CN113822832A - Natural resource multi-source vector data fusion method - Google Patents

Natural resource multi-source vector data fusion method Download PDF

Info

Publication number
CN113822832A
CN113822832A CN202111034005.2A CN202111034005A CN113822832A CN 113822832 A CN113822832 A CN 113822832A CN 202111034005 A CN202111034005 A CN 202111034005A CN 113822832 A CN113822832 A CN 113822832A
Authority
CN
China
Prior art keywords
data
spot
image
fusion
spots
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111034005.2A
Other languages
Chinese (zh)
Inventor
董雅雯
师静
曾明宇
郑红
杨宁
贺东北
李世杰
肖玲
曹霸
彭检贵
陈兴亚
黄冰倩
汪帅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guizhou Forestry Investigation And Planning Institute
Central South Investigation Planning And Design Institute Of State Forestry And Grassland Administration
Original Assignee
Guizhou Forestry Investigation And Planning Institute
Central South Investigation Planning And Design Institute Of State Forestry And Grassland Administration
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guizhou Forestry Investigation And Planning Institute, Central South Investigation Planning And Design Institute Of State Forestry And Grassland Administration filed Critical Guizhou Forestry Investigation And Planning Institute
Priority to CN202111034005.2A priority Critical patent/CN113822832A/en
Publication of CN113822832A publication Critical patent/CN113822832A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Geometry (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Processing Or Creating Images (AREA)

Abstract

The invention discloses a natural resource multi-source vector data fusion method, which comprises the steps of selecting basic data and thematic data in natural resource multi-source vector data, carrying out spatial topology inspection and spot rationality inspection, and then matching the boundary of the two graphs with the geometric characteristics of the graphs to obtain first fusion data; carrying out image spot rationality check on the first fusion data, then removing unreasonable image spots in the basic data, and carrying out image geometric adjustment on the screened unreasonable image spots to obtain second fusion data; fusing the basic data attribute and the thematic data attribute of the second fused data to obtain a corresponding fused data attribute; and finally, combining the adjacent image spots with similar attributes in the second fusion data to obtain third fusion data. The invention can realize the linkage adjustment of the geometric characteristic fusion result and the attribute characteristic fusion result, and respectively adjust the unreasonable image spots, thereby effectively reducing the data redundancy in the fusion result.

