A method and device for evaluating quality of cultivated land data based on geospatial

Abstract

The application relates to the technical field of cultivated land evaluation, in particular to a cultivated land data quality evaluation method and device based on geographic space, which comprises the following steps: collecting cultivated land space data and cultivated land attribute data in a current time period, and acquiring reference cultivated land space data in a previous time period; based on the cultivated land space data in the current time period and the reference cultivated land space data, a spatial consistency index is acquired; wherein the spatial consistency index comprises a spatial change value of each plot; geographic space data in the current time period is acquired, and based on the cultivated land attribute data and the geographic space data in the current time period, an attribute accuracy index is acquired. The application can comprehensively evaluate the quality of cultivated land data from multiple dimensions by combining the spatial consistency and attribute accuracy indexes, guarantees the comprehensiveness of the evaluation result, and can accurately evaluate each plot according to the comprehensive quality score of each plot by combining the spatial change value, the attribute matching value and historical data.

Description

Technical Field

[0001] This invention relates to the field of arable land assessment technology, specifically to a method and apparatus for assessing the quality of arable land data based on geospatial data. Background Technology

[0002] Traditional methods typically rely on single assessment indicators, such as focusing only on the spatial consistency or attribute accuracy of arable land, lacking multi-dimensional comprehensive assessment. This can result in incomplete assessment results that fail to accurately reflect the actual quality of arable land data. Furthermore, traditional methods often analyze data based on only a specific time period, neglecting historical data or changes over time. This limits the assessment results to those from a single period and may not accurately reflect trends in arable land quality. Additionally, traditional methods often provide coarse assessments of spatial consistency, lacking detailed multi-scale analysis and easily overlooking subtle spatial variations between different plots, leading to insufficient accuracy in spatial quality assessment. Moreover, traditional methods often lack effective spatial continuity indicators or remote sensing image verification, potentially failing to identify fragmented areas in arable land data. This oversight reduces the reliability of the assessment results, especially when arable land is fragmented or scattered. Finally, traditional methods often lack flexibility to adapt to different regions and arable land conditions and cannot integrate multiple data sources for comprehensive analysis, limiting their application scope and failing to meet the complex and ever-changing needs of arable land data quality assessment. Summary of the Invention

[0003] To achieve the above objectives, the present invention provides the following technical solution: a geospatial-based method for assessing the quality of cultivated land data, comprising:

[0004] Collect farmland spatial data and farmland attribute data for the current time period, and obtain reference farmland spatial data for the previous time period;

[0005] Based on the farmland spatial data for the current time period and the reference farmland spatial data, a spatial consistency index is obtained; wherein, the spatial consistency index includes the spatial change value of each plot.

[0006] Obtain geospatial data for the current time period, and based on the cultivated land attribute data and geospatial data for the current time period, obtain attribute accuracy indicators; wherein, the attribute accuracy indicators include the attribute matching value for each plot.

[0007] Traverse each plot of land and obtain a comprehensive quality score for each plot of land based on the spatial variation value and attribute matching value of each plot of land.

[0008] The overall quality level of arable land data is determined based on the comprehensive quality score of all plots.

[0009] Preferably, based on the arable land spatial data of the current time period and the reference arable land spatial data, a spatial consistency index is obtained, including:

[0010] Extract the first plot boundary of the cultivated land spatial data for the current time period and the second plot boundary of the reference cultivated land spatial data;

[0011] The overlap rate of the land parcels is calculated based on the overlapping area of ​​the first land parcel boundary and the second land parcel boundary.

[0012] The overlap rate of the plots is checked using the first spatial weight to obtain the first spatial enhancement value;

[0013] The overlap rate of the land parcels is processed using a second spatial weight kernel to obtain a second spatial enhancement value; wherein, the first spatial weight kernel and the second spatial weight kernel are used to focus on different land parcels;

[0014] The first spatial enhancement value and the second spatial enhancement value are fused to obtain the spatial enhancement overlap rate;

[0015] The spatial consistency index is obtained based on the plot overlap rate and the spatial enhancement overlap rate.

[0016] Preferably, the spatial consistency index is obtained based on the plot overlap rate and the spatially enhanced overlap rate, including:

[0017] A multi-scale buffer analysis was performed on the overlap rate of the aforementioned land parcels to obtain the buffer overlap rate.

[0018] The overlapping rates of the land parcels and the buffer overlapping rates are weighted and fused to obtain the fused overlapping rate.

[0019] The final overlap rate is obtained by weighted fusion based on the fusion overlap rate and the spatial enhancement overlap rate.

[0020] Based on the final overlap rate, the spatial consistency index is obtained.

[0021] Preferably, each plot of land is traversed, and a comprehensive quality score for each plot is obtained based on its spatial variation value and attribute matching value, including:

[0022] In response to the spatial change value of the land parcel being greater than a first threshold and the attribute matching value being less than a second threshold, a first quality adjustment value is determined;

[0023] In response to the spatial change value of the land parcel being less than or equal to a first threshold, or the attribute matching value being greater than or equal to a second threshold, a second quality adjustment value is determined;

[0024] Wherein, the first quality adjustment value is a decrease in the score value, and the second quality adjustment value is a maintenance of the score value; the absolute value of the first quality adjustment value is greater than the absolute value of the second quality adjustment value.

[0025] Preferably, the method of traversing each plot of land and obtaining a comprehensive quality score for each plot based on its spatial variation value and attribute matching value further includes:

[0026] Obtain historical cultivated land data for multiple consecutive time periods preceding the current time period;

[0027] Obtain the historical comprehensive quality score of each plot of land in the historical cultivated land data of the aforementioned multiple consecutive time periods;

[0028] For each plot of land, calculate the comprehensive quality score for the current time period based on the first or second quality adjustment value of each plot of land and the historical comprehensive quality score.

