1: 2000 topographic map terrain recognition and extraction method based on deep learning

Abstract

The invention relates to a 1: 2000 topographic map terrain recognition and extraction method based on deep learning, and relates to the technical field of geographic information, and the method comprises the steps: obtaining a first terrain symbol of a 1: 2000 topographic map in response to a delineated region input by a user side; obtaining a first ground feature type matching result; when the first ground feature type matching result is empty, obtaining a second ground feature type matching result; when the maximum size of the ground feature symbol size identifier is greater than or equal to 0.09 cm, carrying out size reduction to obtain the actual size of the ground feature; when the maximum size of the ground feature symbol size identifier is smaller than 0.09 cm, marking the actual size of the ground feature as empty; and sending the second ground object type matching result, the actual ground object size and the ground object symbol position identifier to the user side. According to the method and the device, the technical problem that the accuracy and the adaptability of existing topographic map terrain recognition and extraction are insufficient is solved, and the technical effect of improving the accuracy and the applicability of topographic map terrain recognition and information extraction is achieved.

Description

Technical Field

[0001] The present invention relates to the field of geographic information technology, and in particular to a method for identifying and extracting features in a 1:2000 topographic map based on deep learning. Background Art

[0002] With the continuous development of geographic information systems, unmanned recognition of terrain map features has been widely used in the fields of traffic navigation, map cultural display, etc. In the prior art, terrain map feature recognition usually adopts a method based on map annotation symbols, that is, a database of annotation symbols and feature information is pre-set, and the map annotation symbols are extracted and combined with the associated database to realize the recognition and extraction of feature type, distribution, size and other information.

[0003] However, the above-mentioned existing technical solutions still have the following problems: First, there are many types of topographic maps, and there are large differences in the representation methods of topographic maps from different sources and different periods, which leads to the insufficient applicability of the recognition method based on a single associated database. Second, even for annotation symbols that represent the same meaning of land objects, there may be slight differences in their size and shape. Such non-standardized symbols are difficult to be accurately matched and recognized by the associated database. Third, due to the establishment cost and update efficiency of the associated database, it is difficult for the existing solutions to cover all possible annotation symbols, and it is impossible to achieve recognition and information extraction when facing unknown symbols. The above problems lead to the lack of accuracy and adaptability of the existing solutions in the recognition and extraction of land objects in topographic maps, and it is difficult to meet the actual application needs of unmanned recognition of land objects in topographic maps. Summary of the invention

[0004] The present invention aims to solve the technical problems of insufficient accuracy and adaptability of existing topographic map feature recognition and extraction methods due to the various types of topographic maps, difficulty in recognizing non-standardized annotation symbols, and inability to recognize unknown symbols. A 1:2000 topographic map feature recognition and extraction method based on deep learning is provided to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides a method for identifying and extracting features in a 1:2000 topographic map based on deep learning, comprising: obtaining a first feature symbol of a 1:2000 topographic map in response to a circled area input by a user terminal, wherein the first feature symbol has a feature symbol size identifier and a feature symbol position identifier; processing the first feature symbol through a symbol description twin network in combination with a symbol description database to obtain a first feature type matching result; when the first feature type matching result is empty, performing adjacent symbol analysis on the first feature symbol in combination with a topographic map sharing platform to obtain a second feature type matching result; when the maximum size of the feature symbol size identifier is greater than or equal to 0.09 cm, restoring the size according to a 1:2000 scale to obtain the actual size of the feature; when the maximum size of the feature symbol size identifier is less than 0.09 cm, marking the actual size of the feature as empty; and sending the second feature type matching result, the actual size of the feature and the feature symbol position identifier to the user terminal.

[0007] The beneficial effects of the present invention are:

[0008] In response to the circled area input by the user, the first land feature symbol of the 1:2000 topographic map is obtained, wherein the first land feature symbol has a land feature symbol size identifier and a land feature symbol position identifier, and the map annotation symbol is obtained from the area circled by the user as the object to be identified, and its size and position information are extracted to provide input data for subsequent identification and information extraction; the first land feature symbol is processed through the symbol description twin network in combination with the symbol description database to obtain the first land feature type matching result, the symbol description twin network is introduced, and the annotation symbol is feature learned and matched by the deep learning algorithm, and the preliminary identification of the land feature type is realized in combination with the symbol description database; when the first land feature type matching result is empty, the first land feature symbol is analyzed for adjacent symbols in combination with the topographic map sharing platform, The second feature type matching result is obtained, and the feature symbols that have not been successfully matched are matched again through adjacent symbol analysis to improve the recognition ability of non-standardized and unknown symbols; when the maximum size of the feature symbol size mark is greater than or equal to 0.09 cm, the size is restored according to the scale of 1:2000 to obtain the actual size of the feature; when the maximum size of the feature symbol size mark is less than 0.09 cm, the actual size of the feature is marked as empty, and the symbol size is restored and calculated according to the map scale to obtain the actual size of the real feature, and the features that are too small are filtered out to improve the accuracy of information extraction; the second feature type matching result, the actual size of the feature and the feature symbol position mark are sent to the user end, thereby improving the accuracy and applicability of topographic map feature recognition and extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1A schematic diagram of the process of the 1:2000 topographic map feature recognition and extraction method based on deep learning provided by the present invention;

[0010] Figure 2 A schematic diagram of the structure of an electronic device provided by the present invention;

[0011] Figure 3 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0012] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0013] Electronic device 100 , memory 110 , processor 120 , first computer program 111 , computer-readable storage medium 200 , second computer program 211 . DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0015] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0016] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0017] Embodiment 1:

[0018] like Figure 1As shown, the embodiment of the present invention provides a 1:2000 topographic map feature recognition and extraction method based on deep learning, including:

[0019] S100: In response to a delimited area input by a user terminal, a first land feature symbol of a 1:2000 topographic map is obtained, wherein the first land feature symbol has a land feature symbol size identifier and a land feature symbol position identifier.

