Territorial space planning and splicing system for multi-zone information processing

Abstract

The invention discloses a territorial space planning and splicing system for multi-zone information processing, and relates to the technical field of territorial space planning, which comprises a data acquisition and preprocessing module, a zone analysis module, a planning classification module and a territorial space planning and splicing module, through the regional analysis module and the planning classification module, the acquired multi-source data information can be converted into the same format and substituted into the same regional coordinate, so that the landmark points of each region can be marked and inquired, and the accuracy of the data processing is improved. Meanwhile, database information is subjected to unified classification and feature extraction through automatic classification and extraction, the purpose of classifying territorial planning is achieved, collected space images are fused more accurately through a territorial space planning and splicing module, in the image splicing and fusion process, fusion gap traces between the images are reduced, and the fusion efficiency is improved. The images are more natural and smooth when being spliced, so that the quality and the stability of the fused images are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of national land space planning, and in particular to a national land space planning splicing system for multi-region information processing. Background Art

[0002] National land space planning splicing is the integration of geographic information data from different data sources, such as satellite images, aerial photography, topographic maps, land use data, planning drawings, etc. Splicing and integrating these data can provide comprehensive and accurate basic information for national land planning. As time goes by, national land planning needs to continuously update data to reflect changes in actual conditions. The national land planning splicing system can easily update and maintain data to ensure that the information based on the planning is always kept up to date.

[0003] Existing splicing processing may not be able to handle complex geographical features and boundary conditions well. Data from different regions may have inconsistent accuracy, inconsistent coordinate systems, incompatible data formats, etc., which will affect the accuracy and reliability of splicing. National land space planning is a dynamic process. Over time, the planning data of various regions will change. If the splicing system produces errors when performing planning classification, the planning results will not be consistent with the actual situation. Unified standards and specifications can improve the reliability and interoperability of the system;

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to obtain the required planned land space image by collecting database information of multiple areas, and at the same time adopt a classification model to classify the attributes of the collected images so as to distinguish and manage the planning map, and then perform image processing on the database images to reduce the obvious splicing marks produced in the fusion process when the images are spliced ​​and fused, so as to establish a complete spatial planning splicing map.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a national land space planning splicing system for multi-region information processing, including a data collection and preprocessing module, a regional analysis module, a planning classification module and a national land space planning splicing module:

[0007] Data collection and preprocessing module: used to collect national land space planning related data from different departments and data sources in different formats, and uniformly process the collected national land planning data, establish a database based on the processed national land planning data, and send the database information to the regional analysis module;

[0008] Regional analysis module: used to receive and analyze database information, convert the coordinate system of the land planning data in each database information, establish a corresponding land information coordinate system, and classify different regional locations through the established land information coordinate system;

[0009] Planning classification module: used to obtain and receive the database information after division, and conduct regional spatial analysis on the database information after division, classify the regional locations into spatial regions, and mark the range and coordinate values ​​between each region;

[0010] National land space planning splicing module: processes the processed database information, determines the boundaries and scope of splicing, coordinates the national land space planning of different regions, resolves spatial conflicts, achieves consistency and coordination of planning, uses appropriate algorithms to splice the planning data of different regions, and generates a complete national land space splicing map.

[0011] Furthermore, the specific steps of constructing the data coordinate system of the regional analysis module are as follows:

[0012] Step 1: Determine the coverage of the spatial area through the acquired database information, divide the area according to the geographical location and administrative boundaries of the spatial area, draw a regional range map, and automatically extract the characteristics of each spatial area including area, shape, center location and other characteristics;

[0013] Step 2: Use the known regional geographical location as the coordinate origin, the east-west direction as the horizontal axis, and the north-south direction as the vertical axis to establish a national land information coordinate system, and at the same time establish a regional geographical location coordinate map, and mark each regional coordinate point and convert it into the same type;

[0014] Step 3: Establish the corresponding parameter table of coordinate points: organize the parameters of each coordinate point into a parameter table for query and application in the coordinate system. The parameter table includes parameter name, parameter value, and parameter description, which are converted and queried according to the interactive software; use the selected conversion tool and conversion parameters to convert the coordinates of each data source from the original coordinate system to the coordinates in the land information coordinate system;

[0015] Step 4: Verify the coordinate point parameter location information, use the established land information coordinate system, collect the coordinate data information of some points with known locations, and convert them into the land information coordinate system for training.