Description

Natural resource multi-source vector data fusion method
Technical Field
The invention relates to the field of spatial data processing, in particular to a natural resource multi-source vector data fusion method.
Background
The natural resources are uniformly managed by a natural resource department adjusted by the organization, and comprise mountains, water, forests, fields, lakes, grasses and sands. The natural resource data come from a plurality of management departments before the adjustment of the mechanism, and the department standards and the side points of each management department have local differences to form natural resource multi-source vector data. In an actual situation, according to needs, basic data and thematic data can be selected from natural resource multi-source vector data, the basic data and the thematic data are fused to obtain fused data meeting requirements, for example, when the actual distribution situation of forest resources of an administrative unit needs to be determined, third national homeland survey data of the administrative unit is selected as the basic data, forest resource management 'map' data of the administrative unit is selected as the thematic data, and the fused data obtained by fusing the basic data and the thematic data can reflect the actual distribution situation of the forest resources of the administrative unit.
The multi-source vector data fusion mainly comprises two aspects of geometric feature fusion and attribute feature fusion. Wherein:
the geometric feature fusion comprises two processes of geometric feature matching and geometric adjustment:
the geometric feature matching comprises the following steps:
based on the matching of the geometric features, whether the geographic features belong to the same element is judged by calculating the similarity of the geometric features (distance, shape, direction and the like) of the multisource geographic elements and setting a threshold.
The point element matching mostly adopts the distance as an index to calculate the similarity between the geographic elements, and the specific calculation method comprises a position nearest algorithm and a mutual position nearest algorithm;
the line element matching adopts indexes such as distance, length, direction, maximum chord, composition area and the like to calculate the similarity between the geographic elements, and the specific calculation method comprises probability calculation, similarity calculation of different road sections, a matching algorithm based on the spatial relationship between a typical feature and a point to be matched, a buffer area overlapping mode, an ant colony algorithm, a multi-source logistic regression model matching algorithm and the like.
The similarity between the geographic elements is calculated by adopting indexes including area, curvature, turning point, invariant distance, solidity and the like in the surface element matching, and the specific calculation method comprises similarity matching based on position proximity, similarity matching based on overlapping area, similarity matching based on shape, surface entity matching based on comprehensive factors and the like.
Based on the matching of the topological features, whether the topological features belong to the same element is judged by calculating the similarity of the topological features (adjacency, association, inclusion and the like) of the multi-source geographic elements and setting a threshold. The method comprises matching based on spatial relationship similarity, face entity matching by using topology and spatial similarity and the like.
Based on the matching of the attribute characteristics, the characteristic that the basic description of the same geographic phenomenon is similar by the multi-source geographic elements is utilized to realize the characteristic matching. The method comprises a Chinese approximate character string matching algorithm, a Chinese place name proper name similarity calculation method and a Chinese place name full name semantic similarity calculation method. The method is carried out by adopting a direct spatial superposition analysis mode.
And the geometric adjustment integrates the matched multi-source vector data through operations such as selection, simplification, updating, relationship coordination and the like, and finally obtains the fused geometric characteristics.
The attribute feature fusion realizes matching between different semantics through application of semantic mapping relation and calculation of attribute information similarity so as to obtain the fused attribute feature. The method comprises two modes of attribute fusion based on mapping conversion rules and attribute fusion based on geographic ontologies.
The attribute fusion based on the mapping conversion rule comprises two modes of character string based and dictionary based. The method comprises the steps of calculating semantic similarity facing a general graph based on attribute fusion of a geographic ontology, utilizing a mixed ontology mode supporting bidirectional mapping, matching based on an object-oriented method, a semantic matching method based on grid service, a semantic similarity calculating method, a Chinese place name similarity matching algorithm considering full name semantics and the like.
The geometric characteristics and the attribute characteristics of the multi-source vector data are different due to the fact that image base maps, geographic information standards, industry identification standards, data processing requirements and data quality inspection standards and the like adopted in the production process of the multi-source vector data of natural resources are different, difficulty of data fusion is greatly increased, the geometric characteristics are different due to the fact that image base maps and artificial division results are different, the geometric characteristics are matched to generate a large number of unreasonable image spots such as fine-breaking image spots, fine-strip image spots and sharp-angle image spots, and the geometric characteristics and the attribute characteristics are split in the fusion process of the multi-source vector data to cause a large amount of data redundancy.
In the prior art, patent CN201010230447 discloses a spatial data fusion method, which first preprocesses data and converts the data into data with a uniform format; then, fusing and edge matching are carried out on the converted data; and finally, merging and outputting the processed data, and hooking the attributes in the original data. The scheme only combines the tiny patches, and fails to consider the solution of attribute feature conflict, and simultaneously the geometric feature fusion and the attribute feature fusion are split from each other. Patent CN201610514743 discloses a method for merging and updating road vector data, which comprises preprocessing data, merging data by geometric matching, semantic matching, topology matching and regularity knowledge matching, selecting high-precision merged data to establish a road network topology relationship, and publishing data. According to the scheme, the fusion of surface data is not considered, the data with high precision is directly selected, so that the key information contained in the data with low precision is lost, the fusion of data fusion results is insufficient, the solution of attribute feature conflict cannot be considered, and meanwhile, the geometric feature fusion and the attribute feature fusion are mutually split. Patent CN201110434191 discloses a method for integrating and fusing sea and land vector map data, which includes data preprocessing, data format conversion, and matching and merging entities with the same name, but the scheme includes matching of graph geometry, geometric adjustment of the graph is not considered, precision of fused data is difficult to guarantee, and a solution of attribute feature conflict cannot be considered, and meanwhile, geometric feature fusion and attribute feature fusion are split.
Therefore, the prior art cannot completely solve the problems existing in the current natural resource multi-source vector data fusion, and a more optimized natural resource multi-source vector data fusion scheme is needed.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides a natural resource multi-source vector data fusion method which can realize linkage adjustment of a geometric feature fusion result and an attribute feature fusion result, and respectively adjust unreasonable pattern spots such as fine-grained pattern spots, long-grained pattern spots, sharp-angle pattern spots and the like in fusion data, thereby effectively reducing data redundancy in the fusion result.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a natural resource multi-source vector data fusion method comprises the following steps:
s1) selecting basic data and thematic data in natural resource multi-source vector data, carrying out spot rationality check on the basic data to obtain a first unreasonable spot and recording the first unreasonable spot, fitting the graph boundary of the basic data and the thematic data, fixing the graph nodes in the basic data, moving the corresponding graph nodes in the thematic data, and fitting the graph nodes of the basic data with the fitting distance smaller than a threshold value and the graph nodes corresponding to the thematic data;
s2) carrying out graph geometric feature matching on the basic data after the graph boundary fitting and the thematic data to obtain first fusion data;
s3) carrying out image spot rationality check on the first fusion data to obtain a second unreasonable image spot, removing the first unreasonable image spot from the second unreasonable image spot to obtain a third unreasonable image spot, and carrying out image geometric adjustment on the third unreasonable image spot in the first fusion data to obtain second fusion data;
s4) extracting the basic data attribute and the thematic data attribute of each graphic element in the second fused data, and combining the basic data attribute and the thematic data attribute of all the graphic elements corresponding to the same graphic spot to obtain the fused data attribute of the graphic spot;
s5) merging the adjacent patches with similar fusion data attributes in the second fusion data to obtain third fusion data.
Further, the steps of step S1) and step S3) including the spot rationality check specifically include:
acquiring the areas of all the image spots in the inspection object, and marking the image spots with the areas smaller than a preset area threshold as fine-crushing image spots in the unreasonable image spots;
obtaining the internal angles of all the image spots in the inspection object, and marking the image spots corresponding to the internal angles smaller than a preset angle threshold value as sharp angle image spots in unreasonable image spots;
acquiring all the image spots in the inspection object, calculating the ratio of the area of each image spot to the length of the central line to obtain the corresponding average thickness, screening the image spots with the average thickness smaller than a preset thickness threshold, calculating the maximum inscribed rectangle of the screened image spots, and marking the image spots with the side length of the maximum inscribed rectangle smaller than the preset side length threshold as long and thin image spots in unreasonable image spots.
Further, the performing of the geometric figure adjustment on the third unreasonable pattern spot in the first fusion data in step S3) includes fine-breaking pattern spot adjustment, which specifically includes:
and for the fine-crushing image spots in the third unreasonable image spots, calculating the length of the common edge of the current fine-crushing image spot and other adjacent image spots, and combining the current fine-crushing image spot and the adjacent image spot with the longest common edge to generate a new image spot.
Further, the performing of the geometric adjustment on the third unreasonable image spot in the first fusion data in step S3) includes adjusting an elongated image spot, which specifically includes:
