Microtopography recognition method, device and equipment and storage medium

Through the mean variable point analysis method and the calculation of the average elevation deviation in the elevation point neighborhood window, combined with the parallelized random forest model, the problems of low efficiency and insufficient accuracy of micro-terrain recognition of transmission lines are solved, and efficient and accurate micro-terrain recognition is achieved .

CN119066531AActive Publication Date: 2024-12-03WUHAN UNIV

Patent Information

Application Number
CN202411037434.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-12-03
Estimated Expiration
2044-07-31

AI Technical Summary

Technical Problem

The existing micro-terrain identification methods for transmission lines are inefficient and insufficiently accurate, making it difficult to meet the needs of large-scale data processing, resulting in challenges in transmission lines planning, maintenance and disaster prevention.

Method used

The mean variable point analysis method is used to find the best calculation window for the terrain position index, and the profile curvature characteristics are calculated based on the average elevation deviation in the neighborhood window of the elevation point, and a micro-terrain classification decision table and sample feature data set are constructed, and the parallelized random forest model is trained. Perform micro-terrain recognition.

Benefits of technology

It realizes efficient and accurate identification of micro-terrain of transmission line corridors, improves identification efficiency and accuracy, makes the boundaries of various terrain clearer, and can quickly identify micro-terrain within a large range.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066531A_ABST
    Figure CN119066531A_ABST
Patent Text Reader

Abstract

The invention provides a microtopography recognition method, device and equipment and a storage medium, and the method comprises the steps: searching an optimal calculation window of a topography position index based on a mean change point analysis method, and obtaining the topography position index features of a target sample based on the optimal calculation window; obtaining the size of a neighborhood window when the average elevation deviation reaches a preset threshold value based on the average elevation deviation counted in the neighborhood window of the elevation points in the target sample, and calculating a section curvature feature based on the obtained size of the neighborhood window; constructing a sample feature data set at least based on the terrain position index features and the section curvature features; training a microtopography recognition model based on the sample feature data; and inputting the elevation data of the to-be-detected area into the trained microtopography recognition model to recognize the microtopography of the to-be-detected area. The optimized characteristic factors and the combination thereof can realize effective distinguishing of the microtopography, so that boundaries of various microtopographies are clearer, and the microtopographies can be quickly identified.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of terrain surveying and mapping technology, and in particular to a micro-terrain recognition method, device, equipment and storage medium. Background Art

[0002] Meteorological disasters are usually closely related to the terrain where the transmission lines are located. The local complex terrain will change meteorological factors such as air humidity and airflow direction in the local area. Under the dual influence of terrain changes and meteorological changes, the strong winds, low temperatures, and high humidity generated in the micro-topography area are likely to cause damage to the transmission lines, leading to accidents such as wind deviation, icing, and line breaks. Therefore, the identification of micro-topography can provide a reference for the planning and layout, daily maintenance, and disaster prevention of transmission lines, and ensure the stable operation of the power transmission system. Compared with macro-topography, micro-topography can express terrain changes more meticulously. At present, the research on the identification of micro-topography of transmission lines is mostly based on rule knowledge and experience. The identification accuracy and efficiency need to be further improved. At present, there is no effective method for the rapid identification of micro-topography of transmission lines in a large area.

[0003] In addition, since the transmission lines span multiple regions and the surrounding terrain is relatively complex, it is often necessary to use DEM data (Digital Elevation Model) over a large area to identify micro-topography. The efficiency issue is one of the difficulties currently faced by transmission line micro-topography identification, and traditional processing methods can hardly meet the needs of massive data processing. Summary of the invention

[0004] The present invention provides a micro-topography recognition method, device, equipment and storage medium for realizing efficient and accurate recognition of micro-topography of a transmission line corridor.

[0005] According to one aspect of the present invention, a micro-terrain recognition method is provided, comprising:

[0006] Finding an optimal calculation window for the terrain position index based on the mean change point analysis method, and obtaining the terrain position index characteristics of the target sample based on the optimal calculation window;

[0007] Based on the statistical average elevation deviation in the neighborhood window of the elevation points in the target sample, the neighborhood window size when the average elevation deviation reaches a preset threshold is obtained, and the profile curvature feature is calculated based on the obtained neighborhood window size;

[0008] Constructing a micro-topography classification decision table based on the terrain position index and the profile curvature, as well as the elevation variation coefficient, terrain slope, and relief characteristics;

[0009] Construct a sample feature data set based at least on the terrain position index feature, the profile curvature feature, and the micro-topography classification decision table;

[0010] Train a micro-topography recognition model based on the sample feature data;

[0011] Input the elevation data of the area to be measured into the trained micro-topography recognition model to identify the micro-topography of the area to be measured.

