Artificial Intelligence-Based Method and System for Land Consolidation Potential Analysis

Abstract

This invention relates to the field of artificial intelligence technology and proposes a method and system for analyzing the potential of comprehensive land consolidation based on artificial intelligence. The method includes: performing spatial overlay analysis on multi-source data of the target area to obtain potential influencing factors; using historical land consolidation patches as positive samples and unconsolidated areas as negative samples, and comparing the feature differences between the positive and negative samples to determine the consolidation intention discrimination rule; performing matching analysis between the consolidation intention discrimination rule and the potential influencing factors to obtain a preliminary consolidation intention value; performing spatial differentiation quantification on high-value cluster areas to obtain spatial weighting factors, and co-calibrating the preliminary consolidation intention value to obtain an optimized consolidation potential value; performing spatial cluster analysis based on the optimized consolidation potential value to obtain consolidation potential zones; evaluating the consolidation benefits of the consolidation potential zones and prioritizing them to obtain a recommended implementation order. This invention can improve the efficiency of land consolidation potential analysis based on artificial intelligence.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method and system for analyzing the potential of land consolidation based on artificial intelligence. Background Technology

[0002] In existing applications of technologies for analyzing the potential of land consolidation, the processing of multi-source land-related data lacks systematic spatial overlay analysis methods, making it difficult to accurately extract effective potential influencing factors. Furthermore, the lack of scientific discrimination rules that combine historical consolidation cases with the characteristic differences of unconsolidated areas results in a lack of objective basis and insufficient accuracy in determining consolidation intention values. Moreover, existing technologies do not perform spatial differentiation quantification and collaborative calibration of consolidation intention values, nor do they conduct professional spatial cluster analysis, making the zoning planning of consolidation potential lack scientific rigor and fail to align with the actual spatial distribution characteristics of the target region.

[0003] Current technologies for analyzing the potential of land consolidation lack standardized and intelligent design. The data processing and analysis results at each stage are poorly integrated, and the assessment of consolidation benefits relies solely on single indicators, making it difficult to generate a comprehensive benefit index. This results in a lack of rational basis for prioritizing consolidation implementation, low overall analysis efficiency, and limited guiding value of the analysis results for actual land consolidation work. Therefore, improving the accuracy and efficiency of land consolidation potential analysis has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for analyzing the potential of land consolidation based on artificial intelligence, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an artificial intelligence-based method for analyzing the potential of comprehensive land consolidation, comprising: Step a: Perform spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region; Step b: Using historical land consolidation plots in the target area as positive samples and unconsolidated areas in the target area as negative samples, determine the consolidation intention discrimination rules for the target area based on the characteristic differences between positive and negative samples in potential influencing factors. Step c: Perform a matching analysis between the rules for judging the willingness to carry out remediation and the potential influencing factors to obtain the preliminary remediation willingness value of the target area; Step d: Spatial differentiation quantification is performed on high-value clusters in the target region to obtain the spatial weight factor of the target region. Based on the spatial weight factor, the preliminary remediation intention value is collaboratively calibrated to obtain the optimization remediation potential value of the target region. Step e: Based on the optimized remediation potential value, perform spatial cluster analysis on the target area to obtain the remediation potential zoning of the target area; Step f: Evaluate the remediation benefits of the potential zones to obtain the expected benefits of the target areas, prioritize them according to the expected benefits, and obtain the recommended implementation order for the target areas.

[0006] In a preferred embodiment, the step of performing spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region includes: Acquire land use status data, territorial spatial planning data, socio-economic statistics data, and high-resolution remote sensing image data of the target area; Geometric correction is performed on high-resolution remote sensing image data to obtain a standardized image base for the target area; The physical properties of the target region are obtained by inverting and analyzing the standardized image base. Spatial processing is performed on land use status data, national spatial planning data, and socio-economic statistics to obtain thematic attribute layers for the target region; Spatially connect the physical attribute and thematic attribute layers to obtain the basic associated attributes of the target region; By performing feature screening on the basic correlation attributes, the potential influencing factors of the target region can be obtained.

[0007] In a preferred embodiment, the step of using historical land consolidation patches in the target area as positive samples and unconsolidated areas in the target area as negative samples, and determining the consolidation intention discrimination rule for the target area based on comparing the characteristic differences between the positive and negative samples in the potential influencing factors, includes: Extract land parcels that have undergone remediation from the historical land consolidation project database of the target region, and use them as positive samples for the target region; From the unremediated areas of the target region, extract unremediated land parcels that match the positive samples in terms of geographical environmental characteristics, and use them as negative samples for the target region; Extract the factor values ​​of land parcels in the potential influence factors from the positive and negative samples to obtain the positive and negative sample feature values ​​of the target region; By comparing the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples, candidate discriminant factors between positive and negative samples can be identified. Based on the candidate discriminant factors, a comprehensive analysis of the central tendency and dispersion in positive and negative samples is conducted to obtain the discriminant direction and discriminant limit of the candidate discriminant factors; Candidate discriminant factors are logically combined according to the discriminant direction and discriminant boundary to obtain the governance intention discrimination rule for the target area.

[0008] In a preferred embodiment, the step of comparing the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples to identify candidate discriminant factors between positive and negative samples includes: The positive sample feature value set and the negative sample feature value set are sorted according to their numerical values ​​to obtain the positive sample value sequence and the negative sample value sequence of the target region. Plotting the positive sample value sequence and the negative sample value sequence on the same numerical axis yields a comparison chart of the value distribution of the target region. If the overall numerical range of the positive sample value sequence is located on one side of the overall numerical range of the negative sample value sequence, then the potential influencing factor is determined to have discriminative power. If the numerical ranges of the positive sample value sequence and the negative sample value sequence intersect, then the median position and dispersion range of the positive sample value sequence and the negative sample value sequence are further compared. If the deviation of the median exceeds the comprehensive width of the dispersion range, then the potential influencing factor is determined to have discriminative power. Potentially influential factors that are deemed to have discriminative power will be used as candidate discriminant factors.

[0009] In a preferred embodiment, the step of matching the remediation intention judgment rule with potential influencing factors to obtain the preliminary remediation intention value of the target area includes: Extract the key discrimination factors and their corresponding discrimination directions and boundaries implicit in the rules for judging the willingness to rectify, and obtain the rule elements of the target region; The potential influencing factors of the target region are extracted and the measured characteristic values ​​of the target region on the key discriminant factors are obtained, thus obtaining the measured values ​​of the plot factors of the target region. The measured values ​​of land parcel factors are compared with the rule elements one by one to determine whether the key discrimination factors meet the discrimination direction and discrimination boundary, and the factor matching identifier of the target region is obtained. Based on factor matching identifiers, key discriminant factors of the target region are aggregated and integrated to obtain the preliminary remediation intention value of the target region.

[0010] In a preferred embodiment, the step of spatially differentiating and quantifying high-value clusters in the target region to obtain spatial weight factors for the target region, and then collaboratively calibrating the preliminary remediation intention value based on the spatial weight factors to obtain the optimized remediation potential value of the target region, includes: Based on the initial willingness to rectify, the spatial distribution of the willingness values ​​in the target area is analyzed to identify high-value clusters in the target area. Geometric features are extracted from the spatial morphology of high-value clusters to obtain the spatial attributes of clusters in the target region; The spatial differentiation degree of the spatial attributes of the cluster area is quantified to obtain the spatial weight factor of the target region. Based on spatial weighting factors, the preliminary remediation intention values ​​within and adjacent areas of high-value clusters are spatially correlated and fused to obtain the revised remediation intention values ​​for the target area. A gradual influence attenuation value is assigned to the buffer zone of the high-value cluster area, and the attenuation coefficient of the buffer zone is determined according to the distance from the center of the cluster area. Based on the attenuation coefficient and spatial weighting factor, the initial remediation intention value in the buffer zone is spatially coordinated and calibrated to obtain the calibrated remediation intention value of the target area. The modified remediation intention value and the calibrated remediation intention value are summarized and calculated to obtain the optimization remediation potential value of the target area.

[0011] In a preferred embodiment, the formula for calculating the optimization and remediation potential value is as follows: ; in, To optimize the potential value for rectification, To adjust the willingness to rectify, Spatial weighting factor, To calibrate the willingness to rectify, The attenuation coefficient is... This is the deviation correction factor. To determine the initial willingness to rectify the situation, The average value of the initial willingness to rectify.

