Method for evaluating potential of newly increased cultivated land in global land comprehensive improvement based on engineering layout

Abstract

The invention provides an engineering layout-based new cultivated land potential evaluation method for global land comprehensive improvement, relates to the technical field of land space planning and geographic information, and realizes semantic unification of multi-source land utilization data through a land type intelligent conversion engine. And performing engineering adaptability screening on the potential land parcels by adopting a dynamic threshold mechanism. Constructing a chain topology erasure model to realize cooperative processing of the multi-stage restricted area; an engineering-oriented potential evaluation model is innovatively provided, and accurate evaluation is realized through comprehensive calculation of'potential plots + engineering radiation cultivated land '. According to the method, the technical bottlenecks of data processing fragmentation, weak engineering relevance and the like in traditional evaluation are broken through, a quantitative association system of engineering layout and resource potential is constructed, the accuracy of cultivated land potential identification is remarkably improved, the application range of the method in land space planning is expanded, scientific decision support is provided for global land comprehensive renovation, and the method is suitable for popularization and application. The method plays an important role in promoting the deep application of the GIS technology in the field of cultivated land protection.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of territorial space planning and geographic information technology, and particularly relates to a method for evaluating the potential of newly added cultivated land based on engineering layout in comprehensive land improvement. BACKGROUND

[0002] The current method for evaluating the potential of newly added cultivated land in comprehensive land improvement has the dual defects of weak engineering orientation and insufficient spatial quantification. In the engineering implementation level, traditional technical means rely on static patch superposition and theoretical area calculation, and cannot effectively integrate the spatial connectivity characteristics of cultivated land resources. The mechanical aggregation of fragmented plots often leads to the discrete distribution of potential areas, which is in sharp contradiction with the requirement of concentrated and contiguous for large-scale agricultural engineering. Such evaluation results often only have statistical significance and are difficult to support the continuity of actual construction layout, resulting in some theoretical newly added cultivated land being forced to give up development due to scattered plots and mechanical access.

[0003] The more deep-seated technical shackles lie in the systematic fragmentation of the evaluation process. The current technical framework separates land class identification, constraint exclusion and potential calculation into independent modules, resulting in cascading errors in data transmission. The land class conversion relies on manual experience to set thresholds, resulting in poor recognition accuracy of pits, ponds and ditches; the constraint condition processing uses simple spatial superposition and fails to establish a dynamic response mechanism for hierarchical erasure, resulting in false rejection of edge plots in ecologically sensitive areas; and most importantly, the analysis of cultivated land proximity is still at the static judgment stage of buffer distance, completely ignoring the engineering transmission effect of the spatial topological relationship of cultivated land resources. SUMMARY

[0004] The application provides a method for evaluating the potential of newly added cultivated land based on engineering layout in comprehensive land improvement, which breaks through the technical bottlenecks of land class identification deviation, spatial constraint fragmentation and engineering connectivity deficiency in traditional cultivated land potential evaluation, and can realize accurate identification of the improvement area and maximize the implementation benefit.

[0005] To solve the above technical problems, the application provides the following technical scheme: a method for evaluating the potential of newly added cultivated land based on engineering layout in comprehensive land improvement, the steps are as follows:

[0006] S1, a dynamic mapping of land survey land class and engineering standard land class is constructed, and the area of each engineering standard land class patch is calculated;

[0007] S2, the target land class is selected based on the engineering standard land class of step S1, and an intelligent constraint mechanism for pit and pond water area is added for further selection;

[0008] S3, the target land type screened out in step S2 performs a vector erasing operation in the order of engineering priority, and a minimum plot area threshold filtering mechanism is implanted;

[0009] S4, the geometric fragments caused by the cultivated land are eliminated through a three-level topology to obtain a three-level cultivated land unit for engineering implementation, and a dynamic buffer zone connectivity model is constructed with a buffer radius parameter as input and a spatially connected unit as output;

[0010] S5, the spatially connected unit is topologically intersected with the three-level cultivated land unit, a spatial contribution correlation is established, and a grouping potential decision algorithm is used in combination with a preset threshold to obtain a newly added cultivated land potential plot.