and for the long and thin image spots in the third unreasonable image spots, selecting a common node of the current long and thin image spot and other adjacent image spots, generating a perpendicular line from the common node to the center line of the current long and thin image spot, dividing the current long and thin image spot by using the perpendicular line to obtain divided image spots, calculating the length of a common edge of the current divided image spot and other adjacent image spots aiming at each divided image spot, and combining the current divided image spot and the adjacent image spot with the longest common edge to generate a new image spot.
Further, the performing of the geometric figure adjustment on the third unreasonable patches in the first fusion data in step S3) includes acute-angle patch adjustment, which specifically includes:
for the sharp corner pattern spots in the third unreasonable pattern spots, sequentially selecting common nodes of the current sharp corner pattern spot and the adjacent pattern spots according to the sequence of the distance from the sharp corner vertex of the current sharp corner pattern spot to the sharp corner vertex of the current sharp corner pattern spot from near to far, calculating the distance from the selected node in the current sharp-angle pattern spot to the perpendicular of the opposite side until the distance of the perpendicular meets the requirement, dividing the current sharp-angle pattern spot by using the perpendicular with the distance meeting the requirement to obtain a first pattern spot retaining the sharp angle and a second pattern spot without the sharp angle, selecting a common node of the first pattern spot and other adjacent pattern spots, generating the perpendicular from the common node to the center line of the first pattern spot, dividing the first pattern spot by using the perpendicular to obtain divided pattern spots, calculating the length of the common edge of the current divided pattern spot and other adjacent pattern spots for each divided pattern spot, and combining the current divided pattern spot and the adjacent pattern spot with the longest common edge to generate a new pattern spot.
Further, steps S1) and S3) further include a step of obtaining a center line of the spot, which specifically includes: and taking the graph nodes of the current image spot as discrete points to generate a Thiessen polygon, and taking the boundary of the Thiessen polygon as the central line of the current image spot.
Further, the extracting of the basic data attribute and the thematic data attribute of each graphic element in the second fused data in step S4) includes:
if the graphic elements of the basic data and the thematic data are the same, directly extracting the basic data attribute field and the thematic data attribute field of each graphic element according to the database structure and the data dictionary of the second fusion data;
if the graphic elements of the basic data and the thematic data are different, setting an attribute mapping transmission scheme, selecting fields required by data fusion in the thematic data of each graphic element, setting a scheme for attribute mapping transmission of the second fusion data and the thematic data to carry out attribute transmission, screening data capable of receiving transmission information through the ground class information and the buffer area of the thematic data in the attribute transmission process, carrying out data correction during data mapping transmission, finally obtaining the attribute fields of the mapped thematic data, and simultaneously extracting the corresponding basic data attribute fields.
Further, the merging of the basic data attributes and the thematic data attributes of all the graphic elements corresponding to the same blob in step S4) specifically includes: and combining the basic data attributes and the thematic data attributes to obtain the fusion data attributes of the image spots, and modifying the fusion data attribute information according to the texture information of the remote sensing image of the image spot area if the basic data attributes and the thematic data attributes of all the graphic elements corresponding to the same image spot conflict.
Further, the specific step of modifying the attribute information of the thematic data according to the texture information of the remote sensing image of the spot area comprises the following steps:
s401) converting the remote sensing image of the second fusion data area into a gray level image;
s402) compressing the gray level of the gray map;
s403), setting observation parameters, wherein the observation parameters comprise the size, the step pitch and the direction of a window;
s404) calculating a texture characteristic value of the center point of the window, and generating a texture characteristic matrix and a texture characteristic image along with the movement of the window;
s405) identifying the pattern spots with conflict between the basic data attribute and the thematic data attribute of the graphic element, and judging a correct result according to the texture feature matrix and the texture feature image;
s406) modifying the image spot conflict information in the fusion data attribute of the image spot according to the correct result.
Further, before the step S1) of fitting the graph boundary of the basic data and the thematic data, the method further includes the following steps:
and (3) open-air inspection: erasing basic data by using thematic data, if the erased data is empty, passing open-air inspection, and otherwise, prompting an error data range;
multi-component inspection: respectively counting the sum of single-side data in each record of the basic data and the thematic data, and prompting modification when the counted number exceeds 1, wherein the multiple parts are provided;
inspecting the surface gap: for vector data of adjacent surface data in the basic data and the thematic data, generating an inspection surface by using the outer boundary of the vector data, erasing the inspection surface by using surface data to be inspected in the vector data, if the erased data is empty, checking a surface gap, and otherwise, prompting a gap pattern spot between surfaces;
and (3) surface overlapping inspection: and extracting the overlapping part of the adjacent surface data in the basic data and the thematic data according to the spatial superposition analysis, and prompting the data of the overlapping part to be modified.
Further, before the step S1) of fitting the graph boundary of the basic data and the thematic data, the method further includes the following steps:
outdoor problem modification: completing the thematic data, and completing the thematic data of the adjacent administrative regions in the basic data range if the administrative regions are crossed;
multi-component problem modification: breaking up basic data or thematic data which are prompted to be modified, and converting multiple parts into single parts;
modification of the surface clearance: combining the gap pattern spots with the pattern spots with longer adjacent edges, thereby eliminating the gap pattern spots;
modification of the face overlap problem: determining an overlapped part to be eliminated according to the remote sensing image and eliminating the overlapped part;
data normalization: and carrying out standardization processing on the thematic data and the basic data according to a database structure, a data dictionary and data conversion logic.
Further, step S5) is followed by a step of quality inspection of the fusion result, which specifically includes:
s601) carrying out topology check on the third fusion data according to a preset topology check rule, and identifying a topology error pattern spot in the third fusion data;
s602) according to a preset logic check rule, performing logic check on the fusion data attribute of each pattern spot in the third fusion data, and identifying an attribute error pattern spot in the third fusion data;
s603) respectively modifying the topology error pattern spots and the attribute error pattern spots in the third fused data, and returning to the step S601) until the pattern spots in the third fused data meet the requirements.
The invention also provides a natural resource multi-source vector data fusion system, which comprises:
the preliminary fusion unit is used for selecting basic data and thematic data in the natural resource multi-source vector data, checking the rationality of the image spots of the basic data to obtain a first unreasonable image spot and recording the first unreasonable image spot, fitting the image boundary of the basic data and the thematic data, fixing the image node in the basic data, moving the corresponding image node in the thematic data, and fitting the image node of the basic data with the fitting distance smaller than a threshold value and the image node corresponding to the thematic data;
the geometric feature matching unit is used for performing graph geometric feature matching on the basic data after the graph boundary fitting and the thematic data to obtain first fusion data;
the geometric adjustment unit is used for carrying out image spot rationality check on the first fusion data to obtain a second unreasonable image spot, removing the first unreasonable image spot from the second unreasonable image spot to obtain a third unreasonable image spot, and carrying out image geometric adjustment on the third unreasonable image spot in the first fusion data to obtain second fusion data;
the attribute feature matching unit is used for extracting the basic data attribute and the thematic data attribute of each graphic element in the second fusion data, and the basic data attribute and the thematic data attribute of all the graphic elements corresponding to the same graphic spot are combined to obtain the fusion data attribute of the graphic spot;
and the linkage adjusting unit is used for combining adjacent image spots with similar fusion data attributes in the second fusion data to obtain third fusion data.
The invention also provides a natural resource multi-source vector data fusion system which comprises computer equipment, wherein the computer equipment is programmed or configured to execute any natural resource multi-source vector data fusion method.
The invention also provides a computer readable storage medium storing a computer program programmed or configured to perform any of the natural resource multi-source vector data fusion methods.
Compared with the prior art, the invention has the advantages that:
1. according to the method, after the fusion result of the geometric characteristics and the fusion result of the attribute characteristics are obtained, the image spots with the similarity of the attribute characteristics are further merged, so that the geometric characteristics and the attribute characteristics of the natural resource multi-source vector data are adjusted in a linkage mode, and the data redundancy of the fusion data is reduced.
2. In the geometric feature fusion process, the invention provides corresponding processing methods for fine-breaking pattern spots, long-thin pattern spots and sharp-acute pattern spots existing after vector data fusion, thereby ensuring the quality and effectiveness of data results.
3. In the attribute feature fusion process, under the condition that the thematic data attribute and the basic data attribute of the image spot conflict, the texture feature of the remote sensing image is used for judging the image spot attribute, the correct result is used for replacing the error information in the fusion data, the correctness of the attribute feature fusion result is ensured, and the interference of human factors is reduced.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Fig. 2 is a schematic diagram of data integration of basic data and topical data according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of selecting a fine-breaking pattern spot from the third unreasonable pattern spots in the first fusion data and performing geometric adjustment according to an embodiment of the present invention.
FIG. 4 is a graph illustrating the effect of the first fused data after the geometric adjustment of the fine pattern spots according to the embodiment of the present invention.
Fig. 5 is a schematic diagram of selecting a thin and long image spot from the third unreasonable image spots in the first fusion data and performing geometric adjustment according to an embodiment of the present invention.