[0012] In a preferred embodiment, the step of performing spatial cluster analysis on the target region based on the optimized remediation potential value to obtain the remediation potential partitions of the target region includes: Spatial distribution analysis of the potential values ​​for optimization and remediation is performed to obtain the initial cohesion base points of the target area; Using the initial agglomeration base point as the center, the spatial proximity of the optimization and remediation potential values ​​of the surrounding plots is determined to obtain the primary clustering map of the target area; Merging degree test is performed on the primary cluster patches to obtain the intermediate cluster of the target region; The distribution range of potential values ​​in intermediate clusters is divided into intervals to obtain the remediation potential zones of the target region.

[0013] In a preferred embodiment, the step of assessing the remediation benefits of potential zones to obtain the expected benefits of the target areas, and prioritizing them according to the expected benefits to obtain a recommended implementation order for the target areas, includes: Attribute information is extracted from zones with potential for improvement to obtain the basic attributes of the target region. By comprehensively calculating the various indicators in the basic attribute set, the expected benefit value of the target region can be obtained. The expected benefit values ​​are standardized to obtain the comprehensive benefit index of the target region; Based on the comprehensive benefit index, the potential zones for remediation are ranked in descending order to obtain the priority remediation sequence for the target areas.

[0014] To address the aforementioned problems, this invention also provides an artificial intelligence-based land consolidation potential analysis system, the system comprising: The overlay analysis module is used to perform spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region. The module for constructing rules for judging the willingness to remediate is used to determine the rules for judging the willingness to remediate in the target area by taking the historical land remediation patches of the target area as positive samples and the unremediated areas of the target area as negative samples, and by comparing the characteristic differences between the positive and negative samples in the potential influencing factors. The matching analysis module is used to perform matching analysis between the rules for judging the willingness to carry out remediation and the potential influencing factors to obtain the preliminary remediation willingness value of the target area; The collaborative calibration module is used to perform spatial differentiation quantification on high-value clusters in the target area to obtain the spatial weight factor of the target area, and based on the spatial weight factor, to perform collaborative calibration on the preliminary remediation intention value to obtain the optimization remediation potential value of the target area. The spatial clustering analysis module is used to perform spatial clustering analysis on the target area based on the optimized remediation potential value, and obtain the remediation potential zoning of the target area; The benefit assessment module is used to assess the benefits of remediation in potential zones, obtain the expected benefits of the target areas, prioritize them according to the expected benefits, and obtain the recommended implementation order for the target areas.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves intelligent analysis of land consolidation potential throughout the entire process by relying on artificial intelligence technology. It accurately extracts potential influencing factors by conducting spatial overlay analysis on multi-source data, constructs scientific rules for judging consolidation intentions by combining the differences in characteristics of positive and negative samples, and then optimizes the results layer by layer through matching analysis, spatial differentiation quantification, and collaborative calibration to obtain accurate optimized consolidation potential values. At the same time, it achieves reasonable zoning of consolidation potential through spatial clustering and prioritizes the implementation order based on benefit assessment. The entire process realizes the digitalization, refinement, and standardization of land consolidation potential analysis, which greatly improves the accuracy and scientific nature of the consolidation potential analysis results and makes the determination of land consolidation potential more in line with the actual geographical and land use characteristics of the target area.

[0016] 2. This invention achieves the technical implementation of each stage of land consolidation potential analysis by establishing a multi-module collaborative analysis system. Each module performs specialized processing for different stages of the analysis process, which can efficiently mine effective information from multi-source data and accurately calibrate the remediation intention value through the quantitative assignment of parameters such as spatial weight factors and attenuation coefficients. At the same time, it realizes the scientific planning of remediation zoning and implementation ranking through cluster analysis and benefit evaluation, which effectively improves the overall efficiency of land consolidation potential analysis. This allows the results of the remediation potential analysis to directly provide clear and feasible decision-making basis for the actual implementation of land consolidation, and improves the pertinence and effectiveness of land consolidation planning and implementation. Attached Figure Description

[0017] Figure 1 A flowchart illustrating an artificial intelligence-based land consolidation potential analysis method according to an embodiment of the present invention; Figure 2 A functional module diagram of an artificial intelligence-based land consolidation potential analysis system provided in an embodiment of the present invention; Figure 3 The experimental curves for land consolidation potential analysis scenario group 1 provided in an embodiment of the present invention are used to demonstrate the experimental effect of the system under the conditions of high spatial weight and clustering influence. Figure 4 The experimental curves for scenario group 2 of the land consolidation potential analysis provided in an embodiment of the present invention are used to demonstrate the experimental effect of the system under conditions of high consolidation willingness and suitability. Figure 5 The experimental curves for scenario group 3 of the land consolidation potential analysis provided in an embodiment of the present invention are used to demonstrate the experimental effect of the system under the conditions of high suitability and clustering influence. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides an artificial intelligence-based method for analyzing the potential of land consolidation. The executing entity of this artificial intelligence-based land consolidation potential analysis method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the artificial intelligence-based land consolidation potential analysis method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a land consolidation potential analysis method based on artificial intelligence, according to an embodiment of the present invention. In this embodiment, the land consolidation potential analysis method based on artificial intelligence includes: Step a: Perform spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region; In this embodiment of the invention, the step of performing spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region includes: Acquire land use status data, territorial spatial planning data, socio-economic statistics data, and high-resolution remote sensing image data of the target area; Geometric correction is performed on high-resolution remote sensing image data to obtain a standardized image base for the target area; The physical properties of the target region are obtained by inverting and analyzing the standardized image base. Spatial processing is performed on land use status data, national spatial planning data, and socio-economic statistics to obtain thematic attribute layers for the target region; Spatially connect the physical attribute and thematic attribute layers to obtain the basic associated attributes of the target region; By performing feature screening on the basic correlation attributes, the potential influencing factors of the target region can be obtained.

[0021] Land use status data, territorial spatial planning data, socio-economic statistics data, and high-resolution remote sensing imagery data are obtained from land management departments, statistical departments, and remote sensing data platforms corresponding to the target region. Among them, the land use status data includes spatial distribution and area information of types such as cultivated land, construction land, and unused land; the territorial spatial planning data includes permanent basic farmland protection red lines, ecological protection red lines, urban development boundaries, and various land use planning layouts; the socio-economic statistics data include indicators such as population density, per capita cultivated land area, and output value of secondary and tertiary industries; and the high-resolution remote sensing imagery data consists of meter- or sub-meter-resolution optical images covering the target region, ensuring complete data coverage and consistent time points.

[0022] Geometric correction is performed on high-resolution remote sensing image data. Ground control points are selected that are evenly distributed within the target area. These ground control points are fixed features that are clearly identifiable in both the image and the actual ground, such as road intersections, field ridges, and corners. Based on the known ground coordinates and the corresponding pixel coordinates in the image, a coordinate transformation relationship is established. A polynomial fitting method is used to adjust the image position and angle pixel by pixel to eliminate geometric distortions caused by sensor attitude, terrain undulations, and atmospheric refraction. The image is then unified to the nationally prescribed plane coordinate system to obtain a standardized image base for the target area.

[0023] The standardized image substrate is inverted and analyzed. Based on the spectral reflectance characteristics of different image bands, the surface cover type, vegetation cover and topographic relief information are extracted pixel by pixel. By identifying the spectral differences of different land features in the visible and near-infrared bands, the cover types such as cultivated land, forest land and construction land are distinguished. The vegetation cover is obtained by calculating the reflectance ratio of the near-infrared and red bands. The topographic elevation data is obtained by solving the parallax information of the image stereo pairs. The spectral signal is converted into quantifiable surface physical information to obtain the physical attributes of the target area.

[0024] Spatialization processing is performed on land use status data, national spatial planning data, and socio-economic statistics. Non-spatial statistical data are matched to the spatial coordinates of corresponding administrative units or geographic grids. Statistical indicators are bound to spatial boundaries one by one. Vector-format planning and status data are unified to the same spatial coordinate system as the standardized image base. Data projection transformation and boundary matching are completed, spatial offset of data from different sources is eliminated, and thematic attribute layers of the target region are obtained.

[0025] Spatially connect physical attribute and thematic attribute layers, using the same geographic grid or administrative unit as the matching benchmark. Aggregate and statistically analyze the surface information extracted pixel by pixel from the physical attributes by spatial unit to obtain the average vegetation cover, average topographic relief, and proportion of surface cover type for each unit. At the same time, associate the land use type planning constraint indicators and socio-economic data of the corresponding spatial units in the thematic attribute layer with the same spatial unit, so as to achieve unit-by-unit binding of physical information and thematic information and obtain the basic association attributes of the target region.

