pair (G1001, G1004) is "South". This forms a spatial combination node pair record with the number "ZB001". The number of combination pairs under all stable segment numbers is automatically summarized and a two-dimensional matrix index table is generated, and the number of spatial combination node pairs is "72".

[0027] S312: Based on the number of spatial combination node pairs, filter node pair combinations that match the spatial location overlap relationship and land use field consistency, extract the raster center coordinates, spatial extension vector and land use field identifier, determine whether there is a continuous path relationship at the regional level, calculate the regional path continuity value, combine the numerical value to determine whether the node connection standard is met, obtain the set of spatial paths that meet the connection conditions, and generate continuous path connection quantity value. The regional path continuity value is a quantitative indicator that measures the degree of continuity of multiple spatial nodes in terms of spatial distribution and land use attributes. It is calculated by comprehensively considering factors such as spatial distance, attribute consistency and hierarchical depth of node combinations. The larger the value, the more stable the path connection and the more complete the structure. It is the core basis for spatial splicing judgment. By rasterizing the data of the area to be analyzed, selecting a suitable raster resolution such as 30m × 30m, and generating center point coordinates for each raster cell, while extracting land use field identifiers such as agricultural land, construction land, and forest land, a basic list of spatial node pair combinations is established. Then, the distance and spatial overlap between the center coordinates of each node pair need to be checked. Buffer zones or spatial intersection rules can be set in the GIS software for filtering. If two center points are within the set buffer zone and the intersection area reaches the specified minimum area, they are considered to be overlapping. The land use field of the filtered node pairs needs to be checked for consistency. Field consistency is confirmed by comparing field values ​​to ensure they belong to either agricultural land or construction land. Node pairs with consistent fields are retained to form new spatial combinations. The center coordinates of each node pair in the combination are extracted for subsequent spatial extension vector analysis. The direction of extension is determined by the relative position between coordinates and represented in GIS visualization. To further determine whether there is a continuous path relationship at the regional level, the filtered node pairs need to be combined to form a node network. The path combinations are traversed using network analysis tools such as ArcGIS Network Analyst or QGIS plugin. Based on the geographic topology, it is determined whether it is sustainable from the starting node to the ending node. If a break in the path is found, the connection conditions of the node pairs need to be adjusted or the adjacency relationship needs to be recalculated. If multiple node pairs can form a continuous path, the coverage and connection density of the paths within the region are statistically analyzed. At the same time, a judgment standard value, such as the average connection density within the region, is set to compare whether a certain path or combination meets the connection standard. After the connection standard is met, the set of spatial paths that meet the conditions can be extracted to generate a continuous path connection value.

[0028] S313: Based on the continuous path connection value, filter the node sequence with continuous paths in the spatial path set, construct the number splicing structure according to the connection number and path direction order, reorganize the node sequence by number and splicing index classification, and generate the spatial affiliation splicing sequence structure. Extract the numbering order of combination pairs that meet the connection criteria from the corresponding spatial path set, and perform directed path splicing according to the numbering sorting rules of combination pairs to form a continuous numbered path structure. Taking the path (G1001, G1002) → (G1002, G1003) → (G1003, G1006) as the clue, extract the raster numbers in sequence, and number each path node in turn. For example, if the path node numbers are P01 to P04, establish mapping relationships in sequence according to the numbers, such as P01→G1001, P02→G1002, etc., and establish a directed index table for the path nodes. Read the land use field values ​​corresponding to each node, such as "forest land", "forest land", "shrubland", "shrubland". Segment and merge the land use fields of continuous nodes. If two or more consecutive nodes have the same land use attribute, a field combination identifier is formed, such as "forest land shrubland", and a spatial affiliation splicing sequence structure is generated.