[0012] Optionally, the method for finding the optimal calculation window of the terrain position index based on the mean change point analysis method and obtaining the terrain position index feature of the target sample includes:

[0013] The calculation formula of the terrain position index TPI is:

[0014]

[0015] Z i is the elevation value of point i, is the average value of all elevation points within the neighborhood range of point i;

[0016] Calculate the range value of the terrain position index under each window;

[0017] Based on the range value of the terrain position index and the window size, obtain the range value of the terrain position index per unit area;

[0018] Construct a data sequence based on the range values of the terrain position index corresponding to each window;

[0019] Calculate the variance of the data sequence, and the calculation formula is:

[0020]

[0021] In the formula, is the average value of the data sequence {x i}, i = 1, 2, 3,..., N, and N is the number of windows used;

[0022] Select x i in turn as the cut-off point to divide the data sequence into two subsequences, and 1 < t < N; calculate the sum of variances S i of the two subsequences:

[0023]

[0024] In the formula, and are the variances of the subsequences and The arithmetic mean of

[0025] In this way, the difference between S and Si is calculated, and the result sequence {SS i}; The maximum value in the result sequence is selected as the truncation point x i The corresponding window is the optimal calculation window;

[0026] The terrain position index feature of the target sample is obtained based on the optimal calculation window.

[0027] Optionally, the average elevation deviation in the neighborhood window of the elevation points in the target sample is obtained, the neighborhood window size is obtained when the average elevation deviation reaches a preset threshold, and the profile curvature feature is calculated based on the obtained neighborhood window size, including:

[0028] Calculate the average elevation deviation DEV in each neighborhood window of the elevation point in turn:

[0029]

[0030]

[0031] In the formula, z0 is the elevation value of a certain point; is the average elevation in the neighborhood; SD is the standard deviation of the elevation in the neighborhood; n is the number of grid points in the neighborhood; z i is a grid point in the neighborhood;

[0032] When the average elevation deviation DEV reaches a preset threshold, the profile curvature feature is calculated based on the neighborhood window size corresponding to the threshold.

[0033] Optionally, before constructing the sample feature data set based at least on the terrain position index feature and the profile curvature feature, the method further includes:

[0034] Calculating the slope characteristics, undulation characteristics and elevation variation coefficient characteristics of the target sample respectively;

[0035] The step of constructing a sample feature data set based at least on the terrain position index feature and the profile curvature feature comprises:

[0036] Constructing a micro-topography classification decision table based on the terrain position index feature, the profile curvature feature, the elevation variation coefficient feature, the slope feature, and the undulation feature;

[0037] Based on the characteristic values ​​of the terrain position index feature, the profile curvature feature, the slope feature, the undulation feature, and the elevation variation coefficient feature of the target sample, a classification label of the target sample is obtained according to the micro-topography classification decision table;

[0038] A sample point is constructed based on the characteristic values ​​of the terrain position index feature, the profile curvature feature, the slope feature, the undulation feature, the elevation variation coefficient feature and the classification label corresponding to the target sample, and a sample feature data set is constructed from several sample points.

[0039] Optionally, the method further includes:

[0040] Construct a parallelized random forest model;

[0041] The parallelized random forest model construction includes:

[0042] Construct several parallel decision trees with self-service sampling, and assign the sample feature data set required for training to each decision tree;

[0043] Parallel training of a plurality of the decision trees, each decision tree being constructed in parallel by the assigned threads;

[0044] Parallelization of classification prediction: parallel prediction of the micro-topography categories of the sample points to be classified, and obtaining the final micro-topography categories of the sample points through voting.

[0045] Optionally, the training of a micro-terrain recognition model based on the sample feature data includes:

[0046] Selecting a training set from the sample feature data set;

[0047] Randomly extract samples with replacement from the training set to construct multiple sub-training sets;

[0048] Features are randomly selected from each sub-training set to train each decision tree.

[0049] Optionally, also include:

[0050] Evaluate the recognition effect of the parallel random forest model;

[0051] The evaluating the recognition effect of the parallel random forest model includes:

[0052] The recognition effect of the parallel random forest model is evaluated based on the recognition indicators of the parallel random forest model; the recognition indicators include one or more of the confusion matrix, classification accuracy, Kappa coefficient, single category precision Precision and recall Recall of the parallel random forest model.

[0053] According to another aspect of the present invention, there is provided a micro-topography recognition device, comprising:

[0054] A first feature acquisition unit is used to find an optimal calculation window of the terrain position index based on a mean change point analysis method, and acquire the terrain position index feature of the target sample based on the optimal calculation window;

[0055] A second feature acquisition unit is used to acquire the neighborhood window size when the average elevation deviation reaches a preset threshold based on the statistical average elevation deviation in the neighborhood window of the elevation points in the target sample, and calculate the profile curvature feature based on the acquired neighborhood window size;

[0056] A data set construction unit, configured to construct a sample feature data set based at least on the terrain position index feature and the profile curvature feature;

[0057] A training unit, used for training a micro-terrain recognition model based on the sample feature data;

[0058] The recognition unit is used to input the elevation data of the area to be measured into the trained micro-topography recognition model to recognize the micro-topography of the area to be measured.

[0059] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0060] at least one processor; and

[0061] a memory communicatively connected to the at least one processor; wherein,

[0062] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the micro-terrain recognition method described in any embodiment of the present invention.

[0063] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the micro-terrain recognition method described in any embodiment of the present invention when executed.