[0026] Feature screening is performed on the basic correlation attributes. All basic correlation attribute fields are traversed, and the correlation between each field and land consolidation potential is analyzed one by one. Redundant fields that are not directly related to consolidation potential are eliminated, and fields that can reflect the degree of planning constraints and development suitability of land use efficiency are retained, including the degree of contiguousness of cultivated land, the degree of intensification of construction land, the ecological protection level, the population carrying capacity, and the planning use control requirements. This ensures that the screened fields can comprehensively and accurately depict the core characteristics of the consolidation potential of the target area and obtain the potential influencing factors of the target area.

[0027] The beneficial effects are as follows: through the systematic acquisition and standardized processing of multi-source data, comprehensive collection and accurate correlation of information related to land consolidation potential are achieved; the geometric correction and inversion analysis ensure the accuracy and reliability of physical attributes; spatial processing and spatial connection realize the organic integration of different types of data; and the feature screening process effectively eliminates redundant information, ensuring that the potential influencing factors can accurately reflect the core consolidation potential characteristics of the target area. This provides a reliable data foundation for the subsequent determination of consolidation intention values ​​and the calculation of consolidation potential values, thereby improving the accuracy and pertinence of land consolidation potential analysis.

[0028] Step b: Using historical land consolidation plots in the target area as positive samples and unconsolidated areas in the target area as negative samples, determine the consolidation intention discrimination rules for the target area based on the characteristic differences between positive and negative samples in potential influencing factors. In this embodiment of the invention, the step of using historical land consolidation patches of the target region as positive samples and unconsolidated areas of the target region as negative samples, and determining the consolidation intention discrimination rule of the target region based on comparing the characteristic differences between the positive and negative samples in the potential influencing factors, includes: Extract land parcels that have undergone remediation from the historical land consolidation project database of the target region, and use them as positive samples for the target region; From the unremediated areas of the target region, extract unremediated land parcels that match the positive samples in terms of geographical environmental characteristics, and use them as negative samples for the target region; Extract the factor values ​​of land parcels in the potential influence factors from the positive and negative samples to obtain the positive and negative sample feature values ​​of the target region; By comparing the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples, candidate discriminant factors between positive and negative samples can be identified. Based on the candidate discriminant factors, a comprehensive analysis of the central tendency and dispersion in positive and negative samples is conducted to obtain the discriminant direction and discriminant limit of the candidate discriminant factors; Candidate discriminant factors are logically combined according to the discriminant direction and discriminant boundary to obtain the governance intention discrimination rule for the target area.

[0029] The comparison of the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples to identify candidate discriminant factors between positive and negative samples includes: The positive sample feature value set and the negative sample feature value set are sorted according to their numerical values ​​to obtain the positive sample value sequence and the negative sample value sequence of the target region. Plotting the positive sample value sequence and the negative sample value sequence on the same numerical axis yields a comparison chart of the value distribution of the target region. If the overall numerical range of the positive sample value sequence is located on one side of the overall numerical range of the negative sample value sequence, then the potential influencing factor is determined to have discriminative power. If the numerical ranges of the positive sample value sequence and the negative sample value sequence intersect, then the median position and dispersion range of the positive sample value sequence and the negative sample value sequence are further compared. If the deviation of the median exceeds the comprehensive width of the dispersion range, then the potential influencing factor is determined to have discriminative power. Potentially influential factors that are deemed to have discriminative power will be used as candidate discriminant factors.

[0030] The process involves extracting land parcels that have undergone remediation from the historical land consolidation project database of the target region. First, all land consolidation projects that have completed project approval, construction, and acceptance by relevant authorities are retrieved from the database. The spatial vector boundary integrity and coordinate accuracy of each land parcel are verified, and invalid parcels with blurred boundaries, offset coordinates, or overlaps are removed. Simultaneously, core attribute information from the project files, such as remediation type, implementation area, project completion time, post-remediation land use status, and acceptance conclusions, is linked to select land parcels that meet the technical standards for land consolidation implementation and have complete attribute information. These are then used as positive samples for the target region. This positive sample set comprehensively covers remediated areas within the target region with different terrain conditions, remediation types, and development intensities, providing a standardized benchmark for subsequent feature difference comparisons.

[0031] From the unremediated areas of the target region, unremediated land parcels matching the positive samples in terms of geographical environmental characteristics are extracted. First, the spatial location information of the positive samples is overlaid with a thematic layer of geographical environmental characteristics of the target region. Core geographical environmental characteristic parameters, such as topographic slope, altitude, basic land use type, soil texture, hydrological conditions, and locational distance to towns and major transportation routes, are extracted for each positive sample to form a positive sample feature parameter library. Then, in the vector layer of the unremediated areas of the target region, a reasonable spatial search range is defined centered on each positive sample, and within this range, a search is conducted... Unremediated plots that highly match the corresponding positive samples in terms of core geographical environmental features such as topographic slope, land use type, soil texture, and hydrological conditions are selected. Then, the feature value similarity of the selected unremediated plots is quantitatively calculated, and plots with similarity that do not meet the preset standard are removed. Finally, unremediated plots with high similarity, reasonable spatial distribution, and complete attribute information are retained as negative samples of the target area. This ensures that the negative samples and positive samples are highly comparable in terms of natural endowment and external geographical environment, and avoids deviations in the subsequent construction of discrimination rules due to large differences in sample features.

[0032] The process involves extracting factor values ​​from land parcels in the potential impact factors from both positive and negative samples. First, the vector boundaries of the land parcels in both samples are precisely registered with the potential impact factor spatial raster layer of the target region, ensuring they are in the same spatial coordinate system. Then, using a single land parcel as the basic analysis unit, all potential impact factor raster cells covered by the parcel's boundary are traversed, and the factor value corresponding to each raster cell is extracted one by one. For cases where a single land parcel covers multiple raster cells, the average value of all raster cell values ​​under the corresponding potential impact factor is calculated and used as the unique factor representation value for that land parcel under that potential impact factor. Simultaneously, a unique spatial location identifier is established for each land parcel, and the identifier is associated with and stored one-to-one with all corresponding potential impact factor values. Finally, the positive sample feature values ​​corresponding to all positive samples and the negative sample feature values ​​corresponding to all negative samples within the target region are summarized, ensuring that the feature value of each sample is accurately bound to its actual spatial attributes.

[0033] By comparing the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples, candidate discriminant factors between positive and negative samples are identified. First, for each potential influencing factor, the set of feature values ​​of all positive samples and the set of feature values ​​of all negative samples under that factor are organized. Then, numerical sorting and distribution visualization analysis are carried out on the two feature value sets. The factor discrimination is judged by combining the numerical interval distribution, the degree of deviation of the median, and other dimensions. Finally, potential influencing factors that can effectively distinguish between positive and negative samples are selected as candidate discriminant factors.

[0034] The positive and negative sample feature values ​​are sorted according to their numerical values. For each potential influencing factor in the target region, the positive and negative sample feature value sets corresponding to that factor are processed separately. All values ​​in the positive sample feature value set are arranged in ascending order to form a non-repeating, ordered sequence of positive sample values. At the same time, all values ​​in the negative sample feature value set are arranged according to the same ascending order to form a sequence of negative sample values. This sorting process covers all potential influencing factors in the target region, ensuring that the positive and negative sample values ​​corresponding to each factor form a regular and ordered numerical sequence, providing a standardized data foundation for subsequent distribution difference comparisons.

[0035] The positive and negative sample value sequences are plotted on the same numerical axis, with the magnitude of the potential impact factor value as the horizontal axis and the proportion of the sample size to the total sample size as the vertical axis. The proportion of the sample size corresponding to each value in the positive and negative sample value sequences is marked on the axis, and the numerical distribution density of the two types of samples is marked point by point. At the same time, the overall numerical range of the positive sample values ​​is outlined with solid lines and the overall numerical range of the negative sample values ​​is outlined with dashed lines. This visually presents the overlap, separation trend, and density concentration areas of the two types of samples in the value of the potential impact factor, resulting in a value distribution comparison map of the target region, providing a clear visual basis for subsequent judgment on whether the factor has discriminative power.

[0036] If the overall numerical range of the positive sample value sequence is located on one side of the overall numerical range of the negative sample value sequence, then the potential impact factor is determined to have discriminative power. First, determine the minimum and maximum values ​​of the positive sample value sequence, which together constitute the overall numerical range of the positive sample. Then, determine the minimum and maximum values ​​of the negative sample value sequence, which together constitute the overall numerical range of the negative sample. Subsequently, accurately compare the spatial distribution of the two numerical ranges. If all values ​​of the positive sample are higher than all values ​​of the negative sample, or all values ​​of the positive sample are lower than all values ​​of the negative sample, that is, the positive sample numerical range and the negative sample numerical range have no overlap and show obvious unilateral distribution characteristics, then it is determined that the potential impact factor can effectively distinguish between the remediated plots and the unremediated matching plots, and the factor is directly determined to have discriminative power.