[0016] Furthermore, the process of automatically extracting the features of each spatial region in step 1 is as follows:

[0017] Step 1: Establish a standard land planning classification data classification model, select training samples, establish a classification model based on the characteristics of the samples, input the database image of the preprocessed database information into the classification model, perform classification operations on the entire area through cluster analysis of the acquired data, and conduct a preliminary evaluation of the classification results to check whether there are obvious errors and unreasonable classifications, and finally obtain a trained data classification model;

[0018] Step 2: Substitute the acquired database information into the trained data classification model, perform automatic extraction and classification of spatial regional features, distinguish landform and terrain features according to the classification recognition model, and obtain the differentiated database information;

[0019] Step 3: Convert the differentiated database information from the image into a three-dimensional image view, perform image processing fusion on the three-dimensional image view, perform geometric correction on the input image, perform contrast and clarity enhancement on the image, and obtain the processed database information;

[0020] Step 4: Integrate the processed database information into images, mark the feature points of each image, detect the feature points of each view, and verify the view feature points using a geometric verification method to ensure that the three-dimensional image view image information is in a unified format. Integrate the processed database information and send it to the national land space planning splicing module.

[0021] Furthermore, the specific process of constructing the data classification model is as follows:

[0022] The database image in the database information is preset as a test sample. If its image feature vector neighbor set is {x1, x2, ..., x k}, the feature vector of the standard training image is {b1, b2, ..., b k}, then the corresponding classification category is {y1, y2, ..., y k}, the standard feature vector classification categories are {C1, C2, ..., C k},

[0023] First, the feature distance d between the feature vector of the i-th test sample and the feature vector of the standard training image is calculated. i , and then through the test sample distance d i Find the weight Finally, the classification result of the test sample x is obtained:

[0024] If the category corresponding to the test sample belongs to a specific standard category, then y i The value of is 1, otherwise it is 0. The category to which the test sample finally belongs is obtained through the distance weight and the neighboring category of the image feature vector.

[0025] Furthermore, the specific process of constructing the national land space planning splicing is as follows;

[0026] Step 1: Obtain the database information sent by the planning classification module, and transform the image of the same area according to the classified image information so that it is aligned with the image of another area in terms of spatial position;

[0027] Step 2: Determine the stitching order of the images according to the image content and feature matching results. For each pair of adjacent images, calculate the transformation matrix between them according to the feature matching results, including translation, rotation and scaling.

[0028] Step 3: After spatially aligning the database information, perform weighted fusion processing on the spatial images of the two regions.

[0029] Furthermore, the specific processing process of image fusion in step 3 is as follows:

[0030] The edge points of the images to be fused are preliminarily determined, the pixel value at the edge is set to 1, and the pixel value at the non-edge is 0. For weak edge pixels, if they are adjacent to a strong edge pixel, they are marked as edge pixels; otherwise, they are marked as non-edge pixels; the overlapping part of the images during the fusion process is the overlapping area, and the pixel value of the image to be spliced ​​and fused at a fixed point coordinate is used, and the corresponding fusion weight coefficient is assigned to the pixel of each image. The weight coefficient is greater than or equal to zero and less than or equal to 1 when taking the value; the pixel value of the fused image is determined by calculating the pixel value of the image to be fused and the weight coefficient, and the fused image is obtained.

[0031] Furthermore, the weighted fusion processing operation process in step 3 is specifically as follows:

[0032] When fusing adjacent images, the preset image edge detection result is H(x,y), and the image fusion weight is as follows:

[0033] If the first image fusion weight is:

[0034] σ is the actual adjustment parameter, and its value range is a number greater than or equal to zero and less than or equal to 1. w0 is the original fusion weight initially assigned;

[0035] Then the second image fusion weight is: w2(x,y)=1-w1(x,y), (w1+w2=1);

[0036] When two images are fused, the final fusion result is: F(x,y)=w iH1(x,y)+w2H2(x,y).

