Land surveying and mapping method and system for cadastral survey
Through drones, grid division and feature extraction, combined with clustering technology and flight altitude adjustment, the high cost and low efficiency problems of traditional land surveying and mapping methods are solved, and high-precision and high-efficiency land use type mapping are achieved.
Patent Information
- Application Number
- CN202411493684.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-24
- Publication Date
- 2025-02-07
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional land surveying and mapping survey methods are costly, time-consuming and difficult to obtain precise land use in remote or complex terrain areas.
UAVs are used for surveying and mapping, and through grid division and preset mapping routes, surveying and mapping images are obtained in real time, and land types are identified through feature extraction and clustering technology, and the drone's flight altitude is adjusted to improve surveying and mapping accuracy and efficiency.
High-precision mapping of land use types is achieved, the cost and time of mapping is reduced, and the mapping efficiency in remote or complex terrain areas is improved.
Smart Images
Figure CN119394271A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of land resource surveying and mapping, and in particular to a land surveying and mapping method and system for cadastral survey. Background Art
[0002] Land surveying and mapping refers to the use of computer technology, optoelectronic technology, network communication technology, space science, and information science as the basis, with global positioning system, remote sensing, and geographic information system as the technical core, to obtain graphics, land use types, and location information reflecting the current status of the ground through measurement methods based on existing characteristic points and boundaries on the ground for the planning, design, and administrative management of engineering construction.
[0004] In the surveying and investigation of land use types, spatial data is the core. Traditional surveying and investigation usually adopts manual investigation and field survey, which is not only costly, time-consuming and labor-intensive, but also difficult to obtain accurate land use conditions in some areas that cannot be directly observed, such as remote areas and complex terrain. Therefore, how to ensure the accuracy of land surveying and mapping data and speed up the efficiency of land resource investigation is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention
[0005] The purpose of the present invention is to provide a land surveying method and system for cadastral survey to solve the following technical problems:
[0006] In the surveying and investigation of land use types, spatial data is the core. Traditional surveying and investigation usually adopts manual investigation and field survey, which is not only costly, time-consuming and labor-intensive, but also difficult to obtain accurate land use conditions in some areas that cannot be directly observed, such as remote areas and complex terrain. Therefore, how to ensure the accuracy of land surveying and mapping data and speed up the efficiency of land resource investigation is a technical problem that technical personnel in this field urgently need to solve.
[0007] The purpose of the present invention can be achieved through the following technical solutions:
[0008] A land surveying method for cadastral survey, characterized in that it comprises the following steps:
[0009] S1, obtaining the area to be surveyed and dividing it into grids to generate a plurality of grid areas; the grid length is 0.6L, the width is 0.6L, L is the survey width of the surveying and mapping device carried by the UAV at a first height; the first height is the flight height of the UAV to meet the minimum target accuracy;
[0010] S2, defining the center point of the wide side of any grid area as a mapping point, taking the line connecting two mapping points of the same grid area as a preset mapping route of the UAV, controlling the UAV to start from the mapping point at a first altitude, flying along the mapping route and acquiring a mapping image in real time; preprocessing the mapping image, evenly dividing the preprocessed mapping image into a plurality of sub-mapping images, extracting features from all the sub-mapping images in turn to obtain color feature vectors and texture feature vectors, generating feature points corresponding to all the sub-mapping images with the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate, clustering all the sub-mapping images to generate a plurality of category clusters;
[0011] S3, selecting the largest category cluster from the plurality of category clusters and marking it as the target category cluster, obtaining the area S1 of the target category cluster, and if S1 is less than a preset threshold, lowering the flight altitude of the current UAV to a second altitude and flying according to a preset mapping route and obtaining a mapping image in real time, thereby obtaining a second mapping image;
[0012] If S1 is greater than or equal to the preset threshold, all areas except the area corresponding to the target category cluster are obtained, with the current mapping route forward direction as the length and the vertical mapping route forward direction as the width, and a minimum bounding rectangle covering all areas outside the area corresponding to the target category cluster is intercepted from the mapping image, the center point of the minimum bounding rectangle is obtained, and the minimum bounding rectangle area S2 is calculated and calibrated as the re-mapping area, and the flight height of the drone is adjusted according to the re-mapping area, and the current drone is controlled to go to the center point of the minimum bounding rectangle to obtain the mapping image, and a second mapping image is obtained; wherein the second height is the flight height corresponding to the mapping width of 0.6L;
[0013] S4, respectively processing the surveying images corresponding to different grid areas to generate the initial land surveying result of the area to be surveyed; and correcting the initial land surveying result according to the second surveying image to obtain the final land surveying result.
