Rural base station planning method and system, electronic device and storage medium

Abstract

The application discloses a rural base station planning method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring a to-be-identified rural remote sensing image; identifying and analyzing the to-be-identified rural remote sensing image by using a preset building identification model to obtain preset building data; performing clustering analysis on the preset building data by using a preset clustering algorithm to obtain preset concentrated area data; performing site planning according to the preset concentrated area data and existing network base station information to generate a desired base station planning site. The embodiment of the application can realize automatic planning of a rural base station, effectively improve the site identification accuracy and planning efficiency. The application can be widely applied to the technical field of wireless communication planning.

Description

Technical Field

[0001] This application relates to the field of wireless communication planning technology, and in particular to a rural base station planning method, system, electronic device and storage medium. Background Technology

[0002] To improve wireless network coverage in rural areas, base station planning and construction are necessary in areas lacking coverage. However, due to the vast area and dispersed population of rural areas, and the lack of concentrated MR grid positioning data similar to that in urban areas, manual site selection is time-consuming, labor-intensive, and prone to omissions. Furthermore, current base station planning methods primarily target urban areas, indoor spaces, and areas with weak coverage; planning for empty areas is relatively simple and coarse, making it difficult to achieve precise planning for rural areas lacking coverage.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this application is to propose a rural base station planning method, system, electronic device, and storage medium, which can realize automatic planning of rural base stations and effectively improve the accuracy of site identification and planning efficiency.

[0005] To achieve the above objectives, one aspect of this application proposes a rural base station planning method, which includes the following steps:

[0006] Acquire remote sensing images of rural areas to be identified;

[0007] The rural remote sensing image to be identified is analyzed by a preset building recognition model to obtain preset building data.

[0008] The preset building data is clustered using a preset clustering algorithm to obtain preset concentrated area data;

[0009] Based on the preset centralized area data and existing network base station information, site planning is performed to generate the desired base station planning site.

[0010] In some embodiments, before performing the identification and analysis of the rural remote sensing image to be identified using a preset building identification model to obtain preset building data, the method further includes:

[0011] A remote sensing image dataset is acquired, and the remote sensing image dataset is subjected to a first preprocessing step to obtain a model training dataset; wherein, the first preprocessing step includes resizing, image normalization, and data augmentation.

[0012] Construct a preset training model; wherein, the preset training model includes a visual converter model;

[0013] The model training dataset is input into the visual converter model to train the model using a backpropagation algorithm and a preset loss function, thereby obtaining the preset building recognition model.

[0014] In some embodiments, the step of identifying and analyzing the rural remote sensing image to be identified using a preset building recognition model to obtain preset building data includes:

[0015] The rural remote sensing image to be identified is subjected to a second preprocessing to obtain an image patch to be identified; wherein, the second preprocessing includes image segmentation and image normalization;

[0016] The image block to be identified is input into the preset building recognition model for feature extraction, generating a feature vector;

[0017] Based on the feature vector, a mask prediction is performed to obtain the image segmentation prediction mask;

[0018] The image segmentation prediction mask is mapped and transformed to obtain a preset mask raster image;

[0019] The preset mask raster image is converted into a preset geographic information system data format to obtain the preset building data.

[0020] In some embodiments, the step of performing cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data includes:

[0021] Construct preset clustering parameters; wherein, the preset clustering parameters include preset neighborhood radius and neighborhood sample threshold;

[0022] Based on the preset clustering parameters, a density-based spatial clustering algorithm is used to identify the regions where buildings are clustered in the preset building data, thereby obtaining the preset concentrated region data.

[0023] In some embodiments, the step of identifying areas where buildings are clustered in the preset building data using a density-based spatial clustering algorithm based on the preset clustering parameters to obtain the preset clustered area data includes:

[0024] Detect the first unmarked building in the preset building data to determine whether the number of second buildings contained in the first building within the preset neighborhood radius is less than the neighborhood sample threshold;

[0025] When it is determined that the number of times the first building contains the second building within the preset neighborhood radius is greater than the neighborhood sample threshold, a preset residential area is constructed, and the second building is added to the preset candidate set;

[0026] Determine whether the number of unmarked third buildings in the preset candidate set that contain fourth buildings within the preset neighborhood radius is greater than the neighborhood sample threshold.