[0029] Preferably, the overall quality level of cultivated land data is determined based on the comprehensive quality score of all plots, including:

[0030] In response to the fact that the overall quality score of the land parcel is greater than a preset score threshold, an expansion operation is performed on the land parcel to expand and generate a high-quality area;

[0031] If the overall quality score of the land parcel is less than or equal to the preset score threshold, the area consisting of the land parcel is designated as a qualified area.

[0032] Areas outside of the high-quality and qualified areas in the cultivated land data are considered unqualified areas.

[0033] Preferably, geospatial data for the current time period is obtained, and based on the cultivated land attribute data and geospatial data for the current time period, attribute accuracy indicators are obtained, including:

[0034] Obtain geospatial data for the current time period; wherein, the geospatial data includes farmland distribution data and remote sensing image data;

[0035] Based on the farmland distribution data, vector information of plot boundaries is extracted; based on the vector information, a farmland spatial continuity index is calculated; based on the spatial continuity index, fragmented areas are detected in the farmland distribution data; wherein, the spatial continuity index includes the minimum and maximum spacing between adjacent plots;

[0036] If such a problem exists, the corresponding land parcel is identified as a problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on the remote sensing image data, and attribute matching values ​​are obtained.

[0037] Preferably, based on the cultivated land distribution data, vector information of plot boundaries is extracted; based on the vector information, a cultivated land spatial continuity index is calculated; and based on the spatial continuity index, fragmented areas are detected in the cultivated land distribution data, including:

[0038] Edge detection is performed on the farmland distribution data to obtain an initial boundary line; the initial boundary line is then smoothed to generate the vector information.

[0039] Based on the vector information, determine the boundary points of adjacent plots;

[0040] Calculate the minimum and maximum distances between the boundary points of adjacent plots, as the spatial continuity index;

[0041] Set the minimum spacing threshold and the maximum spacing threshold;

[0042] If the minimum distance between the boundary points of adjacent plots is less than the minimum spacing threshold, or the maximum distance is greater than the maximum spacing threshold, then a fragmented area is determined to exist.

[0043] Preferably, if such a problem exists, the corresponding land parcel is determined as a problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on the remote sensing image data, and attribute matching values ​​are obtained, including:

[0044] Based on the location information of the fragmented areas, locate the corresponding plots in the cultivated land distribution data;

[0045] Mark the located plot of land as a problem plot;

[0046] Based on the location information of the problematic land parcel, local images of the corresponding area are extracted from the remote sensing image data;

[0047] Match the attribute matching values ​​between the problem plot and the local image.

[0048] A geospatial-based farmland data quality assessment device, applicable to the aforementioned geospatial-based farmland data quality assessment method, comprising:

[0049] The data acquisition unit is used to collect farmland spatial data and farmland attribute data for the current time period, as well as obtain reference farmland spatial data for the previous time period.

[0050] The change calculation unit is used to obtain a spatial consistency index based on the cultivated land spatial data of the current time period and the reference cultivated land spatial data; wherein, the spatial consistency index includes the spatial change value of each plot;

[0051] An attribute matching unit is used to acquire geospatial data for the current time period, and to acquire attribute accuracy indicators based on the cultivated land attribute data and geospatial data for the current time period; wherein, the attribute accuracy indicators include the attribute matching value for each plot.

[0052] The comprehensive scoring unit is used to traverse each plot of land and obtain a comprehensive quality score for each plot of land based on the spatial change value and attribute matching value of each plot of land.

[0053] The quality assessment unit is used to determine the overall quality level of arable land data based on the comprehensive quality score of all plots.

[0054] Compared with the prior art, the beneficial effects of the present invention are:

[0055] (1) By combining two indicators, spatial consistency and attribute accuracy, this invention can comprehensively evaluate the quality of cultivated land data from multiple dimensions, ensuring the comprehensiveness and accuracy of the evaluation results. Furthermore, by adopting multi-scale buffer analysis and weighted fusion of spatially enhanced overlap rate, the calculation of spatial consistency index is more refined and accurate, which can effectively identify spatial changes and overlap between different plots, further improving the evaluation level of the spatial quality of cultivated land data.

[0056] (2) By combining spatial change values, attribute matching values, historical data and other information, this invention can accurately evaluate each plot based on its comprehensive quality score. This scoring mechanism not only considers the data of the current period, but also refers to historical data, thus avoiding the limitations of data from a single time period. Furthermore, based on the comprehensive quality score, the method can clearly distinguish between high-quality areas, qualified areas and unqualified areas, which helps to make targeted decisions in the management and optimization of arable land resources.

[0057] (3) By calculating the spatial continuity index, this invention can effectively identify fragmented areas in cultivated land distribution data and verify the authenticity through remote sensing images, thereby improving the authenticity and reliability of the data, especially in the case of fragmented cultivated land.

[0058] (4) This invention has strong flexibility and can adapt to the analysis of cultivated land data in different time periods. It can also combine multiple data sources (such as remote sensing images, geospatial data, etc.) to achieve multi-dimensional quality assessment. It is suitable for quality assessment needs in different regions and under different cultivated land conditions. By continuously monitoring changes in cultivated land quality and assessment results, it can provide data support and decision-making basis for the long-term management of cultivated land resources, thereby promoting the sustainable use of cultivated land resources. Attached Figure Description

[0059] Figure 1 This is a schematic flowchart of the overall method in one embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the overall device architecture in one embodiment of the present invention.

[0061] In the diagram: 1. Data acquisition unit; 2. Change calculation unit; 3. Attribute matching unit; 4. Comprehensive scoring unit; 5. Quality assessment unit. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Example 1, please refer to Figure 1 This invention provides a technical solution: a method for assessing the quality of arable land data based on geospatial data, comprising:

[0064] S1. Collect farmland spatial data and farmland attribute data for the current time period, and obtain reference farmland spatial data for the previous time period.