[0014] As a further solution of the present invention: in S2, when the drone completes the re-mapping of the target grid area, it automatically rises from the current flight altitude to the first altitude and goes to the next grid area according to the preset mapping route.
[0015] As a further solution of the present invention: In S2, the specific process of feature extraction is:
[0016] The mapping image is evenly divided into a number of grid areas to obtain a number of sub-mapping images, any of the sub-mapping images is converted into an HSV color space to obtain an HSV color matrix and perform a Haar transformation, and the matrix after the Haar transformation is normalized to obtain a 256-dimensional color feature vector S;
[0017] The sub-mapping image is subjected to grayscale processing to obtain a grayscale image, and texture features are extracted from the grayscale image to obtain roughness, contrast and orientation values of the grayscale image and normalize the image to obtain a texture feature vector N.
[0018] As a further solution of the present invention: in S2, the specific process of clustering is:
[0019] Calculate the Euclidean distance I between all feature points, generate the Euclidean distance set U, set the clustering control radius R, take any feature point as the center, calculate the feature point density P within the control radius R, obtain the mean feature point density within the control radius R of all feature points, and mark the mean feature point density as MinP; for any feature point, if P is greater than MinP, generate a category cluster with the feature point as the core point, and classify all adjacent feature points within the radius R into the category cluster to obtain several category clusters.
[0020] As a further solution of the present invention: if there is a non-core feature point whose density P within the control radius R is also greater than MinP, the category cluster generated by the non-core point is merged with the original category cluster to generate several category clusters.
[0021] As a further solution of the present invention: the process of setting the cluster control radius R is:
[0022] The sum of each Euclidean distance data in the set U is obtained to obtain u, and the control radius R is obtained based on the value u. The calculation formula is as follows:
[0023]
[0024] Among them, u is the sum of all Euclidean distance data values in set U.
[0025] As a further solution of the present invention: In S3, the specific process of adjusting the flight altitude is:
[0026] If the circumscribed rectangle S2 is greater than or equal to the preset threshold, the current UAV's flight altitude is lowered to a second altitude and the UAV is flown according to a preset mapping route and a mapping image is acquired in real time;
[0027] If the area S2 of the circumscribed rectangle is less than a preset threshold, the center point of the minimum circumscribed rectangle is obtained, and the center point is connected according to the position of the current UAV to obtain a reference route, and a plane vertical triangle is constructed with the reference route as the hypotenuse. The right-angled side of the plane vertical triangle is used as the surveying route, and the flight altitude of the current UAV is lowered to a second altitude and the UAV is flown according to the surveying route and the surveying image is obtained in real time to obtain a second surveying image.