[0027] When it is determined that the number of fourth buildings contained within the preset neighborhood radius of the third building is greater than the neighborhood sample threshold, the fourth building is added to the preset candidate set;

[0028] When it is determined that the third building is not marked in the residential area, the third building is added to the preset residential area, and the process returns to the step of determining whether the number of fourth buildings contained in the preset neighborhood radius of the unmarked third building in the preset candidate set is greater than the neighborhood sample threshold, until the preset candidate set is traversed.

[0029] When it is determined that the preset candidate set has been traversed, the step of returning to detect the first unmarked building in the preset building data, and determining whether the number of second buildings contained in the first building within the preset neighborhood radius is less than the neighborhood sample threshold, continues until all the first buildings in the preset building data are marked, and the preset concentrated area data is obtained.

[0030] In some embodiments, the existing network base station information includes data on the coverage area of ​​existing network base stations;

[0031] The step of planning site locations based on the preset centralized area data and existing network base station information to generate desired base station planning sites includes:

[0032] Acquire preset operating parameter data to calculate the coverage area data of the existing network base station; wherein, the preset operating parameter data includes base station identifier, base station coordinate data, base station azimuth angle, base station frequency band, and base station transmit power;

[0033] Based on the existing network base station coverage area data, the preset centralized area data is filtered to obtain several candidate buildings;

[0034] The candidate building with the largest number of neighboring buildings among the candidate buildings is selected as the target building, and the desired base station planning site is generated based on the target building.

[0035] In some embodiments, the step of filtering buildings in the preset centralized area data based on the existing network base station coverage area data to obtain a number of candidate buildings includes:

[0036] The coverage radius of the existing network base station is determined based on the preset operating parameter data;

[0037] Based on the coverage radius of the existing network base stations, the covered buildings in the preset centralized area data are removed to obtain the uncovered buildings;

[0038] The number of buildings within a preset coverage radius of the uncovered buildings is counted to obtain the number of neighboring buildings; wherein, the preset coverage radius includes the expected coverage radius of the preset planned base station;

[0039] The uncovered buildings whose number of neighboring buildings is less than a preset threshold are removed to obtain dense buildings;

[0040] The base station distance between the preset existing base station and the preset planned base station is calculated based on the coverage radius of the existing base station and the preset coverage radius. The range of the planned base station is determined by the base station distance, and the dense buildings located within the range of the planned base station are eliminated to obtain a number of candidate buildings.

[0041] To achieve the above objectives, another aspect of this application proposes a rural base station planning system, the system comprising:

[0042] The first module is used to acquire remote sensing images of rural areas to be identified;

[0043] The second module is used to identify and analyze the rural remote sensing image to be identified using a preset building identification model to obtain preset building data.

[0044] The third module is used to perform cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data;

[0045] The fourth module is used to plan the site based on the preset centralized area data and the existing network base station information, and generate the desired base station planning site.

[0046] To achieve the above objectives, another aspect of this application provides an electronic device, the electronic device comprising:

[0047] At least one processor;

[0048] At least one memory for storing at least one program;

[0049] When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.

[0050] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0051] The embodiments of this application include at least the following beneficial effects: This application provides a rural base station planning method, system, electronic device, and storage medium. This scheme acquires remote sensing images of rural areas to be identified, and then uses a preset building identification model to identify and analyze the images to obtain preset building data. Then, the embodiments of this invention perform cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data. Based on the preset concentrated areas and existing network base station information, site planning is performed to generate desired base station planning sites, thus achieving automatic planning of rural base stations. It is easily understood that the embodiments of this invention extract rural building information through the fusion of remote sensing imagery and deep learning technology to obtain preset building data, supplementing the rural building data and providing a basis for assessing the deployment value of rural base stations. Simultaneously, the preset clustering algorithm obtains rural concentrated areas, i.e., preset concentrated area data, which is then combined with existing network base station information for rural base station planning, thereby effectively improving the accuracy of site identification and planning efficiency. Attached Figure Description

[0052] Figure 1 This is a flowchart of the steps of the rural base station planning method provided in the embodiments of the present invention;

[0053] Figure 2 This is a schematic diagram of a sample of a rural remote sensing image to be identified, provided in an embodiment of the present invention.