[0029] Please see Figure 5 The specific steps for obtaining the index table of structural paragraphs related to land use rate changes are as follows: S411: Call the land utilization rate field data corresponding to the path chain in the spatial affiliation splicing sequence structure, establish a one-to-one correspondence between the path chain number and the land identification value, arrange them in time order, and generate a land utilization rate time series value group. The path chain is defined as a set of connections established according to the spatial distribution of land parcels in a geographic information system. Path chain numbers are constructed by parsing the boundary connections in the GIS layer. Path chain P001 represents a chain structure formed by sequentially connecting land parcels A, B, and C. The process of calling this path chain requires extracting the corresponding land parcel identifier and utilization rate field information from the land parcel attribute database. Attribute association is used for pairing, mapping the path chain number to the corresponding land parcel identifier value. Then, the utilization rate value of each land parcel at different time points needs to be extracted, such as 2010, 2015, and 2020. This can be done based on existing land survey data or remote sensing interpretation results. The database is then filtered and sorted by the time field, and the utilization rate value of each land parcel is arranged in chronological order. The utilization rate of land parcel A at the three time points is set to 50%, 53%, and 55%, generating a time series value group of land parcel utilization rates.

[0030] S412: Based on the time series value group of land utilization rate, extract the land utilization rate field value under adjacent time nodes, calculate the increase or decrease trend direction mark between adjacent field values, scan the continuous and consistent segments in the trend direction sequence, detect the position of the direction turning point and mark the trend node, and generate the trend node marked coordinate sequence. The implementation process requires extracting the utilization rate values ​​of adjacent time nodes sequentially and judging the direction of change by comparison. The numerical comparison between each pair of adjacent time points is judged manually or by program as "increase", "decrease" or "stable". If the utilization rate of block A in a certain path chain is 50% in 2010 and 55% in 2015, then the interval is judged as "increase" and the direction is recorded as +1; if it is 45% to 40%, it is recorded as -1; if it is stable, it is recorded as 0. The complete trend direction sequence is obtained by repeating this operation. The direction sequence can be stored in the form of an array or a string. For example, [+1, +1, -1] means that the first two time periods are increasing and the last time period is decreasing. The sequence is continuously scanned, and the position where the trend direction changes is set as the turning point. For example, if +1 changes to -1, it means that the trend has reversed. This position needs to be marked as a trend node, and a trend node marking coordinate sequence is generated.

[0031] S413: Call the trend node annotation coordinate sequence and the trend direction sequence of the corresponding path chain, divide the segments according to the trend node position, identify the segments with the same direction as a group of structural segments, and assign start and end position indexes and direction information to each group of structural segments to generate a land use rate change structural segment index table. The time nodes marked by trend nodes should be used as dividing points to divide the overall time series of the path chain into several continuous segments. If a node is marked in 2015 and 2020, it should be divided into time periods such as 2010–2015, 2015–2020, and 2020–2025. Each segment is a sub-interval with a unified direction. The consistency of the trend direction markings within each segment should be checked. If the trend direction in a certain segment is always +1, it can be identified as a "growth segment" and assigned three basic information: start time, end time, and direction. This corresponds to a structural segment in the path chain structure. Each segment should be assigned a unique index number to ensure the structural clarity during data processing. The information of this structural segment can be organized into a table to summarize and sort the land use trends in the region in segments and time series hierarchy, generating a land use rate change structural segment index table.

[0032] Please see Figure 6 The specific steps for obtaining the adjacent region land class fusion structure layer set are as follows: S511: Call the trend change paragraph in the land use rate change structure paragraph indexing table, load the adjacent direction identifier and use field value of the plot in the cloud platform layer service, match the use field difference value of the adjacent plot according to the plot number corresponding to the paragraph number, and record the number of adjacent field directions that are consistent with the use change direction of the paragraph, and generate the adjacent use direction matching number. Extract the trend change paragraph number, the corresponding land parcel number, and the change information of the land use field. Combine this with the spatial layer data of the land parcels loaded by the cloud platform layer service to identify attribute records containing adjacent direction identifiers and land use fields. Based on the main land parcel number corresponding to the paragraph number, identify the land use field of each adjacent land parcel in the eight adjacent directions. Calculate the difference value of adjacent land use fields, i.e., compare the difference between the land use of the adjacent land parcel and the main land parcel, using Boolean difference (0 for the same, 1 for the different) as the discrimination standard. At the same time, record the adjacent land parcel direction that is consistent with the change direction of the main land parcel's land use. For example, if the main land parcel changes from agricultural land to construction land, count the number of adjacent directions with the real-time land use as construction land. If the land use of a certain main land parcel A changes from "agricultural land" to "construction land", and its north, west, and east adjacent land parcels are currently all "construction land", and its south side is "agricultural land", then the number of adjacent land use direction matches is 3, and the number of adjacent land use direction matches is generated.