[0065] S2. Based on the current time period's farmland spatial data and reference farmland spatial data, obtain spatial consistency indicators; among which, spatial consistency indicators include the spatial change value of each plot;

[0066] S3. Obtain geospatial data for the current time period. Based on the farmland attribute data and geospatial data for the current time period, obtain attribute accuracy indicators. Among them, the attribute accuracy indicators include the attribute matching value of each plot.

[0067] S4. Traverse each plot of land and obtain the comprehensive quality score for each plot of land based on the spatial change value and attribute matching value of each plot of land.

[0068] S5. Determine the overall quality level of cultivated land data based on the comprehensive quality score of all plots.

[0069] In an optional embodiment, a spatial consistency index is obtained based on the current time period's arable land spatial data and reference arable land spatial data, including:

[0070] Extract the first plot boundary from the current time period's farmland spatial data, and the second plot boundary from the reference farmland spatial data;

[0071] Calculate the land overlap rate based on the overlapping area of ​​the first land parcel boundary and the second land parcel boundary;

[0072] The first spatial enhancement value is obtained by checking the overlap rate of the land parcels using the first spatial weight.

[0073] The overlap rate of land parcels is processed using a second spatial weight kernel to obtain a second spatial enhancement value; wherein, the first spatial weight kernel and the second spatial weight kernel are used to focus on different land parcels;

[0074] The first spatial enhancement value and the second spatial enhancement value are fused to obtain the spatial enhancement overlap rate;

[0075] Spatial consistency indicators are obtained based on the plot overlap rate and spatial enhancement overlap rate.

[0076] It should be noted that "first plot boundary" refers to the boundary of a plot in the current time period's cultivated land spatial data; "second plot boundary" refers to the boundary of another plot in the cultivated land spatial data. Boundary extraction means identifying the outline of each plot, which can be represented by vector data. Based on the overlapping area of ​​the first and second plot boundaries, the plot overlap rate is calculated. The plot overlap rate is the proportion of the area of ​​the intersection of two plot boundaries to the area of ​​one of the plots. The overlap rate is obtained by calculating the overlapping area of ​​the first and second plots and dividing it by the area of ​​the first plot. The overlap rate is an indicator of the similarity between two plots; the higher the value, the greater the overlap. The first spatial weight kernel is used to process the plot overlap rate to obtain the first spatial enhancement value. The first spatial weight kernel is a weight matrix or kernel used to enhance the focus on the spatial characteristics of the first plot. The spatial weight kernel usually assigns different weights based on the relative position of the plot, the surrounding environment, and other factors. By combining the overlap rate with the first spatial weight kernel and performing weighted processing, the first spatial enhancement value is obtained, which is the value adjusted after considering the influence of spatial characteristics. Overlap rate; the overlap rate of plots is processed using a second spatial weight kernel to obtain a second spatial enhancement value; similarly, the second spatial weight kernel is a weight kernel for the spatial characteristics of the second plot; its design is also adjusted according to the spatial characteristics of the second plot; the overlap rate is weighted with the second spatial weight kernel to obtain the second spatial enhancement value; in this way, the spatial characteristics of the second plot are also taken into consideration, further refining the overlap rate; the first spatial enhancement value and the second spatial enhancement value are fused to obtain the spatial enhancement overlap rate; the fusion of the first spatial enhancement value and the second spatial enhancement value can be achieved through weighted averaging or other fusion methods; the purpose of fusion is to comprehensively consider the spatial characteristics of the two plots to obtain a more comprehensive spatial consistency index, namely the spatial enhancement overlap rate; based on the plot overlap rate and the spatial enhancement overlap rate, the spatial consistency index is obtained; the spatial consistency index is obtained by comparing the plot overlap rate and the spatial enhancement overlap rate; generally, the higher the spatial consistency index, the stronger the consistency between the current cultivated land spatial data and the reference cultivated land spatial data; if the difference between the spatial enhancement overlap rate and the plot overlap rate is small, it indicates that the consistency of cultivated land space is relatively high.

[0077] In an optional embodiment, a spatial consistency index is obtained based on the plot overlap rate and the spatially enhanced overlap rate, including:

[0078] Multi-scale buffer analysis was performed on the overlap rate of land parcels to obtain the buffer overlap rate;

[0079] The overlapping rate of land parcels and the buffer overlapping rate are weighted and merged to obtain the merged overlapping rate.

[0080] The final overlap rate is obtained by weighted fusion based on the fusion overlap rate and the spatially enhanced overlap rate.

[0081] Based on the final overlap rate, a spatial consistency index is obtained.