[0037] If the numerical ranges of the positive and negative sample value sequences intersect, the median position and dispersion range of the positive and negative sample value sequences are further compared. If the deviation of the median exceeds the comprehensive width of the dispersion range, the potential impact factor is determined to have discriminative power. First, the median of the positive and negative sample value sequences are calculated separately. The median is the value in the middle of the corresponding ordered sequence, directly reflecting the core central tendency of the sample values. Then, the dispersion range of the positive and negative sample value sequences are calculated separately. The dispersion range is obtained by calculating the difference between the maximum and minimum values ​​in the corresponding sequence, intuitively reflecting the degree of dispersion of the sample values. Subsequently, the numerical difference between the two medians is calculated, which is the deviation of the median. Then, the dispersion ranges of the positive and negative samples are added together to obtain the comprehensive width of the dispersion range. Finally, the deviation of the median and the comprehensive width of the dispersion range are compared numerically. If the value of the deviation of the median is greater than the value of the comprehensive width of the dispersion range, the potential impact factor is determined to have discriminative power.

[0038] Potential impact factors deemed to have discriminative power are used as candidate discriminant factors. After the discriminative power of all potential impact factors in the target region is determined, all determination results are systematically reviewed, and potential impact factors that do not have discriminative power and cannot effectively distinguish between positive and negative samples are eliminated. All potential impact factors that have been determined to have discriminative power are summarized and organized to form a set of candidate discriminant factors for the target region, providing a core factor basis for subsequent determination of the discrimination direction and discrimination boundary.

[0039] Based on candidate discriminant factors, a comprehensive analysis of the central tendency and dispersion in positive and negative samples is conducted to obtain the discriminant direction and discriminant limit of the candidate discriminant factors. For each factor in the candidate discriminant factor set, the positive and negative sample feature values ​​under that factor are extracted, and the mean, median, and interquartile range of the positive sample feature values ​​are calculated. At the same time, the mean, median, and interquartile range of the negative sample feature values ​​are also calculated. By comparing the relative positions of the mean and median of the positive and negative samples, the discriminant direction of the candidate discriminant factor is determined, that is, it is clear what kind of change trend the plot of land meets on this factor to be more likely to have remediation attributes. By analyzing the overlap range of the interquartile range of the positive and negative samples, combined with the numerical distribution characteristics, the discriminant limit of the candidate discriminant factor is determined, that is, it is clear what threshold the plot of land meets the remediation judgment requirements when the value of this factor reaches. Finally, a corresponding discriminant direction and discriminant limit are matched for each candidate discriminant factor.

[0040] Candidate discriminant factors are logically combined according to their discriminant direction and boundary to obtain the remediation intention judgment rules for the target area. First, all candidate discriminant factors and their corresponding discriminant directions and boundaries are sorted out. Combining the actual implementation requirements of comprehensive land remediation and the land use planning principles of the target area, the logical relationships between each candidate discriminant factor are determined. Using "AND" and "OR" as the core logical connectors, the judgment rules of individual candidate discriminant factors are systematically linked to form a comprehensive judgment logic that can simultaneously meet multiple factor constraints and fit the actual situation of the target area. This logic clarifies the various discriminant factor requirements that the plot must meet, and finally forms the remediation intention judgment rules for the target area, providing a clear and unified judgment basis for the subsequent determination of remediation intention values.

[0041] Extract land parcels that have undergone remediation from the historical land consolidation project database of the target region. These parcels are all remediation sites that have completed construction and passed acceptance, and contain clear remediation boundaries, remediation types, and implementation time information, serving as positive samples for the target region.

[0042] From the unremediated areas of the target region, unremediated land parcels that match the positive samples in terms of geographical environmental characteristics are extracted. During the matching process, the topography, slope, land use type and location conditions of the positive samples are compared to screen out unremediated land parcels that are spatially adjacent and have similar natural endowments, which are then used as negative samples for the target region.

[0043] Extract the factor values ​​of land parcels in the potential influence factors from the positive and negative samples. Traverse the spatial boundary of each land parcel and match it pixel by pixel with the spatial layer of the potential influence factors. Summarize the values ​​of all potential influence factors corresponding to each land parcel to obtain the positive and negative sample feature values ​​of the target region.

[0044] The positive and negative sample feature values ​​are sorted according to their numerical values. All positive sample feature values ​​under the same potential influencing factor are arranged from low to high, and all negative sample feature values ​​are also arranged from low to high, thus obtaining the positive and negative sample value sequences of the target region.

[0045] Plot the positive and negative sample value sequences on the same numerical axis, with the numerical value as the horizontal axis and the sample number as the vertical axis, to mark the numerical distribution range and density of the two types of samples, thus obtaining a comparison map of the value distribution of the target region.

[0046] If the overall numerical range of the positive sample value sequence is located on one side of the overall numerical range of the negative sample value sequence, then the potential influencing factor is determined to have discriminative power, that is, the values ​​of the positive and negative samples on this factor do not overlap, and the samples can be directly distinguished.

[0047] If the numerical ranges of the positive and negative sample value sequences intersect, then the median positions and dispersion ranges of the positive and negative sample value sequences are further compared. First, the median of the two types of sample value sequences is calculated separately, and then the combined width of the dispersion range of the positive and negative samples is calculated. The difference between the medians is compared with the combined width. If the deviation of the median exceeds the combined width of the dispersion range, then the potential influencing factor is determined to have discriminative power.

[0048] Potentially discriminating factors that are deemed to have discriminative power are selected as candidate discriminant factors. All potentially discriminating factors that can effectively distinguish between positive and negative samples are retained, while factors that cannot distinguish between samples are removed.

[0049] Based on the candidate discriminant factors, a comprehensive analysis of the central tendency and dispersion of positive and negative samples is conducted. The mean, median and interquartile range of positive and negative samples are calculated for each candidate discriminant factor. The discriminant direction is determined based on the relative position of the mean and median, and the discriminant limit is determined based on the overlap range of the interquartile range. Thus, the discriminant direction and discriminant limit of the candidate discriminant factors are obtained.

[0050] Candidate discrimination factors are logically combined according to discrimination direction and discrimination boundary. The discrimination rules of multiple candidate discrimination factors are linked together using AND-OR logic to form a judgment logic that can simultaneously satisfy the constraints of multiple factors, thus obtaining the discrimination rules for the remediation intention of the target area.

[0051] The beneficial effects are as follows: a comparative sample set is constructed through precise matching of positive and negative samples; candidate discriminant factors with discriminative power are intuitively identified based on numerical sorting and distribution visualization; scientific discriminant direction and discriminant boundary are determined by combining the analysis of central tendency and dispersion; and finally, rigorous rules for judging remediation intention are formed through logical combination, which can accurately identify plots with remediation intention, providing a reliable rule basis for the subsequent calculation of remediation intention value and the generation of remediation potential value, thereby improving the objectivity and accuracy of land remediation potential analysis.

[0052] Step c: Perform a matching analysis between the rules for judging the willingness to carry out remediation and the potential influencing factors to obtain the preliminary remediation willingness value of the target area; In this embodiment of the invention, the step of matching the remediation intention judgment rule with potential influencing factors to obtain the preliminary remediation intention value of the target area includes: Extract the key discrimination factors and their corresponding discrimination directions and boundaries implicit in the rules for judging the willingness to rectify, and obtain the rule elements of the target region; The potential influencing factors of the target region are extracted and the measured characteristic values ​​of the target region on the key discriminant factors are obtained, thus obtaining the measured values ​​of the plot factors of the target region. The measured values ​​of land parcel factors are compared with the rule elements one by one to determine whether the key discrimination factors meet the discrimination direction and discrimination boundary, and the factor matching identifier of the target region is obtained. Based on factor matching identifiers, key discriminant factors of the target region are aggregated and integrated to obtain the preliminary remediation intention value of the target region.

[0053] The text and logical structure of the rules for judging the willingness to rectify were comprehensively deconstructed and sorted out. The core potential influencing factors that were clearly pointed to in each rule were extracted as key judgment factors. At the same time, the judgment trend corresponding to each key judgment factor was accurately extracted as the judgment direction and the judgment value threshold as the judgment boundary. Each key judgment factor was bound to its corresponding judgment direction and judgment boundary one by one to form a structured, list-style information set. After the extraction and integration of all key information, the rule elements of the target region were obtained.

[0054] Using independent plots within the target region as the basic analysis unit, the full potential impact factors corresponding to each plot are systematically screened. Based on the names and attributes of key discriminant factors in the rule elements, precise information matching is performed. The actual measured statistical values ​​of each plot on all key discriminant factors are selected from the full potential impact factors as measured representation values. A unique spatial identification information is established for each plot. The spatial identification is associated with and stored with the corresponding measured representation values ​​of key discriminant factors. After completing the information extraction and storage of all plots within the target region, the measured values ​​of plot factors in the target region are obtained.