[0033] S512: Based on the number of adjacent use direction matching, extract the spatial location combination of adjacent plots in the coordinate system for each plot, compare the consistency direction of use field changes within the direction combination, and generate a list of continuous use change combinations. Extract the actual location coordinates of the main plot and its adjacent plots in a coordinate system, such as using the WGS84 coordinate system or the local projection coordinate system (set to CGCS2000) to represent the coordinates of the center point or polygon boundary vertex of the main plot and its adjacent plots. Use a GIS platform (such as ArcGIS or QGIS) to perform layer analysis and form a spatial location combination matrix. Set the center of the main plot as the origin, and the eight adjacent plots in each direction constitute relative location information units. Each combination contains attributes such as plot number, use field, and change direction. Compare the consistency of the change of use field in each combination in each direction, that is, determine whether the use of each adjacent plot is consistent with the change direction of the main plot. For example, if the main plot changes from agricultural land to construction land, the direction identifier of the use field of the adjacent plot is "construction land", which is the consistent direction. If the combination is a continuous adjacent direction, it is considered a continuous combination, and a list of continuous combination of use changes is generated.

[0034] S513: Based on the list of continuous combinations of land use changes, analyze the combinations of land use changes that simultaneously possess directional consistency and concentrated continuity within the spatial range, extract the land parcel segment numbers with combination characteristics, and generate a set of land use clustering structure layers for adjacent areas. Analyze whether each combination simultaneously possesses directional consistency and concentrated continuity in spatial structure. Using GIS layer topology analysis tools, determine whether the combination direction exhibits a continuous regional expansion structure, such as forming a fan-shaped, strip, or block-shaped concentrated structure. Extract the main plot segment numbers that meet the conditions, generate structural feature identifiers, evaluate the correspondence between the segment number and the layer block index, that is, establish a one-to-one mapping relationship between the segment number and the layer data index field. Realize the synchronous extraction of layer identifier information through spatial connection query method, set the output as a vector layer, and generate a set of adjacent area land use clustering structure layers.

[0035] The cloud-based land resource management data analysis system is used to execute the aforementioned cloud-based land resource management data analysis method. The system includes: The spatial trajectory extraction module obtains the sequence of land parcel number field, spatial location matrix and use field recorded under different time nodes in the cloud platform, constructs the number evolution chain of the same land parcel number field in the time series, and generates a dynamic mapping table of land parcel numbers. The land use change identification module calls the land parcel number dynamic mapping table, extracts the land category field value of the corresponding time node of the number record, divides the continuous repetition of field value into segments, and obtains the land category change stability label table. The attribution path construction module calls the land use change stability labeling table, collects the spatial location matrix field and adjacent direction field of the land parcels corresponding to the stable segment, compares the spatial overlap between adjacent land parcels and the consistency of the land use field, and obtains the spatial attribution splicing sequence structure when the path is continuous and the attributes are consistent. The utilization rate trend segmentation module calls the spatial attribution splicing sequence structure, extracts the land utilization rate field value of the plots in the path, analyzes the direction of change by time, divides the trend segment at the position where the direction changes, and generates a land utilization rate change structure segment index table. The fusion region filtering module uses the land use rate change structure paragraph index table to load the adjacent direction field and use field of the corresponding land parcels in the trend paragraph, analyzes the frequency of change and spatial concentration distribution of the use field in the same direction, and obtains the land use fusion structure layer set of adjacent regions.