[0028] A land surveying and mapping system for cadastral survey, characterized in that it comprises:
[0029] The data acquisition module is used to acquire the area to be surveyed and divide it into grids to generate a plurality of grid areas; the grid length is 0.6L, the width is 0.6L, and L is the survey width of the surveying and mapping device carried by the UAV at a first height; the first height is the flight height of the UAV to meet the minimum target accuracy;
[0030] The data analysis module is used to define the center point of the wide side of any grid area as a surveying point, use the line connecting two surveying points in the same grid area as a preset surveying route of the UAV, control the UAV to start from the surveying point at a first altitude, fly along the surveying route and obtain surveying images in real time; pre-process the surveying image, evenly divide the pre-processed surveying image into a number of grid areas to obtain a number of sub-surveying images, perform feature extraction on all the sub-surveying images in turn to obtain color feature vectors and texture feature vectors, use the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate to generate feature points corresponding to all the sub-surveying images, cluster all the sub-surveying images to generate a number of category clusters;
[0031] A data optimization module is used to select the largest category cluster from the plurality of category clusters and mark it as the target category cluster, obtain the area S1 of the target category cluster, and if S1 is less than a preset threshold, reduce the flight altitude of the current UAV to a second altitude and fly according to a preset mapping route and obtain a mapping image in real time to obtain a second mapping image; wherein the second altitude is the flight altitude corresponding to a mapping width of 0.6L;
[0032] If S1 is greater than or equal to the preset threshold, all areas except the area corresponding to the target category cluster are obtained, with the direction of the current mapping route as the length and the direction perpendicular to the mapping route as the width, a minimum bounding rectangle covering all areas outside the area corresponding to the target category cluster is intercepted from the mapping image, the center point of the minimum bounding rectangle is obtained, and the minimum bounding rectangle area S2 is calculated and marked as the re-mapping area, the flight altitude of the drone is adjusted according to the re-mapping area, and the current drone is controlled to go to the center point of the minimum bounding rectangle to obtain the mapping image, and the second mapping image is obtained;
[0033] The result generation module is used to process the surveying images corresponding to different grid areas respectively to obtain the initial land surveying results; and to correct the initial land surveying results according to the second surveying images to obtain the final land surveying results.
[0034] Beneficial effects of the present invention:
[0035] The present invention performs grid processing on the area to be surveyed and maps, and plans the surveying route based on the grid, obtains the surveying image in real time, and evenly divides the preprocessed surveying image into a number of grid areas by preprocessing the surveying image, so as to obtain a number of sub-surveying images, extracts features from all the sub-surveying images in turn, obtains color feature vectors and texture feature vectors, generates feature points corresponding to all the sub-surveying images with the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate, clusters all the sub-surveying images, and generates a number of category clusters. It can be understood that clustering can help identify abnormal images that are significantly different from other images, distinguish different land types, and select from the several category clusters. Take the largest category cluster and mark it as the target category cluster, obtain the area S1 of the target category cluster, if S1 is less than the preset threshold, lower the current UAV's flight altitude to the second altitude and fly according to the preset surveying route and obtain the surveying image in real time to obtain the second surveying image. It can be understood that for land surveying based on the captured images, the image resolution and coverage area are two important parameters. The higher the resolution, the richer the details that can be captured, but the corresponding coverage area will be reduced. Conversely, if a larger area needs to be covered, a certain resolution may need to be sacrificed. When the area S1 of the target category cluster is less than the preset threshold, it indicates that there are multiple land types in the current surveying image, and the coverage area of each type is small. In this case, a single height and viewing angle may not be able to fully capture the detailed features of various land types, so it is necessary to adjust the flight altitude of the drone; when the area S1 of the target category cluster is greater than or equal to the preset threshold, a minimum circumscribed rectangle covering all areas outside the corresponding area of the target category cluster is constructed, the center point of the minimum circumscribed rectangle is obtained, and the minimum circumscribed rectangle area S2 is calculated and calibrated as the re-survey area. If S2 is greater than or equal to the preset threshold, it means that the land type in the area is relatively single, and the remaining land types are scattered. If each land type area is re-surveyed separately, the actual surveying efficiency will be low. Therefore, the next grid area is surveyed along the current surveying route; if the circumscribed rectangle area M2 is small, it means that the remaining land types in the current area are more concentrated, then the current drone's flight altitude is lowered to the second altitude to obtain the surveying image in real time, thereby achieving high-precision surveying of the land type while improving the surveying efficiency. By processing the surveying images of different grid areas, the final land surveying result is obtained. The present invention greatly improves the efficiency and accuracy of land surveying. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] The present invention will be further described below in conjunction with the accompanying drawings.
[0037] Figure 1 The invention discloses a flow chart of a land surveying and mapping method for cadastral survey. DETAILED DESCRIPTION
[0038] 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.
