AI-assisted natural boundary growth method and system

Through the AI-assisted boundary natural growth method, the neural deep convolutional learning model is used to automatically identify and expand land-like maps, solving the limitations of traditional systems in massive data analysis and flexible expansion, and achieving high-quality intelligent expansion of land space boundaries .

CN118035372BActive Publication Date: 2024-11-08HAINAN KAIPU DIGITAL TECH CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410245595.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-05
Publication Date
2024-11-08
Estimated Expiration
2044-03-05

AI Technical Summary

Technical Problem

Traditional natural growth systems of boundary are difficult to adapt to massive data analysis, and are limited to fixed expansion or manual identification within the region, so they cannot achieve flexible and intelligent expansion of the boundaries of the land space.

Method used

Using AI-assisted boundary natural growth method, we can establish a neural deep convolutional learning model by obtaining comprehensive scores of the importance, adaptability and change possibility dimensions of the geographic map, and automatically identify and expand the neighboring geographic map with the highest comprehensive score. 斑,实现边界的自然生长和灵活扩展。

Benefits of technology

It improves the intelligent expansion of land-type map spots, obtains interactive target boundary calculation results, effectively retains land-type map spots with high comprehensive scores, improves the quality of planning results, and adapts to different scenarios and needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118035372B_ABST
    Figure CN118035372B_ABST
Patent Text Reader

Abstract

The present invention relates to an AI-assisted natural growth method and system for boundaries, and belongs to the field of geographic information technology. Among them, the method includes: obtaining a comprehensive score of land class spots in a selected area. Obtaining existing areas and existing boundaries in the selected area, preferentially expanding the land class spot recognition model to identify the adjacent land class spots with the highest comprehensive score as the first priority expanded land class spots. Extending the first priority expanded land class spots to the existing area to obtain the expanded existing area. Setting the target area ratio to obtain the target area, obtaining the area of ​​the existing area after expansion, calculating the target achievement degree, setting the target achievement degree threshold, the target achievement degree is less than the target achievement degree threshold, continuing to expand, and the target achievement degree reaches the target achievement degree threshold to end the expansion to obtain the target area and target boundary. The system can intelligently and selectively expand, help achieve planning goals, and provide strong support for planning boundary demarcation decisions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of geographic information technology, and specifically relates to an AI-assisted boundary natural growth method and system. Background Art

[0002] In the fields of national land planning and land use planning, the intelligent expansion of land class map boundaries is a key task (such as basic farmland planning, high-standard farmland demarcation, development boundary flexible boundary demarcation, etc.). To achieve the conditional intelligent expansion of existing boundaries in national land space planning. The present invention provides an AI-assisted boundary natural growth method, which can more flexibly and intelligently meet the problem of national land space boundary demarcation. According to the established goals, it relies on the natural growth algorithm of computer automatic analysis, analyzes the comprehensive evaluation results of the spots around the existing boundaries, and takes the boundary expansion goal as the guide. The priority expansion land class map recognition model is used to automatically identify the adjacent land class spots with the highest comprehensive score, expand them, and realize the natural growth of the boundary. This method has broad application prospects in geographic information systems (GIS) and planning fields. Traditional boundary natural growth systems are often limited to fixed expansion or manual identification within the region, and it is difficult to adapt to the actual needs of massive data analysis. Summary of the invention

[0003] In order to solve the above problems existing in the prior art, in the first aspect, the present invention provides an AI-assisted boundary natural growth method, which is implemented by the following steps:

[0004] S1: Obtaining the importance dimension, adaptability dimension, and change possibility dimension of the land type spots in the region, assigning values ​​to the importance dimension, the adaptability dimension, and the change possibility dimension, and obtaining a comprehensive score of the land type spots through comprehensive calculation according to the assigned values;