[0020] Specifically, the user can open the image browsing interface storing the 1:2000 topographic map to be identified through a user-end device, such as a smart phone, tablet computer or PC, and then circle the area of ​​interest on the image browsing interface manually or through other human-computer interaction methods to obtain a circled area. Subsequently, in response to the user's circled area, the image content in the circled area is used as the target data for subsequent identification. Next, the feature symbols are obtained from the circled area in sequence, and the feature symbols obtained each time are used as the first feature symbols. The so-called feature symbol refers to the symbolic expression of various types of feature entities (such as buildings, roads, water systems, etc.) in the 1:2000 topographic map using specific graphics. Different types of features are usually represented by different symbols.

[0021] When extracting the first feature symbol, two attributes associated with it are also obtained, namely, the feature symbol size identifier and the feature symbol position identifier. Among them, the feature symbol size identifier represents the size information of the symbol in the graphic, and the feature symbol position identifier represents the spatial position information of the symbol. These two attribute information will be used to restore the actual size of the feature and determine its spatial position in the subsequent steps.

[0022] By responding to the user's circle input, the first feature symbol representing the feature entity is obtained from the target area of ​​the 1:2000 topographic map, and the size and position attribute information related to the symbol is extracted, preparing for the subsequent further identification of the feature type represented by the symbol and obtaining the real information of the feature.

[0023] S200: Processing the first land feature symbol through the symbol description twin network in combination with the symbol description database to obtain a first land feature type matching result.

[0024] Specifically, since different 1:2000 topographic maps may be drawn by different cartographers, even if they represent the same type of landform, the shapes and colors of the landform symbols used may be different. In order to improve the accuracy and robustness of landform symbol recognition, the first landform symbol is processed by the symbol description twin network.

[0025] The so-called twin network is a neural network structure that contains two sub-networks with exactly the same structure and parameters. For example, the symbol description twin network is set to include two parallel feature extraction channels and a post-feature comparison channel. The first feature symbol to be identified and the reference feature symbol randomly selected from the symbol description database are input into the two feature extraction channels respectively, and their respective shape and color features are extracted. The feature comparison channel then calculates the similarity of the two sets of features. If the shape and color similarities exceed the preset threshold, the category of the reference feature symbol and the first feature symbol is used as the recognition result, that is, the first feature type matching result.

[0026] Among them, the symbol description database is a data set that stores the shapes, colors and corresponding feature category annotation information of various standard feature symbols. By continuously searching the symbol samples in the symbol description database, the one that is most similar to the first feature symbol can be found, so as to know the feature type represented by the first feature symbol. If no matching item that meets the requirements of shape and color similarity is found after traversing the symbol description database, the first feature type matching result is set to empty, indicating that the semantic category of the first feature symbol cannot be directly identified at present.

[0027] By utilizing the symbol description twin network, the reference symbol with the most similar shape and color is matched from the symbol description database, and the feature type corresponding to the first feature symbol is obtained as the first feature type matching result. This can effectively overcome the problem of symbol expression differences in different topographic maps and improve the generalization ability of feature symbol recognition.

[0028] S300: When the first feature type matching result is empty, combining with the topographic map sharing platform, performing adjacent symbol analysis on the first feature symbol to obtain a second feature type matching result.

[0029] Specifically, if the first feature type matching result is empty after being processed by the symbol description twin network, it means that the feature category to which the feature symbol belongs cannot be determined based on the symbol shape and color alone. At this time, the first feature symbol is analyzed for adjacent symbols in combination with the topographic map sharing platform to obtain the second feature type matching result.

[0030] Since 1:2000 topographic maps usually cover landform information within a certain area, the types of features represented by feature symbols at adjacent locations in the map usually have a certain correlation. For example, there are usually bridges near rivers, and there are usually roads around buildings. Based on this, the contextual semantic information formed by the adjacent symbols can be used to indirectly infer the feature symbols that cannot be directly identified at present.

[0031] To achieve the above goals, a topographic map sharing platform is introduced. The topographic map sharing platform is a topographic map data sharing and service platform built based on cloud technology. It stores a large amount of 1:2000 topographic map data drawn in different regions and at different times, and properly classifies and organizes these data and semantically annotates them. Through learning and analyzing massive topographic map data, the topographic map sharing platform has accumulated prior knowledge about the spatial association laws between different types of landforms.

[0032] In specific implementation, the first feature symbol is submitted to the topographic map sharing platform, and several adjacent symbols with similar shapes and colors to the first feature symbol are retrieved in the topographic map sharing platform, and the feature type information of the adjacent symbols is summarized and counted, and the feature type with the highest frequency is taken as the second feature type matching result. For example, if the symbol A to be identified is similar to symbols B, C, and D in the topographic map sharing platform, and the most common feature in the neighborhood of these three symbols is the "road" type feature, it can be inferred that symbol A is also likely to represent a road.

[0033] By making full use of the topographic map sharing platform, the symbolic semantic information around the first land feature symbol is analyzed and mined, and then a reasonable inference is made about the land feature type of the symbol itself, which solves the problem of accurate identification based on the symbol itself and improves the accuracy of land feature symbol recognition as a whole.

[0034] S400: When the maximum size of the ground feature symbol size mark is greater than or equal to 0.09 cm, the size is restored according to a scale of 1:2000 to obtain the actual size of the ground feature.

[0035] Specifically, in a 1:2000 topographic map, the symbols of various features are usually drawn at a certain scale so that they can be clearly expressed on the map. However, in the application stage, people often pay more attention to the actual size information of the features. Therefore, it is necessary to restore the symbol size to the actual size according to the drawing scale of the topographic map.