[0064] The technical solution of the embodiment of the present invention is to find the optimal calculation window of the terrain position index based on the mean change point analysis method, and obtain the terrain position index characteristics of the target sample with the optimal calculation window; the average elevation deviation statistically calculated in the neighborhood window of the elevation point in the target sample is used to obtain the neighborhood window size when the average elevation deviation reaches a preset threshold, and the profile curvature characteristics are calculated; and then a sample feature data set is constructed from the terrain position index characteristics and profile curvature characteristics obtained by optimization calculation to serve as a training sample for a micro-terrain recognition model. The optimized feature factors and their combinations of the present invention can effectively distinguish micro-terrain, make the boundaries of various types of terrain clearer, and quickly identify micro-terrain.

[0065] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0067] Figure 1 is a flow chart of a micro-terrain recognition method provided by an embodiment of the present invention;

[0068] Figure 2 A schematic diagram of a sub-process of a micro-terrain recognition method provided by an embodiment of the present invention;

[0069] Figure 3 A schematic diagram of another sub-process of a micro-terrain recognition method provided by an embodiment of the present invention;

[0070] Figure 4 A schematic diagram of the architecture of a parallel random forest model of a micro-terrain recognition method provided by an embodiment of the present invention;

[0071] Figure 5 A schematic diagram of another sub-process of a micro-terrain recognition method provided by an embodiment of the present invention;

[0072] Figure 6 A schematic diagram of the effect of using the method of an embodiment of the present invention in identifying two types of micro-topography, namely, a pass and a canyon;

[0073] Figure 7 A schematic diagram of micro-topography recognition effect obtained by using the model of the present invention;

[0074] Figure 8 A schematic diagram of the structure of a micro-topography recognition device provided by an embodiment of the present invention;

[0075] Fig. 9 It is a schematic diagram of the structure of an electronic device for implementing the micro-topography recognition method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0076] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0077] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0078] The terms in the embodiments of the present invention are explained as follows:

[0079] Topographic Position Index (TPI).

[0080] Digital Elevation Model (DEM).

[0081] Figure 1 A flowchart of a micro-terrain recognition method is provided for an embodiment of the present invention. Figure 1 As shown, the method includes:

[0082] S101. Find an optimal calculation window for a terrain position index based on a mean change point analysis method, and obtain terrain position index characteristics of the target sample based on the optimal calculation window.

[0083] The terrain position index (TPI) refers to the difference between the elevation value of a point in a digital elevation model and the average elevation value of other points in its neighborhood. It is used to reflect the relative position relationship between a point and other points in its neighborhood on the terrain, and is thus used for terrain morphological analysis. For example, TPI can be used to distinguish ridges, valleys, ups and downs, flat slopes, etc.

[0084] In one embodiment, large TPI and small TPI are defined according to the size of the window radius. The calculation window radius of small TPI is generally less than 600 meters. When the window radius exceeds 600 meters, TPI can still capture relatively detailed micro-topographic features and is suitable for describing local terrain changes. However, when the window radius exceeds 2000 meters, TPI gradually loses its sensitivity to elevation changes, thereby ignoring the detailed information of micro-topography. The TPI calculated when the window radius is set to 600 meters to 2000 meters is called large TPI. In this embodiment, the resolution of the DEM grid data is 30 meters, and the radius of the optimal calculation window of large TPI can be set to search in the range of 20 pixels to 70 pixels.

[0085] This embodiment can be based on different window sizes, for example, 20×20 pixels, 21×21 pixels, ..., 69×69 pixels, 70×70 pixels windows can be obtained to calculate the range value of TPI under each window, and the mean change point analysis method is used to find the best calculation window of the terrain position index, and the terrain position index characteristics of the target sample are obtained based on the best calculation window. Among them, the target sample can be the DEM data of a transmission line crossing area.

[0086] S102, based on the statistical average elevation deviation in the neighborhood window of the elevation points in the target sample, obtain the neighborhood window size when the average elevation deviation reaches a preset threshold, and calculate the profile curvature feature based on the obtained neighborhood window size.

[0087] The calculation window of profile curvature can generally be set to 3 pixels × 3 pixels. A smaller window can provide sufficient resolution to analyze micro-topographic features and help to accurately calculate local profile curvature. Considering the accuracy requirements of micro-topography, it is more appropriate that the maximum calculation window of profile curvature does not exceed 500 meters × 500 meters. Exceeding this maximum value may over-smooth the elevation data, causing important terrain details to be ignored or blurred. Therefore, the minimum neighborhood window size in the adaptive neighborhood method can be set to 3 pixels × 3 pixels, and the maximum neighborhood window can be set to 15 pixels × 15 pixels (450 meters × 450 meters). The average elevation deviation DEV in the neighborhood window of 15 pixels × 15 pixels to 3 pixels × 3 pixels is calculated in order from large to small. This indicator is used to evaluate the complexity of the terrain in the window. When the average elevation deviation reaches the preset threshold, the neighborhood window size is used to calculate the profile curvature characteristics based on the obtained neighborhood window size.