[0054] Figure 3 This is a flowchart illustrating the steps of training a preset building recognition model according to an embodiment of the present invention.

[0055] Figure 4 This is a schematic diagram of a sample remote sensing image dataset provided in an embodiment of the present invention;

[0056] Figure 5 This is a flowchart of the steps provided in this embodiment of the invention to identify and analyze a rural remote sensing image to be identified using a preset building recognition model to obtain preset building data.

[0057] Figure 6 This is a schematic diagram of building identification provided in an embodiment of the present invention;

[0058] Figure 7 This is a schematic diagram of base station planning provided in an embodiment of the present invention;

[0059] Figure 8 This is a schematic diagram of the overall technical architecture of the rural base station planning method provided in the embodiments of the present invention;

[0060] Figure 9 This is a schematic diagram of the structure of the rural base station planning system provided in an embodiment of the present invention;

[0061] Figure 10This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0063] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0064] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0066] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0067] Geographic Information System (GIS): A system used to capture, store, manage, analyze, and display geospatial data. GIS combines maps, databases, and analytical tools.

[0068] The Vision Transformer (ViT) is a deep learning model based on the Transformer architecture used to process computer vision tasks. Instead of the traditional convolutional neural networks (CNNs) typically used for computer vision tasks, ViT applies the Transformer model to visual data.

[0069] A Convolutional Neural Network (CNN) is a deep learning model used to process data with a grid structure, such as images. Accordingly, a CNN is a feedforward neural network that can be applied to image recognition, object detection, image segmentation, or other visual tasks.

[0070] U-Net neural network: A convolutional neural network with an encoder-decoder structure, used for image segmentation. It can effectively utilize contextual information and accurately locate targets.

[0071] SegNet neural network: a deep convolutional neural network for image segmentation that can improve segmentation accuracy while maintaining computational efficiency.

[0072] Dice loss is a set similarity metric function used to evaluate model performance in image segmentation tasks. It is based on the Dice coefficient, which is commonly used to measure the similarity between two samples.

[0073] DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm with noise that can discover clusters of arbitrary shapes without specifying the number of clusters in advance and can effectively handle noisy data.

[0074] To improve wireless network coverage in rural areas, base station planning and construction are necessary in areas lacking coverage. However, rural areas are vast, with dispersed residents and a lack of concentrated MR grid positioning data similar to urban areas. Manual site selection is time-consuming, labor-intensive, and prone to omissions. Current automatic base station planning methods primarily target urban areas, indoor spaces, and areas with weak coverage. Some algorithms for empty areas are simplistic, failing to consider user density and resulting in coarse site selection. They lack precise automatic planning for rural areas lacking coverage. Furthermore, building layer data mainly comes from purchases from traditional map providers and acquisitions from open-source databases, both of which significantly lack data on rural buildings, failing to meet the needs of rural planning and network analysis.

[0075] In view of this, this invention provides a rural base station planning method, system, electronic device, and storage medium. This scheme acquires remote sensing images of rural areas to be identified, analyzes these images using a preset building identification model, and obtains preset building data. Then, this invention uses a preset clustering algorithm to perform cluster analysis on the preset building data, obtaining preset concentrated area data. Based on the preset concentrated areas and existing network base station information, site planning is performed to generate desired base station planning sites, achieving automatic rural base station planning and effectively improving site identification accuracy and planning efficiency.

[0076] The rural base station planning method provided in this application relates to the field of wireless communication planning technology. The rural base station planning method provided in this application can be applied to a terminal, a server, or software running on a terminal or server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle-mounted terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the rural base station planning method, but is not limited to the above forms.