[0039] See also Figure 1 As shown, the present invention is a land surveying method for cadastral survey, comprising the following steps:
[0040] S1, obtaining the area to be surveyed and dividing it into grids to generate a plurality of grid areas; the grid length is 0.6L, the width is 0.6L, L is the survey width of the surveying and mapping device carried by the UAV at a first height; the first height is the flight height of the UAV to meet the minimum target accuracy;
[0041] S2, defining the center point of the wide side of any grid area as a mapping point, taking the line connecting two mapping points of the same grid area as a preset mapping route of the UAV, controlling the UAV to start from the mapping point at a first altitude, flying along the mapping route and acquiring a mapping image in real time; preprocessing the mapping image, evenly dividing the preprocessed mapping image into a plurality of sub-mapping images, extracting features from all the sub-mapping images in turn to obtain color feature vectors and texture feature vectors, generating feature points corresponding to all the sub-mapping images with the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate, clustering all the sub-mapping images to generate a plurality of category clusters;
[0042] S3, selecting the largest category cluster from the plurality of category clusters and marking it as the target category cluster, obtaining the area S1 of the target category cluster, and if S1 is less than a preset threshold, lowering the flight altitude of the current UAV to a second altitude and flying according to a preset mapping route and obtaining a mapping image in real time, thereby obtaining a second mapping image;
[0043] If S1 is greater than or equal to the preset threshold, all areas except the area corresponding to the target category cluster are obtained, with the current mapping route forward direction as the length and the vertical mapping route forward direction as the width, and a minimum bounding rectangle covering all areas outside the area corresponding to the target category cluster is intercepted from the mapping image, the center point of the minimum bounding rectangle is obtained, and the minimum bounding rectangle area S2 is calculated and calibrated as the re-mapping area, and the flight height of the drone is adjusted according to the re-mapping area, and the current drone is controlled to go to the center point of the minimum bounding rectangle to obtain the mapping image, and a second mapping image is obtained; wherein the second height is the flight height corresponding to the mapping width of 0.6L;
[0044] S4, respectively processing the surveying images corresponding to different grid areas to generate the initial land surveying result of the area to be surveyed; and correcting the initial land surveying result according to the second surveying image to obtain the final land surveying result.
[0045] The present invention performs grid processing on the area to be surveyed and maps, and plans the surveying route based on the grid, obtains the surveying image in real time, and evenly divides the preprocessed surveying image into a number of grid areas by preprocessing the surveying image, so as to obtain a number of sub-surveying images, extracts features from all the sub-surveying images in turn, obtains color feature vectors and texture feature vectors, generates feature points corresponding to all the sub-surveying images with the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate, clusters all the sub-surveying images, and generates a number of category clusters. It can be understood that clustering can help identify abnormal images that are significantly different from other images, distinguish different land types, and select from the several category clusters. Take the largest category cluster and mark it as the target category cluster, obtain the area S1 of the target category cluster, if S1 is less than the preset threshold, lower the current UAV's flight altitude to the second altitude and fly according to the preset surveying route and obtain the surveying image in real time to obtain the second surveying image. It can be understood that for land surveying based on the captured images, the image resolution and coverage area are two important parameters. The higher the resolution, the richer the details that can be captured, but the corresponding coverage area will be reduced. Conversely, if a larger area needs to be covered, a certain resolution may need to be sacrificed. When the area S1 of the target category cluster is less than the preset threshold, it indicates that there are multiple land types in the current surveying image, and the coverage area of each type is small. In this case, a single height and viewing angle may not be able to fully capture the detailed features of various land types, so it is necessary to adjust the flight altitude of the drone; when the area S1 of the target category cluster is greater than or equal to the preset threshold, the minimum circumscribed rectangle covering all areas outside the corresponding area of the target category cluster is constructed and the minimum circumscribed rectangle area S2 is calculated and calibrated as the re-survey area. If S2 is greater than or equal to the preset threshold, it means that the land type in the area is relatively single, and the remaining land types are scattered. If each land type area is re-surveyed separately, the actual surveying efficiency will be low. Therefore, the next grid area is surveyed along the current surveying route; if the circumscribed rectangle area M2 is small, it means that the remaining land types in the current area are more concentrated, then the current drone's flight altitude is lowered to the second altitude to obtain surveying images in real time, thereby achieving high-precision surveying of land use types while improving surveying efficiency. By processing surveying images of different grid areas, the final land surveying result is obtained. The present invention greatly improves the efficiency and accuracy of land surveying.