[0088] In this embodiment, the preset threshold can be obtained based on experimental data and empirical data. When the average elevation deviation DEV reaches the preset threshold, the neighborhood window size corresponding to the preset threshold can be selected to calculate the profile curvature characteristics. When the average elevation deviation DEV is greater than the preset threshold, the neighborhood window size corresponding to the DEV that is greater than the preset threshold can also be used to calculate the profile curvature characteristics.

[0089] S103: Construct a sample feature data set based at least on the terrain position index feature and the profile curvature feature.

[0090] The present invention can use the optimized terrain position index feature and profile curvature feature to construct a sample feature data set. For example, a sample of P can be constructed. i =(F1,F2,..F n ,L), where F1,F2,..F n is the characteristic value of the sample point, F1, F2, .. F n includes at least the characteristic values ​​of the optimized terrain position index feature and the profile curvature feature, and L is the micro-topography label of the sample point.

[0091] S104: training a micro-terrain recognition model based on the sample feature data.

[0092] The embodiments of the present invention can be combined with a neural network model to identify micro-topography. For example, the present application can use a random forest tree as a micro-topography recognition model. Of course, other neural network architectures can also be used to identify micro-topography.

[0093] S105 , inputting the elevation data of the area to be measured into the trained micro-topography recognition model to identify the micro-topography of the area to be measured.

[0094] After the micro-terrain recognition model is trained, the large-scale digital elevation model of the power transmission line to be tested can be input into the recognition model, so as to quickly identify the micro-terrain in the large-scale digital elevation model of the power transmission line and classify it into a map.

[0095] The technical solution of the embodiment of the present invention is to find the optimal calculation window of the terrain position index based on the mean change point analysis method, and obtain the terrain position index characteristics of the target sample with the optimal calculation window; the average elevation deviation statistically calculated in the neighborhood window of the elevation point in the target sample is used to obtain the neighborhood window size when the average elevation deviation reaches a preset threshold, and the profile curvature characteristics are calculated; and then a sample feature data set is constructed from the terrain position index characteristics and profile curvature characteristics obtained by optimization calculation to serve as a training sample for a micro-terrain recognition model. The optimized feature factors and their combinations of the present invention can effectively distinguish micro-terrain, make the boundaries of various types of terrain clearer, and quickly identify micro-terrain.

[0096] In one embodiment, step S101 is to find the optimal calculation window of the terrain position index based on the mean change point analysis method, and obtain the terrain position index feature of the target sample based on the optimal calculation window. Specifically:

[0097] The calculation formula of the terrain position index TPI is:

[0098]

[0099] Z i is the elevation value of point i, is the average value of all elevation points within the neighborhood range of point i.

[0100] Calculate the range value of the terrain position index under each window.

[0101] Based on the range value of the terrain position index and the window size, obtain the range value of the terrain position index per unit area.

[0102] Based on the range values of the terrain position index corresponding to each window, form a data sequence; the window can be selected from 20 pixels × 20 pixels to 70 pixels × 70 pixels.

[0103] Calculate the variance of the data sequence. The calculation formula is:

[0104]

[0105] In the formula, is the average value of the data sequence {x i}, i = 1, 2, 3,..., N, and N is the number of windows used.

[0106] Select x i in turn as the truncation point to divide the data sequence into two subsequences, and 1 < t < N; calculate the sum of variances S i of the two subsequences.

[0107]

[0108] In the formula, and are the arithmetic means of the subsequences and respectively.

[0109] Calculate the difference between S and Si to obtain the result sequence {S - S i}; the window corresponding to the truncation point x i selected as the maximum value in the result sequence is the optimal calculation window.

[0110] The terrain position index feature of the target sample is obtained based on the optimal calculation window.

[0111] In this embodiment, the range value of the terrain position index can be the difference between the maximum and minimum TPI values ​​within the window range. The TPI range per unit area can be obtained by dividing the TPI range by the corresponding window area. The logarithm of the range value of the terrain position index corresponding to each window can be taken to form a data sequence.

[0112] In one embodiment, step S102, based on the average elevation deviation counted in the neighborhood window of the elevation points in the target sample, obtain the neighborhood window size when the average elevation deviation reaches a preset threshold, and calculate the profile curvature feature based on the obtained neighborhood window size, including:

[0113] Calculate the average elevation deviation DEV in each neighborhood window in turn:

[0114]

[0115]

[0116] In the formula, z0 is the elevation value of a certain point; is the average elevation in the neighborhood; SD is the standard deviation of the elevation in the neighborhood; n is the number of grid points in the neighborhood; z i is a grid point in the neighborhood;

[0117] When the average elevation deviation DEV reaches a preset threshold, the profile curvature feature is calculated based on the neighborhood window size corresponding to the threshold.

[0118] Among them, the average elevation deviation DEV can be calculated by calculating the neighborhood window size from 15×15 pixels to 3×3 pixels in order from large to small. In this embodiment, the preset threshold can be obtained based on experimental data and empirical data. When the average elevation deviation DEV reaches the preset threshold, the neighborhood window size corresponding to the preset threshold can be selected to calculate the profile curvature feature. When the average elevation deviation DEV is greater than the preset threshold, the neighborhood window size corresponding to the DEV greater than the preset threshold can also be used to calculate the profile curvature feature.