[0082] It should be noted that buffer analysis is a common operation in spatial analysis, aiming to generate buffer zones around the boundaries of land parcels. The size of the buffer zones is usually set at multiple scales, either as a fixed distance or based on the actual characteristics of the land parcel. Multi-scale buffer analysis means analyzing different buffer ranges of the same land parcel, considering spatial overlap at different scales. Typically, these buffer zones can be generated at multiple scales, for example, generating buffer zones with different radii of 1 km, 5 km, 10 km, etc., for a single land parcel. The buffer overlap rate refers to the proportion of the area of ​​overlap between the land parcel and other land parcels within the buffer zone to the total area of ​​the buffer zone. This indicator reflects the spatial similarity and influence range of the land parcel and its surrounding area at different scales. The land parcel overlap rate is an indicator calculated in the previous steps, representing the degree of overlap between the current land parcel and the reference land parcel. The buffer overlap rate, however, considers the impact of the buffer zone on land parcel overlap, emphasizing spatial consistency near the land parcel boundaries. The weighted fusion method combines the overlap rate of land parcels and the overlap rate of buffers by assigning different weights. The weighting method can be based on experience, spatial characteristics, or other practical considerations. For example, it can be considered that the closer the buffer is, the greater its impact on spatial consistency, or different weights can be assigned to the two according to actual needs. The spatially enhanced overlap rate is the overlap rate after considering spatial weights, which further emphasizes the spatial relationship between the land parcel and the reference land parcel. The weighted fusion step combines the fused overlap rate and the spatially enhanced overlap rate, aiming to comprehensively consider the direct overlap of land parcels, the impact of buffers, and spatial enhancement factors. This step also uses a weighted average method to weight and fuse the two indicators to obtain a new overlap rate. The spatial consistency index reflects the spatial consistency or similarity between two land parcels (current cultivated land and reference cultivated land). The final overlap rate provides a comprehensive spatial consistency measure by combining multiple factors (land parcel overlap, buffer overlap, and spatial enhancement).

[0083] In an optional embodiment, each plot of land is traversed, and a comprehensive quality score for each plot is obtained based on its spatial variation value and attribute matching value, including:

[0084] In response to a spatial change value of a land parcel exceeding a first threshold and an attribute matching value falling below a second threshold, a first quality adjustment value is determined.

[0085] A second quality adjustment value is determined in response to the spatial change value of the land parcel being less than or equal to a first threshold, or the attribute matching value being greater than or equal to a second threshold;

[0086] The first quality adjustment value is to reduce the score, and the second quality adjustment value is to maintain the score; the absolute value of the first quality adjustment value is greater than the absolute value of the second quality adjustment value.

[0087] It should be noted that spatial change value typically refers to the spatial changes of a plot of land within a certain time period, which may include changes in land use, shape, area, etc. This value reflects the spatial dynamics of the plot. The first threshold is a preset standard used to judge whether the spatial change value of a plot is too large. If the spatial change value of a plot exceeds this threshold, it indicates that the plot has undergone significant spatial changes. The second threshold is another preset standard used to judge whether the attribute matching value of a plot is too low. If the attribute matching value is less than this threshold, it indicates that the attribute of the plot has a poor match with the reference standard. Based on the comparison of the spatial change value and attribute matching value of the plot with the thresholds, a quality adjustment value is determined for each plot. This adjustment value affects the overall quality score. If the spatial change value is greater than the first threshold and the attribute matching value is less than the second threshold, in this case, the spatial change of the plot is large, and its attributes do not match the reference standard. Therefore, the quality score of the plot needs to be reduced, and the first quality adjustment value is to reduce the score. Specifically, the plot's score will be subject to a negative adjustment. Quality adjustment lowers the plot's score; a larger absolute value of this adjustment means that if this condition is triggered, the decrease in the plot's score will be more significant; spatial change value is less than or equal to the first threshold, or attribute matching value is greater than or equal to the second threshold; this indicates that the plot's spatial change is small, or its attributes match the reference standard well, therefore the plot's quality score remains unchanged, and the second quality adjustment value is the "maintain score" value; if the conditions are met, the plot's quality score does not change significantly, and the score remains unchanged; here, the adjustment value is positive, indicating that the plot's score will not decrease; since the absolute value of the first quality adjustment value is greater than the absolute value of the second quality adjustment value, the impact of maintaining the score value is relatively small; based on the calculation of the above two quality adjustment values, the plot's overall quality score will be adjusted accordingly; if the plot meets the condition of the first quality adjustment value, the score will decrease; while if the plot meets the condition of the second quality adjustment value, the score remains unchanged; through this method, the overall quality score of the plot can be flexibly adjusted for different spatial changes and attribute matching situations.

[0088] In an optional embodiment, traversing each plot of land and obtaining a comprehensive quality score for each plot based on its spatial variation value and attribute matching value, further includes:

[0089] Retrieve historical cultivated land data for multiple consecutive time periods preceding the current time period;

[0090] Obtain the historical comprehensive quality score of each plot of farmland in historical farmland data across multiple consecutive time periods;

[0091] For each plot of land, calculate the overall quality score for the current time period based on the first or second quality adjustment value of each plot of land and the historical overall quality score.

[0092] It should be noted that when calculating the comprehensive quality score for the current time period, historical data needs to be referenced to obtain more comprehensive assessment information. Before the current assessment point in time, historical farmland data for multiple consecutive time periods needs to be accessed and collected. This data typically includes information such as spatial changes, attribute matching values, and historical quality scores for different plots in the past. This historical data can provide the performance of each plot in different time periods, providing background reference for the comprehensive quality score of the current time period. Each plot will have a historical comprehensive quality score over multiple past time periods, which reflects the quality change trend of the plot in the past. By reviewing these historical scores, we can understand the quality changes of the plot, thus providing a basis for the current assessment. After obtaining the historical data and the historical comprehensive quality score of each plot, the next step is to use this data to calculate the comprehensive quality score for the current time period. This involves traversing each plot to obtain information such as spatial change values, attribute matching values, and historical comprehensive quality scores for each plot; calculating the comprehensive quality score for the current time period; and based on each... If the spatial variation value of a land parcel exceeds a certain threshold and the attribute matching value is low, the score is typically adjusted to a negative value, i.e., the score is reduced. If the spatial variation value is small, or the attribute matching is good, the score remains unchanged or is slightly adjusted. The quality score for the current time period is corrected by considering the comprehensive quality score in historical data (which may be the average of multiple past time periods or the trend of recent time periods). For example, if a land parcel has consistently maintained a high score in the past, a positive adjustment may be added to the current score; conversely, if the historical score is low, a moderate downward adjustment may be made to the current score. The comprehensive quality score for the current time period may be calculated by combining the historical comprehensive quality score with the current quality adjustment value. The specific calculation method may involve a weighted average or some kind of correction factor, and the weighting of historical scores is determined based on the first or second quality adjustment value. Some smoothing mechanism may be used (e.g., historical scores over longer periods may have a larger weighting, while scores from recent periods may have a smaller impact).