[0037] Furthermore, when unreasonable classification judgments are made during the automatic extraction of features of various spatial regions, the images classified into a certain category are checked to determine whether they should indeed belong to the category, and specific image category attributes are set for comparison with the images in the database information. If the image comparison is a correct classification, the image comparison operation is continued; if the image comparison is an incorrect classification, the image is marked; all marked images are processed again through the classification model, and the reclassified database images are subjected to a secondary classification comparison verification, if the database image classification is correct.

[0038] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0039] The national land space planning splicing system for multi-region information processing, through the regional analysis module and the planning classification module, enables the collected multi-source data information to be converted into the same format and substituted into the same regional coordinates, so as to facilitate the marking query of landmark points in each region, and at the same time, the database information is uniformly classified and extracted features through automatic classification extraction, so as to achieve the classification purpose of national land planning, and through the national land space planning splicing module, the collected spatial images are more accurately fused, so that the gap traces between the images are reduced during the splicing and fusion process, and the images are more natural and smooth when the splicing is completed, and the real-time updating of weights can make the fusion algorithm better adapt to the dynamically changing scenes, improve the quality and stability of the fused images, and provide more complete information for image analysis and understanding during use, thereby improving the accuracy of the analysis, and by integrating the national land space planning information of different regions, a more complete splicing system diagram is obtained for use and display. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of the overall external system structure flow of the present invention is shown. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Example:

[0043] like Figure 1As shown, a national land space planning splicing system for multi-region information processing includes a data acquisition and preprocessing module, a regional analysis module, a planning classification module and a national land space planning splicing module:

[0044] Data collection and preprocessing module: used to collect national land space planning related data from different departments and data sources in different formats, and uniformly process the collected national land planning data, establish a database based on the processed national land planning data, and send the database information to the regional analysis module;

[0045] In the process of data collection, geographic information data from different data sources can be integrated, including but not limited to satellite images, aerial photography, topographic maps, land use data, and planning drawings. The splicing and integration of these data can provide comprehensive and accurate basic information for national land planning.

[0046] Regional analysis module: used to receive and analyze database information, convert the coordinate system of the land planning data in each database information, establish a corresponding land information coordinate system, and classify different regional locations through the established land information coordinate system;

[0047] Planning classification module: used to obtain and receive the database information after division, and conduct regional spatial analysis on the database information after division, classify the regional locations into spatial regions, and mark the range and coordinate values ​​between each region;

[0048] National land space planning splicing module: Process the processed database information, determine the boundaries and scope of splicing, coordinate the national land space planning of different regions, resolve spatial conflicts, achieve consistency and coordination of planning, use appropriate algorithms to splice planning data of different regions, and generate a complete national land space planning map;

[0049] After the database images within the database information are fused, the resulting splicing system map can be fused with the established land information coordinate system map, which is convenient for users to consult and analyze. By marking the coordinates on the image, the distance, direction and other information between each location can be clearly seen, which helps users better understand the spatial layout. The image fused with coordinate information can simultaneously display the visual content of the image itself and the corresponding geographic location information, enriching the level of information display. Users can obtain more data background information while viewing the image, thereby improving the readability and usability of the information.

[0050] The specific steps of constructing the data coordinate system of the regional analysis module are as follows:

[0051] Step 1: Determine the coverage of the spatial area through the acquired database information, divide the area according to the geographical location and administrative boundaries of the spatial area, draw a regional range map, and automatically extract the characteristics of each spatial area including area, shape, center location and other characteristics;

[0052] Step 2: Use the known regional geographical location as the coordinate origin, the east-west direction as the horizontal axis, and the north-south direction as the vertical axis to establish a national land information coordinate system, and at the same time establish a regional geographical location coordinate map, and mark each regional coordinate point and convert it into the same type;

[0053] Step 3: Establish the corresponding parameter table of coordinate points: organize the parameters of each coordinate point into a parameter table for query and application in the coordinate system. The parameter table includes parameter name, parameter value, and parameter description, which are converted and queried according to the interactive software; use the selected conversion tool and conversion parameters to convert the coordinates of each data source from the original coordinate system to the coordinates in the land information coordinate system;