[0036] At a scale of 1:2000, a length of 1 cm on the map represents an actual distance of 20 meters in the field, that is, the ratio between the map size and the actual size is 1:2000. Combined with practical experience, 0.09 cm is used as the judgment threshold, that is, when the maximum size of the feature symbol size (referring to the maximum value of the length, width, height and other geometric dimensions of the symbol) is greater than or equal to 0.09 cm, it can be considered that the real feature size corresponding to the symbol is large enough and worth restoring; when the maximum size is less than 0.09 cm, it can be ignored and no restoration is performed. In this way, a balance can be achieved between the integrity of the restored information and the efficiency of calculation. For example, if the width of the first feature symbol to be processed is 0.13 cm and the height is 0.09 cm, its maximum size (0.13 cm) exceeds the judgment threshold and needs to be restored. Multiply 0.13 by 2000 to obtain the actual width of the feature, which is 260 cm, or 2.6 meters. Since the height size is equal to the threshold, it can also be restored to 1.8 meters according to the above method.

[0037] By multiplying the symbols of features whose sizes are greater than a certain threshold by the inverse of the scale, the actual size of the features can be calculated and restored. On the one hand, the real geometric information of the features contained in the symbols can be accurately obtained. On the other hand, by setting the threshold, the slight errors that may exist due to the limitation of drawing accuracy can be eliminated, making the restoration result more reliable.

[0038] S500: When the maximum size of the ground feature symbol size identifier is less than 0.09 cm, the actual size of the ground feature is marked as empty.

[0039] Specifically, as mentioned above, 0.09 cm is set as the threshold for determining whether the size of a feature symbol is restored. When the maximum size of a feature symbol is smaller than the threshold, it can be considered that the real feature corresponding to the symbol is either very small or difficult to distinguish from adjacent features. In this case, calculating and restoring its real size is meaningless. Therefore, the actual size information of such feature symbols is directly marked as a null value and no restoration is performed.

[0040] For example, if the width of the feature symbol to be processed is 0.08 cm and the height is 0.06 cm, then its maximum size (0.08 cm) is less than the judgment threshold of 0.09 cm, and the actual width and height of the feature symbol are marked as empty. The mark as empty can be represented by a specific character such as "null" or a conventional value such as "-1", as long as it can be distinguished from a valid size value.

[0041] For the sake of reliability and practicality, for ground feature symbols that are too small, we do not restore them to their actual size, but only mark them with empty values. This can effectively balance information quality and processing efficiency and better support subsequent applications.

[0042] S600: Sending the second feature type matching result, the actual size of the feature and the feature symbol position identifier to the user terminal.

[0043] Specifically, the second feature type matching result represents the real feature type represented by the first feature symbol, the feature actual size reflects the real geometric size of the feature, and the feature symbol position identifier indicates its spatial position. These three items constitute a comprehensive and accurate semantic description of a feature symbol, including key information such as feature type, size, and location that users are concerned about.

[0044] The above three items obtained by identification and extraction are packaged and sent to the user end that initiated the identification request in real time through network transmission. On the user side, the received semantic information of the land object can be presented to the user through a specially designed information display interface. For example, in the topographic map browsing interface, the land object symbol is covered or side-marked with the identified land object type text; the length, width, height and other size information of the land object is displayed in the form of pop-up windows, dialog boxes, etc.; the center point position of the land object symbol is accurately marked in the spatial coordinate system of the drawing, and so on. Users can combine the recognition results to intuitively and accurately grasp the semantic connotation of the original topographic map, greatly improving the readability and usability of the graphic information. Furthermore, in some application scenarios, the user end can also convert the received land object semantic information into other common data formats, such as XML, JSON, etc., to facilitate sharing and pushing between different systems and platforms.

[0045] By sending the results of feature symbol recognition and size restoration to the user end through the network and displaying them, the technical effect of improving the accuracy and applicability of topographic map feature recognition and information extraction is achieved.

[0046] Furthermore, the embodiment of the present application also includes:

[0047] S110: Segmenting the 1:2000 topographic map according to the demarcated area to obtain a segmented digital image;

[0048] S120: performing semantic segmentation on the segmented digital image through a multi-layer convolutional symbol detection model, and extracting a list of ground object symbols, a list of ground object symbol size identifiers, and a list of ground object symbol position identifiers;

[0049] S130: The first geographical feature symbol belongs to the geographical feature symbol list, the geographical feature symbol size identifier belongs to the geographical feature symbol size identifier list, and the geographical feature symbol position identifier belongs to the geographical feature symbol position identifier list.

[0050] In a feasible implementation, the specific implementation of S100 can be further refined into steps S110 to S130, so as to illustrate how to obtain the first land feature symbol and its related attributes from the 1:2000 topographic map according to the area defined by the user.

[0051] First, the area of ​​interest that the user has circled manually or in other human-computer interaction modes on the terminal device is obtained, that is, the circled area. Then, the image segmentation operation is performed on the complete 1:2000 topographic map with the circled area as the cutting boundary to obtain a sub-image, which is called a segmented digital image. The segmented digital image only contains the topographic map information within the circled area, which reduces the data size of the subsequent recognition processing. Then, a multi-layer convolutional neural network model based on deep learning is used to detect and identify the ground object symbols in the segmented digital image. The multi-layer convolutional neural network model takes the segmented digital image as input and accurately segments the graphic area of ​​each ground object symbol from the complex image background through a series of operations such as convolution, pooling, and activation. Furthermore, the model can also calculate the geometric size parameters (such as length, width, and height) of each symbol area, as well as the spatial position parameters in the image coordinate system. Finally, the model outputs three lists: a ground object symbol list, a ground object symbol size identification list, and a ground object symbol position identification list, which respectively store the detected ground object symbol graphics, corresponding size information, and position information, in preparation for subsequent processing.

[0052] Any feature symbol in the obtained feature symbol list can be used as the first feature symbol for subsequent processing. The first feature symbol can be specified by the user, or can be selected by the system randomly or according to certain rules. At the same time, the feature symbol size identifier of the first feature symbol can be extracted from the feature symbol size identifier list, and the feature symbol position identifier of the first feature symbol can be extracted from the feature symbol position identifier list.