[0046] In a preferred implementation of the present invention, in S2, when the drone completes the re-mapping of the target grid area, it automatically rises from the current flight altitude to the first altitude and goes to the next grid area according to the preset mapping route.
[0047] In another preferred implementation of the present invention, in S2, the specific process of feature extraction is:
[0048] The mapping image is evenly divided into a number of grid areas to obtain a number of sub-mapping images, any of the sub-mapping images is converted into an HSV color space to obtain an HSV color matrix and perform a Haar transformation, and the matrix after the Haar transformation is normalized to obtain a 256-dimensional color feature vector S;
[0049] The sub-mapping image is subjected to grayscale processing to obtain a grayscale image, and texture features are extracted from the grayscale image to obtain roughness, contrast and orientation values of the grayscale image and normalize the image to obtain a texture feature vector N.
[0050] In another preferred implementation of the present invention, in S2, the specific process of clustering is:
[0051] Calculate the Euclidean distance I between all feature points, generate the Euclidean distance set U, set the clustering control radius R, take any feature point as the center, calculate the feature point density P within the control radius R, obtain the mean feature point density within the control radius R of all feature points, and mark the mean feature point density as MinP; for any feature point, if P is greater than MinP, generate a category cluster with the feature point as the core point, and classify all adjacent feature points within the radius R into the category cluster to obtain several category clusters.
[0052] In another preferred implementation of the present invention, if there is a non-core feature point whose density P within the control radius R is also greater than MinP, the category cluster generated by the non-core point is merged with the original category cluster to generate several category clusters.
[0053] In another preferred implementation of the present invention, the process of setting the cluster control radius R is:
[0054] The sum of each Euclidean distance data in the set U is obtained to obtain u, and the control radius R is obtained based on the value u. The calculation formula is as follows:
[0055]
[0056] Among them, u is the sum of all Euclidean distance data values in set U.
[0057] In another preferred implementation of the present invention, in S3, the specific process of adjusting the flight altitude is:
[0058] If the circumscribed rectangle S2 is greater than or equal to the preset threshold, the current UAV's flight altitude is lowered to a second altitude and the UAV is flown according to a preset mapping route and a mapping image is acquired in real time;
[0059] If the area S2 of the circumscribed rectangle is less than a preset threshold, the center point of the minimum circumscribed rectangle is obtained, and the center point is connected according to the position of the current UAV to obtain a reference route, and a plane vertical triangle is constructed with the reference route as the hypotenuse. The right-angled side of the plane vertical triangle is used as the surveying route, and the flight altitude of the current UAV is lowered to a second altitude and the UAV is flown according to the surveying route and the surveying image is obtained in real time to obtain a second surveying image.