[0119] In one embodiment, the method further includes: respectively calculating the slope characteristics, the undulation characteristics and the elevation variation coefficient characteristics of the target sample.

[0120] It should be noted that the slope characteristics, undulation characteristics and elevation variation coefficient characteristics of the target sample calculated in the present invention can be calculated according to the existing slope, undulation and elevation variation coefficient calculation methods. In addition, other characteristic data can also be calculated, and this embodiment is only an exemplary description.

[0121] In one embodiment, it also includes: constructing a micro-topography classification decision table based on the terrain position index feature, the profile curvature feature, the elevation variation coefficient feature, the slope feature and the undulation feature.

[0122] In this embodiment, a micro-terrain classification decision table can be constructed based on the terrain position index feature, profile curvature feature, elevation variation coefficient feature, slope feature, and undulation feature. There are many ways to construct the micro-terrain classification decision table. For example, the decision method of the terrain can be set manually for different micro-terrains based on experience; or the final micro-terrain classification decision table can be determined by selecting the decision method with the best classification effect through a large number of experiments.

[0123] In this embodiment, the optimal large TPI features, profile curvature features and elevation variation coefficients are selected as the classification decision feature factors for the two types of micro-topography, namely, the pass type and the canyon wind channel type; the optimal large TPI features, profile curvature features, slope features and undulation features are selected as the classification decision feature factors for the three types of micro-topography, namely, the alpine watershed type, the terrain uplift type and the atypical micro-topography. Specifically, taking the classification threshold of the optimal large TPI as an example, the standardized TPI usually uses -1 and 1 as the classification threshold, which is usually used for the identification of mountain micro-topography. The TPI classification threshold is set to -0.8 and 0.8, which is more suitable for the transmission line micro-topography identification of the present invention, and the identification result is more accurate. The thresholds of other characteristic factors can also be set in combination with experiments and personal experience. The classification decision table for the extraction of pass-type and canyon wind channel type features can be shown in Table 1, and the decision table for the alpine watershed type, the terrain uplift type and the atypical micro-topography can be shown in Table 2, as follows:

[0124] Table 1 Classification decision table of pass type and canyon wind channel type

[0125]

[0126] Table 2 Decision table for classification of other types of micro-terrain

[0127]

[0128] Step S103, constructing a sample feature data set based at least on the terrain position index feature and the profile curvature feature, comprising:

[0129] S1031. Based on the characteristic values ​​of the terrain position index feature, the profile curvature feature, the slope feature, the undulation feature, and the elevation variation coefficient feature of the target sample, obtain the classification label of the target sample according to the micro-topography classification decision table.

[0130] S1032, constructing a sample point based on the characteristic values ​​of the terrain position index feature, the profile curvature feature, the slope feature, the undulation feature, the elevation variation coefficient feature and the classification label corresponding to the target sample, and constructing a sample feature data set from the plurality of sample points.

[0131] In one embodiment, a sample data may be constructed as:

[0132] P = (large TPI, slope, undulation, profile curvature, elevation variation coefficient, classification label), where the sample labels can be divided into five types: four typical micro-topography and one atypical micro-topography. The four typical micro-topography types are: high mountain watershed type, pass type, canyon wind channel type and terrain uplift type.

[0133] In this embodiment, the classification label of the target sample can be obtained based on the constructed classification decision table. For example, the classification label of the target sample can be obtained according to the characteristic values ​​of the terrain position index characteristics, profile curvature characteristics, slope characteristics, undulation characteristics, and elevation variation coefficient characteristics according to the micro-topography classification decision table in Tables 1 and 2 above.

[0134] If the obtained sample feature data set categories are unbalanced, the downsampling method can be used to ensure the balance of the categories. Table 3 shows some sample points of the feature sample set. When dividing the data set, 70% of the data is used as a training set for training the model, and 30% of the data is used as a test set for evaluating the model.

[0135] Table 3 Sample point example table

[0136]

[0137] In one embodiment, the method further includes: constructing a parallelized random forest model, including:

[0138] S201, constructing several self-sampling parallel decision trees, and allocating a sample feature data set required for training to each decision tree.

[0139] In this embodiment, a kernel function can be constructed to realize GPU parallelization of the random forest model, and the GPU parallelization of the model can be realized through three CUDA kernel functions.

[0140] The kernel-1 function can be used to implement the parallelization of bootstrap sampling, and the training set sample data required for training is allocated to each decision tree. Bootstrap sampling is the basic step for the random forest model to generate training sets for each decision tree.

[0141] Specifically, the original training set data can be copied from the host memory to the global memory of the GPU to ensure that the GPU can quickly access this data; then, a thread block can be assigned to each decision tree, and the threads in each thread block jointly generate a training set for a decision tree to simultaneously process the generation of training sets for multiple decision trees; in addition, within each thread block, the threads perform sampling operations with replacement in parallel, randomly selecting samples from the original training set to construct a self-service sample set. These sample sets are used for subsequent decision tree training.

[0142] S202: Parallel training of a plurality of the decision trees, wherein each decision tree is constructed in parallel by the allocated threads.

