Method and system for collecting and plotting land survey data
The drone image data and deep learning model process the spot block data, and the neural network model is constructed based on groundwater distribution coefficient and settlement parameters, which solves the problems of low efficiency and poor accuracy of traditional field mapping, and achieves efficient and accurate three-dimensional mapping.
Patent Information
- Application Number
- CN202510504279.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-21
- Publication Date
- 2025-08-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional field mapping operation methods are inefficient and have poor accuracy, making it difficult to consider three-dimensional factors, resulting in high work intensity and high cost.
UAV image data acquisition and deep learning model are used to process spot block data through graph segmentation line determination method and machine learning algorithm, and build a neural network model with groundwater distribution coefficient and settlement parameters to realize intelligent tuning and drawing of three-dimensional data.
It improves the accuracy and efficiency of field mapping, reduces the work intensity of mapping personnel, reduces the dependence on business literacy, and reduces costs.
Smart Images
Figure CN120403564A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of land surveying and mapping, and particularly relates to a method and system for collecting and mapping land survey data. Background Art
[0002] Engineering geological mapping is to conduct instrument-based measurement on key surface geology sites on the basis of investigation, and put forward corresponding maps and reports. There are two operation modes, namely indoor work (operation in an indoor environment) and outdoor work (operation in an outdoor environment).
[0003] At present, it mainly relies on on-site geological mapping work outdoors, but the efficiency is low, and it is easy to cause one-sided understanding and judgment errors under complex topographic and geological conditions. The indoor work mainly focuses on route planning, and does not make full use of existing data to carry out analysis to reduce the work intensity outdoors, resulting in repetition in outdoor work, thus increasing the work intensity of outdoor workers.
[0004] Traditional outdoor mapping work uses paper maps as working maps. Before outdoor work, sufficient indoor preparation work needs to be done, and then various digital mapping products prepared are printed into maps for outdoor workers to use; during outdoor work, data information needs to be investigated and drawn, and hand-drawn results and result tracing need to be represented by symbols stipulated by the state. The accuracy of outdoor mapping is the key to achieving the purpose of outdoor investigation. However, the traditional outdoor mapping operation mode is difficult to ensure the accuracy of the investigation results, and the labor intensity of pure manual operation is very high, the work efficiency is very low, and the professional quality requirements for operators are also high. The operation cost is very high, and the operators cannot consider three-dimensional factors for outdoor mapping. There is an urgent need for a big data intelligent mapping key generation model that can conduct mapping based on past outdoor mapping experience and consider three-dimensional data to assist mapping personnel in screening key change areas for mapping work.
[0005] Therefore, the prior art has the following problems:
[0006] The traditional outdoor mapping operation mode can no longer meet the current new requirements. In the prior art, although there are also software for editing electronic maps, it can only correct the shape of patches on the electronic map, but the position and size of patch correction need to be determined through paper materials, and the data considered is not from the three-dimensional aspect, which greatly affects the work efficiency and mapping accuracy of outdoor mapping, resulting in a greater work intensity for mappers. Summary of the Invention
[0007] To solve the above technical problems, the present invention proposes a method and system for collecting and mapping land survey data.
[0008] In the first aspect of the present invention, a method for collecting and mapping land survey data is provided, characterized in that the method includes:
[0009] By processing the first UAV image data collected through on-site inspection of the UAV, obtaining the processed first underground geological correction parameters and the first geological surveying and mapping data for feature processing, and obtaining the first collection and mapping feature;
[0010] Construct a land mapping method generation model according to the first collection and mapping feature and the first collection and mapping method set by the mapping personnel;
[0011] By processing the second UAV image data collected through on-site inspection of the UAV, obtaining the processed second underground geological correction parameters and the second geological surveying and mapping data for feature processing, and obtaining the second collection and mapping feature;
[0012] Input the second collection and mapping feature into the land mapping method generation model to obtain the second collection and mapping method of the second geological surveying and mapping data, and assist the mapping personnel to conduct land data mapping according to the second collection and mapping method.
[0013] Further, the first UAV image data or the second UAV image data is processed by a graphic segmentation line determination method to obtain patch segmentation data.
[0014] Further, the patch segmentation data includes control point coordinates, boundary point coordinates, and patch area, and both the control point coordinates and the boundary point coordinates are on the graphic segmentation line.
[0015] Further, the graphic segmentation line determination method uses a machine learning algorithm to perform segmentation line determination classification, and the machine learning algorithm is trained using the gray value difference in the first UAV image data or the second UAV image data.
[0016] Further, the first underground geological correction parameter or the second underground geological correction parameter is obtained by processing the groundwater distribution coefficient and the settlement parameter of the mapped land.
[0017] Further, the groundwater distribution coefficient is calculated by the number of main groundwater veins, the number of tributaries, and the water flow velocity.
[0018] Further, the patch segmentation data, the underground geological correction parameter, and the geological surveying and mapping data are horizontally spliced by features to obtain the collection and mapping feature.
[0019] Further, the activation function of the land mapping method generation model adopts a neural network model activation function improved based on the groundwater distribution coefficient.
[0020] A collection and mapping system for land survey data is also provided, including a UAV image data acquisition and processing module, an underground geological correction parameter acquisition module, an original geological survey data calling module, a land mapping method generation model construction module, and an assisting mapping module, characterized in that:
[0021] The UAV image data acquisition and processing module: is used to acquire first UAV image data, and is also used to acquire second UAV image data, and processes the first UAV image data or the second UAV image data through a graphic segmentation line determination method to obtain patch segmentation data;
[0022] The underground geological correction parameter acquisition module: acquires first underground geological correction parameters, and is also used to acquire second underground geological correction parameters;
[0023] The original geological survey data calling module: calls the original geological survey data of the land to be mapped by connecting to a server with survey data, including first geological survey data and second geological survey data;
[0024] The land mapping method generation model construction module: horizontally splices the patch segmentation data, underground geological correction parameters, and geological survey data at the feature level to obtain collection and mapping features, and constructs a land mapping method generation model according to the first collection and mapping features and the first collection and mapping method set by the mapping personnel;
[0025] The assisting mapping module: receives the second UAV image data collected by the UAV field investigation, acquires and processes the second underground geological correction parameters and the second geological survey data for feature processing to obtain second collection and mapping features, inputs the second collection and mapping features into the land mapping method generation model to generate a second collection and mapping method, and assists the mapping personnel in mapping land data according to the second collection and mapping method.
[0026] Further, the first UAV image data or the second UAV image data is processed through a graphic segmentation line determination method to obtain patch segmentation data;
[0027] The graphic segmentation line determination method uses a machine learning algorithm to perform segmentation line determination classification, and the machine learning algorithm is trained using the gray value differences in the first UAV image data or the second UAV image data.
[0028] The present invention conducts field survey mapping considering three-dimensional factors, trains through a deep learning model using past field mapping experience, and uses a big data intelligent mapping key generation model that considers three-dimensional data to assist surveyors in screening key change areas for mapping work. It can correct the shape of patches on the electronic map more precisely, and the position and size of patch correction are automatically determined through electronic devices instead of paper materials. The present invention constructs a neural network model for mapping based on the mapping correction coefficient generated by calling groundwater system data and settlement data and land image data, enhancing the accuracy for different land mapping and reducing the workload of personnel.
[0029] More embodiments and improvement effects of the present invention will be further introduced in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 is a flowchart of a method for collecting and mapping land survey data of the present invention;
[0031] Figure 2 is a schematic diagram of a system for collecting and mapping land survey data of the present invention;
[0032] Figure 3 is a schematic diagram of an on-site inspection image of an unmanned aerial vehicle in the present invention;
[0033] Figure 4 is a schematic diagram of a patch in the present invention;
[0034] Figure 5 is a schematic diagram of the principle of the neural network model in the present invention;
[0035] Figure 6 is a schematic diagram of a mapping method generated by the neural network model in the present invention;
[0036] Figure 7 is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0037] Next, in combination with the accompanying drawings and specific embodiments, a further description of the invention will be made.
[0038] In the first aspect of the present invention, a method and system for collecting and mapping land survey data are provided. The original land surveying and mapping data in the present invention is retrieved from units such as the government.
[0039] In the first aspect of the present invention, a method for collecting and mapping land survey data is provided, characterized in that the method includes:
[0040] By performing feature processing on the first underground geological correction parameters and the first geological surveying data obtained from the first drone image data collected through on-site inspection of the drone, the first collection and mapping feature is obtained;
[0041] According to the first collection and mapping feature and the first collection and mapping method set by the mapping personnel, a land mapping method generation model is constructed;
[0042] By performing feature processing on the second underground geological correction parameters and the second geological surveying data obtained from the second drone image data collected through on-site inspection of the drone, the second collection and mapping feature is obtained;
[0043] The second collection and mapping feature is input into the land mapping method generation model to obtain the second collection and mapping method for the second geological surveying data, and the mapping personnel are assisted in mapping the land data according to the second collection and mapping method.
[0044] Furthermore, the first drone image data or the second drone image data is processed by a graphic segmentation line determination method to obtain patch segmentation data.
[0045] In this embodiment, the coordinate refers to a Cartesian coordinate system with the lower left corner of the surveying and mapping map as the origin, and the scale unit is set by those skilled in the art according to needs.
[0046] Furthermore, the patch segmentation data includes control point coordinates, boundary point coordinates, and patch area, and both the control point coordinates and the boundary point coordinates are on the graphic segmentation line.
[0047] In this embodiment, the control point coordinates are important coordinate points for patch demarcation, and the boundary point coordinates are important turning points in patch demarcation, which are set by technicians according to experience.
[0048] Since the gray value at the segmentation line is different from that at other positions of the image, the segmentation line is determined by using the method of combining coordinates and gray value determination.
[0049] Furthermore, the graphic segmentation line determination method uses a machine learning algorithm to classify the determination of the segmentation line. The machine learning algorithm is trained using the gray value difference in the first drone image data or the second drone image data, and the calculation formula of the machine learning algorithm is:
[0050] F(X) = W T X
[0051] X is the gray value feature of the coordinate points of the drone image data, F(X) is the determined parameter of the segmentation point output, and W Tis the normal vector perpendicular to the hyperplane. When the value of F(X) is greater than 0, the coordinate point of the input UAV image data belongs to a point on the graphic segmentation line.
[0052] After determining the segmentation line in this embodiment, the patch segmentation data includes the determination of control point coordinates, boundary point coordinates, and patch area, which will not be elaborated here.
[0053] Further, the first underground geological correction parameter or the second underground geological correction parameter is obtained by processing the groundwater distribution coefficient and the settlement parameter of the surveyed land, and its calculation formula is;
[0054]
[0055] In the formula, S G is the first underground geological correction parameter or the second underground geological correction parameter, W d is the groundwater distribution coefficient, S y is the annual settlement distance of the surveyed land, in micrometers, n k is the type of surface element attributes such as roads, water systems, buildings, and vegetation, and m represents the number of times of annotating surface element attributes.
[0056] During the surveying and mapping, underground geology can help with surveying and mapping corrections, but the use of such data has not yet emerged, and the groundwater system distribution has a significant impact on subsequent land data collection and corresponding surveying and mapping.
[0057] Further, the groundwater distribution coefficient is calculated from the number of main groundwater veins, the number of tributaries, and the water flow velocity:
[0058]
[0059] In the formula, W d is the groundwater distribution coefficient, U m is the number of main groundwater veins, U t is the number of tributaries of the groundwater vein, and v represents the groundwater flow velocity.
[0060] Further, the patch segmentation data, the underground geological correction parameter, and the geological surveying and mapping data are horizontally spliced at the feature level to obtain the acquisition and surveying and mapping features. In this embodiment, horizontal splicing refers to horizontally linking the features of the above three to form a feature value, which is a common means for technicians to process features and will not be elaborated here.
[0061] Further, the activation function calculation formula of the neural network model improved based on the groundwater distribution coefficient is adopted for the land surveying and mapping method generation model as follows:
[0062]
[0063] where δ(X) is the activation function value, and W d is the groundwater distribution coefficient, and X is obtained by weighted addition of the bias of the first acquisition and mapping feature or the second acquisition and mapping feature through neurons.
[0064] A collection and mapping system for land survey data is also provided, including a UAV image data acquisition and processing module, an underground geological correction parameter acquisition module, an original geological mapping data calling module, a land mapping method generation model construction module, and an assisting mapping module, characterized in that:
[0065] The UAV image data acquisition and processing module: is used to acquire first UAV image data, and is also used to acquire second UAV image data, and processes the first UAV image data or the second UAV image data through a graphic segmentation line determination method to obtain patch segmentation data;
[0066] The underground geological correction parameter acquisition module: acquires first underground geological correction parameters, and is also used to acquire second underground geological correction parameters;
[0067] The original geological mapping data calling module: calls the original geological mapping data of the land to be mapped by connecting to a server with mapping data, including first geological mapping data and second geological mapping data;
[0068] The land mapping method generation model construction module: horizontally splices the patch segmentation data, underground geological correction parameters, and geological mapping data at the feature level to obtain acquisition and mapping features, and constructs a land mapping method generation model according to the first acquisition and mapping feature and the first acquisition and mapping method set by the mapping personnel;
[0069] The assisting mapping module: receives the second UAV image data collected by the UAV field investigation, acquires and processes the second underground geological correction parameters and the second geological mapping data for feature processing to obtain second acquisition and mapping features, inputs the second acquisition and mapping features into the land mapping method generation model to generate a second acquisition and mapping method, and assists the mapping personnel to perform land data mapping according to the second acquisition and mapping method.
[0070] Furthermore, the first UAV image data or the second UAV image data is processed through a graphic segmentation line determination method to obtain patch segmentation data;
[0071] The graphic segmentation line determination method uses a machine learning algorithm to perform segmentation line determination classification. The machine learning algorithm is trained using the gray value difference in the first UAV image data or the second UAV image data. The calculation formula of the machine learning algorithm is:
[0072] F(X) = W T X
[0073] X is the grayscale value feature of the coordinate points of the UAV image data, F(X) is the determined parameter of the segmentation point output, and W T is the normal vector perpendicular to the hyperplane. When the value of F(X) is greater than 0, the coordinate point of the input UAV image data belongs to a point on the graphic segmentation line.
[0074] The present invention takes into account three-dimensional factors for field surveying and mapping. Through a deep learning model, it is trained using past field mapping experience and considers three-dimensional data for surveying and mapping. The key generation model for big data intelligent surveying and mapping assists surveying and mapping personnel in screening key change areas for surveying and mapping work, making the correction of the patch shape on the electronic map more accurate. The position and size of the patch correction do not require paper materials but are automatically determined through electronic devices. The present invention constructs a neural network model for surveying and mapping according to the surveying and mapping correction coefficient generated by calling groundwater system data and settlement data and land image data, enhancing the accuracy for different land surveying and mapping and reducing the workload of personnel.
[0075] Of course, it can be understood that each embodiment of the present invention can achieve one of the effects alone, and the combination of multiple embodiments of the present invention can achieve all the above effects. However, it is not required that each embodiment of the present invention achieves all the above advantages and effects, because each embodiment of the present invention can form an independent technical solution and make one or more contributions to the prior art.
[0076] For the part of the module structure not specifically defined in the present invention, it shall be subject to the content recorded in the prior art. The prior art mentioned in the foregoing background art part and the specific embodiment part of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters. The protection scope of the present invention shall be subject to the content actually recorded in the claims.