[0053] The original topographic map is first segmented according to the area circled by the user, and then a targeted multi-layer convolutional symbol detection model is used to recognize the symbols of the segmented sub-images. The graphic, size and position information of the symbols of the objects are extracted, and any symbol is determined as the target symbol for subsequent processing, providing a basis for subsequent object recognition and extraction.

[0054] Furthermore, the embodiment of the present application also includes:

[0055] S121: Setting a symbol detection size threshold of the 1:2000 topographic map, wherein the symbol detection size threshold is less than or equal to 0.33 mm;

[0056] S122: Initializing the feature pyramid topology according to the symbol detection size threshold to obtain a plurality of initial feature pyramid topologies;

[0057] S123: training the plurality of initial feature pyramid topologies to obtain a plurality of ground object symbol segmentation accuracy rates, wherein the ground object symbol segmentation accuracy rates are characterized by similarity between ground object segmentation symbols and ground object preset supervision symbols;

[0058] S124: When the plurality of ground object symbol segmentation accuracy rates are all less than or equal to a convergence accuracy rate threshold, performing feature pyramid topology optimization on the plurality of initial feature pyramid topologies according to the plurality of ground object symbol segmentation accuracy rates to obtain a feature pyramid target topology greater than the convergence accuracy rate threshold;

[0059] S125: when the initial feature pyramid topologies of the plurality of ground object symbol segmentation accuracies are greater than the convergence accuracy threshold, set them as the feature pyramid target topology;

[0060] S126: Generate the multi-layer convolutional symbol detection model according to the feature pyramid target topology.

[0061] In a preferred embodiment, detailed steps for constructing a multi-layer convolutional symbol detection model, namely steps S121 to S126, are provided to illustrate how to design and optimize the network structure and parameters of the detection model for a 1:2000 topographic map to achieve high-precision and robust symbol recognition.

[0062] First, the symbol detection size threshold of the 1:2000 topographic map is set, where the symbol detection size threshold is less than or equal to 0.33 mm. According to relevant standards, the minimum size requirement of the feature symbol in the 1:2000 topographic map is 0.33 mm. Based on this, a symbol detection size threshold specifically for the topographic map of this scale is set to provide prior knowledge for the subsequent training network. Then, according to the symbol detection size threshold, the feature pyramid topology is initialized to obtain several initial feature pyramid topologies. The so-called feature pyramid is a multi-scale, hierarchical convolutional neural network structure that can extract image features at different receptive fields and resolutions. With the symbol detection size threshold as a constraint, a set of initial feature pyramid network topologies is randomly generated as the starting point for subsequent structure search. Subsequently, several initial feature pyramid topologies are trained to obtain several feature symbol segmentation accuracies, where the feature symbol segmentation accuracy is characterized by the similarity between the feature segmentation symbol and the preset supervisory symbol of the feature. Specifically, each initial topological structure is trained using a manually annotated topographic map dataset, and its symbol segmentation accuracy is evaluated on the validation set. Among them, the definition of segmentation accuracy is based on the similarity measurement between the segmented symbols and the manually annotated supervision symbols.

[0063] When the segmentation accuracy of several ground object symbols is less than or equal to the convergence accuracy threshold, several initial feature pyramid topologies are optimized according to the segmentation accuracy of several ground object symbols to obtain a feature pyramid target topology greater than the convergence accuracy threshold. Specifically, if the segmentation accuracy of the existing topological structures is not ideal, the hyperparameters such as the number of network layers, filter size and number of each topology will be iteratively optimized until the accuracy of a certain structure exceeds the preset convergence threshold, and it will be determined as the feature pyramid target topology to adapt to the characteristics of the current data and tasks. When several ground object symbol segmentation accuracies have an initial feature pyramid topology greater than the convergence accuracy threshold, it is set as the feature pyramid target topology. Specifically, if the segmentation accuracy of a certain initial topology has reached the convergence threshold, there is no need for structural optimization, and it is directly determined as the feature pyramid target topology to save computational costs. Afterwards, the obtained feature pyramid target topology is used to generate the corresponding multi-layer convolutional neural network model to complete the construction of the detection model.

[0064] By initializing, training, evaluating and dynamically optimizing the feature pyramid topology under the constraints of a specially set size threshold, the optimal network structure is automatically searched for and a customized detection model is generated. This ensures detection accuracy while automating the model design process to the maximum extent possible and reducing the cost of manual trial and error.

[0065] Furthermore, the embodiment of the present application also includes:

[0066] S1241: constructing a feature pyramid topology distance evaluation function, wherein the feature pyramid topology distance evaluation function is used to evaluate the distance characteristic values ​​of any two feature pyramid topologies;

[0067] S1242: performing cluster analysis on the plurality of initial feature pyramid topologies according to the feature pyramid topology distance evaluation function and based on a topology distance threshold, to obtain multiple clusters of initial feature pyramid topologies;

[0068] S1243: extracting the highest accuracy initial feature pyramid topologies of the multiple clusters of initial feature pyramid topologies according to the segmentation accuracy rates of the plurality of ground feature symbols, respectively, to obtain a plurality of highest accuracy initial feature pyramid topologies;

[0069] S1244: setting the multiple highest accuracy initial feature pyramid topologies as multiple guiding feature pyramid topologies, performing guided mutation on the initial feature pyramid topologies of non-cluster to obtain updated feature pyramid topologies, wherein the guided mutation refers to shortening the distance between the guided feature pyramid topology and the guiding feature pyramid topology;

[0070] S1245: training the updated feature pyramid topology to obtain an updated ground feature symbol segmentation accuracy;

[0071] S1246: When the updated ground feature symbol segmentation accuracy is greater than the convergence accuracy threshold, the updated feature pyramid topology is set as the feature pyramid target topology, otherwise a feature pyramid topology optimization loop is executed.

[0072] In a preferred embodiment, the specific implementation of optimizing several initial feature pyramid topologies can be further refined into steps S1241 to S1246, by introducing mechanisms such as feature pyramid topology distance evaluation function, clustering analysis and guided mutation, so as to search for the optimal network structure more intelligently and efficiently.