[0011] Further, the step S1 of the foregoing constructs a dynamic mapping of land survey land types and engineering standard land types, specifically: a land type name conversion engine is created, and the intelligent conversion of land type names in the original three DLTB layers and land use sea land type names is realized through a dynamic mapping dictionary, and the mapping dictionary is derived from a preset engineering standard conversion table.

[0012] Further, in the step S1 of the foregoing, the area of each engineering standard land type plot is calculated, specifically: an automatic spatial area calibration mechanism is integrated, and a planar rectangular coordinate system topological area calculation model is used to generate the area of each land type plot, wherein the area calculation uses geometric attribute accurate quantification in the planar rectangular coordinate system.

[0013] Further, the target land types of the step S2 of the foregoing include: forest land, garden land, grassland, and pit pond water surface.

[0014] Further, the foregoing method for evaluating the newly added cultivated land potential based on the overall land comprehensive regulation of the entire region, characterized in that, further comprising further screening ditches on the four target land types of forest land, garden land, grassland, and pit pond water surface.

[0015] Further, the step S3 of the foregoing performs a vector erasing operation in the order of engineering priority: permanent basic farmland, ecological protection red line, and urban development boundary, and gradually strips off the restrictive areas.

[0016] Further, in the step S4 of the foregoing, the geometric fragments caused by the cultivated land are eliminated through a three-level topology to obtain a three-level cultivated land unit for engineering implementation, and the steps are as follows:

[0017] S4.1, primary extraction: extracting the original cultivated land elements from the three DLTB layers after the land type name conversion;

[0018] S4.2, secondary fusion: using a vector fusion method of canceling line segmentation to eliminate the geometric fragments caused by the cultivated land.

[0019] S4.3, three-level decomposition: generating the engineering implemented three-level farmland unit through the multi-component to single-component method.

[0020] Further, in the aforementioned step S4, the dynamic buffer zone connection model is specifically constructed by taking the buffer radius parameter as input, performing overall buffer zone fusion on the potential land plot, and then decomposing to form a spatially connected unit through the multi-component to single-component method.

[0021] Further, in the aforementioned step S5, the spatially connected unit and the three-level farmland unit are topologically intersected to establish a spatial contribution association, which is specifically: through the topological intersection detection to establish a quantitative relationship model, dynamically calculating the area superposition effect between the spatially connected unit, i.e., the buffer zone grouping, and the three-level farmland unit.

[0022] Spatial contribution association = ,

[0023] Wherein, G i represents the i-th spatially connected unit, i.e., the buffer zone grouping, F j represents the j-th three-level farmland unit, represents the area of farmland F j , and n represents the total number of farmland units intersecting G i .

[0024] Further, in the aforementioned step S5, the grouping potential decision algorithm is calculated as follows:

[0025] ,

[0026] Wherein, Q 组 is the merged buffer zone group intersecting the farmland, S 潜力地块 is the sum of the areas of all original potential land plots in the zone group, and S 相邻耕地 is the sum of the areas of all farmland polygons intersecting the zone group buffer zone.

[0027] Compared with the prior art, the beneficial technical effects of the above technical solutions are as follows:

[0028] (1) An intelligent mapping system between land survey land types and engineering implementation standards is creatively constructed, completely solving the technical bottleneck of traditional methods relying on manual experience to set thresholds, resulting in land class recognition deviation. By developing a land class name conversion engine, accurate conversion of three-investigation land class names to "land and sea land class names is achieved, and a spatial area automatic calibration mechanism is simultaneously integrated. This technical solution uses the geometric calculation capability of the geographic information system bottom layer, accurately quantifies the engineering implementation area using the topological properties in the plane rectangular coordinate system, eliminates the area distortion problem caused by projection distortion, builds a millimeter-level precision data base for subsequent engineering decision-making, and significantly improves the engineering applicability of the land feature recognition result.

[0029] (2) For the traditional constraint processing in simple space superposition caused by the cascade error problem, the dual response mechanism of hierarchical erasing and threshold filtering is creatively established. By designing the priority sequence of erasing restrictive elements, the iterative space difference algorithm is used to realize the sequential processing of constraint elements. In each level of erasing operation, the minimum spot area threshold filtering is implanted synchronously, and the lower limit standard of 400 square meters of engineering scale is set. This technical path successfully solves the problem of mis-removal of the edge land of the ecological sensitive area, and makes the spatial precision of the potential land completely meet the requirements of mechanized construction.