[0055] Using a single plot of land within the target region as an independent comparison unit, the measured values ​​of the plot factors corresponding to that plot are matched and verified one by one with the key discriminant factors in the rule elements. Strictly following the established discrimination direction in the rule elements, the measured values ​​of the plot factors are verified to see if they meet the numerical requirements set by the discrimination boundary. The comparison results of each key discriminant factor are clearly marked with both text and symbols. Those that meet the discrimination direction and boundary requirements are marked with a matching mark, and those that do not meet the discrimination direction and boundary requirements are marked with a non-match mark. After completing the comparison of all key discriminant factors of all plots within the target region, the factor matching marks of the target region are obtained.

[0056] A systematic summary and statistical analysis of all factor matching identifiers corresponding to each plot within the target area is conducted. The proportion of the number of matching identifiers in each plot to the total number of key discriminant factors for that plot is calculated. In conjunction with the actual assessment requirements of comprehensive land consolidation, this proportion is converted into a specific value that can intuitively reflect the degree of willingness to remediate the plot. After calculating and assigning this value to all plots within the target area, the preliminary remediation willingness value of the target area is obtained.

[0057] The beneficial effects are as follows: by accurately decomposing and extracting standardized rule elements from the rules for judging remediation intentions, the judgment criteria are clarified and structured. Potential influencing factors are extracted and measured values ​​are screened on an independent plot basis, ensuring the accuracy and relevance of the measured values ​​of plot factors. Objective determination of factor matching identifiers is achieved through item-by-item comparison. Finally, the quantitative assignment of preliminary remediation intention values ​​is completed based on the aggregation and fusion of matching identifiers. The entire process forms a standardized and procedural remediation intention judgment system, effectively improving the objectivity and accuracy of the preliminary remediation intention value judgment. This provides reliable and regionally relevant basic data for subsequent optimization of remediation potential value calculation. Furthermore, this judgment process aligns with the quantitative assessment needs of the land planning field, providing accurate numerical references for the preliminary assessment of comprehensive land remediation.

[0058] Step d: Spatial differentiation quantification is performed on high-value clusters in the target region to obtain the spatial weight factor of the target region. Based on the spatial weight factor, the preliminary remediation intention value is collaboratively calibrated to obtain the optimization remediation potential value of the target region. In this embodiment of the invention, the step of spatially differentiating and quantifying high-value clusters in the target region to obtain spatial weight factors for the target region, and then collaboratively calibrating the preliminary remediation intention value based on the spatial weight factors to obtain the optimized remediation potential value of the target region, includes: Based on the initial willingness to rectify, the spatial distribution of the willingness values ​​in the target area is analyzed to identify high-value clusters in the target area. Geometric features are extracted from the spatial morphology of high-value clusters to obtain the spatial attributes of clusters in the target region; The spatial differentiation degree of the spatial attributes of the cluster area is quantified to obtain the spatial weight factor of the target region. Based on spatial weighting factors, the preliminary remediation intention values ​​within and adjacent areas of high-value clusters are spatially correlated and fused to obtain the revised remediation intention values ​​for the target area. A gradual influence attenuation value is assigned to the buffer zone of the high-value cluster area, and the attenuation coefficient of the buffer zone is determined according to the distance from the center of the cluster area. Based on the attenuation coefficient and spatial weighting factor, the initial remediation intention value in the buffer zone is spatially coordinated and calibrated to obtain the calibrated remediation intention value of the target area. The modified remediation intention value and the calibrated remediation intention value are summarized and calculated to obtain the optimization remediation potential value of the target area.

[0059] The formula for calculating the potential value of optimization and remediation is as follows: ; in, To optimize the potential value for rectification, To adjust the willingness to rectify, Spatial weighting factor, To calibrate the willingness to rectify, The attenuation coefficient is... This is the deviation correction factor. To determine the initial willingness to rectify the situation, The average value of the initial willingness to rectify.

[0060] Based on the initial remediation intention values, the spatial distribution of intention values ​​in the target area is analyzed. First, the initial remediation intention values ​​of all plots in the target area are precisely bound to the corresponding spatial vector boundaries to form a spatial distribution layer of initial remediation intention values ​​covering the entire target area. Then, a full-area spatial trend analysis is carried out on this layer to identify the distribution areas of plots with high initial remediation intention values. Adjacent high-value plots are spatially aggregated to delineate continuous and clearly defined high-value plot clusters. After completing the identification and delineation of all high-value clusters, the high-value cluster area of ​​the target area is obtained.

[0061] Geometric features of the spatial morphology of high-value clusters are extracted. Taking each independent high-value cluster as an analysis unit, the spatial geometry of the cluster is extracted in all dimensions. The planar geometric area, perimeter of the outer contour, shape regularity, spatial center coordinates and connectivity of the plots within the cluster are accurately calculated. At the same time, the spatial adjacency relationship between the cluster and the surrounding geographical elements is extracted. All extracted geometric features and spatial relationship information are systematically integrated to establish a unique spatial feature information set for each high-value cluster. After the feature extraction of all high-value clusters is completed, the spatial attributes of the clusters in the target region are obtained.

[0062] The spatial differentiation degree of the cluster area is quantified by first standardizing the various geometric characteristic indicators in the spatial attributes of the cluster area to eliminate the differences in the dimensions of different indicators. Then, according to the spatial planning requirements of comprehensive land consolidation, a fixed weight ratio is assigned to each standardized geometric characteristic indicator. Subsequently, the indicators are comprehensively calculated according to the weight ratio to obtain the spatial differentiation quantification value of each high-value cluster area. This value directly reflects the spatial distribution characteristics of the cluster area and the spatial suitability of consolidation and development. This value is used as the spatial quantification basis for the corresponding high-value cluster area and its surrounding areas to obtain the spatial weight factor of the target region.

[0063] Based on spatial weighting factors, the preliminary remediation intention values ​​within and adjacent to high-value clusters are spatially correlated and fused. First, the core internal range and direct adjacent range of each high-value cluster are delineated. Spatial weighting factors are assigned to the corresponding plots within the core internal range and direct adjacent range respectively. Then, the preliminary remediation intention value of each plot is spatially correlated with the corresponding spatial weighting factor. At the same time, combined with the spatial adjacency relationship between plots, the preliminary remediation intention values ​​of adjacent plots are mutually corrected to eliminate the spatial isolation of individual plot values. After completing the numerical fusion and correction of all plots within and adjacent to high-value clusters, the corrected remediation intention value of the target area is obtained.

[0064] A progressive influence attenuation assignment is applied to the buffer zones of high-value clusters. The attenuation coefficient of the buffer zones is determined based on their distance from the center of the clusters. First, using the spatial center coordinates of each high-value cluster as the center, multi-level buffer zones are delineated according to a preset distance gradient, clarifying the spatial range and boundaries of each level of buffer zone. Then, based on the spatial radiation effect law of land consolidation, a progressive influence attenuation assignment is applied to the buffer zones. The closer the buffer zone level is to the spatial center of the cluster, the higher the influence assignment is assigned, and the farther the buffer zone level is from the spatial center, the lower the influence assignment is assigned. The influence assignment of each level is converted into the quantitative coefficient of the corresponding buffer zone to obtain the attenuation coefficient of the target area.

[0065] Based on the attenuation coefficient and spatial weighting factor, the preliminary remediation intention value within the buffer zone is spatially coordinated and calibrated. First, the attenuation coefficient and spatial weighting factor of each level of buffer zone are integrated to obtain the comprehensive calibration coefficient corresponding to each buffer zone plot. Then, the preliminary remediation intention value of each plot within the buffer zone is correlated with the corresponding comprehensive calibration coefficient. According to the calculation results, the original preliminary remediation intention value is adjusted plot by plot. At the same time, the spatial correlation between the buffer zone plots and the core plots of the high-value cluster area is taken into account. The adjusted values ​​are then checked a second time. After completing the numerical calibration and verification of all buffer zone plots, the calibrated remediation intention value of the target area is obtained.

[0066] The revised remediation intention value and the calibrated remediation intention value are summarized and calculated. First, the attribute of each plot in the target area is clarified, and plots with revised remediation intention values ​​in the high-value cluster area and its adjacent areas and plots with calibrated remediation intention values ​​in the buffer zone are distinguished. Then, the two types of values ​​are integrated across the entire region, and the revised remediation intention values ​​or calibrated remediation intention values ​​of all plots are uniformly incorporated into the overall numerical system of the target region. After the numerical summary and integration calculation of all plots in the region is completed, the optimization remediation potential value of the target region is obtained.