[0073] First, a feature pyramid topological distance evaluation function is constructed to quantitatively characterize the distance characteristic value between any two topological structures and characterize the degree of difference between the two topological structures. The feature pyramid topological distance evaluation function considers factors such as the number of layers of the topological structure, the number of types of filters in each layer, and the size of the receptive field, and calculates the comprehensive difference value by weighted summation, providing a distance metric for subsequent clustering analysis. Then, using the constructed feature pyramid topological distance evaluation function, several initial feature pyramid topologies are clustered with the topological distance threshold as the radius, and topologies with similar structures are divided into the same cluster to obtain multiple clusters of initial feature pyramid topologies. The multi-cluster initial feature pyramid topology is a coarse-grained grouping of the initial topology, and the topologies within the cluster share the local optimal substructure, which helps to narrow the scope of subsequent optimization search and improve the optimization efficiency. Subsequently, according to the segmentation accuracy of several ground feature symbols, the topology with the highest segmentation accuracy is selected from each cluster of the initial feature pyramid topology as the initial feature pyramid topology with the highest accuracy in the corresponding cluster, and multiple initial feature pyramid topologies with the highest accuracy are obtained, which represent the high-quality topological structures in each cluster and can guide the optimization direction of other topologies.

[0074] Then, multiple initial feature pyramid topologies with the highest accuracy are used as multiple guided feature pyramid topologies, and multiple guided feature pyramid topologies are used as references to adjust the structural parameters of the initial feature pyramid topologies of non-cluster, so that they gradually approach the guided feature pyramid topology under the definition of the feature pyramid topology distance evaluation function, and complete the iterative optimization of the structure. Among them, the step size of the guided mutation can be dynamically controlled to balance the optimization speed and the breadth of exploration. After the guided mutation, a set of updated feature pyramid topologies is generated. After that, the accuracy of the ground feature symbol segmentation of the updated feature pyramid topology is re-evaluated. Once the accuracy of a certain updated topology exceeds the preset convergence accuracy threshold, it is determined as the final feature pyramid target topology, and the optimization process ends; otherwise, the clustering, guided mutation and other steps will continue to be iteratively executed until the convergence condition is met or the maximum number of iterations is reached.

[0075] By integrating distance measurement, cluster analysis, and guided mutation mechanisms, an effective feature pyramid topology optimization strategy is formed, which can quickly lock the optimal network structure in the search space while taking into account the tuning of local details, providing a strong guarantee for the high-precision and adaptive construction of multi-layer convolutional symbol detection models.

[0076] Furthermore, the feature pyramid topological distance evaluation function is:

[0077] X=[(x 11 ,x 12 ),…,(x i1 ,x i2 ),…,(x N1 ,x N2 )]

[0078] Y=[(y 11 ,y 12 ),…,(y i1 ,y i2 ),…,(y M1 ,y M2 )]

[0079]

[0080] Among them, D(X,Y) represents the distance between two feature pyramid topologies, X represents the first feature pyramid topology, Y represents the second feature pyramid topology, x 11 , x 12 Characterizes the dimensions of the first convolution kernel in both directions of the first feature pyramid topology, x i1 , x i2 Characterizes the dimensions of the i-th convolution kernel in both directions of the first feature pyramid topology, x N1 , x N2 Represents the dimensions of the Nth convolution kernel in the first feature pyramid topology in two directions. N represents the total number of layers in the first feature pyramid topology. 11 ,y 12 Characterizes the dimensions of the first layer of convolution kernel in two directions of the second feature pyramid topology, y i1 ,y i2 Characterizes the dimensions of the convolution kernel of the i-th layer in the second feature pyramid topology in two directions, y M1 ,y M2 Represents the dimensions of the Mth convolution kernel in two directions of the second feature pyramid topology, and M represents the total number of layers of the second feature pyramid topology.

[0081] In a preferred embodiment, the constructed feature pyramid topological distance evaluation function can be in the following form:

[0082] X=[(x 11 ,x12 ),…,(x i1 ,x i2 ),…,(x N1 ,x N2 )]

[0083] Y=[(y 11 ,y 12 ),…,(y i1 ,y i2 ),…,(y M1 ,y M2 )]

[0084]

[0085] First, define two feature pyramid topologies X and Y to be compared, where X consists of N layers of convolution kernels and Y consists of M layers of convolution kernels. The two dimensions of the i-th convolution kernel in X are represented by x i1 and x i2 Indicates that the two dimensions of the i-th convolution kernel in Y are represented by y i1 and i2 Then, the distance D(X,Y) between feature pyramid topologies X and Y is defined as:

[0086]

[0087] Among them, f(M,N) is a binary function, that is:

[0088]

[0089] When M is not equal to N, the value is MN, and when M is equal to N, the value is 1. min(M,N) means taking the smaller value of M and N.

[0090] The core idea of ​​the above feature pyramid topology distance evaluation function is: first, use f(M,N) to measure the difference in the number of layers between two feature pyramid topologies. The greater the difference in the number of layers, the farther the distance between the two topologies. Secondly, for two topologies with the same number of layers, calculate the Euclidean distance of the difference in the convolution kernel dimension layer by layer, and accumulate to obtain the total distance metric. Therefore, the difference in the number of layers determines the order of magnitude of the topological distance, while the difference in the convolution kernel dimension determines the fine sorting between topologies with the same number of layers.

[0091] The above feature pyramid topology distance evaluation function is only a possible implementation method, and its specific form can be adjusted according to actual needs. However, no matter what form is adopted, the basic starting point is the same, that is, to comprehensively consider the number of layers and convolution kernel dimensions of the feature pyramid topology, quantitatively characterize the degree of difference of the feature pyramid topology in the structural feature space, and provide an operational distance metric for clustering analysis and guided variation.

[0092] By providing a feature pyramid topological distance evaluation function, the hierarchical structure information of the feature pyramid topology is fully utilized, and the topological comparison is transformed into a comprehensive measurement of the number of layers and convolution kernel dimensions, laying the foundation for building a multi-layer convolutional symbol detection model.