[0030] (3) In order to solve the engineering unfeasible dilemma caused by the fragmentation of cultivated land, a three-level topological processing flow is developed. This technical system combines the dynamic buffer zone construction method to create a spatially connected unit based on the user input radius parameter, so as to improve the efficiency of cultivated land connection and fundamentally change the distribution dilemma of star point distributed cultivated land. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 is the schematic diagram of the present application for constructing different types of image segmentation model and training building semantic segmentation network.

[0032] Figure 2 is the schematic diagram of the present application for building semantic segmentation and raster vectorization processing of visible light remote sensing image, and obtaining three kinds of building space objects.

[0033] Figure 3 is the schematic diagram of the present application for constructing difference building space object set and information vector set.

[0034] Figure 4 is the schematic diagram of the present application for constructing difference building analysis model and training difference building effectiveness judgment network.

[0035] Figure 5 is the schematic diagram of the present application for judging the effectiveness of difference building and performing spatial fusion processing with the reference building space object. DETAILED DESCRIPTION

[0036] In order to better understand the technical content of the present application, specific embodiments are described below with the aid of the accompanying drawings.

[0037] Aspects of the present application are described in the context of the drawings, which show numerous illustrative embodiments. Embodiments of the present application are not limited to the drawings described. It should be understood that the present application is realized by any of the various concepts and embodiments introduced above, as well as any of the concepts and embodiments described in detail below, since the concepts and embodiments disclosed are not limited to any embodiment. In addition, some aspects disclosed by the present application can be used alone, or in any appropriate combination with other aspects disclosed by the present application.

[0038] As Figure 5 shown, the embodiment takes a city district 7 administrative village global land comprehensive regulation project as an example, and uses ArcGIS Pro 3.4.0 platform to execute the method: The project area is located in the junction of hilly and mountainous area and Yangtze River alluvial plain, showing the typical geomorphic characteristics of "north forest and south farmland" and rich water system. The region is an important spatial carrier to implement the "Yangtze River protection" strategy and promote rural revitalization, but also faces systematic challenges such as fragmentation of land use layout, fragmentation of farmland, low efficiency of industrial land and local ecological degradation, and is a typical region to carry out global land comprehensive regulation.

[0039] S1, construct the dynamic mapping of land survey land class and engineering standard land class, and calculate the area of each engineering standard land class map.

[0040] Referring to Figure 1 , in the embodiment, the user inputs the land class map (DLTB) layer in the third national land survey data, and the user continues to input the prepared land class conversion table (Excel format). Contains two columns: "land class name" (such as "paddy field"...), "land use and sea land class name" (such as "farmland"...).

[0041] In this step:

[0042] # Call land class conversion method

[0043] converted_fc = convert_landuse_class(

[0044] input_dltb="three land class map",

[0045] excel_path="land class conversion table.xlsx")

[0046] The system automatically reads the Excel content to create a mapping dictionary: {paddy field":"farmland", "watered land":"farmland"...}

[0047] Add "land use and sea land class name" field in output feature class

[0048] Update field values ​​when traversing map patches: When DLMC="paddy field", the new field value is "arable land".

[0049] Add an "Area" field and calculate the geometric area (unit: square meters).

[0050] S2. Based on the engineering standard land categories in step S1, the target land categories are selected, and an intelligent constraint mechanism for the surface area of ​​pits and ponds is added for further selection.

[0051] like Figure 2 As shown, users can select whether to include ditches and set the maximum area of ​​the pond surface.

[0052] In this step:

[0053] # Perform filtering

[0054] layer1 = filter_target_classes(

[0055] input_fc=converted_fc,

[0056] include_ditch=True,

[0057] max_pond_area_value=10,

[0058] area_unit="mu",

[0059] output_path="target location type.shp")

[0060] # Constructing composite query conditions

[0061] "Land Use Category Name IN ("Woodland", "Orchard", "Grassland", "Ponds and Water Surfaces") AND (Land Use Category Name <> "Ponds and Water Surfaces" OR Area <= (the maximum area of ​​ponds and water surfaces entered by the user))

[0062] Output land parcels containing woodlands, gardens, grasslands, and ponds / water surfaces (≤ user input parameters).