[0067] The revised remediation intention value is obtained by matching and analyzing the characteristic differences between historical remediation cases and unremediated areas. The spatial weight factor is obtained by quantifying and assigning values ​​based on the spatial neighborhood relationships of the target region. The calibrated remediation intention value is obtained by performing spatial differentiation quantification and collaborative calibration on the preliminary remediation intention value. The attenuation coefficient is obtained by calculating and assigning values ​​based on the spatial influence range of the remediation intention value. The deviation correction coefficient is obtained by selecting values ​​within a preset range of 0.1 to 0.3 based on actual analysis needs. The preliminary remediation intention value is obtained by constructing discrimination rules after extracting potential influencing factors through spatial overlay analysis of multi-source land data. The average value of the preliminary remediation intention value is obtained by calculating the arithmetic mean of all preliminary remediation intention values.

[0068] The optimized land consolidation potential value is calculated by summing the product of the corrected consolidation intention value and the spatial weight factor, the product of the calibrated consolidation intention value and the attenuation coefficient, and the product of the deviation correction coefficient and the difference between the initial consolidation intention value and its average value. This achieves precise quantification and calibration of the land consolidation potential.

[0069] The optimization potential value increases when the product of the revised remediation intention value and the spatial weight factor increases. The optimization potential value also increases when the product of the calibrated remediation intention value and the attenuation coefficient increases. The optimization potential value increases when the difference between the initial remediation intention value and its average value widens and the deviation correction coefficient is positive. Conversely, the optimization potential value decreases when the difference between the initial remediation intention value and its average value narrows and the deviation correction coefficient is positive.

[0070] The beneficial effects are as follows: by analyzing the spatial distribution of the initial land consolidation intention value, high-value clusters are accurately identified; by extracting and quantifying the spatial attributes of the clusters, scientific spatial weighting factors are obtained, realizing a quantitative assessment of the spatial dimensions of land consolidation potential; then, a corrected land consolidation intention value is obtained through spatial correlation fusion; and the calibration of the consolidation intention value for the buffer zone is completed by combining progressive attenuation assignment. Finally, the optimized land consolidation potential value is obtained through summarization calculation. The entire process fully combines the spatial distribution characteristics and radiation effects of land consolidation potential, realizing multi-dimensional and refined spatial collaborative calibration of the initial land consolidation intention value. This effectively improves the spatial fit and accuracy of the land consolidation potential value, allowing the optimized land consolidation potential value to more realistically reflect the actual spatial distribution characteristics of the land consolidation potential in the target area, and providing a more realistic numerical basis for subsequent land consolidation potential zoning and implementation planning.

[0071] Step e: Based on the optimized remediation potential value, perform spatial cluster analysis on the target area to obtain the remediation potential zoning of the target area; In this embodiment of the invention, the step of performing spatial cluster analysis on the target region based on the optimized remediation potential value to obtain the remediation potential partition of the target region includes: Spatial distribution analysis of the potential values ​​for optimization and remediation is performed to obtain the initial cohesion base points of the target area; Using the initial agglomeration base point as the center, the spatial proximity of the optimization and remediation potential values ​​of the surrounding plots is determined to obtain the primary clustering map of the target area; Merging degree test is performed on the primary cluster patches to obtain the intermediate cluster of the target region; The distribution range of potential values ​​in intermediate clusters is divided into intervals to obtain the remediation potential zones of the target region.

[0072] To analyze the spatial distribution of optimization and remediation potential values, the optimization and remediation potential values ​​of all plots within the target area are first precisely bound to the spatial vector boundaries of the plots, generating a spatial distribution layer of optimization and remediation potential values ​​covering the entire target area. A global spatial kernel density analysis is then performed on this layer to identify core areas of plots with high optimization and remediation potential values ​​and high spatial distribution density. The geometric center of each core area is determined as the starting point for spatial aggregation. At the same time, the stability and spatial representativeness of the optimization and remediation potential values ​​of the starting points are verified. Invalid starting points with large numerical fluctuations and weak correlation with surrounding plots are eliminated, while valid starting points with stable values ​​and strong spatial correlation with surrounding plots are retained, thus obtaining the initial aggregation base points of the target area.

[0073] Using the initial agglomeration base point as the center, the spatial proximity of the optimization and remediation potential values ​​of the surrounding plots is determined. First, a reasonable spatial search neighborhood is defined for each initial agglomeration base point. This neighborhood is defined as a continuous plot coverage area centered on the agglomeration base point and according to the spatial adjacency relationship of the plots. Then, a dual judgment of optimization and remediation potential value and spatial location is carried out on all plots within the neighborhood. Plots with optimization and remediation potential values ​​similar to those of the initial agglomeration base point and directly adjacent to each other in spatial location are selected. These plots are then spatially aggregated with the initial agglomeration base point to form aggregated patches with continuous boundaries and similar potential values ​​of the internal plots. After completing the neighborhood judgment and plot aggregation of all initial agglomeration base points, the primary clustering patch of the target area is obtained.

[0074] The merging degree test is performed on the primary cluster patches. First, the core criteria for the merging degree test are determined, including the difference in the optimization and remediation potential values ​​of adjacent primary cluster patches, the spatial adjacency between patches, and the connectivity of the plots within the patches. Then, all adjacent primary cluster patches in the target area are tested one by one. If there is no significant difference in the optimization and remediation potential values ​​of two adjacent primary cluster patches, and they are directly adjacent in space and their internal plots can form a continuous and connected whole after merging, then the merging degree of the two patches is determined to meet the requirements, and they are spatially merged. If the patches do not meet the above criteria, they are retained in their original independent state. After completing the merging degree test and merging operation of all adjacent patches, the intermediate cluster of the target area is obtained.

[0075] The potential value distribution range of intermediate clusters is divided into intervals. First, the overall optimization and remediation potential value of all intermediate clusters in the target area is extracted. The maximum and minimum potential values ​​of intermediate clusters in the whole area are statistically obtained. Combined with the actual implementation needs of land consolidation, the numerical range of the whole area is divided into multiple continuous and non-overlapping numerical intervals according to the order of potential value from high to low. Each numerical interval corresponds to a land consolidation potential level. Then, each intermediate cluster is matched to the corresponding numerical interval according to the average optimization and remediation potential value of its internal plots. All intermediate clusters in each numerical interval are spatially integrated to delineate a continuous area with clear boundaries and uniform potential level. After completing the interval matching and spatial integration of all intermediate clusters, the remediation potential zoning of the target area is obtained.

[0076] The beneficial effects are as follows: by analyzing the spatial distribution of optimization and remediation potential values, the initial agglomeration base points are accurately determined, laying a scientific starting foundation for spatial clustering. Spatial proximity discrimination based on the agglomeration base points enables the precise aggregation of plots with similar potential values. The resulting primary clustering patches ensure the spatial correlation and numerical convergence of the clusters. The merging degree test further optimizes the rationality of the clustering results, eliminating fragmented and invalid patches. Finally, the scientific delineation of remediation potential zones is achieved through potential value interval division. The entire spatial clustering process fully combines the numerical characteristics of optimization and remediation potential values ​​with the spatial distribution characteristics of the plots. The delineated remediation potential zones have clear boundaries and highly similar internal potential values, which can accurately reflect the spatial differentiation characteristics of land remediation potential in the target area. This provides a clear spatial division basis for subsequent differentiated land comprehensive remediation planning and implementation.

[0077] Step f: Evaluate the remediation benefits of the potential zones to obtain the expected benefits of the target areas, prioritize them according to the expected benefits, and obtain the recommended implementation order for the target areas.

[0078] In this embodiment of the invention, the step of evaluating the remediation benefits of potential remediation zones to obtain the expected benefits of the target areas, and prioritizing them according to the expected benefits to obtain the recommended implementation order for the target areas, includes: Attribute information is extracted from zones with potential for improvement to obtain the basic attributes of the target region. By comprehensively calculating the various indicators in the basic attribute set, the expected benefit value of the target region can be obtained. The expected benefit values ​​are standardized to obtain the comprehensive benefit index of the target region; Based on the comprehensive benefit index, the potential zones for remediation are ranked in descending order to obtain the priority remediation sequence for the target areas.