[0093] Further, the symbol indicates that the twin network includes a first feature extraction channel, a second feature extraction channel and a feature comparison channel, and the first feature extraction channel and the second feature extraction channel have the same structural topology and parameters. Obtaining the first ground feature type matching result includes:

[0094] S210: inputting the first land feature symbol into the first feature extraction channel to obtain a first land feature symbol shape feature and a first land feature symbol color feature;

[0095] S220: inputting the first reference ground object symbol randomly extracted from the symbol description database into the second feature extraction channel to obtain a second ground object symbol shape feature and a second ground object symbol color feature;

[0096] S230: Inputting the first land feature symbol shape feature, the first land feature symbol color feature, the second land feature symbol shape feature and the second land feature symbol color feature into the feature comparison channel to obtain shape similarity and color similarity;

[0097] S240: when the shape similarity is greater than or equal to the shape similarity threshold, and the color similarity is greater than or equal to the color similarity threshold, setting the associated feature type of the first reference feature symbol to the first feature type matching result; otherwise, updating the first reference feature symbol according to the symbol description database for cyclic matching;

[0098] S250: When the symbol description database does not have the first land feature type matching result, the first land feature type matching result is set to empty.

[0099] In a preferred embodiment, through steps S210 to S250, the internal structure and working principle of the twin network are explained in detail with symbols.

[0100] First, the symbol description twin network consists of three sub-networks, namely the first feature extraction channel, the second feature extraction channel and the feature comparison channel. Among them, the first feature extraction channel and the second feature extraction channel use exactly the same structural topology and parameters. Then, the first feature symbol to be identified is input into the first feature extraction channel, and after a series of convolution, pooling and other operations, the shape features and color features of the symbol are extracted as the first feature symbol shape features and the first feature symbol color features. Among them, the shape features describe the geometric appearance of the symbol, and the color features reflect the color attributes of the symbol. At the same time, a reference feature symbol is traversed and selected from the symbol description database, called the first reference feature symbol, and is input into the second feature extraction channel. Similarly, the second feature extraction channel outputs the shape features and color features of the reference symbol as the second feature symbol shape features and the second feature symbol color features.

[0101] Subsequently, the shape features and color features of the first feature symbol and the first reference feature symbol are summarized, namely, the shape features of the first feature symbol, the color features of the first feature symbol, the shape features of the second feature symbol, and the color features of the second feature symbol, and are input into the feature comparison channel. The feature comparison channel adopts a similarity metric to calculate the degree of similarity in shape between the shape features of the first feature symbol and the shape features of the second feature symbol, and calculates the degree of similarity in shape between the color features of the first feature symbol and the color features of the second feature symbol, and obtains two indicators, namely, shape similarity and color similarity. Then, a threshold judgment is performed on the shape similarity and color similarity. If both the shape similarity and the color similarity exceed the preset threshold, namely, the shape similarity is greater than or equal to the shape similarity threshold, and the color similarity is greater than or equal to the color similarity threshold, then it is considered that the first feature symbol and the first reference feature symbol belong to the same category, and at this time, the associated feature type indicated by the first reference feature symbol in the symbol description database is output as the first feature type matching result. Otherwise, it is considered that the current first reference feature symbol is not similar enough, and the first reference feature symbol needs to be updated, and steps S220 to S240 are repeated until a sufficiently similar matching symbol is found, or the entire symbol description database is traversed. If there is no reference feature symbol in the symbol description database that can match the first feature symbol, that is, no valid first feature type matching result is finally obtained, it is set to a null value, indicating that there is no category matching the first feature symbol in the symbol description database.

[0102] By using the symbol description twin network to identify the types of ground object symbols, we can effectively deal with the diversity and complexity of ground object symbols and greatly improve the accuracy and efficiency of symbol recognition.

[0103] Furthermore, the implementation of this application also includes:

[0104] S310: Setting a shape similarity adjacency threshold and a color similarity adjacency threshold;

[0105] S320: Matching, from the topographic map sharing platform, a plurality of feature types of a plurality of adjacent feature symbols for the first feature symbol based on the shape similarity adjacency threshold and the color similarity adjacency threshold;

[0106] S330: Extract the highest frequency ground feature type from the plurality of ground feature types, and set it as the second ground feature type matching result.

[0107] In a preferred embodiment, steps S310 to S330 further illustrate how to use the topographic map sharing platform to perform further adjacent symbol analysis on the first feature symbol to obtain the second feature type matching result when the first feature type matching result is empty.

[0108] First, two threshold parameters are set, namely, the shape similarity adjacency threshold and the color similarity adjacency threshold. These two thresholds are used to retrieve adjacent symbols similar to the first feature symbol in the topographic map sharing platform, but their values ​​are usually greater than the shape similarity threshold and the color similarity threshold. This is because, in the absence of direct matching, appropriately relaxing the similarity requirements of adjacent symbols helps to obtain more reference information. Subsequently, based on the above two adjacent thresholds, adjacent feature symbols with similar shapes and colors to the first feature symbol are retrieved in the topographic map sharing platform. The adjacent feature symbols obtained by the query and their corresponding feature types will be provided as candidate matches to the next step. After that, the frequency of occurrence of each feature type in all adjacent feature symbols is counted, and the feature type with the highest frequency is determined as the second feature type matching result. Here, the prior assumption is used, that is, the feature symbols that are spatially adjacent and have similar features have a consistent tendency to belong to the feature type. Frequency statistics can eliminate the interference of individual abnormal matches to a certain extent and improve the confidence of the matching results.

[0109] When direct matching fails, the type of the first feature symbol is inferred by making full use of the massive feature symbol data and implicit spatial topological relationships contained in the topographic map sharing platform. This inference mechanism based on adjacency analysis makes up for the shortcomings of single symbol shape and color features, and can mine matching clues in a larger range and improve the success rate of recognition.