[0093] In one optional embodiment, the overall quality level of the cultivated land data is determined based on the comprehensive quality score of all plots, including:

[0094] If the overall quality score of a plot of land exceeds a preset score threshold, an expansion operation is performed on the plot to generate a high-quality area.

[0095] If the overall quality score of a plot of land is less than or equal to a preset score threshold, the area consisting of the plots will be considered a qualified area.

[0096] Areas outside of the high-quality and qualified areas in the cultivated land data will be classified as unqualified areas.

[0097] It's important to note that a preset scoring threshold needs to be set first. This threshold serves as a reference standard to distinguish different quality levels. The threshold setting can be determined based on specific needs and data, typically set based on certain quality standards. For plots with a comprehensive quality score higher than the threshold, they are considered high-quality. For plots with a comprehensive quality score lower than or equal to the threshold, they are considered low-quality and belong to the qualified area. For plots with scores higher than the preset threshold (i.e., high-quality plots), an expansion operation is performed, which involves extending the area surrounding the plot into a high-quality area. This expansion operation can be implemented using spatial algorithms, such as neighborhood-based expansion, which considers plots with scores higher than the threshold and plots within a certain radius around them as high-quality areas. The purpose of this operation is to ensure a high quality level is maintained within a larger area. This expansion operation helps reflect that there may also be good-quality areas around a high-quality plot, thus improving the overall regional quality assessment. For example... Sub-category: Assuming a plot of land has a comprehensive quality score of 90, which is higher than the threshold of 80, then adjacent plots around this plot may be included in the high-quality area according to certain rules (such as spatial distance, score range, etc.). For plots with scores less than or equal to the preset score threshold, these plots do not meet the criteria for a high-quality area, but are still considered as qualified areas. These areas, although of lower quality than high-quality areas, still meet the minimum quality requirements and are therefore classified as qualified areas. The scores of these areas do not meet the criteria for high-quality areas, but are still within a relatively reasonable quality range. For example, if a plot of land has a comprehensive quality score of 70, and the score threshold is 80, then this plot will be classified as a qualified area. Any plot that does not belong to a high-quality area or a qualified area will be classified as an unqualified area. These areas are of poor quality, and their scores may be far below the threshold due to drastic spatial changes, attribute mismatches, or other reasons. The comprehensive quality scores of these unqualified areas are significantly lower than the preset score threshold, which usually represents poor-quality farmland and may require further repair or adjustment.

[0098] In an optional embodiment, geospatial data for the current time period is obtained, and based on the cultivated land attribute data and geospatial data for the current time period, an attribute accuracy index is obtained, including:

[0099] Obtain geospatial data for the current time period; the geospatial data includes farmland distribution data and remote sensing image data;

[0100] Based on the farmland distribution data, vector information of plot boundaries is extracted; based on the vector information, farmland spatial continuity index is calculated; based on the spatial continuity index, fragmented areas are detected in the farmland distribution data; among which, the spatial continuity index includes the minimum and maximum distance between adjacent plots.

[0101] If such a problem exists, the corresponding plot of land is identified as the problem plot based on the location information of the fragmented area; the authenticity of the problem plot is verified based on remote sensing image data, and attribute matching values ​​are obtained.

[0102] It should be noted that geospatial data is used to describe the spatial location and related attributes of a geographic region. Geospatial data for the current time period includes the following two types: Farmland distribution data reflects the spatial distribution of farmland in a specific area, usually represented by Geographic Information System (GIS) data, such as vector data (e.g., polygons) or raster data (e.g., imagery); Remote sensing imagery data is ground information obtained through remote sensing technologies (e.g., satellite imagery, aerial photography), reflecting ground features such as land cover, vegetation type, and soil conditions. Vector information represents the spatial distribution of geographic objects using coordinate points, lines, and polygons. The boundaries of farmland can be extracted from raster images (e.g., remote sensing imagery data) using vectorization methods. This process generates polygonal boundary data representing cultivated land areas. By analyzing cultivated land distribution data, the boundaries of each cultivated land plot are identified and extracted. These boundaries typically represent the outline of the cultivated land, reflecting its spatial distribution. Spatial continuity indicators are used to measure the continuity and connectivity of plot distribution within cultivated land areas, usually including the minimum and maximum spacing between adjacent plots. The minimum spacing refers to the minimum distance between adjacent plots, measuring whether the plots are close or compact. A large minimum spacing may indicate significant gaps in the cultivated land distribution, suggesting spatial discontinuity. The maximum spacing refers to the maximum distance between adjacent plots, measuring the distribution between the farthest plots. An excessively large maximum spacing may indicate an incoherent or fragmented cultivated land distribution. Other types of land divide the land; these indicators reflect the spatial structure of arable land, especially whether there is fragmentation; the presence of obvious gaps or discontinuous areas in the distribution of arable land usually means that the arable land area has been cut by other types of land (such as urban construction, roads, water bodies, etc.), forming multiple discontinuous block areas; these fragmented areas affect the utilization efficiency and quality of arable land; these fragmented areas can be detected based on spatial continuity indicators (minimum spacing and maximum spacing); if the distance between adjacent plots is too large, or there are obvious gaps and divisions, fragmented areas can be identified; problem plots are plots in the identified fragmented areas, which are usually problem areas in arable land management; fragmentation not only affects Land use efficiency can also affect agricultural mechanization, irrigation, and management. Analyzing the location of fragmented areas can identify specific plots requiring special attention, designated as problem plots. Remote sensing imagery data provides up-to-date information on the land surface, helping to verify the authenticity and validity of farmland distribution and plot attributes. Analyzing remote sensing images of problem plots allows for the assessment of their actual condition and verification of compliance with farmland use regulations. Comparing remote sensing images with existing farmland data reveals potential quality issues, changes in land type, or deviations in land use. Attribute matching values, based on the authenticity verification of remote sensing images, can calculate the degree of matching between plot attribute data and actual image data.These attributes may include land use type (e.g., whether it is arable land), land quality, and the presence of other land use changes (e.g., buildings or roads). A higher matching value indicates a greater consistency between the remote sensing image and the actual attributes of the land parcel, resulting in more accurate verification results. A lower matching value suggests potential data errors or discrepancies between the actual situation of the land parcel and expectations.