[0005] S2: obtaining an existing area in the selected area, obtaining an existing boundary of the existing area, obtaining adjacent land class patches adjacent to the existing boundary according to the existing boundary, obtaining a comprehensive score of the adjacent land class patches, and establishing a priority extended land class patch recognition model based on a neural deep convolutional learning model, wherein the priority extended land class patch recognition model automatically recognizes the adjacent land class patch with the highest comprehensive score to obtain a first priority extended land class patch;

[0006] S3: Acquire the first priority extended land type map spot, and extend the first priority extended land type map spot to the existing area to obtain an expanded existing area;

[0007] S4: Set the target area ratio, obtain the target area according to the target area ratio and the area of ​​the selected area, obtain the area of ​​the existing area after expansion, calculate the target achievement degree according to the target area and the area of ​​the existing area after expansion, set the target achievement degree threshold, the target achievement degree is less than the target achievement degree threshold, the boundary of the existing area after expansion continues to be expanded, the target achievement degree reaches the target achievement degree threshold, the boundary expansion ends to obtain the target area, and the target boundary is obtained according to the target area.

[0008] Specifically, the specific steps of the comprehensive calculation in S1 are:

[0009] The weights of the importance dimension, the adaptability dimension and the change possibility dimension are obtained through the hierarchical analysis method, and the comprehensive score is obtained through comprehensive calculation based on the weights and the assignments. The comprehensive calculation formula is: y=w1x1+w2x2+w3x3, wherein w1 represents the importance dimension weight, x1 represents the importance assignment, w2 represents the adaptability dimension weight, x2 represents the adaptability assignment, w3 represents the change possibility dimension weight, x3 represents the change possibility assignment, and y represents the comprehensive score.

[0010] Specifically, the specific steps of S3 are:

[0011] The first priority extended land category icon is obtained, and the spatial similarity between the boundary of the priority extended land category icon and the existing boundary is obtained. The boundary whose spatial similarity is less than the overlap spatial similarity threshold is an overlap boundary, and the overlap boundary is eliminated to obtain the expanded existing area.

[0012] Specifically, the target area of ​​S4 is the product of the area of ​​the selected region and the target area ratio.

[0013] Specifically, the area of ​​the existing region after expansion in S4 is the sum of the area of ​​the existing region and the area of ​​the first priority expanded land type patch.

[0014] Specifically, the target achievement degree in S4 is the quotient of the area of ​​the existing region after the expansion and the target area.

[0015] Specifically, the specific steps of continuing to expand the boundary in S4 are:

[0016] Step 1, traversing the adjacent land class spots, and selecting the adjacent land class spot with the highest comprehensive score as the second priority expansion land class spot;

[0017] Step 2: Merging the second priority extended land type map patch into the expanded existing area to obtain the first expanded existing area;

[0018] Step three, calculating the target achievement degree, and if the target achievement degree is less than the target achievement degree threshold, repeating steps one and two.

[0019] In a second aspect, the present invention provides an AI-assisted natural growth system for borders, which is operated using the method described above and includes the following modules:

[0020] Land use pattern scoring module, boundary growth direction selection module, intelligent expansion module, and target demand achievement module;

[0021] The land type spot scoring module obtains the importance dimension, adaptability dimension, and change possibility dimension of the land type spots in the region, assigns values ​​to the importance dimension, the adaptability dimension, and the change possibility dimension, and obtains the comprehensive score of the land type spots through comprehensive calculation according to the assigned values;

[0022] The module for selecting the boundary growth direction obtains an existing area in the selected area, obtains an existing boundary of the existing area, obtains an adjacent land class patch adjacent to the existing boundary according to the existing boundary, obtains a comprehensive score of the adjacent land class patch, and establishes a priority expansion land class patch recognition model based on a neural deep convolutional learning model. The priority expansion land class patch recognition model automatically recognizes the adjacent land class patch with the highest comprehensive score to obtain a first priority expansion land class patch.