[0054] Step 4: Verify the coordinate point parameter location information, use the established land information coordinate system, collect the coordinate data information of some points with known locations, and convert them into the land information coordinate system for training;

[0055] The process of automatically extracting the features of each spatial region in step 1 is as follows:

[0056] Step 1: Establish a standard land planning classification data classification model, select training samples, establish a classification model based on the characteristics of the samples, input the database image of the preprocessed database information into the classification model, perform classification operations on the entire area through cluster analysis of the acquired data, and conduct a preliminary evaluation of the classification results to check whether there are obvious errors and unreasonable classifications, and finally obtain a trained data classification model;

[0057] Step 2: Substitute the acquired database information into the trained data classification model, perform automatic extraction and classification of spatial regional features, distinguish landform and terrain features according to the classification recognition model, and obtain the differentiated database information;

[0058] Step 3: Convert the differentiated database information from the image into a three-dimensional image view, perform image processing fusion on the three-dimensional image view, perform geometric correction on the input image, perform contrast and clarity enhancement on the image, and obtain the processed database information;

[0059] Step 4: Integrate the processed database information into images, mark the feature points of each image, detect the feature points of each view, and verify the view feature points using a geometric verification method to ensure that the three-dimensional image view image information is in a unified format. Integrate the processed database information and send it to the national land space planning splicing module.

[0060] The specific process of constructing the data classification model is as follows:

[0061] The data image in the database information is preset as a test sample. If its image feature vector neighbor set is {x1, x2, ..., x k}, the feature vector of the standard training image is {b1, b2, ..., b k}, then the corresponding classification category is {y1, y2, ..., y k},

[0062] First, the feature distance d between the feature vector of the i-th test sample and the feature vector of the standard training image is calculated. i , and then through the test sample distance d i Find the weight Finally, the classification result of the test sample x is obtained:

[0063] If the category corresponding to the test sample belongs to a specific standard category, then y i The value of is 1, otherwise it is 0. The category to which the test sample finally belongs is obtained through the distance weight and the neighboring category of the image feature vector;

[0064] In the process of testing samples, when a new test image appears, it is only necessary to calculate its feature distance with the image in the training set to classify it. The classification process includes ecological zones, industrial zones, residential areas, environmental zones, and areas to be developed. It is not limited to the above regional classification planning. Each planning area has specific classification distinctions, which can compare and summarize the collected database information.

[0065] The specific process of constructing the national land space planning splicing is as follows;

[0066] Step 1: Obtain the database information sent by the planning classification module, and transform the image of the same area according to the classified image information so that it is aligned with the image of another area in terms of spatial position;

[0067] Step 2: Determine the stitching order of the images based on the image content and feature matching results. For each pair of adjacent images, calculate the transformation matrix between them based on the feature matching results, including translation, rotation, and scaling.

[0068] Step 3: After spatially aligning the database information, perform weighted fusion processing on the spatial images of the two regions;

[0069] The specific processing process is as follows: preliminarily determine the edge points of the images to be fused, set the pixel value at the edge to 1, and the pixel value at the non-edge to 0. For weak edge pixels, if they are adjacent to a strong edge pixel, they are marked as edge pixels; otherwise, they are marked as non-edge pixels; the overlapping part of the images during the fusion process is the overlapping area, and the pixel value of the image to be spliced ​​and fused at a fixed point coordinate is used, and the corresponding fusion weight coefficient is assigned to the pixel of each image. The weight coefficient is greater than or equal to zero and less than or equal to 1 when taking the value; the pixel value of the fused image is determined by calculating the pixel value of the image to be fused and the weight coefficient, and the fused image is obtained;

[0070] The stitched image is a stitched image with a coordinate system that is exchanged and integrated with the established land information coordinate system.