[0110] Furthermore, the embodiment of the present application also includes:

[0111] S321: matching a primary adjacent feature symbol for the first feature symbol from the topographic map sharing platform based on the shape similarity adjacent threshold and the color similarity adjacent threshold;

[0112] S322: When the data amount of the first-level adjacent feature symbol is less than or equal to the data amount threshold, based on the shape similarity adjacency threshold and the color similarity adjacency threshold, traverse the first-level adjacent feature symbols from the topographic map sharing platform to match the second-level adjacent feature symbols;

[0113] S323: Add the second-level adjacent feature symbol and the first-level adjacent feature symbol into the plurality of adjacent feature symbols.

[0114] In a preferred implementation, by introducing primary adjacency and secondary adjacency, the matching range of the adjacency symbol is expanded in a hierarchical and progressive manner.

[0115] First, based on the shape similarity adjacency threshold and the color similarity adjacency threshold, the feature symbols that meet the shape similarity adjacency threshold and the color similarity adjacency threshold with the first feature symbol are retrieved in the topographic map sharing platform, which are called first-level adjacent feature symbols. Then, based on the first-level adjacency matching, the second-level adjacency is introduced. When the number of first-level adjacent feature symbols is small, that is, less than a preset data volume threshold, the first-level adjacent feature symbols will be used as the basis to further retrieve the feature symbols that meet the shape similarity adjacency threshold and the color similarity adjacency threshold with the first-level adjacent feature symbols in the topographic map sharing platform to obtain the second-level adjacent feature symbols, so as to achieve the expansion of feature symbols when there are few first-level adjacent feature symbols to obtain richer information. Subsequently, the obtained first-level adjacent feature symbols and the obtained second-level adjacent feature symbols are merged as several adjacent feature symbols of the first feature symbol, in preparation for the subsequent frequency statistical analysis.

[0116] By adopting a hierarchical adjacent symbol matching strategy from near to far and from direct to indirect, the robustness and adaptability of dealing with complex land feature symbols are enhanced.

[0117] The 1:2000 topographic map feature recognition and extraction method based on deep learning provided by the embodiment of the present invention has at least the following technical effects:

[0118] In response to the circled area input by the user, the first feature symbol of the 1:2000 topographic map is obtained, wherein the first feature symbol has a feature symbol size identifier and a feature symbol position identifier, providing a basis for subsequent recognition and information extraction. Through the symbol description twin network, the first feature symbol is processed in combination with the symbol description database to obtain the first feature type matching result, thereby improving the recognition ability of complex and non-standardized symbols. When the first feature type matching result is empty, the first feature symbol is analyzed for adjacent symbols in combination with the topographic map sharing platform to obtain the second feature type matching result, further improving the recall rate of recognition and making up for the shortcomings of single feature matching. When the maximum size of the feature symbol size identifier is greater than or equal to 0.09 cm, the size is restored according to the 1:2000 scale to obtain the actual size of the feature. When the maximum size of the feature symbol size identifier is less than 0.09 cm, the actual size of the feature is marked as empty, and valuable size information is extracted, thereby improving the accuracy of the information extraction result. The second feature type matching result, the actual size of the feature and the feature symbol location identifier are sent to the user end, thereby improving the accuracy and adaptability of topographic map feature recognition and information extraction.

[0119] Embodiment 2:

[0120] See also Figure 2 , Figure 2 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 2 As shown, an electronic device 100 provided by an embodiment of the present invention includes a memory 110, a processor 120, and a first computer program 111 stored in the memory 110 and executable on the processor 120. When the processor 120 executes the first computer program 111, a 1:2000 topographic map feature recognition and extraction method based on deep learning is implemented.

[0121] Embodiment three:

[0122] See also Figure 3 , Figure 3 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 3 As shown, this embodiment provides a computer-readable storage medium 200 on which a second computer program 211 is stored. When the second computer program 211 is executed by a processor, a method for identifying and extracting features in a 1:2000 topographic map based on deep learning is implemented.

[0123] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0124] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0126] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0128] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0129] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A 1:2000 topographic map feature recognition and extraction method based on deep learning, characterized in that: include: In response to the delimited area input by the user terminal, a first land feature symbol of a 1:2000 topographic map is obtained, wherein the first land feature symbol has a land feature symbol size identifier and a land feature symbol position identifier; Processing the first ground feature symbol through the symbol description twin network in combination with the symbol description database to obtain a first ground feature type matching result; When the first feature type matching result is empty, combining with the topographic map sharing platform, performing adjacent symbol analysis on the first feature symbol to obtain a second feature type matching result; When the maximum size of the feature symbol size mark is greater than or equal to 0.09 cm, the size is restored according to the 1:2000 scale to obtain the actual size of the feature; When the maximum size of the feature symbol size mark is less than 0.09 cm, the actual size mark of the feature is marked as empty; The second feature type matching result, the actual size of the feature and the feature symbol position identifier are sent to the user terminal.

2. The method according to claim 1, characterized in that In response to the delimited area input by the user terminal, a first land feature symbol of a 1:2000 topographic map is obtained, including: According to the demarcated area, the 1:2000 topographic map is segmented to obtain a segmented digital image; Performing semantic segmentation on the segmented digital image through a multi-layer convolutional symbol detection model to extract a list of ground object symbols, a list of ground object symbol size identifiers, and a list of ground object symbol position identifiers; The first geographical feature symbol belongs to the geographical feature symbol list, the geographical feature symbol size identifier belongs to the geographical feature symbol size identifier list, and the geographical feature symbol position identifier belongs to the geographical feature symbol position identifier list.