[0036] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A cloud computing-based land resource management data analysis method, characterized in that, The following steps are involved: S1: Obtain the sequence of land parcel number field, spatial location matrix and use field recorded under different time nodes in the cloud platform, and perform joint comparison based on the spatial coordinate overlap of land parcel records in different time periods, number structure code and use field continuity to generate a dynamic mapping table of land parcel numbers; S2: Call the land parcel trajectory record identified in the dynamic mapping table of land parcel numbers, extract the value sequence of the land category field corresponding to the time node, identify the differences in the time sequence dimension based on the continuous repetition duration of the field value and the interval of the change node, and generate a land category change stability labeling table. S3: Call the stable segment records of the land parcels in the land category change stability labeling table, make a judgment at the regional level based on the spatial location overlap relationship and the consistency of the land category field, construct a numbered splicing logical path between cloud data nodes, and generate a spatial affiliation splicing sequence structure; S4: Call the spatial attribution splicing sequence structure, arrange the field values ​​in time series, determine whether there is a turning point in the direction continuity, mark the trend node at the position of the turning point relationship, and generate a land use rate change structure segment index table.

2. The land resource management data analysis method based on cloud computing according to claim 1, characterized in that, The dynamic mapping table for land parcel numbers includes land parcel time series identifiers, number evolution chain structure, spatial location mapping relationship, and land use attribute mapping index. The land use change stability labeling table includes stable segment time range, land use field stability level, change node time interval, and field continuity assessment value. The spatial attribution splicing sequence structure includes number path chain, spatial continuity judgment result, land use consistency identifier, and regional level attribution node. The land use rate change structure segment indexing table includes change trend segment number, trend node location index, fluctuation direction sequence group, and structure division segment index.

3. The land resource management data analysis method based on cloud computing according to claim 1, characterized in that, The specific steps for obtaining the dynamic mapping table of land parcel numbers are as follows: S111: Obtain the land parcel number field, spatial location matrix, and use field sequence recorded under different time nodes in the cloud platform, extract the spatial coordinate group and use classification value corresponding to each land parcel number in the different time period, group them according to the combination characteristics of the number field and coordinate matrix, determine whether there is a continuous structure of use field sequence within the same spatial range, and generate spatial use sequence group value. S112: Call the spatial use sequence group value, perform character-level comparison and semantic encoding judgment on the number field under the time node in the group, identify whether there is a number change trend in the records in the group, and generate a number evolution comparison sequence value by combining the sorting difference of the number under the time node. S113: Based on the numbering evolution comparison sequence value, jointly determine whether the numbering matching chain under the same spatial coordinates and the time period of the different numbering has the characteristics of continuous use field and numbering structure inheritance, calculate the numbering continuity difference value, and combine the matching numbering path within the difference range to obtain the dynamic mapping table of land parcel numbering.

4. The land resource management data analysis method based on cloud computing according to claim 3, characterized in that, The specific steps for obtaining the land use change stability labeling table are as follows: S211: Call the land parcel trajectory record identified in the land parcel number dynamic mapping table, extract the land category field value sequence recorded under continuous time nodes, classify and statistically analyze the same field values ​​based on the distribution of land category field values ​​between continuous time nodes and mark the corresponding time period, and obtain the field value repetition segment information; S212: Based on the duration of each land category field value in the repeated segment information of the field value and the interval time of the subsequent change node, call the time series annotation information to judge the change amplitude between adjacent segments, and identify the node whose change degree exceeds the set land category change judgment threshold, and obtain the land category change interval intensity value. S213: Based on the coverage ratio of the intensity value of the land use change interval and the multi-field value in the whole time series, extract the land use time series stability information of the land parcel, calculate the stability assessment value of the land parcel, and obtain the land use change stability labeling table.

5. The land resource management data analysis method based on cloud computing according to claim 4, characterized in that, The specific steps for obtaining the spatial attribution splicing sequence structure are as follows: S311: Call the stable segment record of the land parcel in the land use change stability labeling table, extract the spatial matrix and adjacent direction identifier fields of multiple segments, and based on the correspondence between the boundary line coordinates, raster code and adjacent fields of the spatial matrix under the same number record, obtain the raster combination matching pair under the same number, construct the corresponding spatial combination node pair, and obtain the number of spatial combination node pairs. S312: Based on the number of spatial combination node pairs, filter node pair combinations that match the spatial location overlap relationship and land use field consistency, extract the grid center coordinates, spatial extension vector and land use field identifier, determine whether there is a continuous path relationship at the regional level, calculate the regional path continuity value, combine the numerical value to determine whether the node connection standard is met, obtain the set of spatial paths that meet the connection conditions, and generate continuous path connection quantity value. S313: Based on the continuous path connection value, select the node sequence with continuous paths in the spatial path set, construct a number splicing structure according to the connection number and path direction order, reorganize the node sequence by number and splicing index classification, and generate a spatial affiliation splicing sequence structure.