FIG. 6 is a graph illustrating the effect of the geometric adjustment of the elongated patch of the first fused data according to the embodiment of the present invention.
Fig. 7 is a schematic diagram illustrating geometric adjustment performed by selecting a sharp-angled image spot from third unreasonable image spots in the first fusion data according to the embodiment of the present invention.
Fig. 8 is an effect diagram of the second fused data obtained finally after geometric adjustment of the sharp-angle image spot of the first fused data in the embodiment of the present invention.
FIG. 9 is a flowchart illustrating a determination process performed when the basic data topic data attributes corresponding to the pattern patches conflict according to an embodiment of the present invention.
Fig. 10 is an effect diagram of the second fused data after merging the attribute similarity patches in the embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the drawings and specific preferred embodiments of the description, without thereby limiting the scope of protection of the invention.
Example one
As shown in fig. 1, the embodiment provides a natural resource multi-source vector data fusion method, which includes six steps of data preprocessing, graph geometric feature matching, graph geometric adjustment, attribute feature fusion, geometric feature and attribute feature linkage adjustment, and quality inspection of fusion results. The data processing comprises two nodes of original data inspection and original data processing; the geometric adjustment of the graph comprises three nodes of adjustment of a fine-crushing graph spot, adjustment of a slender graph spot and adjustment of a sharp-angle graph spot; the attribute feature fusion comprises four nodes of attribute feature extraction of the same elements, attribute feature extraction of different elements, attribute feature fusion of the elements and inconsistent pattern spot judgment based on a gray level co-occurrence matrix; the quality inspection of the fusion result comprises three nodes of space inspection, attribute inspection and quality inspection after modification. The method specifically comprises the following steps:
s1) data preprocessing: selecting basic data and thematic data in natural resource multi-source vector data, carrying out spot reasonability check on the basic data to obtain a first unreasonable spot and recording the first unreasonable spot, fitting the graph boundary of the basic data and the thematic data, fixing the graph node in the basic data, moving the corresponding graph node in the thematic data, and enabling the graph node of the basic data with the fitting distance smaller than a threshold value and the graph node corresponding to the thematic data to be smaller than the threshold value;
s2) geometric feature matching: carrying out graph geometric feature matching on the basic data after the graph boundary fitting and the thematic data to obtain first fusion data;
s3) geometry adjustment of the pattern: checking the reasonability of the image spots of the first fusion data to obtain second unreasonable image spots, removing the first unreasonable image spots from the second unreasonable image spots to obtain third unreasonable image spots, and carrying out geometric figure adjustment on the third unreasonable image spots in the first fusion data to obtain second fusion data;
s4) attribute feature fusion: extracting basic data attributes and thematic data attributes of each graphic element in the second fusion data, and combining the basic data attributes and the thematic data attributes of all the graphic elements corresponding to the same graphic spot to obtain the fusion data attributes of the graphic spot;
s5) adjusting the geometric characteristics and attribute characteristics in a linkage manner: merging adjacent patches with similar fusion data attributes in the second fusion data to obtain third fusion data;
s6) quality inspection of fusion results: and checking and modifying the topology error pattern spots and the attribute error pattern spots in the third fusion data until the pattern spots in the modified third fusion data meet the requirements.
In step S1) of this embodiment, basic data and topic data in the natural resource multi-source vector data are selected and then data inspection and data processing are performed on the basic data and the topic data, where:
1) data inspection, mainly carry out the space topology inspection to thematic data and basic data, avoid the data after the integration to produce the topological problem, include:
and (3) open-air inspection: erasing basic data by using thematic data, if the erased data is empty, passing open-air inspection, and otherwise, prompting an error data range; in this embodiment, the basic data is used as a "data base map", and when the topical data is smaller than the basic data range, the topical information cannot be acquired in a part of the area, so that the open-air inspection mainly ensures that the topical data range is larger than or equal to the basic data range;
multi-component inspection: respectively counting the sum of single-side data in each record of the basic data and the thematic data, and prompting modification when the counted number exceeds 1, wherein the multiple parts are provided; in this embodiment, the multi-component check avoids the situation where a plurality of graphic elements are included in one record of the basic data and the thematic data, so as to avoid errors in data statistics and spatial position acquisition.
Inspecting the surface gap: for vector data including adjacent surface data in basic data and thematic data, generating a large surface by using the outer boundary of the vector data, wherein the large surface is called an inspection surface in the embodiment, erasing the inspection surface by using surface data to be inspected in the vector data, if the erased data is empty, passing the inspection of a surface gap, otherwise, prompting a gap pattern spot between the surfaces; in this embodiment, the surface gap inspection is to inspect whether there is a gap between adjacent surface elements in the basic data and the thematic data;
and (3) surface overlapping inspection: extracting the overlapping part of the adjacent surface data in the basic data and the thematic data according to the spatial superposition analysis, and prompting the data of the overlapping part to be modified; in this embodiment, the surface overlapping check is to check whether there is an overlapping phenomenon between adjacent surface elements in the basic data and the thematic data;
and carrying out a spot reasonableness check on the basic data, wherein the spot reasonableness check comprises the following steps:
fine pattern spot inspection: acquiring the areas of all the pattern spots in the basic data, and marking the pattern spots with the areas smaller than a preset area threshold as fine-crushing pattern spots in the first unreasonable pattern spots;
inspection of sharp-angle pattern spots: obtaining the internal angles of all the image spots in the basic data, and marking the image spots corresponding to the internal angles smaller than a preset angle threshold value as sharp angle image spots in the first unreasonable image spots;
and (3) inspecting a slender pattern spot: acquiring all the image spots in the basic data, regarding each image spot, taking an image node of the image spot as a discrete point to generate a Thiessen polygon, taking the boundary of the Thiessen polygon as a central line of the image spot, calculating the ratio of the area of each image spot to the length of the central line to obtain a corresponding average thickness, screening the image spots with the average thickness smaller than a preset thickness threshold, calculating a maximum inscribed rectangle of the screened image spots, and marking the image spots with the side length of the maximum inscribed rectangle smaller than the preset side length threshold as long and thin image spots in first unreasonable image spots;
in the embodiment, the first unreasonable pattern spot checked by the basic data is recorded, and the first unreasonable pattern spot is not adjusted in the subsequent geometric adjustment of the graph, so that the basic data in the final fusion result is ensured to be complete and unchanged;
2) data processing, mainly including topical data and basic data topology problem modification, data standardization and data integration, wherein:
the topology problem modification comprises the following steps:
outdoor problem modification: completing the thematic data, and completing the thematic data of the adjacent administrative regions in the basic data range if the administrative regions are crossed;
multi-component problem modification: breaking up basic data or thematic data which are prompted to be modified, and converting multiple parts into single parts;
modification of the surface clearance: combining the gap pattern spots with the pattern spots with longer adjacent edges, thereby eliminating the gap pattern spots;
modification of the face overlap problem: determining an overlapped part to be eliminated according to the remote sensing image and eliminating the overlapped part;
the data standardization specifically comprises: standardizing thematic data and basic data according to a database structure, a data dictionary and data conversion logic;
and finally, data integration mainly uses basic data as a limit to perform thematic data node fitting, after the basic data is fitted with the graph boundary of the thematic data, the distance between the thematic data node and the basic data node is within a certain threshold, the basic data node is fixed, the thematic data node moves, the spatial positions of the thematic data node and the basic data node are overlapped, and further the geometric primary adjustment of the multi-source data is realized.
As shown in fig. 2, in this embodiment, third national homeland survey data (hereinafter, homeland third-key data) of a village-level administration is selected from natural resource multi-source vector data as basic data, which includes 349 patches, forest resource management "one-map" data (hereinafter, forest resource management "one-map" data) of the village-level administration is selected as topic data, which includes 95 patches, the basic data is shown in the upper left side of fig. 2, the topic data is shown in the upper right side of fig. 2, the basic data and the thematic data are sequentially subjected to open-air inspection, multi-component inspection, surface gap inspection and surface overlapping inspection, and carrying out unreasonable pattern spot inspection on the basic data, wherein open-air inspection, multi-component inspection, surface gap inspection and surface overlapping inspection are all passed through without finding problems, and fine and thin pattern spots and sharp angle pattern spots obtained by unreasonable pattern spot inspection are all stored in a new pattern layer.
Since the data inspection result has no open-air, multi-component, surface gap and surface overlapping data topology problems, no modification is needed. Therefore, the thematic data and the basic data are standardized directly according to the database structure, the data dictionary and the data conversion logic, then data integration is started, thematic data node fitting is performed by taking the basic data as a limitation, after the basic data and the graph boundary of the thematic data are fitted, in the embodiment, when the node is within a threshold value of 0.001m, the data nodes of the three local tones are fixed, and the data nodes of the forest resource management ' one graph ' are moved, so that the data nodes of the three local tones are fitted with the corresponding data nodes of the forest resource management ' one graph ', tadpole-shaped ' graph spots in the range of 0.001m in the fused data are solved, and the data integration result is shown in the lower part of fig. 2.
In step S2) of this embodiment, the geometric feature matching of the graph is analyzed on the basis of direct spatial superposition, and the specific process is as follows:
s21) setting basic data not participating in cutting, and setting conditions not participating in cutting for partial data not needing to be refined according to thematic data in the natural resource basic data, in this embodiment, setting slender patches such as country roads, trunk canals and the like in the homeland three-tone data not to be cut;