[0023] The intelligent expansion module obtains the first priority expansion land type map spot, and expands the first priority expansion land type map spot to the existing area to obtain an expanded existing area;

[0024] The target demand achievement module sets a target area ratio, obtains a target area based on the target area ratio and the area of ​​the selected area, obtains the area of ​​the existing area after expansion, calculates a target achievement degree based on the target area and the area of ​​the existing area after expansion, sets a target achievement degree threshold, and when the target achievement degree is less than the target achievement degree threshold, the boundary of the existing area after expansion continues to be expanded, and when the target achievement degree reaches the target achievement degree threshold, the boundary expansion ends to obtain the target area, and a target boundary is obtained based on the target area.

[0025] The beneficial effects of the present invention are:

[0026] (1) The intelligence of positive and negative extension of land type patches is improved to obtain interactive target-boundary calculation results;

[0027] (2) By selectively increasing the boundary based on the comprehensive score, the land use patterns with high comprehensive scores are effectively retained, thus improving the quality of the planning results.

[0028] (3) The algorithm is flexible and can adapt to different scenarios and needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0030] Figure 1 The figure is a flow chart of an AI-assisted natural boundary growth method in the present invention. DETAILED DESCRIPTION

[0031] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0032] See also Figure 1 , a process of an AI-assisted boundary natural growth method, including the following steps:

[0033] S1: Obtaining the importance dimension, adaptability dimension, and change possibility dimension of the land type spots in the region, assigning values ​​to the importance dimension, the adaptability dimension, and the change possibility dimension, and obtaining a comprehensive score of the land type spots through comprehensive calculation according to the assigned values;

[0034] S2: obtaining an existing area in the selected area, obtaining an existing boundary of the existing area, obtaining adjacent land class patches adjacent to the existing boundary according to the existing boundary, obtaining a comprehensive score of the adjacent land class patches, and establishing a priority extended land class patch recognition model based on a neural deep convolutional learning model, wherein the priority extended land class patch recognition model automatically recognizes the adjacent land class patch with the highest comprehensive score to obtain a first priority extended land class patch;

[0035] S3: Acquire the first priority extended land type map spot, and extend the first priority extended land type map spot to the existing area to obtain an expanded existing area;

[0036] S4: Set the target area ratio, obtain the target area according to the target area ratio and the area of ​​the selected area, obtain the area of ​​the existing area after expansion, calculate the target achievement degree according to the target area and the area of ​​the existing area after expansion, set the target achievement degree threshold, the target achievement degree is less than the target achievement degree threshold, the boundary of the existing area after expansion continues to be expanded, the target achievement degree reaches the target achievement degree threshold, the boundary expansion ends to obtain the target area, and the target boundary is obtained according to the target area.

[0037] Specifically, the specific steps of the comprehensive calculation in S1 are:

[0038] The weights of the importance dimension, the adaptability dimension and the change possibility dimension are obtained through the hierarchical analysis method, and the comprehensive score is obtained through comprehensive calculation based on the weights and the assignments. The comprehensive calculation formula is: y=w1x1+w2x2+w3x3, wherein w1 represents the importance dimension weight, x1 represents the importance assignment, w2 represents the adaptability dimension weight, x2 represents the adaptability assignment, w3 represents the change possibility dimension weight, x3 represents the change possibility assignment, and y represents the comprehensive score.

[0039] Specifically, the specific steps of S3 are:

[0040] The first priority extended land category icon is obtained, and the spatial similarity between the boundary of the priority extended land category icon and the existing boundary is obtained. The boundary whose spatial similarity is less than the overlap spatial similarity threshold is an overlap boundary, and the overlap boundary is eliminated to obtain the expanded existing area.

[0041] Specifically, the target area of ​​S4 is the product of the area of ​​the selected region and the target area ratio.

[0042] Specifically, the area of ​​the existing region after expansion in S4 is the sum of the area of ​​the existing region and the area of ​​the first priority expanded land type patch.

[0043] Specifically, the target achievement degree in S4 is the quotient of the area of ​​the existing region after the expansion and the target area.