[0063] S3. For the target land types selected in step S2, perform vector erasure operations according to engineering priority, and implement a minimum patch area threshold filtering mechanism.

[0064] refer to Figure 3 Users need to configure multi-level spatial constraint parameters and import layers such as permanent basic farmland, ecological red line and urban development boundary in order of project priority.

[0065] In this step S3:

[0066] # Erase the restrictive layers one by one for i, restrict_fc in enumerate(restriction_datasets): erase_output =os.path.join(self.temp_workspace, f"Erased_{i}") arcpy.Erase_analysis(current_layer, restrict_fc, erase_output) current_layer = erase_outputarcpy.AddMessage(f"Erased layer: {os.path.basename(restrict_fc)}")

[0067] After all constraints are processed, an area field is added to filter and retain patches with an area ≥ 400㎡. Users can see specific operations through progress prompts, such as "Layer erased: Basic Farmland.shp". The results are saved and output.

[0068] S4. Eliminate geometric fragments caused by cultivated land through three-level topology to obtain three-level cultivated land units for engineering implementation, and then construct a dynamic buffer zone connectivity model with buffer radius parameters as input and spatial connectivity units as output.

[0069] In the core engineering connectivity analysis phase, the user sets the buffer distance parameter, see... Figure 4 Then it proceeds to the third-level farmland treatment process.

[0070] # Three-level farmland treatment process

[0071] farmland_layer1 = Extracts original farmland (Land Use / Sea Use Class Name = 'Farmland')

[0072] farmland_layer2 = Merge Farmland(farmland_layer1, Cancel Line Splitting=True)

[0073] farmland_layer3 = Multi-component decomposition (farmland_layer2)

[0074] Potential land parcels will be handled as follows:

[0075] # Buffer Construction

[0076] buffer_fused = Creates an overall fusion buffer (potential plots, buffer_distance=200).

[0077] buffer_singleparts = Converts multiple parts to a single part (buffer_fused)

[0078] # Screening areas intersecting with farmland

[0079] `buffer_intersected` = filters intersecting regions (buffer_singleparts, farmland_layer3).

[0080] S5. Perform topological intersection detection between spatially connected units and tertiary cultivated land units to establish spatial contribution relationships, and use a grouping potential decision algorithm combined with a preset threshold to obtain potential new cultivated land plots. Specifically, the spatial contribution relationship is established by using a quantitative relationship model established through topological intersection detection to dynamically calculate the area superposition effect between spatially connected units (i.e., buffer grouping) and tertiary cultivated land units.

[0081] Spatial contribution correlation = ,

[0082] Among them, G i F represents the i-th spatially connected unit, i.e., the buffer group. j This represents the j-th tertiary arable land unit. Indicates arable land F j The area, n represents the area with respect to G. i The total number of intersecting farmland units.

[0083] In this step: The group potential decision algorithm is calculated as follows: , Among them, Q 组 To select the merged buffer groups that intersect with cultivated land, S 潜力地块 S is the sum of the areas of all original potential land parcels within the district group. 相邻耕地 It is the sum of the areas of all cultivated land patches that intersect with the buffer zone of the block group.

[0084] The group's potential is assessed as follows:

[0085] # Traverse each connected unit

[0086] for group in buffer_intersected:

[0087] # Calculate the total area of ​​potential land parcels within the group (S1)

[0088] s1_total = Calculates the potential area within a group.

[0089] # Calculate the total area of ​​adjacent cultivated land (S2)

[0090] s2_total = Calculate the area of ​​intersecting farmland (group, farmland_layer3)

[0091] # Comprehensive Potential Assessment

[0092] if (s1_total + s2_total) >= 10000: # 15 acres

[0093] # All plots in the marked group meet the standard

[0094] Output

[0095] # Extracting qualified land parcels

[0096] `output_layer` = Extracts qualified land parcels (potential land parcels, set of tag IDs).

[0097] Save it as a shapefile("final potential plot.shp").

[0098] Users can observe real-time changes in farmland elements: from initially extracted scattered plots to integrated units after eliminating field ridges, ultimately forming standardized contiguous engineering units. The script toolbox synchronously constructs a connected network of potential plots, dynamically displaying the accumulation process of the potential value of each connected unit through the group calculation module. When the total potential value of the group reaches 15 mu, the qualified plot group is marked, and users can view detailed spatial analysis data in the log.