s22), cutting data, wherein the pattern spots are automatically scattered into multiple parts during cutting;
s23), area calculation and adjustment, in this example, adjustment of the data of the small shift area.
Through the steps, the basic data and the thematic data are subjected to graph geometric feature matching to obtain first fusion data which comprise 1070 pattern spots, wherein the first fusion data are 3.1 times of homeland three-tone data and 11.3 times of forest resource management 'one-pattern' data.
In step S3) of the present embodiment, adjustment of the fine-breaking pattern spot is first performed:
acquiring the areas of all the image spots in the first fusion data, marking the image spots with the areas smaller than a preset area threshold as fine-breaking image spots in the second unreasonable image spots, if the fine-breaking image spots belong to the second unreasonable image spots and do not belong to the first unreasonable image spots, the image spots are fine-breaking image spots in the third unreasonable image spots, and performing geometric figure adjustment on the third unreasonable image spots in the first fusion data, including adjustment of the fine-breaking image spots, specifically comprising: and for the fine-crushing image spots in the third unreasonable image spots, calculating the length of the common edge of the current fine-crushing image spot and other adjacent image spots, and combining the current fine-crushing image spot and the adjacent image spot with the longest common edge to generate a new image spot.
As shown in fig. 3, in this embodiment, the first fusion data includes three patches, and the area of the patch a is 500m2Area of the pattern spot B is 2000m2Area of the pattern spot C is 100m2. In this embodiment, the area threshold is set to 400m2Selecting less than 400m2As a fine pattern spot; if the fine-crushing pattern spot is not the fine-crushing pattern spot in the basic data recorded in the step S1), respectively calculating the lengths of the fine-crushing pattern spot C and the adjacent common edges of the pattern spot A and the pattern spot B, wherein the length (5m) of the adjacent common edge of the fine-crushing pattern spot C and the pattern spot A is less than the length (10m) of the adjacent common edge of the pattern spot B, combining the fine-crushing pattern spot C and the pattern spot B to generate a new pattern spot D with the area of 2100m2. As shown in fig. 4, in this embodiment, after the fine patch adjustment is performed on the first fused data, the number of patches is reduced from 1070 to 594, and is reduced by 44.5%.
Step S3) of the present embodiment, adjustment of the elongated pattern spot is then performed:
obtaining all the image spots in the first fusion data, regarding each image spot, taking an image node of the image spot as a discrete point to generate a Thiessen polygon, taking the boundary of the Thiessen polygon as a central line of the image spot, calculating the ratio of the area of each image spot to the length of the central line to obtain a corresponding average thickness, screening the image spots with the average thickness smaller than a preset thickness threshold value, calculating the maximum inscribed rectangle of the screened image spots, and marking the image spots with the side length of the maximum inscribed rectangle smaller than the preset side length threshold value as long and thin image spots in the second unreasonable image spots. If the elongated pattern spot belongs to the second unreasonable pattern spot and does not belong to the first unreasonable pattern spot, the elongated pattern spot is a elongated pattern spot in a third unreasonable pattern spot, and the geometric adjustment of the third unreasonable pattern spot in the first fusion data includes adjustment of the elongated pattern spot, which specifically includes:
and for the long and thin image spots in the third unreasonable image spots, selecting a common node of the current long and thin image spot and other adjacent image spots, generating a perpendicular line from the common node to the center line of the current long and thin image spot, dividing the current long and thin image spot by using the perpendicular line to obtain divided image spots, calculating the length of a common edge of the current divided image spot and other adjacent image spots aiming at each divided image spot, and combining the current divided image spot and the adjacent image spot with the longest common edge to generate a new image spot.
As shown in FIG. 5, the first fused data includes four spots, and the area of the spot A is 2000m2Area of the pattern spot B is 1000m2Area of the pattern spot C is 1000m2And the area of the pattern spot D is 500m2. In this embodiment, the screening rule of the slender image spot is set to be 30m in average thickness and 20m in maximum inscribed rectangle side length; aiming at the four pattern spots, respectively taking the pattern node of each pattern spot as a discrete point to generate a corresponding Thiessen polygon, taking the boundary of the corresponding Thiessen polygon as the central line of each pattern spot, wherein the central line of each pattern spot is 40m, the central line of each pattern spot is 25m, the central line of each pattern spot is 80m, and the central line of each pattern spot is 100 m; calculating the average thickness by utilizing the ratio of the area to the length of the central line, wherein the average thickness of the pattern spot A is 50m, the average thickness of the pattern spot B is 40m, the average thickness of the pattern spot C is 12.5m, and the average thickness of the pattern spot D is 5 m; the pattern spots C and D with the average thickness less than 30m are used as the primary screened slender pattern spots; the image spots C and D generate the maximum inscribed rectangle, in this embodiment, the maximum inscribed rectangle is the maximum inscribed square, the side length of the maximum inscribed square of the image spot C is 30m, the side length of the maximum inscribed square of the image spot D is 5m, and the image spot D is a slender image spot. If the elongated spot is not the elongated spot in the basic data recorded in step S1), a common intersection point, i.e., a point, of the elongated spot D with its plurality of neighboring spots is selectedABDAnd pointBCD(ii) a Generating a perpendicular line to the central line of the elongated pattern spot D through the common intersection point of the adjacent multiple pattern spotsABDAnd the vertical lineBCD(ii) a According to the vertical lineABDAnd the vertical lineBCDSegmenting the elongated pattern spot; and merging the divided image spots into the adjacent image spots with the longest common edge to generate a new image spot E, a new image spot F and a new image spot G. As shown in fig. 6, in this embodiment, after the fine patch adjustment is further performed on the first fused data, the number of patches is reduced from 594 to 471, which is reduced by 20.7%.
Step S3) of the present embodiment, adjustment of the acute angle pattern spot is finally performed:
obtaining the inner angles of all the image spots in the first fusion data, marking the image spots corresponding to the inner angles smaller than a preset angle threshold value as sharp angle image spots in the second unreasonable image spots, if the sharp angle image spots belong to the second unreasonable image spots and do not belong to the first unreasonable image spots, then the sharp angle image spots in the third unreasonable image spots are marked, and the geometric adjustment of the third unreasonable image spots in the first fusion data comprises sharp angle image spot adjustment, which specifically comprises the following steps:
for the sharp corner pattern spots in the third unreasonable pattern spots, sequentially selecting common nodes of the current sharp corner pattern spot and the adjacent pattern spots according to the sequence of the distances from the sharp corner nodes of the current sharp corner pattern spot to the sharp corner nodes of the current sharp corner pattern spot from near to far, calculating the distance from the selected node in the current sharp-angle pattern spot to the perpendicular of the opposite side until the distance of the perpendicular meets the requirement, dividing the current sharp-angle pattern spot by using the perpendicular with the distance meeting the requirement to obtain a first pattern spot retaining the sharp angle and a second pattern spot without the sharp angle, selecting a common node of the first pattern spot and other adjacent pattern spots, generating the perpendicular from the common node to the center line of the first pattern spot, dividing the first pattern spot by using the perpendicular to obtain divided pattern spots, calculating the length of the common edge of the current divided pattern spot and other adjacent pattern spots for each divided pattern spot, and combining the current divided pattern spot and the adjacent pattern spot with the longest common edge to generate a new pattern spot.
As shown in FIG. 7, the first fusion data includes three patches, and the area of the patch A is 2000m2Area of the pattern spot B is 1000m2Area of the pattern spot C is 1000m2. In this embodiment, the acute and acute pattern spot is set to have an acute and acute angle within 10 degrees, and the pattern spot C is an acute angle pattern spot. If the acute angle pattern spot is not the acute angle pattern spot in the basic data recorded in the step S1), setting the length range of the acute angle division vertical line to be 1.5m to 10m, and selecting the node with the closest distance from the peak of the acute angleAGenerating a vertical lineAIf the length is 1m and is not in the set range, returning to the vertex of the acute angle, and selecting the second nearest nodeBGenerating a vertical lineBThe length is 2m, and the length is reserved within a set range; then according to the vertical lineABDThe sharp-angle pattern spot C is divided to obtain a pattern spot C2 containing a sharp angle and a pattern spot C1 not containing the sharp angle, aiming at the pattern spots A, B and C2, the pattern nodes of each pattern spot are respectively used as discrete points to generate corresponding Thiessen polygons, the boundaries of the corresponding Thiessen polygons are used as the central lines of each pattern spot, the common intersection points and the points of the pattern spots C2 and a plurality of adjacent pattern spots are selectedABC2And pointAC1C2(ii) a Generating a perpendicular to the centerline of spot C2 through the common intersection of adjacent multiple spotsABC2And the vertical lineAC1C2(ii) a According to the vertical lineABC2And the vertical lineAC1C2Segmenting the pattern spot C2; and merging the divided spots into adjacent spots with the longest common edge to generate new spots E and F, so that the original spot A, B, C is adjusted to obtain spot E, F, C1. As shown in fig. 8, in the present embodiment, the second fusion data is finally obtained after the first fusion data is further subjected to the sharp-angle speckle adjustment, and the number of speckles remains unchanged, but the segmentation is more reasonable.
In step S4) of this embodiment, considering that the graphic elements of different data in the natural resource multi-source vector data may be the same or different, different extraction methods are proposed for extracting the basic data attribute and the thematic data of each graphic element in the second fused data:
if the graphic elements of the basic data and the thematic data are the same, directly extracting the basic data attribute field and the thematic data attribute field of each graphic element according to the database structure and the data dictionary of the second fusion data; in the embodiment, the three-tone data of the state and the soil and the data of one picture of the forest resource management are both surface data, so the extraction mode is adopted;