[0044] Specifically, the specific steps of continuing to expand the boundary in S4 are:

[0045] Step 1, traversing the adjacent land class spots, and selecting the adjacent land class spot with the highest comprehensive score as the second priority expansion land class spot;

[0046] Step 2: Merging the second priority extended land type map patch into the expanded existing area to obtain the first expanded existing area;

[0047] Step three, calculating the target achievement degree, and if the target achievement degree is less than the target achievement degree threshold, repeating steps one and two.

[0048] In a second aspect, an embodiment of the present invention provides an AI-assisted natural boundary growth system, including the following modules:

[0049] Land use pattern scoring module, boundary growth direction selection module, intelligent expansion module, and target demand achievement module;

[0050] The land type spot scoring module obtains the importance dimension, adaptability dimension, and change possibility dimension of the land type spots in the region, assigns values ​​to the importance dimension, the adaptability dimension, and the change possibility dimension, and obtains the comprehensive score of the land type spots through comprehensive calculation according to the assigned values;

[0051] The module for selecting the boundary growth direction obtains an existing area in the selected area, obtains an existing boundary of the existing area, obtains an adjacent land class patch adjacent to the existing boundary according to the existing boundary, obtains a comprehensive score of the adjacent land class patch, and establishes a priority expansion land class patch recognition model based on a neural deep convolutional learning model. The priority expansion land class patch recognition model automatically recognizes the adjacent land class patch with the highest comprehensive score to obtain a first priority expansion land class patch.

[0052] The intelligent expansion module obtains the first priority expansion land type map spot, and expands the first priority expansion land type map spot to the existing area to obtain the expanded existing area;

[0053] The target demand achievement module sets a target area ratio, obtains a target area based on the target area ratio and the area of ​​the selected area, obtains the area of ​​the existing area after expansion, calculates a target achievement degree based on the target area and the area of ​​the existing area after expansion, sets a target achievement degree threshold, and when the target achievement degree is less than the target achievement degree threshold, the boundary of the existing area after expansion continues to be expanded, and when the target achievement degree reaches the target achievement degree threshold, the boundary expansion ends to obtain the target area, and a target boundary is obtained based on the target area.

[0054] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An AI-assisted boundary natural growth method, characterized in that: The steps include: S1: Obtaining the importance dimension, adaptability dimension, and change possibility dimension of the land type spots in the region, assigning values ​​to the importance dimension, the adaptability dimension, and the change possibility dimension, and obtaining a comprehensive score of the land type spots through comprehensive calculation according to the assigned values; S2: obtaining an existing area in the selected area, obtaining an existing boundary of the existing area, obtaining adjacent land class patches adjacent to the existing boundary according to the existing boundary, obtaining a comprehensive score of the adjacent land class patches, and establishing a priority extended land class patch recognition model based on a neural deep convolutional learning model, wherein the priority extended land class patch recognition model automatically recognizes the adjacent land class patch with the highest comprehensive score to obtain a first priority extended land class patch; S3: Acquire the first priority extended land type map spot, and extend the first priority extended land type map spot to the existing area to obtain an expanded existing area; S4: Set the target area ratio, obtain the target area according to the target area ratio and the area of ​​the selected area, obtain the area of ​​the existing area after expansion, calculate the target achievement degree according to the target area and the area of ​​the existing area after expansion, set the target achievement degree threshold, if the target achievement degree is less than the target achievement degree threshold, continue to expand the boundary, if the target achievement degree reaches the target achievement degree threshold, end the boundary expansion to obtain the target area, and obtain the target boundary according to the target area.

2. The AI-assisted natural boundary growth method according to claim 1, characterized in that: The specific steps of the comprehensive calculation in S1 are: The weights of the importance dimension, the adaptability dimension and the change possibility dimension are obtained through the hierarchical analysis method, and the comprehensive score is obtained through comprehensive calculation based on the weights and the assignments. The comprehensive calculation formula is: y=w1x1+w2x2+w3x3, wherein w1 represents the importance dimension weight, x1 represents the importance assignment, w2 represents the adaptability dimension weight, x2 represents the adaptability assignment, w3 represents the change possibility dimension weight, x3 represents the change possibility assignment, and y represents the comprehensive score.