3. The method according to claim 2, characterized in that The multi-layer convolutional symbol detection model construction step includes: Setting a symbol detection size threshold of the 1:2000 topographic map, wherein the symbol detection size threshold is less than or equal to 0.33 mm; Initializing the feature pyramid topology according to the symbol detection size threshold to obtain a plurality of initial feature pyramid topologies; Training the plurality of initial feature pyramid topologies to obtain a plurality of ground object symbol segmentation accuracy rates, wherein the ground object symbol segmentation accuracy rates are characterized by the similarity between the ground object segmentation symbols and the ground object preset supervision symbols; When the plurality of ground object symbol segmentation accuracy rates are all less than or equal to a convergence accuracy rate threshold, performing feature pyramid topology optimization on the plurality of initial feature pyramid topologies according to the plurality of ground object symbol segmentation accuracy rates to obtain a feature pyramid target topology greater than the convergence accuracy rate threshold; When the initial feature pyramid topologies of the plurality of ground object symbol segmentation accuracies are greater than the convergence accuracy threshold, they are set as the feature pyramid target topology; The multi-layer convolutional symbol detection model is generated according to the feature pyramid target topology.

4. The method according to claim 3, characterized in that The method further comprises: performing feature pyramid topology optimization on the plurality of initial feature pyramid topologies according to the plurality of ground object symbol segmentation accuracy rates to obtain a feature pyramid target topology greater than the convergence accuracy rate threshold value, including: Constructing a feature pyramid topology distance evaluation function, wherein the feature pyramid topology distance evaluation function is used to evaluate the distance characteristic values ​​of any two feature pyramid topologies; According to the feature pyramid topology distance evaluation function and based on a topology distance threshold, cluster analysis is performed on the plurality of initial feature pyramid topologies to obtain multiple clusters of initial feature pyramid topologies; According to the plurality of ground feature symbol segmentation accuracies, respectively extracting the highest accuracy initial feature pyramid topologies of the plurality of clusters of initial feature pyramid topologies to obtain a plurality of highest accuracy initial feature pyramid topologies; The multiple highest accuracy initial feature pyramid topologies are set as multiple guiding feature pyramid topologies, and the initial feature pyramid topologies of non-cluster are guided to mutate to obtain updated feature pyramid topologies, wherein the guided mutation refers to shortening the distance between the guided feature pyramid topology and the guiding feature pyramid topology; Training the updated feature pyramid topology to obtain an updated ground feature symbol segmentation accuracy; When the updated ground feature symbol segmentation accuracy is greater than the convergence accuracy threshold, the updated feature pyramid topology is set to the feature pyramid target topology, otherwise a feature pyramid topology optimization cycle is executed.

5. The method according to claim 4, characterized in that The feature pyramid topological distance evaluation function is: X=[(x 11 ,x 12 ),…,(x i1 ,x i2 ),…,(x N1 ,x N2 )] And=[(and 11 ,and 12 ),…,(and i1 ,and i2 ),…,(and M1 ,and M2 )] Among them, D(X,Y) represents the distance between two feature pyramid topologies, X represents the first feature pyramid topology, Y represents the second feature pyramid topology, x 11 , x 12 Characterizes the dimensions of the first convolution kernel in both directions of the first feature pyramid topology, x i1 , x i2 Characterizes the dimensions of the i-th convolution kernel in both directions of the first feature pyramid topology, x N1 , x N2 Represents the dimensions of the Nth convolution kernel in the first feature pyramid topology in two directions. N represents the total number of layers in the first feature pyramid topology. 11 ,y 12 Characterizes the dimensions of the first layer of convolution kernel in two directions of the second feature pyramid topology, y i1 ,y i2 Characterizes the dimensions of the convolution kernel of the i-th layer in the second feature pyramid topology in two directions, y M1 ,y M2 Represents the dimensions of the Mth convolution kernel in two directions of the second feature pyramid topology, and M represents the total number of layers of the second feature pyramid topology.

6. The method according to claim 1, characterized in that The symbol description twin network includes a first feature extraction channel, a second feature extraction channel and a feature comparison channel. The first feature extraction channel and the second feature extraction channel have the same structural topology and parameters. The first feature symbol is processed by the symbol description twin network in combination with the symbol description database to obtain a first feature type matching result, including: Inputting the first land feature symbol into the first feature extraction channel to obtain a shape feature of the first land feature symbol and a color feature of the first land feature symbol; Inputting the first reference ground object symbol randomly extracted from the symbol description database into the second feature extraction channel to obtain the second ground object symbol shape feature and the second ground object symbol color feature; Inputting the first geographical feature symbol shape feature, the first geographical feature symbol color feature, the second geographical feature symbol shape feature and the second geographical feature symbol color feature into the feature comparison channel to obtain shape similarity and color similarity; When the shape similarity is greater than or equal to a shape similarity threshold, and the color similarity is greater than or equal to a color similarity threshold, setting the associated feature type of the first reference feature symbol as the first feature type matching result, otherwise, updating the first reference feature symbol according to the symbol description database for cyclic matching; When the symbol indicates that there is no matching result of the first land feature type in the database, the matching result of the first land feature type is set to empty.

7. The method according to claim 1, characterized in that When the first feature type matching result is empty, combining with the topographic map sharing platform, performing adjacent symbol analysis on the first feature symbol to obtain a second feature type matching result includes: Set the shape similarity adjacency threshold and color similarity adjacency threshold; Based on the shape similarity adjacency threshold and the color similarity adjacency threshold, matching the first feature symbol with a plurality of feature types of adjacent feature symbols from the topographic map sharing platform; The highest frequency ground feature type among the plurality of ground feature types is extracted and set as the second ground feature type matching result.

8. The method according to claim 7, characterized in that Based on the shape similarity adjacency threshold and the color similarity adjacency threshold, matching the first feature symbol with a plurality of feature types of adjacent feature symbols from the topographic map sharing platform, including: Based on the shape similarity adjacency threshold and the color similarity adjacency threshold, matching a primary adjacent feature symbol for the first feature symbol from the topographic map sharing platform; When the data amount of the first-level adjacent feature symbol is less than or equal to the data amount threshold, based on the shape similarity adjacency threshold and the color similarity adjacency threshold, traverse the first-level adjacent feature symbols from the topographic map sharing platform to match the second-level adjacent feature symbols; The secondary adjacent feature symbol and the primary adjacent feature symbol are added to the plurality of adjacent feature symbols.

Images