[0060] A land surveying and mapping system for cadastral survey, comprising:
[0061] The data acquisition module is used to acquire the area to be surveyed and divide it into grids to generate a plurality of grid areas; the grid length is 0.6L, the width is 0.6L, and L is the survey width of the surveying and mapping device carried by the UAV at a first height; the first height is the flight height of the UAV to meet the minimum target accuracy;
[0062] The data analysis module is used to define the center point of the wide side of any grid area as a surveying point, use the line connecting two surveying points in the same grid area as a preset surveying route of the UAV, control the UAV to start from the surveying point at a first altitude, fly along the surveying route and obtain surveying images in real time; pre-process the surveying image, evenly divide the pre-processed surveying image into a number of grid areas to obtain a number of sub-surveying images, perform feature extraction on all the sub-surveying images in turn to obtain color feature vectors and texture feature vectors, use the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate to generate feature points corresponding to all the sub-surveying images, cluster all the sub-surveying images to generate a number of category clusters;
[0063] A data optimization module is used to select the largest category cluster from the plurality of category clusters and mark it as the target category cluster, obtain the area S1 of the target category cluster, and if S1 is less than a preset threshold, reduce the flight altitude of the current UAV to a second altitude and fly according to a preset mapping route and obtain a mapping image in real time to obtain a second mapping image; wherein the second altitude is the flight altitude corresponding to a mapping width of 0.6L;
[0064] If S1 is greater than or equal to the preset threshold, all areas except the area corresponding to the target category cluster are obtained, with the direction of the current mapping route as the length and the direction perpendicular to the mapping route as the width, a minimum bounding rectangle covering all areas outside the area corresponding to the target category cluster is intercepted from the mapping image, the center point of the minimum bounding rectangle is obtained, and the minimum bounding rectangle area S2 is calculated and marked as the re-mapping area, the flight altitude of the drone is adjusted according to the re-mapping area, and the current drone is controlled to go to the center point of the minimum bounding rectangle to obtain the mapping image, and the second mapping image is obtained;
[0065] The result generation module is used to process the surveying images corresponding to different grid areas respectively to obtain the initial land surveying results; and to correct the initial land surveying results according to the second surveying images to obtain the final land surveying results.
[0066] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. A land surveying method for cadastral survey, characterized in that: The following steps are involved: S1, obtaining the area to be surveyed and dividing it into grids to generate a plurality of grid areas; the grid length is 0.6L, the width is 0.6L, L is the survey width of the surveying and mapping device carried by the UAV at a first height; the first height is the flight height of the UAV to meet the minimum target accuracy; S2, defining the center point of the wide side of any grid area as a mapping point, taking the line connecting two mapping points of the same grid area as a preset mapping route of the UAV, controlling the UAV to start from the mapping point at a first altitude, flying along the mapping route and acquiring a mapping image in real time; preprocessing the mapping image, evenly dividing the preprocessed mapping image into a plurality of sub-mapping images, extracting features from all the sub-mapping images in turn to obtain color feature vectors and texture feature vectors, generating feature points corresponding to all the sub-mapping images with the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate, clustering all the sub-mapping images to generate a plurality of category clusters; S3, selecting the largest category cluster from the plurality of category clusters and marking it as the target category cluster, obtaining the area S1 of the target category cluster, and if S1 is less than a preset threshold, lowering the flight altitude of the current UAV to a second altitude and flying according to a preset mapping route and obtaining a mapping image in real time, thereby obtaining a second mapping image; If S1 is greater than or equal to the preset threshold, all areas except the area corresponding to the target category cluster are obtained, with the current mapping route forward direction as the length and the vertical mapping route forward direction as the width. A minimum bounding rectangle covering all areas outside the area corresponding to the target category cluster is intercepted from the mapping image, the center point of the minimum bounding rectangle is obtained, and the minimum bounding rectangle area S2 is calculated and calibrated as the re-mapping area. The flight height of the drone is adjusted according to the re-mapping area to obtain a second mapping image; wherein the second height is the flight height corresponding to the drone when the mapping width is 0.6L; S4, respectively processing the surveying images corresponding to different grid areas to generate the initial land surveying result of the area to be surveyed; and correcting the initial land surveying result according to the second surveying image to obtain the final land surveying result.
2. A land surveying method for cadastral survey according to claim 1, characterized in that: In S2, the specific process of feature extraction is: The mapping image is evenly divided into a number of grid areas to obtain a number of sub-mapping images, any of the sub-mapping images is converted into an HSV color space to obtain an HSV color matrix and perform a Haar transformation, and the matrix after the Haar transformation is normalized to obtain a 256-dimensional color feature vector M; The sub-mapping image is subjected to grayscale processing to obtain a grayscale image, and texture features are extracted from the grayscale image to obtain roughness, contrast and orientation values of the grayscale image and normalize the image to obtain a texture feature vector N.