[0079] Attribute information is extracted from the potential zones for remediation. First, each potential zone is treated as an independent analysis unit. The spatial vector boundary and the whole-domain geographic information database corresponding to each zone are retrieved. The full-dimensional attribute information is extracted for each zone one by one. The extracted content covers the zone's land use structure, plot area size, mean and distribution characteristics of optimization and remediation potential value, topographic and geomorphological conditions, land use control requirements of land spatial planning, and surrounding transportation and infrastructure. At the same time, relevant planning indicators such as the increase in arable land, intensive use of construction land, and improvement of ecological environment that can be achieved after the zone is remediated are extracted. All the extracted information is systematically classified and organized according to the zones. A complete attribute information set is established for each potential zone. After the information extraction and integration of all zones are completed, the basic attributes of the target area are obtained.

[0080] The expected benefits of the target region are obtained by comprehensively calculating various indicators from the basic attribute set. First, the various indicators from the basic attribute set are classified and sorted into three major categories: economic benefits, social benefits, and ecological benefits. The economic benefit indicators include indicators related to improved land use efficiency and increased land asset value after remediation; the social benefit indicators include indicators related to increased arable land area and optimized urban and rural construction land layout; and the ecological benefit indicators include indicators related to increased ecological land ratio and enhanced land ecosystem stability. Then, in accordance with the core requirements of land comprehensive remediation benefit assessment, each type of indicator is quantitatively calculated one by one to obtain the corresponding quantitative benefit value. Subsequently, the quantitative values ​​of the three types of indicators are comprehensively summarized and the integrated calculation of various benefits is completed according to the established integration method. A quantitative value that can comprehensively reflect the overall benefits after remediation is obtained for each remediation potential zone. After completing the calculation of all zones, the expected benefit value of the target region is obtained.

[0081] The expected benefit values ​​are standardized to obtain the comprehensive benefit index of the target area. First, the maximum and minimum values ​​of the expected benefit values ​​of all potential remediation zones within the target area are statistically analyzed to determine the numerical distribution range of the expected benefit values ​​across the entire area. Then, a unified numerical standardization method is used to transform the expected benefit value of each potential remediation zone into a standardized value within a fixed numerical range, eliminating the differences in dimensions and numerical spans between different zones. This allows for a direct comparison of the benefit values ​​of all zones. At the same time, the standardized values ​​are verified to ensure that the numerical conversion process is unbiased and the results are accurate and effective. After completing the standardization and verification of all expected benefit values, the comprehensive benefit index of the target area is obtained.

[0082] Based on the comprehensive benefit index, the potential remediation zones are sorted in descending order to obtain the priority remediation sequence for the target area. First, all potential remediation zones are linked and bound to their corresponding comprehensive benefit indices to form a correspondence table between zones and benefit indices. Then, all potential remediation zones are sorted in descending order of comprehensive benefit index. During the sorting process, the criteria for determining the index value are strictly followed, with zones having higher index values ​​being placed earlier and zones having lower index values ​​being placed later. At the same time, the sorting results are verified across the entire region to ensure that no zones are omitted and no sorting errors occur. After sorting and verification, a complete remediation implementation sequence table arranged in descending order of benefit is formed, resulting in the priority remediation sequence for the target area.

[0083] The beneficial effects are as follows: by extracting basic attributes from all dimensions of the potential zones for remediation, comprehensive collection of relevant information on the remediation of each zone is achieved, laying a complete and accurate data foundation for benefit assessment. The expected benefit values ​​obtained through classification calculation and comprehensive summarization can reflect the comprehensive economic, social, and ecological benefits of the remediation of potential zones in multiple dimensions. The standardized comprehensive benefit index eliminates the barriers to numerical comparison, making the benefits of each zone directly comparable. Finally, the priority remediation sequence obtained by arranging the comprehensive benefit index in descending order can accurately reflect the remediation benefits of each potential zone, providing a scientific and objective sequential basis for the actual implementation planning of land consolidation, effectively improving the rationality and pertinence of land consolidation implementation planning, ensuring that remediation resources are tilted towards zones with higher benefits, and maximizing the overall benefits of land consolidation.

[0084] like Figure 2 The diagram shown is a functional module diagram of a land consolidation potential analysis system based on artificial intelligence, provided in an embodiment of the present invention.

[0085] The artificial intelligence-based land consolidation potential analysis system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the artificial intelligence-based land consolidation potential analysis system 100 may include an overlay analysis module 101, a consolidation intention judgment rule construction module 102, a matching analysis module 103, a collaborative calibration module 104, a spatial clustering analysis module 105, and a benefit evaluation module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0086] In this embodiment, the functions of each module / unit are as follows: The overlay analysis module 101 is used to perform spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region. The remediation intention judgment rule construction module 102 is used to determine the remediation intention judgment rule of the target area by taking the historical land remediation map patches of the target area as positive samples and the unremediated areas of the target area as negative samples, and by comparing the characteristic differences between the positive and negative samples in the potential influencing factors. The matching analysis module 103 is used to perform matching analysis on the rules for judging the willingness to rectify and the potential influencing factors to obtain the preliminary remediation willingness value of the target area; The collaborative calibration module 104 is used to perform spatial differentiation quantification on high-value clusters in the target area to obtain the spatial weight factor of the target area, and based on the spatial weight factor, to perform collaborative calibration on the preliminary remediation intention value to obtain the optimization remediation potential value of the target area. The spatial clustering analysis module 105 is used to perform spatial clustering analysis on the target area based on the optimized remediation potential value, and obtain the remediation potential partition of the target area; The benefit assessment module 106 is used to assess the benefits of remediation in potential zones, obtain the expected benefits of the target areas, prioritize them according to the expected benefits, and obtain the recommended implementation order for the target areas.

[0087] Figure 3 The experimental curves in Scenario 1 for land consolidation potential analysis, provided in an embodiment of the present invention, are used to demonstrate the experimental performance of the system under conditions of high spatial weighting and clustering effects. Scenario 1 comprises two parts: an optimized consolidation potential curve and a coupled residual curve. The upper curve shows the distribution of consolidation potential values ​​for 40 plots after spatial collaborative calibration, with the mean of its core parameter being... =0.65、 =1.12、 =0.62、 =0.28、 =0.14, the potential value fluctuates between 0.78 and 1.06, and obvious peaks are formed at plots 20 and 24, reflecting a strong spatial clustering effect and differences in remediation potential; the residual curve below has a mean residual of 0.0096 and a standard deviation of 0.015, and the residuals are distributed between -0.025 and 0.035 without systematic shift, proving that the potential value calculation model has high fitting accuracy and stable and reliable results.

[0088] Figure 4 The experimental curves in Scenario 2 of the land consolidation potential analysis provided in an embodiment of the present invention are used to demonstrate the experimental effects of the system under conditions of high consolidation willingness and suitability. In Scenario 2, the upper optimized consolidation potential curve is shown as... =0.67、 =1.09、 =0.65、 =0.25、 With a base parameter of 0.12, the potential values ​​of the 40 plots fluctuated between 0.79 and 1.00, with peak values ​​appearing in plots 13 and 23. The fluctuation range was milder than that of group 1, reflecting that the differences in remediation potential among the plots in this group were relatively mild and the spatial attenuation effect was slightly weaker. The mean residual of the coupling residual curve below was -0.002 and the standard deviation was 0.021. The residuals were distributed between -0.07 and 0.025. Although there were a few extreme residual points, there was no systematic bias overall, and the model fitting accuracy was still within an acceptable range, verifying the reliability of the potential value calculation.

[0089] Figure 5 The experimental curves in Group 3, provided as an embodiment of the present invention, illustrate the experimental performance of the system under conditions of high suitability and clustering effects. The optimized remediation potential curve at the top of Group 3 is based on... =0.59、 =1.05、 =0.70、 =0.30、 Calculated using a value of 0.15, the potential values ​​of the 40 plots fluctuated between 0.74 and 0.96, with an overall level slightly lower than Group 1 and Group 2. The peak values ​​appeared in plots 27 and 31, while the troughs were deeper, indicating that there are still significant differences in the potential of the plots despite the support of the high suitability factor for remediation. The mean residual value of the coupled residual curve below is 0.000, and the standard deviation is 0.021. The residuals are distributed between -0.04 and 0.06, with no systematic shift. Although there are a few extreme positive residual points, the overall fitting accuracy is consistent with Group 2, ensuring the reliability of the potential value results.

[0090] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0091] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0093] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0094] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A land consolidation potential analysis method based on artificial intelligence, characterized in that, The method includes: Step a: Perform spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region; Step b: Using historical land consolidation plots in the target area as positive samples and unconsolidated areas in the target area as negative samples, determine the consolidation intention discrimination rules for the target area based on the characteristic differences between positive and negative samples in potential influencing factors. Step c: Perform a matching analysis between the rules for judging the willingness to carry out remediation and the potential influencing factors to obtain the preliminary remediation willingness value of the target area; Step d: Spatial differentiation quantification is performed on high-value clusters in the target region to obtain the spatial weight factor of the target region. Based on the spatial weight factor, the preliminary remediation intention value is collaboratively calibrated to obtain the optimization remediation potential value of the target region. Step e: Based on the optimized remediation potential value, perform spatial cluster analysis on the target area to obtain the remediation potential zoning of the target area; Step f: Evaluate the remediation benefits of the potential zones to obtain the expected benefits of the target areas, prioritize them according to the expected benefits, and obtain the recommended implementation order for the target areas.