[0103] In one optional embodiment, vector information of plot boundaries is extracted based on cultivated land distribution data; a cultivated land spatial continuity index is calculated based on the vector information; and the presence of fragmented areas in the cultivated land distribution data is detected based on the spatial continuity index, including:

[0104] Edge detection is performed on the cultivated land distribution data to obtain the initial boundary line; the initial boundary line is then smoothed to generate vector information.

[0105] Based on vector information, determine the boundary points of adjacent plots;

[0106] Calculate the minimum and maximum distances between the boundary points of adjacent plots as an indicator of spatial continuity;

[0107] Set the minimum spacing threshold and the maximum spacing threshold;

[0108] If the minimum distance between the boundary points of adjacent plots is less than the minimum spacing threshold, or the maximum distance is greater than the maximum spacing threshold, then a fragmented area is determined to exist.

[0109] It's important to note that edge detection is a crucial step in image processing, aiming to extract plot boundaries from farmland distribution data. Edge detection algorithms (such as Canny edge detection and the Sobel operator) analyze intensity variations in the data to identify the contours of farmland areas, i.e., plot boundaries. Edge detection yields preliminary boundary lines; these line segments mark the boundaries of farmland areas, typically acting as an edge separating farmland from non-farmland. These initial boundary lines may have some unevenness due to noise or irregularities in the detection algorithm; smoothing processes aim to eliminate these imperfections. Irregularities make boundary lines smoother and more natural, reducing overly sharp or unnecessary bends. Common smoothing techniques include Bézier curve smoothing and spline smoothing. After smoothing, the boundary lines can be converted into vector information, typically representing the spatial extent of farmland plots as polygons. Vector data describes the shape of plots through coordinate points; each plot's boundary is formed by connecting multiple coordinate points to create a polygon. Through vectorization, we obtain boundary information for multiple farmland plots. Next, we need to identify the adjacency relationships between these plots, i.e., which plot boundaries are in contact. This is done through calculations... The boundaries of these plots determine which plots share a boundary line; adjacent plots share a common boundary line, and the points on this common boundary line are called "boundary points"; these boundary points indicate the precise locations where two plots of farmland meet; the minimum distance threshold is a preset value, indicating that when the minimum distance between adjacent plots is below this threshold, the farmland plots are considered to be sufficiently close and spatially continuous; the maximum distance threshold is another preset value, indicating that when the maximum distance between adjacent plots is above this threshold, the farmland plots are considered to have a large gap, possibly indicating fragmentation or discontinuity. In the following situations, when there are large gaps or divisions in arable land areas, we call them fragmented areas. Fragmentation not only affects land use efficiency but may also affect the management and operational efficiency of agricultural production. If the minimum distance between adjacent plots is less than the set minimum spacing threshold, it means that the gap between plots is too small, which may indicate that they are connected or very close to each other, and fragmentation may exist. If the maximum distance between adjacent plots is greater than the set maximum spacing threshold, it means that the interval between plots is large and may be separated by other areas (such as roads, water bodies, urban construction, etc.), which also means that arable land is fragmented.

[0110] In an optional embodiment, if such a problem exists, the corresponding land parcel is determined as the problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on remote sensing image data, and attribute matching values ​​are obtained, including:

[0111] Based on the location information of fragmented areas, locate the corresponding plots in the farmland distribution data;

[0112] Mark the located plot of land as a problem plot;

[0113] Based on the location information of the problematic plot, extract local images of the corresponding area from remote sensing image data;

[0114] Match the attribute values ​​between the problem plot and the local image.

[0115] It should be noted that fragmented areas in farmland distribution are detected by calculating minimum and maximum distances. These fragmented areas typically refer to plots whose boundary spacing exceeds a set threshold, exhibiting obvious gaps or isolation. Based on the location information of these fragmented areas (e.g., through geographic coordinates or spatial coordinate systems), the plots corresponding to these fragmented areas can be accurately located in the farmland distribution data. Farmland distribution data is usually vector data, containing the spatial location and shape of each farmland plot. After locating the plots corresponding to the fragmented areas, these plots are considered to have fragmentation problems and therefore need to be marked as "problem plots." Marking these plots is a crucial step for further verification and processing. Remote sensing imagery, usually from satellites, drones, or aerial photography, provides a wealth of surface information. Remote sensing imagery data provides plots with rich spectral, textural, and other features, which can help us analyze the specific conditions of the plots, such as vegetation cover, soil type, and topographic features. Based on the location (coordinate range) of the marked problem plots, the plots corresponding to these fragmented areas are extracted from the remote sensing imagery data. These are local area images corresponding to certain land parcels; local images refer to the remote sensing image portions covering these land parcels, typically high-resolution images that can finely reflect the specific characteristics of the land parcels; attribute matching values ​​refer to comparing some features in the remote sensing image (such as color, texture, vegetation index, etc.) with the attributes of the problem land parcel (such as land parcel type, land use, etc.) to verify the authenticity of the problem land parcel; these attributes may include: vegetation index (such as NDVI): used to determine whether the land parcel is arable land, whether there is over-cultivation or changes in vegetation cover; land cover type: determining whether the land parcel meets the expected arable land type, such as farmland, paddy field, etc.; land parcel shape and area: calculating the area and shape of the land parcel through remote sensing imagery and comparing it with the expected values ​​in arable land distribution data; image processing and spatial analysis techniques, such as image registration, feature extraction, and similarity calculation, can be used to match the features in the remote sensing imagery with the land parcel attributes in the arable land distribution data; if the attributes of the remote sensing imagery match the expected attributes of the land parcel, then the land parcel is considered authentic and without problems.