[0071] The specific operation process of weighted fusion processing in step 3 is as follows:

[0072] When fusing adjacent images, the preset image edge detection result is H(x,y), and the image fusion weight is as follows:

[0073] If the first image fusion weight is:

[0074] σ is the actual adjustment parameter, and its value range is a number greater than or equal to zero and less than or equal to 1. w0 is the original fusion weight initially assigned;

[0075] Then the second image fusion weight is: w2(x,y)=1-w1(x,y), (w1+w2=1);

[0076] When two images are fused, the final fusion result is: F(x,y)=w i H1(x,y)+w2H2(x,y);

[0077] The fusion weight coefficient determines the relative contribution of each image in the fusion result. If w1 is larger, it means that the first image contributes more to the fusion result, and the fused image is closer to the pixel value of the first image at that pixel position. On the contrary, if W2 is larger, the fused image is closer to the pixel value of the second image. Therefore, when adjacent images are fused, the stitching traces between the two images can be reduced, making the stitching image more complete and smooth when displayed.

[0078] When judging the unreasonable classification in step 1, by checking the image classified into a certain category, it is judged whether it should really belong to this category, and by setting a specific image category attribute to compare with the image in the database information, if the image comparison is correctly classified, the image comparison operation is continued; if the image comparison is incorrectly classified, the image is marked; all marked images are processed again through the classification model, and the database images that have been reclassified will be subjected to secondary classification comparison verification. If the database image classification is correct;

[0079] The generated system diagram can provide a multi-person collaborative work environment, making it convenient for planners from different departments to participate in land planning projects together. It can realize functions such as data sharing, task allocation, and progress tracking, improve work efficiency and coordination, and share and publish the results of land planning in various forms, such as web maps, mobile applications, etc. It is convenient for the public to query and understand land planning information, improve the transparency of planning and public participation, and can view urban planning maps through mobile applications to understand surrounding construction projects and development plans.

[0080] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0081] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0082] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware; whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution; professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0083] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A national land space planning splicing system for multi-region information processing, characterized in that: It includes data collection and preprocessing module, regional analysis module, planning classification module and national land space planning splicing module: Data collection and preprocessing module: used to collect national land space planning related data from different departments and data sources in different formats, and uniformly process the collected national land planning data, establish a database based on the processed national land planning data, and send the database information to the regional analysis module; Regional analysis module: used to receive and analyze database information, convert the coordinate system of the land planning data in each database information, establish a corresponding land information coordinate system, and classify different regional locations through the established land information coordinate system; Planning classification module: used to obtain and receive the database information after division, and conduct regional spatial analysis on the database information after division, classify the regional locations into spatial regions, and mark the range and coordinate values ​​between each region; National land space planning splicing module: processes the processed database information, determines the boundaries and scope of splicing, coordinates the national land space planning of different regions, resolves spatial conflicts, achieves consistency and coordination of planning, uses appropriate algorithms to splice the planning data of different regions, and generates a complete national land space splicing map.

2. The national land space planning splicing system for multi-region information processing according to claim 1 is characterized in that: The specific steps of constructing the data coordinate system of the regional analysis module are as follows: Step 1: Determine the coverage of the spatial area through the acquired database information, divide the area according to the geographical location and administrative boundaries of the spatial area, draw a regional range map, and automatically extract the characteristics of each spatial area including area, shape, center location and other characteristics; Step 2: Use the known regional geographical location as the coordinate origin, the east-west direction as the horizontal axis, and the north-south direction as the vertical axis to establish a national land information coordinate system, and at the same time establish a regional geographical location coordinate map, and mark each regional coordinate point and convert it into the same type; Step 3: Establish the corresponding parameter table of coordinate points: organize the parameters of each coordinate point into a parameter table, which includes parameter name, parameter value, and parameter description, and convert and query them according to the interactive software; use the selected conversion tool and conversion parameters to convert the coordinates of each data source from the original coordinate system to the coordinates in the land information coordinate system; Step 4: Verify the coordinate point parameter location information, use the established land information coordinate system, collect the coordinate data information of some points with known locations, and convert them into the land information coordinate system for training.