if the graphic elements of the basic data and the thematic data are different, setting a scheme of attribute mapping transmission, selecting fields required by data fusion in the thematic data of each graphic element, and sets the project of attribute mapping transmission of the second fusion data and the thematic data to transmit the attributes, and screens the data which can accept the transmitted information through the ground class information of the thematic data and the buffer area with a certain threshold value in the process of transmitting the attributes, and data correction is carried out during data mapping and transmission, and a great amount of errors are caused by data deviation caused by image base maps, geographic information standards and artificial zoning errors and by directly utilizing a certain spatial characteristic value to carry out data transmission, therefore, when data mapping transmission is carried out, intervention is carried out in a manual mode, batch transmission is carried out, finally, mapped thematic data attribute fields are obtained, and meanwhile, corresponding basic data attribute fields are extracted.
And then, combining the basic data attributes and the thematic data attributes to obtain the fusion data attributes of the image spots, wherein the attribute field of the fusion data can establish a logical relationship between the basic data and the thematic data, establish a new fusion field and perform data mining. The attribute logic conversion includes single-field based logic conversion and multi-field based logic conversion. The specific flow is firstly setting logic conversion relation, and then processing attribute logic conversion.
If the basic data attributes and the thematic data attributes of all the graphic elements corresponding to the same image spot conflict with each other, the information of the same attribute in the basic data and the thematic data may be inconsistent, for example, the land type in the basic data is covered by vegetation, and the land type in the thematic data is not covered by vegetation, in order to modify the attribute of the fused data where the conflict occurs, in step S3), correct information is determined according to the texture information of the remote sensing image of the image spot region, and information where the error is obviously present in the attribute of the fused data is modified, as shown in fig. 9, the specific steps include:
s401) converting the remote sensing image of the second fusion data area into a gray-scale image, wherein a single wave band can be directly selected to be converted into the gray-scale image in the conversion process, and the three wave bands of red, green and blue of the true color remote sensing image can be converted into the gray-scale image after weighted calculation;
s402), compressing the gray levels of the gray level image, compressing the 256 gray levels into 8-level, 16-level and 32-level gray levels by using an interval combination mode under the condition of not influencing texture information embodied by the gray level image according to actual needs, and determining according to image texture characteristics;
s403), setting observation parameters, wherein the observation parameters comprise the size, the step pitch and the direction of a window;
s404) calculating a texture characteristic value of the center point of the window, and generating a texture characteristic matrix and a texture characteristic image along with the movement of the window;
s405) identifying the pattern spots with conflict between the basic data attribute and the thematic data attribute of the graphic element, and judging a correct result according to the texture feature matrix and the texture feature image;
s406) modifying the image spot conflict information in the fusion data attribute of the image spot according to the correct result.
In this embodiment, after the attribute field is directly extracted from the three tone data of homeland and forest resource management "one image" data, a logical relationship is set to perform logical attribute conversion, and for an image spot with a conflict in the attribute field, the texture information of the image in the image spot region is analyzed through the gray level co-occurrence matrix according to steps S401) to S406), whether the image is covered by vegetation is determined, and information that an error is obviously identified in the fusion data attribute of the image spot is modified.
In step S5) of this embodiment, a mode of merging similar attributes of the fusion result is adopted, so that on one hand, the integrity of the important attribute information of the basic data and the thematic data is preserved, and on the other hand, the graphics with high similarity of the important attributes of the thematic data information part in the fusion result are merged, thereby reducing the data size. The specific process comprises the following steps:
s501) setting a merged grouping field, selecting a unique identification code of a basic data information part and important attributes of a thematic data information part in second fused data as the merged grouping field, and selecting a spot number of homeland third-tone data of the second fused data and important attributes of a forest resource management 'one-picture' data part in the embodiment, wherein the important attributes comprise a land type, a forest type, a case right level, a protection level and the like as the merged grouping field;
s502) setting fields to be accumulated, mainly fields of numerical modes such as area, etc., and in this embodiment, setting a spot area, a small shift area, etc., as fields to be accumulated;
s503) merging the adjacent patches with high similarity of the fused data attributes to carry out data merging processing, wherein conditions are required to be set in the data merging process to limit generation of multi-component topology errors and the like.
As shown in fig. 10, adjacent patches with similar fusion data attributes in the second fusion data are merged to obtain third fusion data, and the number of the patches in the third fusion data is reduced from 471 to 462, which is reduced by 2%.
Step S6) of this embodiment specifically includes:
s601) carrying out topology check on the third fusion data according to a preset topology check rule, and identifying a topology error pattern spot in the third fusion data; in this embodiment, according to the data inspection in step S1), the third fused data is subjected to an open-air inspection (for example, erasing the fused data by using thematic data), a multi-component inspection, a surface gap inspection, a surface overlap inspection, a fine-particle pattern inspection, a sharp-angle pattern inspection, a slender pattern inspection, or the like, and the inspection result is identified;
s602) according to a preset logic check rule, performing logic check on the fusion data attribute of each pattern spot in the third fusion data, and identifying an attribute error pattern spot in the third fusion data; in the embodiment, whether the fusion data attribute of each pattern spot in the third fusion data accords with the logical relationship is manually checked, and the pattern spots which do not accord with the logical relationship are identified;
s603) respectively modifying the topology error pattern spots and the attribute error pattern spots in the third fusion data, in the embodiment, the topology error pattern spots are modified according to the data processing in the step S1) and the graph geometry adjustment on the unreasonable pattern spots in the step S3), a batch modification mode of manual intervention is adopted for the attribute error pattern spots, which of the relevant attribute fields with logic errors is reasonable is manually determined by self, the system automatically modifies other relevant attributes in batches, and the step 101) is returned until the pattern spots in the third fusion data meet the requirements.
The advantages of this embodiment are as follows:
1. and a complete and reasonable data preprocessing, graph geometric feature matching, graph geometric adjustment, attribute feature fusion, geometric feature and attribute feature linkage adjustment and fusion result quality inspection operation standard and system are formed preliminarily. Therefore, the problem of difference of geometric features and attribute features of multi-source vector data in the production process of the multi-source vector data of natural resources is solved, the difficulty of data fusion is reduced, the real operability is improved, the quality of the fused data is improved, and the goal that a fused result is used as a basic base map is achieved. The data is divided into basic data and thematic data, and the main characteristics of the basic data and the thematic data are reserved.
2. And a scientific and effective solution is provided for the situation of geometric inconsistency in the geometric fusion process of the graph. The method aims at the situation of geometric inconsistency, utilizes space superposition analysis to carry out geometric matching of the graphs, and provides an automatic processing method of fine-crushing pattern spots, slender pattern spots and sharp-angle pattern spots. And aiming at the situation of inconsistent attribute features, providing a processing flow of extracting attribute features of the same element, extracting attribute features of different elements, fusing the attribute features of the elements and judging inconsistent attribute patches based on a gray level co-occurrence matrix. And a mode of merging similar attributes of the fusion results is adopted to realize the linkage adjustment of the geometric features and the attribute features. Therefore, the aims of reducing interference of human factors, reducing the data production period and improving the efficiency are achieved.
Example two
This embodiment provides a natural resources multisource vector data fusion system, includes:
the preliminary fusion unit is used for selecting basic data and thematic data in the natural resource multi-source vector data, checking the rationality of the image spots of the basic data to obtain a first unreasonable image spot and recording the first unreasonable image spot, fitting the image boundary of the basic data and the thematic data, fixing the image node in the basic data, moving the corresponding image node in the thematic data, and fitting the image node of the basic data with the fitting distance smaller than a threshold value and the image node corresponding to the thematic data;
the geometric feature matching unit is used for performing graph geometric feature matching on the basic data after the graph boundary fitting and the thematic data to obtain first fusion data;
the geometric adjustment unit is used for carrying out image spot rationality check on the first fusion data to obtain a second unreasonable image spot, removing the first unreasonable image spot from the second unreasonable image spot to obtain a third unreasonable image spot, and carrying out image geometric adjustment on the third unreasonable image spot in the first fusion data to obtain second fusion data;
the attribute feature matching unit is used for extracting the basic data attribute and the thematic data attribute of each graphic element in the second fusion data, and the basic data attribute and the thematic data attribute of all the graphic elements corresponding to the same graphic spot are combined to obtain the fusion data attribute of the graphic spot;
and the linkage adjusting unit is used for combining adjacent image spots with similar fusion data attributes in the second fusion data to obtain third fusion data.
EXAMPLE III
The embodiment also provides a natural resource multi-source vector data fusion system, which includes a computer device programmed or configured to execute the natural resource multi-source vector data fusion method described in the first embodiment.
Example four
The present embodiment provides a computer-readable storage medium storing a computer program programmed or configured to execute the natural resource multi-source vector data fusion method according to the first embodiment.
The foregoing is considered as illustrative of the preferred embodiments of the invention and is not to be construed as limiting the invention in any way. Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention should fall within the protection scope of the technical scheme of the present invention, unless the technical spirit of the present invention departs from the content of the technical scheme of the present invention.