[0099] This embodiment demonstrates that the method effectively solves the technical challenges faced in comprehensive land consolidation across the entire region, such as "inconsistent land category standards, complex constraints, and weak spatial correlation," through three major technological innovations: intelligent conversion, multi-level constraints, and spatial correlation. It provides a replicable and scalable technical paradigm for comprehensive land consolidation in similar areas of the Yangtze River Economic Belt.

[0100] Through the above steps, an assessment of the potential for newly added arable land through comprehensive land consolidation based on engineering layout can be achieved. It can be seen that each step achieves the following:

[0101] (1) The standardized conversion from the land category names of the Third National Land Survey to the land use and sea use land category names was realized, and the land category name system was unified.

[0102] (2) It has achieved precise screening of target land types, including area control of ponds and water surfaces and inclusion of ditches as needed.

[0103] (3) It has achieved the step-by-step erasure of multi-level spatial constraints, effectively avoiding restricted areas such as permanent basic farmland and ecological protection red lines.

[0104] (4) The potential assessment based on engineering connectivity is realized, and the newly added cultivated land potential area with a connected scale up to standard is scientifically identified through cultivated land fusion and grouped potential calculation.

[0105] In the implementation of the global land comprehensive improvement project, 7 administrative villages with a total area of 60.31 square kilometers are selected as typical experimental areas, and the technical scheme is used to carry out empirical research. The experimental data is from the 2023 national land change survey database, which contains 12,570 land class polygons. Through systematic testing and verification, the technical scheme shows significant technical superiority and engineering application value.

[0106] (I) Technical precision breakthrough

[0107] Through the intelligent conversion engine of land class, the accurate mapping of three-investigation land class names to land use and sea land class names is realized. Test data shows:

[0108] The land class conversion accuracy rate is 100%, which is greatly improved compared with the traditional manual interpretation accuracy rate. The pit pond water area constraint mechanism effectively excludes 127 over-limit pits and ponds (average area 18.2 mu), and the multi-level constraint erasing model greatly reduces the error rejection rate of the edge of the ecological sensitive area. The minimum polygon filtering (≥400㎡) eliminates 2,386 fragmented land blocks, and the potential land block aggregation degree is improved by 37.6%. Especially in the cultivated land connectivity processing link, the three-level topology technology realizes a major breakthrough, the dike line elimination efficiency is improved by 8.1 times, the cultivated land unit number is optimized from 8,421 to 1,037, the connected rate is significantly improved, the average single land block area is increased from 1.8 mu to 14.3 mu, and the spatial topological relationship is kept intact.

[0109] (II) Significant engineering implementation benefits

[0110] Through the grouped potential evaluation model, the accurate identification of newly added cultivated land is realized. There are 890 groups of standard land blocks, the minimum potential value is 20164.9㎡ (30.2 mu), and 4,418 potential land blocks that can be implemented are identified. The total area of newly added cultivated land is 22537.3 mu (1502.4 hectares).

[0111] (III) Outstanding ecological synergy effect

[0112] Through the multi-level constraint chain response mechanism, the deep synergy between ecological protection and cultivated land development is realized. The ecological protection red line is avoided, the original forest land in the ecological sensitive area is reserved, the zero damage of forest resources is realized, and the infrastructure corridor is protected intact.

[0113] (IV) Verification of technical innovation value

[0114] The experimental data fully proves the technical value of the three core innovations:

[0115] The engineering semantic dynamic mapping mechanism solves the industry deviation of land classification and greatly reduces the traditional error rate.

[0116] The multi-level constraint chain response model breaks through the contradiction of "rigid constraint-flexible development", and realizes the newly added cultivated land under the premise of strictly protecting the ecological red line.

[0117] The cultivated land connectivity topology processing technology solves the governance dilemma of cultivated land fragmentation, greatly improves the connectivity rate, and lays a foundation for large-scale mechanical operation.

[0118] (Five) Conclusion and application prospect

[0119] The technical scheme successfully builds a good foundation for the integrated technical system of "evaluation-design-implementation" of the global land comprehensive improvement project, and provides a new generation of engineering technical paradigm for global land comprehensive improvement. In the follow-up application in Jiangsu Province, the precision evaluation of newly added cultivated land can be realized, the engineering cost can be greatly saved, and it has great practical value for strictly guarding the restrictive elements.