[0077] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0078] Figure 1 This is an optional flowchart of the rural base station planning method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S110 to S140.

[0079] Step S110: Acquire the rural remote sensing image to be identified.

[0080] Step S120: Use a preset building recognition model to identify and analyze the rural remote sensing image to be identified, and obtain preset building data.

[0081] Step S130: Perform cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data.

[0082] Step S140: Based on the preset centralized area data and existing network base station information, site planning is performed to generate the desired base station planning site.

[0083] In this specific embodiment, the invention first acquires remote sensing images of rural areas to be identified, and then uses a preset building recognition model to analyze and identify these images to obtain preset building data. Specifically, in this embodiment, the remote sensing images of rural areas to be identified refer to remote sensing images of rural areas where base station planning and deployment are required, such as... Figure 2 As shown in the figure. The rural remote sensing image to be identified includes images of the rural areas to be identified and corresponding geographic location information, used to identify the outlines of rural buildings. Simultaneously, the preset building recognition model in this embodiment is a pre-trained deep learning image segmentation and recognition model, such as the ViT neural network model, CNN neural network model, U-Net model, or SegNet model. This embodiment inputs the rural remote sensing image to be identified into the trained preset building recognition model to identify the outlines of rural buildings in the image, obtaining preset building data. Further, this embodiment uses a preset clustering algorithm to perform cluster analysis on the preset building data to obtain preset concentrated area data. Specifically, after identifying the locations of rural buildings, this embodiment uses a preset clustering algorithm to divide the buildings into different residential areas according to their spatial density, that is, to assign the same label to buildings that are relatively close in location, identifying the areas where buildings are clustered, i.e., residential areas, thereby obtaining preset concentrated area data. Finally, this embodiment performs site planning based on the preset concentrated area data and existing network base station information to generate the desired base station planning sites. Specifically, in this embodiment, the existing network base station information refers to the parameter information of each base station in the existing network, such as base station location, base station identifier, coverage area, and performance indicators. After identifying rural residential areas, it is necessary to select appropriate locations within the corresponding areas to plan base stations in order to cover as many buildings as possible. Therefore, this embodiment of the invention filters buildings in a preset concentrated area based on the deployment status and corresponding parameter information of each base station in the existing network, i.e., the existing network base station information, and then selects the corresponding building locations to generate the desired base station planning sites, so as to maximize the coverage of the base stations and reduce unnecessary overlapping coverage.

[0084] Reference Figure 3 In some embodiments of the present invention, before performing identification and analysis on the rural remote sensing image to be identified using a preset building identification model to obtain preset building data, the rural base station planning method provided by the embodiments of the present invention further includes, but is not limited to, the following steps:

[0085] Step S210: Obtain the remote sensing image dataset and perform a first preprocessing step on the remote sensing image dataset to obtain the model training dataset. The first preprocessing step includes resizing, image normalization, and data augmentation.

[0086] Step S220: Construct a preset training model. The preset training model includes a visual converter model.

[0087] Step S230: Input the model training dataset into the visual converter model to train the model using the backpropagation algorithm and a preset loss function to obtain a preset building recognition model.

[0088] In this specific embodiment, the present invention first acquires a remote sensing image dataset and performs a first preprocessing on the remote sensing image dataset to obtain a model training dataset. Specifically, in this embodiment, the remote sensing image dataset includes remote sensing images and labeled building outlines (pixels belonging to buildings in the image are marked as 1, and others are marked as 0), for use in training a building segmentation and recognition model, such as... Figure 4 As shown. Accordingly, this embodiment of the invention performs a first preprocessing on the images in the remote sensing image dataset to increase the robustness of the model. Specifically, this embodiment adjusts the image size (resize adjustment) to uniformly adjust images of different sizes to the same size, such as cropping training images to a 1024*1024 format, to alleviate the bias problem caused by differences in image size and facilitate input into the model. Simultaneously, this embodiment standardizes pixel values, for example, scaling pixel values ​​to the range of [0,1] or [-1,1] by subtracting the mean and dividing by the standard deviation, to mitigate the impact of different feature value scales. Furthermore, this embodiment performs various random transformations on the original images, such as random cropping, flipping, rotation, and scaling, to generate additional training samples to achieve data augmentation and improve model robustness.