Claims (15)

1. A natural resource multi-source vector data fusion method is characterized by comprising the following steps:
s1) selecting basic data and thematic data in natural resource multi-source vector data, carrying out spot rationality check on the basic data to obtain a first unreasonable spot and recording the first unreasonable spot, fitting the graph boundary of the basic data and the thematic data, fixing the graph nodes in the basic data, moving the corresponding graph nodes in the thematic data, and fitting the graph nodes of the basic data with the fitting distance smaller than a threshold value and the graph nodes corresponding to the thematic data;
s2) carrying out graph geometric feature matching on the basic data after the graph boundary fitting and the thematic data to obtain first fusion data;
s3) carrying out image spot rationality check on the first fusion data to obtain a second unreasonable image spot, removing the first unreasonable image spot from the second unreasonable image spot to obtain a third unreasonable image spot, and carrying out image geometric adjustment on the third unreasonable image spot in the first fusion data to obtain second fusion data;
s4) extracting the basic data attribute and the thematic data attribute of each graphic element in the second fused data, and combining the basic data attribute and the thematic data attribute of all the graphic elements corresponding to the same graphic spot to obtain the fused data attribute of the graphic spot;
s5) merging the adjacent patches with similar fusion data attributes in the second fusion data to obtain third fusion data.
2. The natural resource multi-source vector data fusion method according to claim 1, wherein the step of checking the reasonableness of the map spots in the step S1) and the step S3) specifically comprises:
acquiring the areas of all the image spots in the inspection object, and marking the image spots with the areas smaller than a preset area threshold as fine-crushing image spots in the unreasonable image spots;
obtaining the internal angles of all the image spots in the inspection object, and marking the image spots corresponding to the internal angles smaller than a preset angle threshold value as sharp angle image spots in unreasonable image spots;
acquiring all the image spots in the inspection object, calculating the ratio of the area of each image spot to the length of the central line to obtain the corresponding average thickness, screening the image spots with the average thickness smaller than a preset thickness threshold, calculating the maximum inscribed rectangle of the screened image spots, and marking the image spots with the side length of the maximum inscribed rectangle smaller than the preset side length threshold as long and thin image spots in unreasonable image spots.
3. The natural resource multi-source vector data fusion method according to claim 1, wherein the performing of the geometric figure adjustment on the third unreasonable pattern spot in the first fusion data in step S3) includes fine-breaking pattern spot adjustment, and specifically includes:
and for the fine-crushing image spots in the third unreasonable image spots, calculating the length of the common edge of the current fine-crushing image spot and other adjacent image spots, and combining the current fine-crushing image spot and the adjacent image spot with the longest common edge to generate a new image spot.
4. The natural resource multi-source vector data fusion method according to claim 1, wherein the performing of the geometric adjustment on the third unreasonable pattern spot in the first fusion data in step S3) includes a slender pattern spot adjustment, and specifically includes:
and for the long and thin image spots in the third unreasonable image spots, selecting a common node of the current long and thin image spot and other adjacent image spots, generating a perpendicular line from the common node to the center line of the current long and thin image spot, dividing the current long and thin image spot by using the perpendicular line to obtain divided image spots, calculating the length of a common edge of the current divided image spot and other adjacent image spots aiming at each divided image spot, and combining the current divided image spot and the adjacent image spot with the longest common edge to generate a new image spot.
5. The natural resource multi-source vector data fusion method according to claim 1, wherein the performing of the geometric figure adjustment on the third unreasonable pattern spot in the first fusion data in step S3) includes acute-angled pattern spot adjustment, and specifically includes:
for the sharp corner pattern spots in the third unreasonable pattern spots, sequentially selecting common nodes of the current sharp corner pattern spot and the adjacent pattern spots according to the sequence of the distance from the sharp corner vertex of the current sharp corner pattern spot to the sharp corner vertex of the current sharp corner pattern spot from near to far, calculating the distance from the selected node in the current sharp-angle pattern spot to the perpendicular of the opposite side until the distance of the perpendicular meets the requirement, dividing the current sharp-angle pattern spot by using the perpendicular with the distance meeting the requirement to obtain a first pattern spot retaining the sharp angle and a second pattern spot without the sharp angle, selecting a common node of the first pattern spot and other adjacent pattern spots, generating the perpendicular from the common node to the center line of the first pattern spot, dividing the first pattern spot by using the perpendicular to obtain divided pattern spots, calculating the length of the common edge of the current divided pattern spot and other adjacent pattern spots for each divided pattern spot, and combining the current divided pattern spot and the adjacent pattern spot with the longest common edge to generate a new pattern spot.
6. The natural resource multi-source vector data fusion method according to any one of claims 1 to 5, wherein the steps S1) and S3) further include a step of obtaining a spot center line, and specifically include: and taking the graph nodes of the current image spot as discrete points to generate a Thiessen polygon, and taking the boundary of the Thiessen polygon as the central line of the current image spot.
7. The natural resource multi-source vector data fusion method of claim 1, wherein the extracting of the basic data attribute and the thematic data attribute of each graphic element in the second fusion data in step S4) comprises:
if the graphic elements of the basic data and the thematic data are the same, directly extracting the basic data attribute field and the thematic data attribute field of each graphic element according to the database structure and the data dictionary of the second fusion data;
if the graphic elements of the basic data and the thematic data are different, setting an attribute mapping transmission scheme, selecting fields required by data fusion in the thematic data of each graphic element, setting a scheme for attribute mapping transmission of the second fusion data and the thematic data to carry out attribute transmission, screening data capable of receiving transmission information through the ground class information and the buffer area of the thematic data in the attribute transmission process, carrying out data correction during data mapping transmission, finally obtaining the attribute fields of the mapped thematic data, and simultaneously extracting the corresponding basic data attribute fields.
8. The natural resource multi-source vector data fusion method according to claim 1, wherein the merging of the basic data attributes and the thematic data attributes of all the graphic elements corresponding to the same blob in step S4) specifically includes: and combining the basic data attributes and the thematic data attributes to obtain the fusion data attributes of the image spots, and modifying the fusion data attribute information according to the texture information of the remote sensing image of the image spot area if the basic data attributes and the thematic data attributes of all the graphic elements corresponding to the same image spot conflict.
9. The natural resource multi-source vector data fusion method according to claim 8, wherein the specific step of modifying the attribute information of the fusion data according to the texture information of the remote sensing image of the spot region comprises:
s401) converting the remote sensing image of the second fusion data area into a gray level image;
s402) compressing the gray level of the gray map;
s403), setting observation parameters, wherein the observation parameters comprise the size, the step pitch and the direction of a window;
s404) calculating a texture characteristic value of the center point of the window, and generating a texture characteristic matrix and a texture characteristic image along with the movement of the window;
s405) identifying the pattern spots with conflict between the basic data attribute and the thematic data attribute of the graphic element, and judging a correct result according to the texture feature matrix and the texture feature image;
s406) modifying the image spot conflict information in the fusion data attribute of the image spot according to the correct result.
10. The natural resource multi-source vector data fusion method according to claim 1, wherein before the step of fitting the graph boundary of the basic data and the thematic data in the step S1), the method further comprises the following steps:
and (3) open-air inspection: erasing basic data by using thematic data, if the erased data is empty, passing open-air inspection, and otherwise, prompting an error data range;
multi-component inspection: respectively counting the sum of single-side data in each record of the basic data and the thematic data, and prompting modification when the counted number exceeds 1, wherein the multiple parts are provided;
inspecting the surface gap: for vector data of adjacent surface data in the basic data and the thematic data, generating an inspection surface by using the outer boundary of the vector data, erasing the inspection surface by using surface data to be inspected in the vector data, if the erased data is empty, checking a surface gap, and otherwise, prompting a gap pattern spot between surfaces;
and (3) surface overlapping inspection: and extracting the overlapping part of the adjacent surface data in the basic data and the thematic data according to the spatial superposition analysis, and prompting the data of the overlapping part to be modified.
11. The natural resource multi-source vector data fusion method according to claim 1, wherein before the step of fitting the graph boundary of the basic data and the thematic data in the step S1), the method further comprises the following steps:
outdoor problem modification: completing the thematic data, and completing the thematic data of the adjacent administrative regions in the basic data range if the administrative regions are crossed;
multi-component problem modification: breaking up basic data or thematic data which are prompted to be modified, and converting multiple parts into single parts;
modification of the surface clearance: combining the gap pattern spots with the pattern spots with longer adjacent edges, thereby eliminating the gap pattern spots;
modification of the face overlap problem: determining an overlapped part to be eliminated according to the remote sensing image and eliminating the overlapped part;
data normalization: and carrying out standardization processing on the thematic data and the basic data according to a database structure, a data dictionary and data conversion logic.
12. The natural resource multi-source vector data fusion method according to claim 1, further comprising a step of quality inspection of fusion results after step S5), and specifically comprising:
s601) carrying out topology check on the third fusion data according to a preset topology check rule, and identifying a topology error pattern spot in the third fusion data;
s602) according to a preset logic check rule, performing logic check on the fusion data attribute of each pattern spot in the third fusion data, and identifying an attribute error pattern spot in the third fusion data;
s603) respectively modifying the topology error pattern spots and the attribute error pattern spots in the third fused data, and returning to the step S601) until the pattern spots in the third fused data meet the requirements.
13. A natural resource multi-source vector data fusion system is characterized by comprising:
the preliminary fusion unit is used for selecting basic data and thematic data in the natural resource multi-source vector data, checking the rationality of the image spots of the basic data to obtain a first unreasonable image spot and recording the first unreasonable image spot, fitting the image boundary of the basic data and the thematic data, fixing the image node in the basic data, moving the corresponding image node in the thematic data, and fitting the image node of the basic data with the fitting distance smaller than a threshold value and the image node corresponding to the thematic data;
the geometric feature matching unit is used for performing graph geometric feature matching on the basic data after the graph boundary fitting and the thematic data to obtain first fusion data;
the geometric adjustment unit is used for carrying out image spot rationality check on the first fusion data to obtain a second unreasonable image spot, removing the first unreasonable image spot from the second unreasonable image spot to obtain a third unreasonable image spot, and carrying out image geometric adjustment on the third unreasonable image spot in the first fusion data to obtain second fusion data;
the attribute feature matching unit is used for extracting the basic data attribute and the thematic data attribute of each graphic element in the second fusion data, and the basic data attribute and the thematic data attribute of all the graphic elements corresponding to the same graphic spot are combined to obtain the fusion data attribute of the graphic spot;
and the linkage adjusting unit is used for combining adjacent image spots with similar fusion data attributes in the second fusion data to obtain third fusion data.
14. A natural resource multi-source vector data fusion system comprising computer equipment programmed or configured to perform the natural resource multi-source vector data fusion method of any one of claims 1-12.
15. A computer-readable storage medium storing a computer program programmed or configured to perform the natural resource multi-source vector data fusion method of any one of claims 1 to 12.
CN202111034005.2A 2021-09-03 2021-09-03 Natural resource multi-source vector data fusion method Pending CN113822832A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111034005.2A CN113822832A (en) 2021-09-03 2021-09-03 Natural resource multi-source vector data fusion method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111034005.2A CN113822832A (en) 2021-09-03 2021-09-03 Natural resource multi-source vector data fusion method