[0120] Although the present application has been described as above with reference to the preferred embodiments, it is not intended to limit the present application. Those skilled in the art can make various modifications and improvements without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be subject to the scope defined by the claims.

Claims

1. A method for evaluating the potential of newly added arable land based on the overall land consolidation of an engineering layout, characterized by, The steps are as follows: S1, construct the dynamic mapping of land survey land class and engineering standard land class, and calculate the area of each engineering standard land class map spot; S2, based on the engineering standard land class of step S1, the target land class is screened out, and an intelligent constraint mechanism of pit pond water area is added for further screening; S3, according to the engineering priority order of step S2, the vector erasing operation is performed on the target land class, and the minimum spot area threshold filtering mechanism is implanted; S4, through three level topological elimination of geometric fragments caused by cultivated land, the three level cultivated land unit for engineering implementation is obtained, and then the dynamic buffer zone connection model is constructed with buffer radius parameter as input and spatial connection unit as output; S5, the spatial connection unit and the three level cultivated land unit are topologically intersected, the spatial contribution association is established, and the grouping potential decision algorithm is used to obtain the newly added cultivated land potential plot combined with the preset threshold.

2. The method according to claim 1, wherein, Step S1 of constructing the dynamic mapping of land survey land class and engineering standard land class is specifically: creating a land class name conversion engine, realizing the intelligent conversion of land class names in the original three DLTB layers and land use sea land class names through a dynamic mapping dictionary, and the mapping dictionary is derived from the preset engineering standard conversion table.

3. The method according to claim 1, wherein, In step S1, the area of each engineering standard land class map spot is calculated, which is specifically: integrating a spatial area automatic calibration mechanism, using a planar rectangular coordinate system topological area calculation model to generate the area of each land class map spot, wherein the area calculation uses geometric attribute accurate quantification in the planar rectangular coordinate system.

4. The method according to claim 1, wherein, The target land class of step S2 includes: forest land, garden land, grassland, and pit pond water surface.

5. The method according to claim 4, wherein, It also includes further screening of ditches on the four target land classes of forest land, garden land, grassland, and pit pond water surface.

6. The method according to claim 1, wherein, In step S3, the vector erasing operation is performed according to the engineering priority order: permanent basic farmland, ecological protection red line, urban development boundary, and the limiting area is peeled off step by step.

7. The method according to claim 2, wherein the method is characterized by, In step S4, the three level topological elimination of geometric fragments caused by cultivated land is used to obtain the three level cultivated land unit for engineering implementation, and the steps are as follows: S4.1, primary extraction: extracting the original cultivated land elements from the three DLTB layers after land class name conversion; S4.2, secondary fusion: using a vector fusion method of canceling line segmentation to eliminate the geometric fragments caused by cultivated land; S4.3, three level decomposition: generating the three level cultivated land unit for engineering implementation through the multi-component to single-component method.

8. The method according to claim 2, wherein the method is characterized by, In step S4, the dynamic buffer zone connection model is constructed, which is specifically: taking buffer radius parameter as input, performing overall buffer zone fusion on the potential plot, and then decomposing to form spatial connection unit through multi-component to single-component method.

9. The method according to claim 1, wherein the method is characterized by, In step S5, the spatial connection unit and the three level cultivated land unit are topologically intersected, and the spatial contribution association is established, which is specifically: through the quantitative relationship model established by topological intersection detection, the area superposition effect between the spatial connection unit, i.e. buffer zone grouping, and the three level cultivated land unit is dynamically calculated; spatial contribution correlation , where G i represents the i-th spatial connectivity unit, i.e. buffer group, F j represents the j-th tertiary farmland unit, represents the area of farmland F j , and n represents the total number of farmland units intersecting G i .

10. The method according to claim 1, wherein the method is characterized by, In step S5, the grouping potential decision algorithm is calculated as follows: ， Where, Q 组 For screening out the combined buffer zone group intersecting with cultivated land, S 潜力地块 For the sum of the areas of all original potential land blocks in the zone group, S 相邻耕地 For the sum of the areas of all cultivated land patches intersecting with the zone group buffer zone.

Images