[0143] This embodiment can use kernel-2 function to implement parallel training of decision trees, and each decision tree is constructed in parallel by its own assigned thread. The construction of decision trees is the core part of random forest model training. Since the training process of each decision tree is independent of each other, it is suitable for parallel processing.

[0144] Specifically, the root node of the decision tree can be initialized for each thread block according to the assigned self-help sample set. This process includes calculating the initial sample distribution and determining the initial state of the root node. At each node of the decision tree, multiple threads calculate all possible split points in parallel and select the best split point. The selection of the split point is based on certain criteria (such as the Gini index or information gain) that measure the improvement of the purity of the node after the split. Child nodes can be generated according to the best split point, and the tree structure of each child node is recursively constructed. This process continues until all nodes become leaf nodes or the preset stop condition is reached.

[0145] S203, parallelization of classification prediction, parallel prediction of the micro-topography categories of the sample points to be classified, and obtaining the final micro-topography categories of the sample points by voting.

[0146] In this embodiment, the kernel-3 function can be used to implement parallel classification prediction, and the micro-terrain category of the points to be classified can be predicted in parallel, and finally the final transmission line micro-terrain category is obtained by voting. The schematic diagram of the parallel random forest model in this embodiment is shown in FIG. Figure 4 shown.

[0147] Specifically, a thread can be assigned to each sample to be predicted, and each thread is independently responsible for predicting a sample. This allocation method ensures that all samples can be processed at the same time; and each thread traverses all decision trees and summarizes the prediction results of each tree. Through parallel execution, each thread completes the prediction of all trees in a short time; voting is performed based on the prediction results of all trees, and the classification category of each sample is ultimately determined by the voting results. The majority voting method counts the number of votes for each category and selects the category with the most votes as the final prediction result.

[0148] In one embodiment, step S105, training a micro-terrain recognition model based on the sample feature data, includes:

[0149] S1051, selecting a training set from the sample feature data set;

[0150] S1052, randomly extracting samples with replacement from the training set to construct multiple sub-training sets;

[0151] A part of the samples can be selected from the sample feature set as a training set and a part of the samples can be selected as a test set to train and test the micro-terrain recognition model.

[0152] For the parallelized random forest model that needs to be trained in this embodiment, samples can be randomly extracted with replacement (Bootstrap Sampling) from the training set to construct multiple sub-training sets for training each decision tree separately. Some samples may appear multiple times in a sub-training set, while some samples may be ignored.

[0153] S1053. Randomly select features from each sub-training set to train each decision tree.

[0154] It should be noted that you can randomly select a portion of the features from the sub-training set to build a decision tree instead of using all the features, so as to increase the diversity of the model and reduce the risk of overfitting.

[0155] In addition, features can be randomly selected from each sub-training set to train the decision tree separately. In the process of building the decision tree, the Gini impurity indicator is used to select the best features for node splitting until the stopping condition is reached (such as reaching the maximum depth max_depth or reaching the minimum number of node samples min_samples_leaf). Among them, the Gini impurity of a node can be calculated by the following formula:

[0156]

[0157] Among them, p i is the sample proportion of the i-th category, and C is the number of categories.

[0158] In this embodiment, the number of decision trees n_estimators of the parallel random forest model is set to 300.

[0159] In one embodiment, the method further includes: evaluating the recognition effect of the parallel random forest model.

[0160] The estimation of the recognition effect of the parallel random forest model includes:

[0161] The recognition effect of the parallel random forest model is evaluated based on the recognition indicators of the parallel random forest model; the recognition indicators include one or more of the confusion matrix, classification accuracy, Kappa coefficient, single category precision Precision and recall Recall of the parallel random forest model.

[0162] In this embodiment, the recognition effect of the parallel random forest model can be evaluated by one or more of confusion matrix, classification accuracy, Kappa coefficient, single category precision Precision and recall Recall.

[0163] Among them, the confusion matrix is ​​a two-dimensional matrix that analyzes the classification results predicted by the model. It is an intuitive representation of the difference between the actual categories and the categories predicted by the classification model.

[0164] The classification accuracy of the model can be expressed as:

[0165]

[0166] In the formula, C ii Represents the sum of the number of samples correctly classified in each category, there are N categories in total; Indicates the total number of samples.

[0167] The Kappa coefficient is a statistical indicator used to measure the consistency and accuracy of classification tasks. It evaluates the performance of a classification model by comparing the consistency between the model's predictions and the actual observations. The Kappa coefficient can be expressed as:

[0168]

[0169] Where r is the number of micro-topography categories; M is the total number of sample points involved in the calculation; x ii is the number of pixels in the ith row and ith column of the confusion matrix (i.e., the number of correctly classified pixels); x i+ 、x +i are the total number of pixels in the i-th row and i-th column respectively.

[0170] Calculate the precision Precision and recall Recall of a single category, which can be expressed as:

[0171]

[0172]

[0173] For a specific category i, TP is the number of samples that are correctly classified into this category, FP is the number of samples that are misclassified into other categories, and FN is the number of samples that are misclassified into other categories.