[0116] Example 2, please refer to Figure 2 This invention provides a technical solution: a geospatial-based farmland data quality assessment device, applicable to the aforementioned geospatial-based farmland data quality assessment method, comprising:

[0117] Data acquisition unit 1 is used to collect farmland spatial data and farmland attribute data for the current time period, as well as obtain reference farmland spatial data for the previous time period.

[0118] The change calculation unit 2 is used to obtain spatial consistency indicators based on the current time period's cultivated land spatial data and reference cultivated land spatial data; wherein, the spatial consistency indicators include the spatial change value of each plot;

[0119] The attribute matching unit 3 is used to obtain geospatial data for the current time period, and to obtain attribute accuracy indicators based on the cultivated land attribute data and geospatial data for the current time period; wherein, the attribute accuracy indicators include the attribute matching value of each plot.

[0120] The comprehensive scoring unit 4 is used to traverse each plot and obtain a comprehensive quality score for each plot based on the spatial change value and attribute matching value of each plot.

[0121] Quality assessment unit 5 is used to determine the overall quality level of cultivated land data based on the comprehensive quality score of all plots.

[0122] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A geospatial-based method for assessing the quality of cultivated land data, characterized in that, include: Collect farmland spatial data and farmland attribute data for the current time period, and obtain reference farmland spatial data for the previous time period; Based on the farmland spatial data for the current time period and the reference farmland spatial data, a spatial consistency index is obtained; wherein, the spatial consistency index includes the spatial change value of each plot. Obtain geospatial data for the current time period, and based on the cultivated land attribute data and geospatial data for the current time period, obtain attribute accuracy indicators; wherein, the attribute accuracy indicators include the attribute matching value for each plot. Traverse each plot of land and obtain a comprehensive quality score for each plot of land based on the spatial variation value and attribute matching value of each plot of land. The overall quality level of the cultivated land data is determined based on the comprehensive quality score of all plots. Based on the farmland spatial data for the current time period and the reference farmland spatial data, spatial consistency indicators are obtained, including: Extract the first plot boundary of the cultivated land spatial data for the current time period and the second plot boundary of the reference cultivated land spatial data; The overlap rate of the land parcels is calculated based on the overlapping area of ​​the first land parcel boundary and the second land parcel boundary. The overlap rate of the plots is checked using the first spatial weight to obtain the first spatial enhancement value; The overlap rate of the land parcels is processed using a second spatial weight kernel to obtain a second spatial enhancement value; wherein, the first spatial weight kernel and the second spatial weight kernel are used to focus on different land parcels; The first spatial enhancement value and the second spatial enhancement value are fused to obtain the spatial enhancement overlap rate; The spatial consistency index is obtained based on the plot overlap rate and the spatial enhancement overlap rate. Obtain geospatial data for the current time period, and based on the cultivated land attribute data and geospatial data for the current time period, obtain attribute accuracy indicators, including: Obtain geospatial data for the current time period; wherein, the geospatial data includes farmland distribution data and remote sensing image data; Based on the farmland distribution data, vector information of plot boundaries is extracted; based on the vector information, a farmland spatial continuity index is calculated; based on the spatial continuity index, fragmented areas are detected in the farmland distribution data; wherein, the spatial continuity index includes the minimum and maximum spacing between adjacent plots; If such a problem exists, the corresponding land parcel is identified as a problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on the remote sensing image data, and attribute matching values ​​are obtained. Based on the farmland distribution data, vector information of plot boundaries is extracted; based on the vector information, a farmland spatial continuity index is calculated; based on the spatial continuity index, the presence of fragmented areas in the farmland distribution data is detected, including: Edge detection is performed on the farmland distribution data to obtain an initial boundary line; the initial boundary line is then smoothed to generate the vector information. Based on the vector information, determine the boundary points of adjacent plots; Calculate the minimum and maximum distances between the boundary points of adjacent plots, as the spatial continuity index; Set the minimum spacing threshold and the maximum spacing threshold; If the minimum distance between the boundary points of adjacent plots is less than the minimum spacing threshold, or the maximum distance is greater than the maximum spacing threshold, then a fragmented area is determined to exist. If such a problem exists, the corresponding land parcel is identified as a problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on the remote sensing image data, and attribute matching values ​​are obtained, including: Based on the location information of the fragmented areas, locate the corresponding plots in the cultivated land distribution data; Mark the located plot of land as a problem plot; Based on the location information of the problematic land parcel, local images of the corresponding area are extracted from the remote sensing image data; Match the attribute matching values ​​between the problem plot and the local image.

2. The method for assessing the quality of cultivated land data based on geospatial data according to claim 1, characterized in that, Based on the plot overlap rate and the spatial enhancement overlap rate, the spatial consistency index is obtained, including: A multi-scale buffer analysis was performed on the overlap rate of the aforementioned land parcels to obtain the buffer overlap rate. The overlapping rates of the land parcels and the buffer overlapping rates are weighted and fused to obtain the fused overlapping rate. The final overlap rate is obtained by weighted fusion based on the fusion overlap rate and the spatial enhancement overlap rate. Based on the final overlap rate, the spatial consistency index is obtained.