6. The land resource management data analysis method based on cloud computing according to claim 5, characterized in that, The specific steps for obtaining the index table of structural segments representing changes in land use rate are as follows: S411: Call the land utilization rate field data corresponding to the path chain in the spatial attribution splicing sequence structure, establish a one-to-one correspondence between the path chain number and the land identification value, arrange them in time order, and generate a land utilization rate time series value group. S412: Based on the time series value group of land utilization rate, extract the land utilization rate field value under adjacent time nodes, calculate the increase / decrease trend direction mark between adjacent field values, scan the continuous and consistent segments in the trend direction sequence, detect the position of the direction turning point and mark the trend node, and generate the trend node mark coordinate sequence. S413: Call the trend node annotation coordinate sequence and the trend direction sequence of the corresponding path chain, divide the segments according to the trend node position, identify the segments with the same direction as a group of structural segments, and assign start and end position indexes and direction information to each group of structural segments to generate a land use rate change structural segment index table.

7. The land resource management data analysis method based on cloud computing according to claim 1, characterized in that, The method further includes step S5: S5: Call the trend change paragraph in the land use rate change structure paragraph indexing table, load the adjacency direction identifier and use field value of the plot in the cloud platform layer service, identify the consistency of the use field change cluster direction and the spatial arrangement concentration, filter the regional paragraphs with change trend clustering characteristics in the adjacency direction relationship, and generate the adjacent regional land type clustering structure layer set. The adjacent area land use clustering structure layer set includes trend clustering segments, spatial distribution concentration index, distribution of usage field change direction, and adjacent direction association information.

8. The land resource management data analysis method based on cloud computing according to claim 7, characterized in that, The specific steps for obtaining the adjacent region land type fusion structure layer set are as follows: S511: Call the trend change paragraph in the land use rate change structure paragraph indexing table, load the adjacent direction identifier and use field value of the plot in the cloud platform layer service, match the use field difference value of the adjacent plot according to the plot number corresponding to the paragraph number, and record the number of adjacent field directions that are consistent with the use change direction of the paragraph, and generate the adjacent use direction matching number. S512: Based on the number of adjacent use direction matching, extract the spatial position combination of adjacent plots in the coordinate system for each plot, compare the consistency direction of use field changes within the direction combination, and generate a list of continuous use change combinations. S513: Based on the list of continuous combinations of land use changes, analyze the combinations of land use changes that simultaneously possess directional consistency and concentrated continuity within the spatial range, extract the land parcel segment numbers with combination characteristics, and generate a set of adjacent area land use clustering structure layers.

9. A cloud-based land resource management data analysis system, characterized in that: The system is used to implement the cloud computing-based land resource management data analysis method according to any one of claims 1-8, and the system includes: The spatial trajectory extraction module obtains the sequence of land parcel number field, spatial location matrix and use field recorded under different time nodes in the cloud platform, constructs the number evolution chain of the same land parcel number field in the time series, and generates a dynamic mapping table of land parcel numbers. The land use change identification module calls the dynamic mapping table of the land parcel number, extracts the land category field value of the corresponding time node of the number record, divides the continuous repetition of field value into segments, and obtains the land category change stability label table. The attribution path construction module calls the land use change stability labeling table, collects the spatial location matrix field and adjacent direction field of the land parcels corresponding to the stable segment, compares the spatial overlap between adjacent land parcels with the consistency of the land use field, and obtains the spatial attribution splicing sequence structure when the path is continuous and the attributes are consistent. The utilization rate trend segmentation module calls the spatial attribution splicing sequence structure, extracts the land utilization rate field value of the plots in the path, analyzes the direction of change by time, divides the trend segment at the position where the direction changes, and generates a land utilization rate change structure segment index table. The fusion region filtering module uses the land use rate change structure segment index table to load the adjacent direction field and use field of the corresponding land parcels in the trend segment, analyzes the frequency of change and spatial concentration distribution of the use field in the same direction, and obtains the land type fusion structure layer set of adjacent areas.

Images