[0174] In this embodiment, a total of 40960 sample points are selected to form a sample feature data set, and the number of samples in each category is 8192. 70% of the sample feature data set is selected as a training set, and the remaining 30% is selected as a test set. The evaluation results of the parallel random forest model are shown in Tables 4 and 5.

[0175] Table 4. Test set classification confusion matrix

[0176]

[0177] Table 5 Classification performance evaluation

[0178]

[0179] In addition, the recognition efficiency of the parallel random forest model can also be evaluated. Specifically, a large number of points to be identified can be selected from a large range of transmission line areas, and these points to be classified can be identified using the current method (CPU) and the parallel random forest model. The recognition efficiency of the parallel random forest model can be evaluated by comparing the time consumption of the two. The results show that the recognition efficiency of the GPU parallel random forest model is significantly improved.

[0180] In the embodiment, 49 lines in the study area were selected for the experiment, with a total length of 1224.53 km. Then, 5×10 5 , 1×10 6 , 5×10 6 , 1×10 7 , 5×10 7 Points to be identified are identified using the current method (CPU) and the GPU parallel random forest model. The time consumption (in seconds) of the two methods for identifying points to be classified is shown in Table 6 below.

[0181] Table 6 Comparison of recognition efficiency of CPU serial and GPU parallel random forest models

[0182]

[0183] Based on the method of the embodiment of the present invention, a parallel random forest model for large-scale transmission line micro-topography identification can be obtained. For areas where transmission line micro-topography needs to be identified, various characteristic values ​​of samples are calculated based on raster DEM data to form a data set, and the corresponding data set is input into the model to quickly realize intelligent identification of transmission line micro-topography in a large area.

[0184] In the embodiment, the effects of the conventional method and the method of the present invention in identifying two types of micro-topography, namely, pass and canyon, are compared. Figure 6 As shown, Figure 6The left side of the middle diagram is a schematic diagram of the micro-topography effect of the traditional method for identifying passes and canyons. Figure 6 The right side view is a micro-topography effect map obtained by using the method of the present invention. Six of the 49 transmission lines in the study area were selected. The micro-topography recognition results obtained based on the model of the present invention are shown in Table 7 below. Figure 7 shown.

[0185] Table 7 Statistics of transmission line micro-topography extraction results

[0186]

[0187] The micro-terrain recognition model used in the embodiment of the present invention can effectively solve the problem of incomplete classification and can greatly improve the recognition efficiency of micro-terrain; and by optimizing the terrain position index characteristics and profile curvature characteristics, and through the decision table, it can quickly and effectively identify the pass-type and canyon wind channel-type micro-terrain, and can make the category boundary contour clearer.

[0188] Figure 8 A micro-terrain recognition device provided in an embodiment of the present invention includes:

[0189] A first feature acquisition unit 301 is used to find an optimal calculation window of the terrain position index based on a mean change point analysis method, and acquire the terrain position index feature of the target sample based on the optimal calculation window;

[0190] The second feature acquisition unit 302 is used to acquire the neighborhood window size when the average elevation deviation reaches a preset threshold based on the statistical average elevation deviation in the neighborhood window of the elevation points in the target sample, and calculate the profile curvature feature based on the acquired neighborhood window size;

[0191] A data set construction unit 303, configured to construct a sample feature data set based at least on the terrain position index feature and the profile curvature feature;

[0192] A training unit 304 is used to train a micro-terrain recognition model based on the sample feature data;

[0193] The identification unit 305 is used to input the elevation data of the area to be measured into the trained micro-topography identification model to identify the micro-topography of the area to be measured.

[0194] The message data processing resource scheduling device provided in the embodiment of the present invention can execute the message data processing resource scheduling device method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0195] Fig. 9A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0196] like Fig. 9 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0197] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0198] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a micro-terrain recognition method.

[0199] In some embodiments, a micro-terrain recognition method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the micro-terrain recognition method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform a micro-terrain recognition method in any other appropriate manner (e.g., by means of firmware).

[0200] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0201] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0202] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0203] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0204] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0205] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0206] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0207] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A micro-topography recognition method, characterized in that: include: Finding an optimal calculation window for the terrain position index based on the mean change point analysis method, and obtaining the terrain position index characteristics of the target sample based on the optimal calculation window; Based on the statistical average elevation deviation in the neighborhood window of the elevation points in the target sample, the neighborhood window size when the average elevation deviation reaches a preset threshold is obtained, and the profile curvature feature is calculated based on the obtained neighborhood window size; constructing a sample feature data set based at least on the terrain position index feature and the profile curvature feature; Training a micro-terrain recognition model based on the sample feature data; The elevation data of the area to be measured is input into the trained micro-topography recognition model to identify the micro-topography of the area to be measured.