3. The method for assessing the quality of cultivated land data based on geospatial data according to claim 2, characterized in that, Traverse each plot of land, and based on the spatial variation value and attribute matching value of each plot, obtain a comprehensive quality score for each plot, including: In response to the spatial change value of the land parcel being greater than a first threshold and the attribute matching value being less than a second threshold, a first quality adjustment value is determined; In response to the spatial change value of the land parcel being less than or equal to a first threshold, or the attribute matching value being greater than or equal to a second threshold, a second quality adjustment value is determined; Wherein, the first quality adjustment value is a decrease in the score value, and the second quality adjustment value is a maintenance of the score value; the absolute value of the first quality adjustment value is greater than the absolute value of the second quality adjustment value.

4. The method for assessing the quality of cultivated land data based on geospatial data according to claim 3, characterized in that, Traversing each plot of land, and based on the spatial variation value and attribute matching value of each plot, obtaining a comprehensive quality score for each plot, which also includes: Obtain historical cultivated land data for multiple consecutive time periods preceding the current time period; Obtain the historical comprehensive quality score of each plot of land in the historical cultivated land data of the aforementioned multiple consecutive time periods; For each plot of land, calculate the comprehensive quality score for the current time period based on the first or second quality adjustment value of each plot of land and the historical comprehensive quality score.

5. The method for assessing the quality of cultivated land data based on geospatial data according to claim 4, characterized in that: Based on the comprehensive quality score of all plots, the overall quality level of the cultivated land data is determined, including: In response to the fact that the overall quality score of the land parcel is greater than a preset score threshold, an expansion operation is performed on the land parcel to expand and generate a high-quality area; If the overall quality score of the land parcel is less than or equal to the preset score threshold, the area consisting of the land parcel is designated as a qualified area. Areas outside of the high-quality and qualified areas in the cultivated land data are considered unqualified areas.

6. A geospatial-based farmland data quality assessment device, applicable to the geospatial-based farmland data quality assessment method according to any one of claims 1-5, characterized in that, include: The data acquisition unit is used to collect farmland spatial data and farmland attribute data for the current time period, as well as obtain reference farmland spatial data for the previous time period. The change calculation unit is used to obtain a spatial consistency index based on the cultivated land spatial data of the current time period and the reference cultivated land spatial data; wherein, the spatial consistency index includes the spatial change value of each plot; An attribute matching unit is used to acquire geospatial data for the current time period, and to acquire attribute accuracy indicators based on the cultivated land attribute data and geospatial data for the current time period; wherein, the attribute accuracy indicators include the attribute matching value for each plot. The comprehensive scoring unit is used to traverse each plot of land and obtain a comprehensive quality score for each plot of land based on the spatial change value and attribute matching value of each plot of land. The quality assessment unit is used to determine the overall quality level of arable land data based on the comprehensive quality score of all plots. Based on the farmland spatial data for the current time period and the reference farmland spatial data, spatial consistency indicators are obtained, including: Extract the first plot boundary of the cultivated land spatial data for the current time period and the second plot boundary of the reference cultivated land spatial data; The overlap rate of the land parcels is calculated based on the overlapping area of ​​the first land parcel boundary and the second land parcel boundary. The overlap rate of the plots is checked using the first spatial weight to obtain the first spatial enhancement value; The overlap rate of the land parcels is processed using a second spatial weight kernel to obtain a second spatial enhancement value; wherein, the first spatial weight kernel and the second spatial weight kernel are used to focus on different land parcels; The first spatial enhancement value and the second spatial enhancement value are fused to obtain the spatial enhancement overlap rate; The spatial consistency index is obtained based on the plot overlap rate and the spatial enhancement overlap rate. Obtain geospatial data for the current time period, and based on the cultivated land attribute data and geospatial data for the current time period, obtain attribute accuracy indicators, including: Obtain geospatial data for the current time period; wherein, the geospatial data includes farmland distribution data and remote sensing image data; Based on the farmland distribution data, vector information of plot boundaries is extracted; based on the vector information, a farmland spatial continuity index is calculated; based on the spatial continuity index, fragmented areas are detected in the farmland distribution data; wherein, the spatial continuity index includes the minimum and maximum spacing between adjacent plots; If such a problem exists, the corresponding land parcel is identified as a problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on the remote sensing image data, and attribute matching values ​​are obtained. Based on the farmland distribution data, vector information of plot boundaries is extracted; based on the vector information, a farmland spatial continuity index is calculated; based on the spatial continuity index, the presence of fragmented areas in the farmland distribution data is detected, including: Edge detection is performed on the farmland distribution data to obtain an initial boundary line; the initial boundary line is then smoothed to generate the vector information. Based on the vector information, determine the boundary points of adjacent plots; Calculate the minimum and maximum distances between the boundary points of adjacent plots, as the spatial continuity index; Set the minimum spacing threshold and the maximum spacing threshold; If the minimum distance between the boundary points of adjacent plots is less than the minimum spacing threshold, or the maximum distance is greater than the maximum spacing threshold, then a fragmented area is determined to exist. If such a problem exists, the corresponding land parcel is identified as a problem land parcel based on the location information of the fragmented area; the authenticity of the problem land parcel is verified based on the remote sensing image data, and attribute matching values ​​are obtained, including: Based on the location information of the fragmented areas, locate the corresponding plots in the cultivated land distribution data; Mark the located plot of land as a problem plot; Based on the location information of the problematic land parcel, local images of the corresponding area are extracted from the remote sensing image data; Match the attribute matching values ​​between the problem plot and the local image.

Images