3. The AI-assisted natural boundary growth method according to claim 1, characterized in that: The specific steps of obtaining the expanded existing area in S3 are: Obtain the first priority extended land category icon, obtain the boundary matrix of the priority extended land category icon and the existing boundary matrix of the existing boundary, obtain the intersection of the boundary matrix and the existing boundary matrix to obtain the overlapping boundary of the priority extended land category icon and the existing area, and eliminate the overlapping boundary to obtain the existing area after expansion.

4. The AI-assisted natural boundary growth method according to claim 1, characterized in that: The target area of ​​S4 is the product of the area of ​​the selected region and the target area ratio.

5. The AI-assisted natural boundary growth method according to claim 1, characterized in that: The area of ​​the existing region after expansion in S4 is the sum of the area of ​​the existing region and the area of ​​the first priority expanded land type map.

6. The AI-assisted natural boundary growth method according to claim 1, characterized in that: The target achievement degree in S4 is the quotient of the area of ​​the existing region after the expansion and the target area.

7. The AI-assisted natural boundary growth method according to claim 1, characterized in that: The specific steps of continuing to expand the boundary in S4 are: Step 1, traversing the adjacent land class spots, and selecting the adjacent land class spot with the highest comprehensive score as the second priority expansion land class spot; Step 2: Merging the second priority extended land type map patch into the expanded existing area to obtain the first expanded existing area; Step three, calculating the target achievement degree, and if the target achievement degree is less than the target achievement degree threshold, repeating steps one and two.

8. An AI-assisted border natural growth system, operated using the method described in any one of claims 1 to 7, characterized in that: Includes the following modules: Land use pattern scoring module, boundary growth direction selection module, intelligent expansion module, and target demand achievement module; The land type spot scoring module obtains the importance dimension, adaptability dimension, and change possibility dimension of the land type spots in the region, assigns values ​​to the importance dimension, the adaptability dimension, and the change possibility dimension, and obtains the comprehensive score of the land type spots through comprehensive calculation according to the assigned values; The module for selecting the boundary growth direction obtains an existing area in the selected area, obtains an existing boundary of the existing area, obtains an adjacent land class patch adjacent to the existing boundary according to the existing boundary, obtains a comprehensive score of the adjacent land class patch, and establishes a priority expansion land class patch recognition model based on a neural deep convolutional learning model. The priority expansion land class patch recognition model automatically recognizes the adjacent land class patch with the highest comprehensive score to obtain a first priority expansion land class patch. The intelligent expansion module obtains the first priority expansion land type map spot, and expands the first priority expansion land type map spot to the existing area to obtain the expanded existing area; The target demand achievement module sets a target area ratio, obtains a target area based on the target area ratio and the area of ​​the selected area, obtains the area of ​​the existing area after expansion, calculates a target achievement degree based on the target area and the area of ​​the existing area after expansion, sets a target achievement degree threshold, and when the target achievement degree is less than the target achievement degree threshold, the boundary of the existing area after expansion continues to be expanded, and when the target achievement degree reaches the target achievement degree threshold, the boundary expansion ends to obtain the target area, and a target boundary is obtained based on the target area.

Citation Information

Patent Citations

  • Remote sensing image segmentation method based on hard boundary constraint and two-stage combination

    CN103208121A

  • Generation method three-dimensional land utilization present situation data model

    CN105719344A

  • Edge completeness-based image segmentation optimal identification method

    CN106846344A

  • City development boundary prediction method based on SLEUTH model

    CN110598513A

  • Land satellite image law enforcement system based on pattern spot dynamic remote sensing monitoring

    CN117274792A