[0089] Next, this embodiment of the invention constructs a preset training model. Specifically, the preset training model in this embodiment includes a Visual Transformer (ViT) model. For example, this embodiment uses the vit-base-patch16-224 large visual model as the preset training model. Then, this embodiment inputs the model training dataset into the Visual Transformer model to train the model using a backpropagation algorithm and a preset loss function to obtain a preset building recognition model. Specifically, this embodiment includes pixel-level cross-entropy loss or Dice loss to measure the difference between the model prediction and the real annotation. For example, this embodiment inputs the model training dataset, uses Adam as the optimizer for training, and sets the learning rate to 2e. 10-5 The model is trained using the Dice loss function and with 64 cores / RTX 4080 hardware resources. Furthermore, this embodiment of the invention monitors the model's performance using a corresponding validation dataset and adjusts hyperparameters to improve model performance, thereby completing the training of the preset building recognition model.

[0090] Reference Figure 5 In some embodiments of the present invention, a preset building recognition model is used to identify and analyze the rural remote sensing image to be identified, thereby obtaining preset building data, including but not limited to the following steps:

[0091] Step S310: Perform a second preprocessing on the rural remote sensing image to be identified to obtain image patches to be identified. The second preprocessing includes image segmentation and image normalization.

[0092] Step S320: Input the image block to be recognized into the preset building recognition model for feature extraction and generate feature vector.

[0093] Step S330: Perform mask prediction based on the feature vector to obtain the image segmentation prediction mask.

[0094] Step S340: Map and transform the image segmentation prediction mask to obtain a preset mask raster image.

[0095] Step S350: Convert the preset mask raster map into a preset geographic information system data format to obtain preset building data.

[0096] In this specific embodiment, the present invention first performs a second preprocessing on the rural remote sensing image to be identified, obtaining image patches to be identified. Specifically, before performing identification analysis, the present invention first performs a second preprocessing on the acquired rural remote sensing image to be identified, including image segmentation and image normalization. Accordingly, the present invention first segments the image into fixed-size patches, for example, segments the image into patches up to 1024*1024, and these patches are usually uniformly divided. Then, each patch is converted into a vector, which is used as input to the ViT model. At the same time, the present invention also performs image normalization to make the image data suitable for a preset building recognition model. Next, the present invention inputs the image patches to be identified into a preset building recognition model library for feature extraction, generating feature vectors. Specifically, the present invention sends the preprocessed image patches to be identified into the input layer of the ViT model (preset building recognition model), and the model generates feature vectors for each image patch to be identified. These feature vectors are processed by the model's multi-layer self-attention mechanism to capture contextual information in the image. Furthermore, in this embodiment of the invention, a mask prediction is performed based on the feature vector to obtain an image segmentation prediction mask. This image segmentation prediction mask is then mapped and transformed to obtain a preset mask raster image, which is then converted into a preset geographic information system data format to obtain preset building data. Specifically, in this embodiment of the invention, a decoder is added to the top of the ViT model to generate an image segmentation mask, i.e., a mask of the same size as the input image, where each pixel corresponds to a predicted semantic category; for example, pixels representing buildings are marked as 1, and others as 0. Correspondingly, this decoder consists of a series of convolutional layers, and the output of the last convolutional layer is the mask prediction, i.e., the image segmentation prediction mask. For example, in this embodiment of the invention, a trained image segmentation model is used for inference, outputting an image segmentation prediction mask of the same size as the input image, where the hyperparameters are box_threshold = 0.19 and text_threshold = 0.1. Next, this embodiment of the invention maps and transforms the image segmentation prediction mask to convert the predicted mask data into a pixel-level grid, i.e., a pixel-level classification mask raster (preset mask raster), to represent the classification information of each pixel. Finally, this embodiment of the invention converts the pixel-level classification mask raster into a preset geographic information system data format, such as a vector shapefile format, to facilitate the storage, display, and analysis of building data. It is readily understood that this embodiment of the invention uses deep learning image algorithms to automatically identify and locate the positions of rural buildings in remote sensing imagery, achieving intelligent and rapid acquisition of rural building layers, and providing more comprehensive and accurate basic data for base station planning.