3. The national land space planning splicing system for multi-region information processing according to claim 2 is characterized in that: The process of automatically extracting the features of each spatial region in step 1 is as follows: Step 1: Establish a standard land planning classification data classification model, select training samples, establish a classification model based on the characteristics of the samples, input the database image of the preprocessed database information into the classification model, perform classification operations on the entire area through cluster analysis of the acquired data, and conduct a preliminary evaluation of the classification results to check whether there are obvious errors and unreasonable classifications, and finally obtain a trained data classification model; Step 2: Substitute the acquired database information into the trained data classification model, perform automatic extraction and classification of spatial regional features, distinguish landform and terrain features according to the classification recognition model, and obtain the differentiated database information; Step 3: Convert the differentiated database information from the image into a three-dimensional image view, perform image processing fusion on the three-dimensional image view, perform geometric correction on the input image, perform contrast and clarity enhancement on the image, and obtain the processed database information; Step 4: Integrate the processed database information into images, mark the feature points of each image, detect the feature points of each view, and verify the view feature points using a geometric verification method to ensure that the three-dimensional image view image information is in a unified format. Integrate the processed database information and send it to the national land space planning splicing module.

4. The national land space planning splicing system for multi-region information processing according to claim 3 is characterized in that: The specific process of constructing the data classification model is as follows: The database image in the database information is preset as a test sample. If its image feature vector neighbor set is {x1, x2, ..., x k }, the feature vector of the standard training image is {b1, b2, ..., b k }, then the corresponding classification category is {y1, y2, ..., y k }, the standard feature vector classification categories are {C1, C2, ..., C k }, First, the feature distance d between the feature vector of the i-th test sample and the feature vector of the standard training image is calculated. i , and then through the test sample distance d i Find the weight Finally, the classification result of the test sample x is obtained: If the category corresponding to the test sample belongs to a specific standard category, then y i The value of is 1, otherwise it is 0. The category to which the test sample finally belongs is obtained through the distance weight and the neighboring category of the image feature vector.

5. The national land space planning splicing system for multi-region information processing according to claim 1 is characterized in that: The specific process of constructing the national land space planning splicing is as follows; Step 1: Obtain the database information sent by the planning classification module, and transform the image of the same area according to the classified image information so that it is aligned with the image of another area in terms of spatial position; Step 2: Determine the stitching order of the images according to the image content and feature matching results. For each pair of adjacent images, calculate the transformation matrix between them according to the feature matching results, including translation, rotation and scaling. Step 3: After spatially aligning the database information, perform weighted fusion processing on the spatial images of the two regions.

6. The national land space planning splicing system for multi-region information processing according to claim 5 is characterized in that: The specific processing process of image fusion in step 3 is as follows: The edge points of the images to be fused are preliminarily determined, the pixel value at the edge is set to 1, and the pixel value at the non-edge is 0. For weak edge pixels, if they are adjacent to a strong edge pixel, they are marked as edge pixels; otherwise, they are marked as non-edge pixels; the overlapping part of the images during the fusion process is the overlapping area, and the pixel value of the image to be spliced ​​and fused at a fixed point coordinate is used, and the corresponding fusion weight coefficient is assigned to the pixel of each image. The weight coefficient is greater than or equal to zero and less than or equal to 1 when taking the value; the pixel value of the fused image is determined by calculating the pixel value of the image to be fused and the weight coefficient, and the fused image is obtained.

7. The national land space planning splicing system for multi-region information processing according to claim 5 is characterized in that: The specific operation process of weighted fusion processing in step 3 is as follows: When fusing adjacent images, the preset image edge detection result is H(x, y), and the image fusion weight is as follows: If the first image fusion weight is: σ is the actual adjustment parameter, and its value range is a number greater than or equal to zero and less than or equal to 1. w0 is the original fusion weight initially assigned; Then the second image fusion weight is: w2(x, y) = 1-w1(x, y), (w1+w2 = 1); When two images are fused, the final fusion result is: F(x, y) = w i H1(x,y)+w2H2(x,y).

8. The national land space planning splicing system for multi-region information processing according to claim 3 is characterized in that: When the unreasonable classification judgment is made in the automatic extraction of each spatial region feature, the image classified into a specific category is checked to determine whether it should indeed belong to the category, and a specific image category attribute is set to compare with the image in the database information. If the image comparison is a correct classification, the image comparison operation is continued; if the image comparison is an incorrect classification, the image is marked; All marked images are processed again through the classification model. The reclassified database images will be subjected to secondary classification comparison verification. If the database images are classified correctly.

Images