2. The micro-topography recognition method according to claim 1, characterized in that: The method of finding the best calculation window of the terrain position index based on the mean change point analysis method and obtaining the terrain position index feature of the target sample based on the best calculation window includes: The calculation formula of the terrain position index TPI is: Z i is the elevation value of point i, is the average value of all elevation points within the neighborhood of point i; Calculating the range value of the terrain position index under each of the windows; Based on the range value of the terrain position index and the window size, obtaining the range value of the terrain position index per unit area; Constructing a data sequence based on the range values ​​of the terrain position index corresponding to each of the windows; Calculate the variance of the data sequence using the following formula: In the formula, For the data sequence {x i }, i = 1, 2, 3, ..., N, N is the number of windows used; Select x sequentially i , and divide the data sequence into two subsequences using it as the truncation point and 1 < t < N; calculate the sum S of the variances of the two subsequences i :[[]] In the formula, and Subsequences and The arithmetic mean of In this way, the difference between S and Si is calculated, and the result sequence {SS i }; The maximum value in the result sequence is selected as the truncation point x i The corresponding window is the optimal calculation window; The terrain position index feature of the target sample is obtained based on the optimal calculation window.

3. The micro-topography recognition method according to claim 1, characterized in that: The method comprises: obtaining a neighborhood window size when the average elevation deviation reaches a preset threshold based on the statistical average elevation deviation in the neighborhood window of the elevation point in the target sample; and calculating a profile curvature feature based on the obtained neighborhood window size, including: Calculate the average elevation deviation DEV in each neighborhood window of the elevation point in turn: In the formula, z0 is the elevation value of a certain point; is the average elevation in the neighborhood; SD is the standard deviation of the elevation in the neighborhood; n is the number of grid points in the neighborhood; z i is a grid point in the neighborhood; When the average elevation deviation DEV reaches a preset threshold, the profile curvature feature is calculated based on the neighborhood window size corresponding to the threshold.

4. The micro-topography recognition method according to claim 1, characterized in that: Before constructing the sample feature data set based at least on the terrain position index feature and the profile curvature feature, the method further includes: Calculating the slope characteristics, undulation characteristics and elevation variation coefficient characteristics of the target sample respectively; Constructing a micro-topography classification decision table based on the terrain position index feature, the profile curvature feature, the elevation variation coefficient feature, the slope feature, and the undulation feature; The step of constructing a sample feature data set based at least on the terrain position index feature and the profile curvature feature comprises: Based on the characteristic values ​​of the terrain position index feature, the profile curvature feature, the slope feature, the undulation feature, and the elevation variation coefficient feature of the target sample, a classification label of the target sample is obtained according to the micro-topography classification decision table; A sample point is constructed based on the characteristic values ​​of the terrain position index feature, the profile curvature feature, the slope feature, the undulation feature, the elevation variation coefficient feature and the classification label corresponding to the target sample, and a sample feature data set is constructed from several sample points.

5. The micro-topography recognition method according to claim 1, characterized in that: The method also includes: Construct a parallelized random forest model; The parallelized random forest model construction includes: Construct several parallel decision trees with self-service sampling, and assign the sample feature data set required for training to each decision tree; Parallel training of a plurality of the decision trees, each decision tree being constructed in parallel by the assigned threads; Parallelization of classification prediction: parallel prediction of the micro-topography categories of the sample points to be classified, and obtaining the final micro-topography categories of the sample points by voting.

6. The micro-topography recognition method according to claim 5, characterized in that: The step of training the micro-terrain recognition model based on the sample feature data comprises: Selecting a training set from the sample feature data set; Randomly extract samples with replacement from the training set to construct multiple sub-training sets; Features are randomly selected from each sub-training set to train each decision tree.

7. The micro-topography recognition method according to claim 5, characterized in that: Also includes: Evaluate the recognition effect of the parallel random forest model; The evaluating the recognition effect of the parallel random forest model includes: Evaluating the recognition effect of the parallel random forest model based on the recognition index of the parallel random forest model; The identification index includes one or more of the confusion matrix, classification accuracy, Kappa coefficient, precision of a single category, and recall rate of the parallel random forest model.

8. A micro-topography recognition device, characterized in that: include: A first feature acquisition unit is used to find an optimal calculation window of the terrain position index based on a mean change point analysis method, and acquire the terrain position index feature of the target sample based on the optimal calculation window; A second feature acquisition unit is used to acquire the neighborhood window size when the average elevation deviation reaches a preset threshold based on the statistical average elevation deviation in the neighborhood window of the elevation points in the target sample, and calculate the profile curvature feature based on the acquired neighborhood window size; A data set construction unit, configured to construct a sample feature data set based at least on the terrain position index feature and the profile curvature feature; A training unit, used for training a micro-terrain recognition model based on the sample feature data; The recognition unit is used to input the elevation data of the area to be measured into the trained micro-topography recognition model to recognize the micro-topography of the area to be measured.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the micro-terrain recognition method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the micro-topography recognition method according to any one of claims 1 to 7 when executed.

Citation Information

Patent Citations

  • Terrainfeature point extraction method based on double TPI parameters

    CN109840447A

  • Power transmission line easy-icing micro-terrain classification method

    CN113688903A

  • Special dangerous terrain recognition method and apparatus for forest fire

    US20230342962A1

Cited By

  • DEM window optimization selection method for power grid microtopography recognition

    CN119622420A

  • Complex mountain landform intelligent classification method and system based on dual-scale TPI optimization

    CN120705669A