3. A land surveying method for cadastral survey according to claim 1, characterized in that: In S2, the specific process of clustering is: Calculate the Euclidean distance I between all feature points, generate the Euclidean distance set U, set the clustering control radius R, take any feature point as the center, calculate the feature point density P within the control radius R, obtain the mean feature point density within the control radius R of all feature points, and mark the mean feature point density as MinP; for any feature point, if P is greater than MinP, generate a category cluster with the feature point as the core point, and classify all adjacent feature points within the radius R into the category cluster to obtain several category clusters.
4. A land surveying method for cadastral survey according to claim 3, characterized in that: It also includes merging the category cluster generated by the non-core point with the original category cluster to generate several category clusters if the density P within the control radius R of the non-core feature point is also greater than MinP.
5. A land surveying method for cadastral survey according to claim 3, characterized in that: The process of setting the cluster control radius R is: The sum of each Euclidean distance data in the set U is obtained to obtain u, and the control radius R is obtained based on the value u. The calculation formula is as follows: Among them, u is the sum of all Euclidean distance data values in set U.
6. A land surveying method for cadastral survey according to claim 1, characterized in that: In S3, the specific process of adjusting the flight altitude is as follows: If the circumscribed rectangle S2 is greater than or equal to a preset threshold, the UAV is controlled to survey the next grid area along the current surveying route; if the area S2 of the circumscribed rectangle is less than the preset threshold, the center point of the minimum circumscribed rectangle is obtained, and the reference route is obtained by connecting it with the center point according to the current position of the UAV. A plane vertical triangle is constructed with the reference route as the hypotenuse, and the right-angled side of the plane vertical triangle is used as the surveying route. The flight altitude of the current UAV is lowered to a second altitude, and the UAV flies according to the surveying route and obtains the surveying image in real time to obtain a second surveying image.
7. A land surveying method for cadastral survey according to claim 1, characterized in that: In S3, when the drone completes the re-mapping, it automatically rises from the current flight altitude to the first altitude and goes to the next grid area according to the preset mapping route.
8. A land surveying and mapping system for cadastral survey, characterized in that: include: The data acquisition module is used to acquire the area to be surveyed and divide it into grids to generate a plurality of grid areas; the grid length is 0.6L, the width is 0.6L, and L is the survey width of the surveying and mapping device carried by the UAV at a first height; the first height is the flight height of the UAV to meet the minimum target accuracy; The data analysis module is used to define the center point of the wide side of any grid area as a surveying point, use the line connecting two surveying points in the same grid area as a preset surveying route of the UAV, control the UAV to start from the surveying point at a first altitude, fly along the surveying route and obtain surveying images in real time; pre-process the surveying image, evenly divide the pre-processed surveying image into a number of grid areas to obtain a number of sub-surveying images, perform feature extraction on all the sub-surveying images in turn to obtain color feature vectors and texture feature vectors, use the color feature vector as the horizontal coordinate and the texture feature vector as the vertical coordinate to generate feature points corresponding to all the sub-surveying images, cluster all the sub-surveying images to generate a number of category clusters; A data optimization module is used to select the largest category cluster from the plurality of category clusters and mark it as the target category cluster, obtain the area S1 of the target category cluster, and if S1 is less than a preset threshold, reduce the flight altitude of the current UAV to a second altitude and fly according to a preset mapping route and obtain a mapping image in real time to obtain a second mapping image; wherein the second altitude is the flight altitude corresponding to a mapping width of 0.6L; If S1 is greater than or equal to the preset threshold, all areas except the area corresponding to the target category cluster are obtained, with the direction of the current mapping route as the length and the direction perpendicular to the mapping route as the width, a minimum bounding rectangle covering all areas outside the area corresponding to the target category cluster is intercepted from the mapping image, the center point of the minimum bounding rectangle is obtained, and the minimum bounding rectangle area S2 is calculated and marked as the re-mapping area, the flight altitude of the drone is adjusted according to the re-mapping area, and the current drone is controlled to go to the center point of the minimum bounding rectangle to obtain the mapping image, and the second mapping image is obtained; The result generation module is used to process the surveying images corresponding to different grid areas respectively to obtain the initial land surveying results; and to correct the initial land surveying results according to the second surveying images to obtain the final land surveying results.