[0097] In some embodiments of the present invention, a preset clustering algorithm is used to perform cluster analysis on preset building data to obtain preset concentrated area data, including but not limited to the following steps:

[0098] Construct preset clustering parameters. These preset clustering parameters include preset neighborhood radius and neighborhood sample threshold.

[0099] Based on preset clustering parameters, a density-based spatial clustering algorithm is used to identify areas where buildings are clustered in the preset building data, thus obtaining preset concentrated area data.

[0100] In this specific embodiment, the present invention sets preset clustering parameters to perform clustering analysis on preset building data using a density-based spatial clustering algorithm, thereby obtaining corresponding preset concentrated area data. Specifically, the present invention uses a density-based spatial clustering algorithm (DBSCAN algorithm) for clustering analysis. The preset clustering parameters include a preset neighborhood radius and a neighborhood sample threshold, such as setting the preset neighborhood radius to 500 meters and the neighborhood sample threshold to 30. Accordingly, in this embodiment, the preset neighborhood radius refers to the neighborhood radius scanned when dividing the buildings into regions, and the neighborhood sample threshold refers to the minimum number of points (minPts) of buildings within the preset radius neighborhood. Then, based on the preset neighborhood radius and neighborhood, the present invention uses the DBSCAN algorithm to divide the buildings into different residential areas using density connections, and identifies noise points, thereby obtaining the concentrated areas of buildings, i.e., the preset concentrated area data. For example, Figure 6 As shown in the figure, the orange blocks represent the identified building areas.

[0101] In some embodiments of the present invention, regions where buildings are clustered in preset building data are identified using a density-based spatial clustering algorithm based on preset clustering parameters to obtain preset concentrated region data, including but not limited to the following steps:

[0102] The system detects the first unlabeled building in the preset building data to determine whether the number of second buildings contained in the first building within the preset neighborhood radius is less than the neighborhood sample threshold.

[0103] When it is determined that the number of second buildings contained within the preset neighborhood radius of the first building is greater than the neighborhood sample threshold, a preset residential area is constructed, and the second building is added to the preset candidate set.

[0104] Determine whether the number of unlabeled third buildings in the preset candidate set that contain fourth buildings within a preset neighborhood radius is greater than the neighborhood sample threshold.

[0105] When it is determined that the number of fourth buildings contained within a preset neighborhood radius of the third building is greater than the neighborhood sample threshold, the fourth building is added to the preset candidate set.

[0106] When it is determined that the third building is not marked in the residential area, the third building is added to the preset residential area, and the process returns to the step of determining whether the number of fourth buildings contained in the preset neighborhood radius of the unmarked third building in the preset candidate set is greater than the neighborhood sample threshold, until the preset candidate set is traversed.

[0107] Once the preset candidate set has been traversed, the process returns to the step of detecting the first unmarked building in the preset building data, and determining whether the number of second buildings contained in the first building within the preset neighborhood radius is less than the neighborhood sample threshold, until all the first buildings in the preset building data are marked, thus obtaining the preset concentrated area data.

[0108] In this specific embodiment, the present invention first detects an unlabeled first building in the preset building data to determine whether the number of second buildings contained within the first building in the preset neighborhood radius is less than a neighborhood sample threshold. When it is determined that the number of second buildings contained within the first building in the preset neighborhood radius is greater than the neighborhood sample threshold, a preset residential area is constructed, and the second building is added to a preset candidate set. Specifically, this embodiment of the present invention first detects unlabeled buildings x. i If a building (x) contains at least minPts (neighborhood sample threshold) of a second building within its radius e, then a residential area C (preset residential area) is established, and all buildings within its neighborhood are added to a candidate set N (preset candidate set). In this embodiment, the first building refers to a building that is not classified into any residential area or is not identified as noise. Accordingly, this embodiment detects this building x... i The number of buildings contained within a radius *e* of a given value. If this number is not less than a neighborhood sample threshold, then building *x* is considered a valid building. i It is a core point, and a new residential area C is built around it. Then, building x... i All buildings within the neighborhood (including building x) i The building itself is added to a candidate set N. These buildings can be core points or edge points (i.e., the number of points in the neighborhood is less than minPts but belongs to the neighborhood of a certain core point).

[0109] Next, this embodiment of the invention determines whether the number of unlabeled third buildings in the preset candidate set that contain fourth buildings within a preset neighborhood radius is greater than a neighborhood sample threshold. If the number of third buildings containing fourth buildings within the preset neighborhood radius is greater than the neighborhood sample threshold, this embodiment of the invention adds the fourth building to the preset candidate set. Specifically, this embodiment of the invention detects unlabeled buildings x in the candidate set N (preset candidate set). j (Third building), when the third building has more than minPts buildings within a preset neighborhood radius, these buildings (fourth building) are added to the candidate set N. Correspondingly, when it is determined that the third building is not marked as a residential area, this embodiment adds the third building to the preset residential area and returns to the step of determining whether the number of unmarked third buildings in the preset candidate set that contain fourth buildings within the preset neighborhood radius is greater than the neighborhood sample threshold, until the preset candidate set is traversed completely. Specifically, after adding the fourth building to the preset candidate set, this embodiment determines whether the third building is classified into the corresponding residential area. Correspondingly, if building x... j If a third building is not assigned to any residential area (i.e., it has not yet been marked as any residential area), then this third building is added to residential area C (the preset residential area) to include more buildings in the current residential area, while simultaneously discovering new core points and expanding the candidate set. Meanwhile, this embodiment of the invention continues to detect other unmarked buildings in the preset candidate set until the preset candidate set is empty, meaning that all buildings in the preset candidate set have been marked. At the boundary of the current residential area C, no more buildings can be added through density connections, completing the iteration.

[0110] Furthermore, after traversing the preset candidate set, this embodiment of the invention returns to the step of detecting unmarked first buildings in the preset building data to determine whether the number of second buildings contained in the first building within a preset neighborhood radius is less than a neighborhood sample threshold, until all first buildings in the preset building data are marked, thus obtaining preset concentrated area data. Specifically, after marking each building in the preset candidate set and completing the construction of a preset residential area, this embodiment of the invention returns to the step of detecting unmarked first buildings in the preset building data to select the next unmarked building to begin the construction process of a new residential area. Accordingly, this process is repeated until all buildings are classified into a certain residential area or marked as noise. Here, noise refers to isolated points that neither belong to any residential area nor meet the core point conditions.

[0111] In some embodiments of the present invention, the existing network base station information includes existing network base station coverage area data. Accordingly, in these embodiments, site planning is performed based on preset centralized area data and existing network base station information to generate desired base station planning sites, including but not limited to the following steps:

[0112] Preset operating parameter data is obtained to calculate the coverage area data of the existing network base stations. This preset operating parameter data includes base station identifier, base station coordinates, base station azimuth, base station frequency band, and base station transmit power.

[0113] Based on the coverage area data of the existing network base stations, buildings are screened in the preset centralized area data to obtain several candidate buildings.

[0114] The candidate building with the largest number of neighboring buildings among several candidate buildings is selected as the target building, and the desired base station planning site is generated based on the target building.