Publications (1)

Publication Number Publication Date
CN113822832A true CN113822832A (en) 2021-12-21

Family

ID=78914081

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111034005.2A Pending CN113822832A (en) 2021-09-03 2021-09-03 Natural resource multi-source vector data fusion method

Country Status (1)

Country Link
CN (1) CN113822832A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114218337A (en) * 2022-02-22 2022-03-22 江苏省基础地理信息中心 Natural resource survey monitoring data identification and fusion updating method
CN114549685A (en) * 2022-04-28 2022-05-27 江苏省基础地理信息中心 Natural resource one-base map construction method based on data fusion
CN115311569A (en) * 2022-08-16 2022-11-08 自然资源部国土卫星遥感应用中心 Remote sensing image-based push-fill change detection method and device and terminal equipment
CN115359149A (en) * 2022-08-24 2022-11-18 中国测绘科学研究院 Continuous pattern spot simplification method under constraint of generalized topological structure
CN115828110A (en) * 2022-11-29 2023-03-21 中国国土勘测规划院 Water system spatial feature similarity detection method, equipment, storage medium and device
CN117056856A (en) * 2023-07-21 2023-11-14 广东省国土资源测绘院 Method and system for monitoring change pattern spot data by fusing multi-source natural resources

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101908215A (en) * 2010-07-13 2010-12-08 中国农业科学院农业资源与农业区划研究所 Spatial data fusing method
CN107845139A (en) * 2017-10-18 2018-03-27 中国测绘科学研究院 A kind of processing method of long and narrow figure spot divisural line
CN110176018A (en) * 2019-04-18 2019-08-27 中国测绘科学研究院 A kind of figure spot merging method keeping structuring atural object contour feature
KR102239562B1 (en) * 2019-12-10 2021-04-13 주식회사 스트리스 Fusion system between airborne and terrestrial observation data

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101908215A (en) * 2010-07-13 2010-12-08 中国农业科学院农业资源与农业区划研究所 Spatial data fusing method
CN107845139A (en) * 2017-10-18 2018-03-27 中国测绘科学研究院 A kind of processing method of long and narrow figure spot divisural line
CN110176018A (en) * 2019-04-18 2019-08-27 中国测绘科学研究院 A kind of figure spot merging method keeping structuring atural object contour feature
KR102239562B1 (en) * 2019-12-10 2021-04-13 주식회사 스트리스 Fusion system between airborne and terrestrial observation data

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
师静 等: "矢量数据融合中图形几何特征调整方法的探讨", 《中南林业调查规划》, 28 February 2022 (2022-02-28) *
曾明宇 等: "自然资源多源矢量数据融合方案的探讨", 《中南林业调查规划》, 30 November 2021 (2021-11-30) *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114218337A (en) * 2022-02-22 2022-03-22 江苏省基础地理信息中心 Natural resource survey monitoring data identification and fusion updating method
CN114549685A (en) * 2022-04-28 2022-05-27 江苏省基础地理信息中心 Natural resource one-base map construction method based on data fusion
CN114549685B (en) * 2022-04-28 2022-07-15 江苏省基础地理信息中心 Natural resource one-base map construction method based on data fusion
CN115311569A (en) * 2022-08-16 2022-11-08 自然资源部国土卫星遥感应用中心 Remote sensing image-based push-fill change detection method and device and terminal equipment
CN115311569B (en) * 2022-08-16 2023-08-22 自然资源部国土卫星遥感应用中心 Remote sensing image-based method and device for detecting change of push-fill soil and terminal equipment
CN115359149A (en) * 2022-08-24 2022-11-18 中国测绘科学研究院 Continuous pattern spot simplification method under constraint of generalized topological structure
CN115828110A (en) * 2022-11-29 2023-03-21 中国国土勘测规划院 Water system spatial feature similarity detection method, equipment, storage medium and device
CN117056856A (en) * 2023-07-21 2023-11-14 广东省国土资源测绘院 Method and system for monitoring change pattern spot data by fusing multi-source natural resources
CN117056856B (en) * 2023-07-21 2024-04-05 广东省国土资源测绘院 Method and system for monitoring change pattern spot data by fusing multi-source natural resources

Similar Documents

Publication Publication Date Title
CN113822832A (en) Natural resource multi-source vector data fusion method
CN107025685B (en) Airborne building roof point cloud modeling method under topology perception
CN106547853B (en) Forestry big data construction method based on one graph
KR20220000898A (en) Method to identify shoreline changes based on multi-factor
Chen et al. Topologically aware building rooftop reconstruction from airborne laser scanning point clouds
CN111598101B (en) Urban area intelligent extraction method, system and equipment based on remote sensing image scene segmentation
CN106548141B (en) A kind of object-oriented farmland information extraction method based on the triangulation network
CN109102193A (en) Geography designs ecological red line and delimit and management system and database, evaluation model
CN116310192A (en) Urban building three-dimensional model monomer reconstruction method based on point cloud
CN103703490A (en) Device for generating three-dimensional feature data, method for generating three-dimensional feature data, and recording medium on which program for generating three-dimensional feature data is recorded
CN110619258B (en) Road track checking method based on high-resolution remote sensing image
CN109766824B (en) Active and passive remote sensing data fusion classification method based on fuzzy evidence theory
CN109598056B (en) Method and system for measuring urban construction land form compactness and storage medium
CN113837892A (en) Slope farmland dividing method based on 3S technology
CN113724279B (en) System, method, equipment and storage medium for automatically dividing traffic cells into road networks
CN113033401B (en) Human activity change recognition and supervision method for ecological protection red line
CN114692236A (en) Big data-oriented territorial space planning base map base number processing method
CN114661744B (en) Terrain database updating method and system based on deep learning
Khayyal et al. Creation and spatial analysis of 3D city modeling based on GIS data
CN110688756B (en) Collection and drawing integrated completion drawing manufacturing method
CN115019163A (en) City factor identification method based on multi-source big data
CN115880427A (en) Three-dimensional construction method and device of natural environment
CN115270904B (en) Method and system for spatialization of proper-age permanent population in compulsory education stage
Li et al. Automatic extraction of urban built-up area based on object-oriented method and remote sensing data
Basaraner Model generalization in gIS

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination