A method and system for determining micro-topography based on inverse distance weighted method and decision tree

Abstract

The invention discloses a micro-topography determination method and system based on an inverse distance weighted method and a decision tree, and relates to the technical field of micro-topography determination. The method comprises: using a filter function and adopting a matrix notation to filter invalid values ​​and abnormal values ​​in original elevation data; calculating large and small window terrain indexes based on a data matrix, and obtaining corresponding weighted terrain indexes according to the large and small window terrain indexes; generating a determination decision tree according to a preset terrain determination rule set, and generating micro-topography classification data of a micro-topography determination area based on a neighborhood terrain reference matrix, a neighborhood terrain slope matrix, and a weighted terrain index by using the determination decision tree; obtaining a terrain classification matrix after classification of the micro-topography determination area by using a preset terrain classifier and the micro-topography classification data; and solving the problems that the original micro-topography classification method is greatly affected by the calculation window radius, the micro-topography classification result is unstable, and the ice thickness prediction accuracy is not high.

Description

Technical Field

[0001] The present invention relates to the technical field of micro-topography determination, and more specifically, it relates to a micro-topography determination method and system based on the inverse distance weighting method and decision tree. Background Art

[0002] Micro-topography refers to the type of surface undulation morphology within a small range. With the in-depth research, micro-topography has been gradually and deeply applied in fields such as landscape classification and terrain feature recognition, and its scientificity and usability have been verified in different specialties; it has been found in the practice of power grid icing observation and icing thickness early warning that the development of icing shows an obvious correlation with micro-topography. Therefore, micro-topography can be used as a feasible index for icing prediction; however, it has been found in the practice of icing prediction that the original micro-topography index cannot well predict the icing value, and there is a large error between the predicted value and the actual value. It is necessary to make appropriate adjustments to the original terrain index for more scientific and reasonable icing prediction. Summary of the Invention

[0003] The purpose of the present invention is to provide a micro-topography determination method and system based on the inverse distance weighting method and decision tree. This determination method aims to solve problems such as the large influence of the original micro-topography classification method by the calculation window radius, unstable micro-topography classification results, and low prediction accuracy of icing thickness. It has been verified that this method can improve the accuracy of micro-topography classification and icing value prediction and can be widely applied in icing value prediction.

[0004] The above technical purpose of the present invention is achieved through the following technical solutions:

[0005] In the first aspect, the present application provides a micro-topography determination method based on the inverse distance weighting method and decision tree, including the following specific steps:

[0006] Obtain the original elevation data of the micro-topography determination area, and use the filtering function and matrix notation to filter out the invalid values and abnormal values in the original elevation data to obtain the data matrix corresponding to the original elevation data;

[0007] Based on the data matrix, calculate the large and small window terrain indices of the micro-topography determination area, and calculate the corresponding weighted terrain index according to the large and small window terrain indices;

[0008] Based on the data matrix and the large and small window terrain indices, calculate the neighborhood terrain reference matrix and neighborhood terrain slope matrix of the micro-topography determination area;

[0009] Generate a decision tree according to the preset terrain determination rule set, and based on the neighborhood terrain reference matrix, neighborhood terrain slope matrix, and the weighted terrain index after normalization processing, use the decision tree to generate the micro-topography classification data of the micro-topography determination area;

[0010] Using a preset terrain classifier and through micro-topography classification data, a terrain classification matrix after classification of the micro-topography determination area is obtained.

[0011] Based on the above technical solution, the present invention can further be improved as follows.

[0012] Furthermore, the above original elevation data is specifically:

[0013] ;

[0014] wherein, and respectively represent the length and width of the micro-topography determination area, is the grid size, and the grid is an equilateral side grid, and respectively represent the number of grid divisions of the micro-topography determination area in the length and width directions, represents the grid height value.

[0015] Furthermore, the above large and small window terrain indices include a large window terrain index and a small window terrain index. The large window terrain index is specifically:

[0016] ;

[0017] wherein, represents the large window terrain index, represents an annular neighborhood window with as the center coordinate, represents the elevation of the window center point, represents the total number of grids within the annular neighborhood window, represents the grid elevation within the window, and , represents radius.

[0018] Furthermore, the above small window terrain index is specifically:

[0019] ;

[0020] wherein, represents the small window terrain index, represents a rectangular window with as the center coordinate, represents the elevation of the window center point, represents the total number of grids within the small window, represents the grid elevation within the window, and , is radius.

[0021] Furthermore, the above-mentioned weighted terrain index is specifically:

[0022] , where , ;

[0023] In the formula, represents the weighted terrain index, represents the terrain index of the large and small windows, represents the weight matrix, takes 2, is the distance between the current point and the specified point.

[0024] Furthermore, the above-mentioned neighborhood terrain reference matrix is specifically:

[0025] , where ;

[0026] In the formula, represents the neighborhood terrain reference matrix, represents the row and column numbers corresponding to the window, represents the elevation value of the cell within the window.

[0027] Furthermore, the above-mentioned neighborhood terrain slope matrix is specifically:

[0028] , where ;

[0029] In the formula, represents the neighborhood terrain slope matrix, represents the row and column numbers corresponding to the window, represents the slope value of the non-core cell relative to the core cell within the window.

[0030] In the second aspect, the present application provides a micro-topography determination system based on the inverse distance weighting method and decision tree, which is applied to a micro-topography determination method based on the inverse distance weighting method and decision tree in any item of the first aspect, including:

[0031] The first module is used to obtain the original elevation data of the micro-topography determination area, filter out the invalid values and outliers in the original elevation data by using the filtering function and adopting the matrix notation, and obtain the data matrix corresponding to the original elevation data;

[0032] The second module is used to calculate the terrain index of the large and small windows of the micro-topography determination area based on the data matrix, and calculate the corresponding weighted terrain index according to the terrain index of the large and small windows;

[0033] The third module is used to calculate the neighborhood terrain reference matrix and the neighborhood terrain slope matrix of the micro-topography determination area based on the data matrix and the size window terrain index;

[0034] The fourth module is used to generate a decision tree according to a preset set of terrain determination rules, and generate micro-topography classification data of the micro-topography determination area by using the decision tree based on the neighborhood terrain reference matrix, the neighborhood terrain slope matrix, and the weighted terrain index after normalization processing;

[0035] The fifth module is used to obtain the classified terrain classification matrix of the micro-topography determination area by using a preset terrain classifier and through the micro-topography classification data.

[0036] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method according to any one of the first aspect is implemented.

[0037] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium, and the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the method according to any one of the first aspect.

[0038] Compared with the prior art, the present invention has at least the following beneficial effects:

[0039] In the present application, through a micro-topography determination method based on the inverse distance weighting method and the decision tree, the influence of invalid values and outliers is eliminated through multiple filtering processes, and the weighted terrain index of the size window is used to weaken the classification influence of the calculation window radius size on the terrain index, which can make the terrain classification effect more reasonable, and solves the problems that the original micro-topography classification method is greatly affected by the calculation window radius, the micro-topography classification result is unstable, and the prediction accuracy of the ice coating thickness is not high. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of this application, and do not limit the embodiments of the present invention. In the drawings:

[0041] Figure 1 is the method flow chart of the micro-topography determination method in the embodiment of the present invention;

[0042] Figure 2 is the flow chart of the determination method in the embodiment of the present invention;

[0043] Figure 3 is the schematic diagram of the original elevation data in the embodiment of the present invention;

[0044] Figure 4Schematic diagram of the large window weighted terrain index in the weighted terrain index of the embodiment of the present invention;

[0045] Figure 5 Schematic diagram of the small window weighted terrain index in the weighted terrain index of the embodiment of the present invention;

[0046] Figure 6 Standard deviation of the large window weighted terrain index in the embodiment of the present invention;

[0047] Figure 7 Standard deviation of the small window weighted terrain index in the embodiment of the present invention;

[0048] Figure 8 Schematic diagram of the slope of the micro-topography determination area in the embodiment of the present invention;

[0049] Figure 9 Schematic diagram of the decision tree in the embodiment of the present invention;

[0050] Figure 10 Schematic diagram of the terrain classification matrix after classification of the micro-topography determination area in the embodiment of the present invention. Detailed implementation manners

[0051] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and illustrated herein generally can be arranged and designed in a variety of different configurations.

[0052] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0053] It should be noted that: like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0054] In the description of the embodiments of the present invention, "a plurality of" represents at least two.

[0055] Embodiment 1: To solve the problems that the original micro-topography classification method is greatly affected by the calculation window radius, the micro-topography classification result is unstable, and the prediction accuracy of the ice coating thickness is not high, etc., this embodiment provides a micro-topography determination method based on the inverse distance weighting method and the decision tree, as Figure 1 andFigure 2 As shown in the figure, it includes the following specific steps:

[0056] S1. Obtain the original elevation data of the micro-topography determination area, filter out the invalid values and abnormal values in the original elevation data by using the filtering function and adopting the matrix notation, and obtain the data matrix corresponding to the original elevation data.

[0057] Optionally, the above original elevation data is specifically:[[]]

[0058] ;

[0059] In the formula, and respectively represent the length and width of the micro-topography determination area, is the grid size, and the grid adopts an equilateral grid, and respectively represent the number of grid divisions of the micro-topography determination area in the length and width directions, represents the grid height value.

[0060] Among them, for the original elevation data of the obtained micro-topography determination area, the grid size can be set to 12.5 m, and the mean filtering function is selected for filtering processing. As Figure 3 shown, it is to obtain the original elevation data after filtering processing.

[0061] S2. Based on the data matrix, calculate the size window topographic index of the micro-topography determination area, and calculate the corresponding weighted topographic index according to the size window topographic index.

[0062] Optionally, the above size window topographic index includes a large window topographic index and a small window topographic index. The large window topographic index is specifically:[[]]

[0063] ;

[0064] In the formula, represents the large window topographic index, represents the annular neighborhood window with as the center coordinate, represents the elevation of the window center point, represents the total number of grids in the annular neighborhood window, represents the grid elevation in the window, and , represents radius.

[0065] Optionally, the above small window topographic index is specifically:[[]]

[0066] ;

[0067] In the formula, represents the small-window topographic index, represents a rectangular window with coordinates centered on ; represents the elevation of the center point of the window, represents the total number of grids within the small window, represents the elevation of the grids within the window, and , is the radius of.

[0068] Among them, in the large-window and small-window topographic position indices, the radius of the large-window circular neighborhood can be set to 10, and the radius of the small-window rectangular neighborhood can be set to 5. Then, the corresponding weighted topographic indices are calculated through the determined weighting function respectively; as shown in Figure 4 and Figure 5 , they respectively represent the processing results after being processed by the weighting function.

[0069] Optionally, the above-mentioned weighted topographic index is specifically:

[0070] , where , ;

[0071] In the formula, represents the weighted topographic index, represents the large-window and small-window topographic index, represents the weight matrix, takes 2, is the distance between the current point and the specified point.

[0072] Specifically, the weighted topographic index can also be normalized, that is, the corresponding standard deviations are calculated. The standard deviations corresponding to the weighted topographic indices are as shown in Figure 6 and Figure 7 .

[0073] S3. Based on the data matrix and the large-window and small-window topographic indices, calculate the neighborhood topographic reference matrix and the neighborhood topographic slope matrix of the micro-topographic determination area.

[0074] Optionally, the above-mentioned neighborhood topographic reference matrix is specifically:

[0075] , where ;

[0076] In the formula, represents the neighborhood topographic reference matrix, represents the row and column numbers corresponding to the window, represents the elevation value of the cells within the window.

[0077] Optionally, the above neighborhood terrain slope matrix is as Figure 8 shown, specifically:

[0078] , where ;

[0079] In the formula, represents the neighborhood terrain slope matrix, represents the row and column numbers corresponding to the window, represents the slope value of the non-core cell relative to the core cell within the window.

[0080] S4. Generate a decision tree based on the preset terrain determination rule set, and generate the micro-topography classification data of the micro-topography determination area by using the decision tree based on the neighborhood terrain reference matrix, the neighborhood terrain slope matrix, and the normalized weighted terrain index.

[0081] Among them, by designing the micro-topography classification determination rule, a decision tree is generated, as Figure 9 shown, that is, using the data obtained above to generate the micro-topography type distribution data.

[0082] S5. Use the preset terrain classifier and through the micro-topography classification data, obtain the terrain classification matrix after classification of the micro-topography determination area; specifically, the final micro-topography classification result is as Figure 10 shown.

[0083] Embodiment 2: To solve the problems that the original micro-topography classification method is greatly affected by the calculation window radius, the micro-topography classification result is unstable, and the prediction accuracy of the ice coating thickness is not high, this embodiment provides a micro-topography determination method based on the inverse distance weighting method and the decision tree, as Figure 2 shown, including the following specific steps:

[0084] S1. If the original elevation data of the obtained micro-topography determination area is , it can be recorded as :

[0085] ;

[0086] In the formula, , respectively represent the length and width of the micro-topography determination area, is the grid size, and the grid uses an equilateral side grid, and respectively represent the number of grid divisions of the micro-topography determination area in the length and width directions, represents the grid height value.

[0087] S2. Select a filtering function and, using matrix notation, perform a filtering operation on the aforementioned which can be denoted as: ;

[0088] where is the data matrix of the gridded DEM after filtering processing, is the filtering function, and is the matrix representation of the aforementioned original elevation dataset.

[0089] S3. Calculate the terrain position indices of the large and small windows and , and the calculation formulas are as follows:

[0090] ;

[0091] In the formula, represents the terrain index of the large window, represents the annular neighborhood window with as the central coordinates, represents the elevation of the window center point, represents the total number of grids within the annular neighborhood window, represents the grid elevation within the window, and , represents the radius; similarly, the specific terrain index of the small window is:

[0092] ;

[0093] In the formula, represents the terrain index of the small window, represents the rectangular window with as the central coordinates, represents the elevation of the window center point, represents the total number of grids within the small window, represents the grid elevation within the window, and , is the radius.

[0094] S4. Determine the weighting function and calculate the weighted terrain index , using matrix notation as follows:

[0095] ;

[0096] where is the weight matrix, which can be calculated by the following formula: ;

[0097] where , Usually take 2, is the distance between the current point and the specified point.

[0098] S5. Calculate the normalized terrain position index: ; where is the normalization function.

[0099] S6. Calculate the neighborhood terrain reference and the neighborhood terrain slope .

[0100] Among them, the above neighborhood terrain reference matrix is specifically:

[0101] , where ;

[0102] In the formula, represents the neighborhood terrain reference matrix, represents the row and column numbers corresponding to the window, represents the elevation value of the cell within the window.

[0103] Optionally, the above neighborhood terrain slope matrix is as Figure 8 shown, specifically:

[0104] , where ;

[0105] In the formula, represents the neighborhood terrain slope matrix, represents the row and column numbers corresponding to the window, represents the slope value of the non-core cell relative to the core cell within the window.

[0106] S7. According to the micro-topography definition, design a set of decision rules and generate a decision tree , such as Figure 9 shown, using the previously obtained , , , use the decision tree to generate micro-topography classification data. It can be expressed as:

[0107] ;

[0108] Among them, is the preset terrain classifier, is the filtering function, is the terrain classification matrix output after classification, such as Figure 10 shown.

[0109] Embodiment 3: The embodiment of the present application provides a microtopography determination system based on the inverse distance weighting method and decision tree, which is applied to a microtopography determination method based on the inverse distance weighting method and decision tree in any one of Embodiment 1, and includes:

[0110] The first module is used to obtain the original elevation data of the microtopography determination area, filter out the invalid values and abnormal values in the original elevation data by using a filtering function and adopting matrix notation, and obtain the data matrix corresponding to the original elevation data.

[0111] The second module is used to calculate the size window topographic index of the microtopography determination area based on the data matrix, and calculate the corresponding weighted topographic index according to the size window topographic index.

[0112] The third module is used to calculate the neighborhood topographic reference matrix and neighborhood topographic slope matrix of the microtopography determination area based on the data matrix and the size window topographic index.

[0113] The fourth module is used to generate a determination decision tree according to a preset set of terrain determination rules, and generate microtopography classification data of the microtopography determination area by using the determination decision tree based on the neighborhood topographic reference matrix, neighborhood topographic slope matrix, and the weighted topographic index after normalization processing.

[0114] The fifth module is used to obtain the terrain classification matrix after classification of the microtopography determination area by using a preset terrain classifier and through the microtopography classification data.

[0115] Embodiment 4: The embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method in any one of Embodiment 1 or Embodiment 2 is implemented.

[0116] Embodiment 5: The embodiment of the present application provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the method in any one of Embodiment 1 or Embodiment 2.

[0117] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A micro-topography determination method based on inverse distance weighted method and decision tree, characterized in that: The specific steps include: Acquire the original elevation data of the micro-topography determination area, filter invalid values ​​and abnormal values ​​in the original elevation data by using a filter function and matrix notation, and obtain a data matrix corresponding to the original elevation data; Based on the data matrix, the large and small window terrain indexes of the micro-topography determination area are calculated, and the corresponding weighted terrain indexes are calculated according to the large and small window terrain indexes; Based on the data matrix and the large and small window terrain indexes, a neighborhood terrain reference matrix and a neighborhood terrain slope matrix of the micro-topography determination area are calculated; Generate a decision tree according to a preset terrain determination rule set, and generate micro-topography classification data of a micro-topography determination area using the decision tree based on the neighborhood terrain reference matrix, the neighborhood terrain slope matrix, and the normalized weighted terrain index; A terrain classification matrix after classification of the micro-topography determination area is obtained by using a preset terrain classifier and the micro-topography classification data.

2. The micro-topography determination method based on inverse distance weighted method and decision tree according to claim 1 is characterized in that: The original elevation data is specifically: ； In the formula, , Respectively represent the length and width of the micro-topography determination area, is the size of the grid. The grid uses an equilateral grid. and Respectively represent the number of grid divisions in the length and width directions of the micro-topography determination area, Representing a grid The height value of .

3. The micro-topography determination method based on inverse distance weighted method and decision tree according to claim 1 is characterized in that: The large and small window terrain indexes include a large window terrain index and a small window terrain index. The large window terrain index is specifically: ； In the formula, represents the large window terrain index, Indicates is the annular neighborhood window with the center coordinates, Indicates the elevation of the window center point. represents the total number of grids in the annular neighborhood window, represents the grid elevation within the window, and , express The radius of .

4. The micro-topography determination method based on inverse distance weighted method and decision tree according to claim 3 is characterized in that: The small window terrain index is specifically: ； In the formula, represents the small window terrain index, Indicates The coordinates of the rectangular window centered at Indicates the elevation of the window center point. Indicates the total number of grids in the small window, represents the grid elevation within the window, and , for The radius of .

5. A micro-topography determination method based on inverse distance weighted method and decision tree according to claim 3 or 4, characterized in that: The weighted terrain index is specifically: ,in, , ; In the formula, represents the weighted terrain index, Represents the size window terrain index, represents the weight matrix, Take 2, The distance between the current point and the specified point.

6. The micro-topography determination method based on inverse distance weighted method and decision tree according to claim 1 is characterized in that: The neighborhood terrain reference matrix is ​​specifically: ,in, ; In the formula, represents the neighborhood terrain reference matrix, Indicates the row and column number corresponding to the window. Indicates the cell elevation value within the window.

7. The micro-topography determination method based on inverse distance weighted method and decision tree according to claim 1 is characterized in that: The neighborhood terrain slope matrix is ​​specifically: ,in, ; In the formula, represents the neighborhood terrain slope matrix, Indicates the row and column number corresponding to the window. Represents the slope value of non-core cells relative to core cells in the window.

8. A micro-terrain determination system based on inverse distance weighted method and decision tree, applied to a micro-terrain determination method based on inverse distance weighted method and decision tree as claimed in any one of claims 1 to 7, characterized in that: include: The first module is used to obtain the original elevation data of the micro-topography determination area, and filter the invalid values ​​and abnormal values ​​in the original elevation data by using a filter function and a matrix notation to obtain a data matrix corresponding to the original elevation data; The second module is used to calculate the large and small window terrain indexes of the micro-topography determination area based on the data matrix, and calculate the corresponding weighted terrain index according to the large and small window terrain indexes; The third module is used to calculate the neighborhood terrain reference matrix and the neighborhood terrain slope matrix of the micro-topography determination area based on the data matrix and the large and small window terrain indexes; The fourth module is used to generate a determination decision tree according to a preset terrain determination rule set, and based on the neighborhood terrain reference matrix and the neighborhood terrain slope matrix, and the normalized weighted terrain index, use the determination decision tree to generate micro-topography classification data of the micro-topography determination area; The fifth module is used to obtain a terrain classification matrix after the micro-terrain determination area is classified by using a preset terrain classifier and the micro-terrain classification data.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the method according to any one of claims 1 to 7 is implemented when the processor executes the computer program.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which enable a computer to execute the method of any one of claims 1-7.

Images