Claims
1. A method for collecting and mapping land survey data, characterized in that, The method includes: Performing feature processing on the first underground geological correction parameters and the first geological mapping data obtained by processing the first drone image data collected through on-site inspection of the drone to obtain the first collection and mapping features; Constructing a land mapping method generation model according to the first collection and mapping features and the first collection and mapping method set by the mapping personnel; Performing feature processing on the second underground geological correction parameters and the second geological mapping data obtained by processing the second drone image data collected through on-site inspection of the drone to obtain the second collection and mapping features; Inputting the second collection and mapping features into the land mapping method generation model to obtain the second collection and mapping method for the second geological mapping data, and assisting the mapping personnel to conduct land data mapping according to the second collection and mapping method.
2. The collection and mapping method for land survey data according to claim 1, characterized in that: The first drone image data or the second drone image data is processed by a graphic segmentation line determination method to obtain patch segmentation data.
3. The collection and mapping method for land survey data according to claim 2, characterized in that: The patch segmentation data includes control point coordinates, boundary point coordinates, and patch area, and both the control point coordinates and the boundary point coordinates are on the graphic segmentation line.
4. The collection and mapping method for land survey data according to claim 2, characterized in that: The graphic segmentation line determination method uses a machine learning algorithm to perform segmentation line determination classification, and the machine learning algorithm is trained using the gray value difference in the first drone image data or the second drone image data.
5. The collection and mapping method for land survey data according to claim 4, characterized in that: The first underground geological correction parameter or the second underground geological correction parameter is obtained by processing the groundwater distribution coefficient and the settlement parameter of the mapped land.
6. The collection and mapping method for land survey data according to claim 5, characterized in that: The groundwater distribution coefficient is calculated by the number of main groundwater veins, the number of tributaries, and the water flow velocity.
7. The collection and mapping method for land survey data according to claim 5, characterized in that: Performing feature horizontal splicing on the patch segmentation data, the underground geological correction parameters, and the geological mapping data to obtain collection and mapping features.
8. The collection and mapping method for land survey data according to claim 7, characterized in that: The activation function of the land mapping method generation model uses a neural network model improved based on the groundwater distribution coefficient.
9. A collection and mapping system for land survey data, including a drone image data acquisition and processing module, an underground geological correction parameter acquisition module, an original geological mapping data calling module, a land mapping method generation model construction module, and an assisting mapping module, characterized in that: The UAV image data acquisition and processing module: is used to acquire the first UAV image data, and is also used to acquire the second UAV image data, and processes the first UAV image data or the second UAV image data through a graphic segmentation line determination method to obtain patch segmentation data; The underground geological correction parameter acquisition module: acquires the first underground geological correction parameter, and is also used to acquire the second underground geological correction parameter; The original geological survey data calling module: calls the original geological survey data of the land to be surveyed and mapped by connecting to a server with survey data, including the first geological survey data and the second geological survey data; The land surveying and mapping method generation model construction module: horizontally splices the patch segmentation data, the underground geological correction parameters, and the geological survey data at the feature level to obtain the acquisition and surveying and mapping features, and constructs a land surveying and mapping method generation model according to the first acquisition and surveying and mapping features and the first acquisition and surveying and mapping method set by the surveying and mapping personnel; The assisted surveying and mapping module: receives the second UAV image data acquired by the UAV field investigation, acquires and processes the second underground geological correction parameter and the second geological survey data for feature processing to obtain the second acquisition and surveying and mapping features, inputs the second acquisition and surveying and mapping features into the land surveying and mapping method generation model to generate the second acquisition and surveying and mapping method, and assists the surveying and mapping personnel to conduct the surveying and mapping of land data according to the second acquisition and surveying and mapping method.
10. A land survey data acquisition and surveying and mapping system according to claim 9, wherein: The first UAV image data or the second UAV image data is processed through a graphic segmentation line determination method to obtain patch segmentation data; The graphic segmentation line determination method uses a machine learning algorithm to determine the classification of the segmentation line, and the machine learning algorithm is trained using the gray value difference in the first UAV image data or the second UAV image data.