2. The land consolidation potential analysis method based on artificial intelligence as described in claim 1, characterized in that, The spatial overlay analysis of multi-source data for the target region yields potential influencing factors for the target region, including: Acquire land use status data, territorial spatial planning data, socio-economic statistics data, and high-resolution remote sensing image data of the target area; Geometric correction is performed on high-resolution remote sensing image data to obtain a standardized image base for the target area; The physical properties of the target region are obtained by inverting and analyzing the standardized image base. Spatial processing is performed on land use status data, national spatial planning data, and socio-economic statistics to obtain thematic attribute layers for the target region; Spatially connect the physical attribute and thematic attribute layers to obtain the basic associated attributes of the target region; By performing feature screening on the basic correlation attributes, the potential influencing factors of the target region can be obtained.

3. The land consolidation potential analysis method based on artificial intelligence as described in claim 1, characterized in that, The process involves using historical land consolidation plots in the target area as positive samples and unconsolidated areas in the target area as negative samples. Based on a comparison of the differences in potential influencing factors between the positive and negative samples, a rule for determining the consolidation intention of the target area is established, including: Extract land parcels that have undergone remediation from the historical land consolidation project database of the target region, and use them as positive samples for the target region; From the unremediated areas of the target region, extract unremediated land parcels that match the positive samples in terms of geographical environmental characteristics, and use them as negative samples for the target region; Extract the factor values ​​of land parcels in the potential influence factors from the positive and negative samples to obtain the positive and negative sample feature values ​​of the target region; By comparing the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples, candidate discriminant factors between positive and negative samples can be identified. Based on the candidate discriminant factors, a comprehensive analysis of the central tendency and dispersion in positive and negative samples is conducted to obtain the discriminant direction and discriminant limit of the candidate discriminant factors; Candidate discriminant factors are logically combined according to the discriminant direction and discriminant boundary to obtain the governance intention discriminant rule for the target area.

4. The land consolidation potential analysis method based on artificial intelligence as described in claim 3, characterized in that, The comparison of the distribution differences between the feature values ​​of positive samples and the feature values ​​of negative samples to identify candidate discriminant factors between positive and negative samples includes: The positive sample feature value set and the negative sample feature value set are sorted according to their numerical values ​​to obtain the positive sample value sequence and the negative sample value sequence of the target region. Plotting the positive sample value sequence and the negative sample value sequence on the same numerical axis yields a comparison chart of the value distribution of the target region. If the overall numerical range of the positive sample value sequence is located on one side of the overall numerical range of the negative sample value sequence, then the potential influencing factor is determined to have discriminative power. If the numerical ranges of the positive sample value sequence and the negative sample value sequence intersect, then the median position and dispersion range of the positive sample value sequence and the negative sample value sequence are further compared. If the deviation of the median exceeds the comprehensive width of the dispersion range, then the potential influencing factor is determined to have discriminative power. Potentially influential factors that are deemed to have discriminative power will be used as candidate discriminant factors.

5. The land consolidation potential analysis method based on artificial intelligence as described in claim 1, characterized in that, The matching analysis of the remediation willingness judgment rules and potential influencing factors yields the preliminary remediation willingness value for the target area, including: Extract the key discrimination factors and their corresponding discrimination directions and boundaries implicit in the rules for judging the willingness to rectify, and obtain the rule elements of the target region; The potential influencing factors of the target region are extracted and the measured characteristic values ​​of the target region on the key discriminant factors are obtained, thus obtaining the measured values ​​of the land parcel factors of the target region. The measured values ​​of land parcel factors are compared with the rule elements one by one to determine whether the key discrimination factors meet the discrimination direction and discrimination boundary, and the factor matching identifier of the target region is obtained. Based on factor matching identifiers, key discriminant factors of the target region are aggregated and integrated to obtain the preliminary remediation intention value of the target region.

6. The land consolidation potential analysis method based on artificial intelligence as described in claim 1, characterized in that, The process involves spatial differentiation and quantification of high-value clusters in the target region to obtain spatial weighting factors for the target region. Based on these spatial weighting factors, the preliminary remediation intention value is collaboratively calibrated to obtain the optimization remediation potential value of the target region, including: Based on the initial willingness to rectify, the spatial distribution of the willingness values ​​in the target area is analyzed to obtain the high-value clusters in the target area; Geometric features are extracted from the spatial morphology of high-value clusters to obtain the spatial attributes of clusters in the target region; The spatial differentiation degree of the spatial attributes of the cluster area is quantified to obtain the spatial weight factor of the target region. Based on spatial weighting factors, the preliminary remediation intention values ​​within and adjacent areas of high-value clusters are spatially correlated and fused to obtain the revised remediation intention values ​​for the target area. A gradual influence attenuation value is assigned to the buffer zone of the high-value cluster area, and the attenuation coefficient of the buffer zone is determined according to the distance from the center of the cluster area. Based on the attenuation coefficient and spatial weighting factor, the initial remediation intention value in the buffer zone is spatially coordinated and calibrated to obtain the calibrated remediation intention value of the target area. The modified remediation intention value and the calibrated remediation intention value are summarized and calculated to obtain the optimization remediation potential value of the target area.

7. The land consolidation potential analysis method based on artificial intelligence as described in claim 6, characterized in that, The formula for calculating the potential value of optimization and remediation is as follows: ； in, To optimize the potential value for rectification, To adjust the willingness to rectify, Spatial weighting factor, To calibrate the willingness to rectify, The attenuation coefficient is... This is the deviation correction factor. To determine the initial willingness to rectify the situation, The average value of the initial willingness to rectify.

8. The land consolidation potential analysis method based on artificial intelligence as described in claim 1, characterized in that, The process involves performing spatial cluster analysis on the target region based on optimized remediation potential values ​​to obtain remediation potential partitions for the target region, including: Spatial distribution analysis of the potential values ​​for optimization and remediation is performed to obtain the initial cohesion base points of the target area; Using the initial agglomeration base point as the center, the spatial proximity of the optimization and remediation potential values ​​of the surrounding plots is determined to obtain the primary clustering map of the target area; Merging degree test is performed on the primary cluster patches to obtain the intermediate cluster of the target region; The distribution range of potential values ​​in intermediate clusters is divided into intervals to obtain the remediation potential zones of the target region.

9. The land consolidation potential analysis method based on artificial intelligence as described in claim 1, characterized in that, The process of evaluating the remediation benefits of potential zones to obtain the expected benefits of the target areas, and prioritizing them according to the expected benefits to obtain the recommended implementation order for the target areas, includes: Attribute information is extracted from zones with potential for improvement to obtain the basic attributes of the target region. By comprehensively calculating the various indicators in the basic attribute set, the expected benefit value of the target region can be obtained. The expected benefit values ​​are standardized to obtain the comprehensive benefit index of the target region; Based on the comprehensive benefit index, the potential zones for remediation are ranked in descending order to obtain the priority remediation sequence for the target areas.

10. A land consolidation potential analysis system based on artificial intelligence, characterized in that: The system for implementing the artificial intelligence-based land consolidation potential analysis method as described in claim 1 includes: The overlay analysis module is used to perform spatial overlay analysis on multi-source data of the target region to obtain the potential influencing factors of the target region. The module for constructing rules for judging the willingness to remediate is used to take historical land remediation patches in the target area as positive samples and unremediated areas in the target area as negative samples, and to determine the rules for judging the willingness to remediate in the target area based on the difference in characteristics between positive and negative samples in potential influencing factors. The matching analysis module is used to perform matching analysis between the rules for judging the willingness to carry out remediation and the potential influencing factors to obtain the preliminary remediation willingness value of the target area; The collaborative calibration module is used to perform spatial differentiation quantification on high-value clusters in the target area to obtain the spatial weight factor of the target area, and based on the spatial weight factor, to perform collaborative calibration on the preliminary remediation intention value to obtain the optimization remediation potential value of the target area. The spatial clustering analysis module is used to perform spatial clustering analysis on the target area based on the optimized remediation potential value, and obtain the remediation potential zoning of the target area; The benefit assessment module is used to assess the benefits of remediation in potential zones, obtain the expected benefits of the target areas, prioritize them according to the expected benefits, and obtain the recommended implementation order for the target areas.

Images