[0115] In this specific embodiment, the present invention first obtains preset engineering parameter data to calculate the existing network container coverage area data. Then, based on the existing network base station coverage area data, it filters buildings in the preset centralized area data to obtain several candidate buildings. Specifically, the preset engineering parameter data in this embodiment includes base station identifier (base station ID), base station coordinate data (base station latitude and longitude), base station azimuth angle, base station frequency band, and base station transmit power. Correspondingly, the present invention calculates the existing network base station coverage area data, i.e., the existing base station coverage area, based on the preset engineering parameter data of each base station in the existing network. The present invention filters each building in the preset centralized area data by combining the coverage area data of each base station in the existing network to identify buildings that do not meet the requirements, obtaining several candidate buildings. Finally, the present invention selects the candidate building with the largest number of neighboring buildings among the candidate buildings as the target building to generate the desired base station planning site based on the target building. Specifically, after filtering, the remaining buildings (candidate buildings) are all potential base station site candidate targets. Accordingly, in this embodiment of the invention, the number of neighboring buildings of each candidate building within a preset distance threshold range, such as a radius of 1.5*r (where r is the coverage radius defined according to the frequency band of the planned station), is compared, and the building with the most neighboring buildings is selected as the final planned site, i.e., the desired base station planned site. Figure 7 As shown, this allows for coverage of the most buildings, achieving optimal coverage. Among them, Figure 7 This is a schematic diagram of base station planning provided in an embodiment of the present invention. In the diagram, orange represents the identified building data, green circles represent the existing base station 1.5km buffer zone, and red represents the base station planning point.

[0116] In some embodiments of the present invention, building screening is performed on a preset centralized area data based on existing network base station coverage area data to obtain a number of candidate buildings, including but not limited to the following steps:

[0117] The coverage radius of the existing network base stations is determined based on the preset engineering parameters.

[0118] Based on the coverage radius of the existing network base stations, the covered buildings in the preset centralized area data are removed to obtain the uncovered buildings.

[0119] The number of buildings within a preset coverage radius of uncovered buildings is counted to obtain the number of neighboring buildings. The preset coverage radius includes the expected coverage radius of the pre-planned base stations.

[0120] Uncovered buildings with fewer than a preset threshold number of neighboring buildings are removed to obtain dense buildings.

[0121] The distance between the existing network base station and the planned base station is calculated based on the coverage radius of the existing base station and the preset coverage radius. The range of the planned base station is determined by the base station distance, and dense buildings located within the range of the planned base station are eliminated to obtain a number of candidate buildings.

[0122] In this specific embodiment, the present invention first determines the coverage radius of the existing network base stations based on preset engineering parameter data. Then, it removes covered buildings from the preset concentrated area data based on the existing base station coverage radius to obtain uncovered buildings. Specifically, in this embodiment, the existing network base station coverage radius refers to the existing network base station coverage radius R. Since different base station engineering parameter configurations, such as frequency band and power, result in different coverage areas, R in this embodiment can be set and calculated based on the base station engineering parameters (i.e., preset engineering parameter data), such as a base station coverage radius R = 1km. Correspondingly, covered buildings in this embodiment refer to buildings already covered by existing base stations (preset existing network base stations). This embodiment uses the existing network base station coverage radius to filter each building in the preset concentrated area data to accurately identify buildings covered by existing base stations, thereby filtering out uncovered buildings. Next, this embodiment counts the number of buildings within the preset coverage radius of the uncovered buildings to obtain the number of neighboring buildings. Then, uncovered buildings with a neighboring building number less than a preset threshold are removed to obtain densely packed buildings. Specifically, in this embodiment of the invention, the preset coverage radius refers to the coverage radius r of the planned base station, defined according to the frequency band of the planned base station, such as r = 1km. After determining the frequency band of the planned base station and the corresponding coverage radius r, for each building in the area, this embodiment of the invention queries all other buildings located within its coverage radius r, and these buildings together constitute a neighborhood. Then, this embodiment of the invention counts how many other buildings are in the neighborhood of each uncovered building, that is, it counts the number of neighborhood buildings, which reflects the potential coverage effect of the location as a base station site. Accordingly, in order to ensure that there are enough buildings around the selected base station site to improve the utilization efficiency and coverage effect of the base station, this embodiment of the invention removes uncovered buildings whose number of buildings in the neighborhood is less than a preset threshold S (such as S = 30), obtaining dense buildings, thereby avoiding the construction of base stations in overly sparse areas and effectively saving resources. Finally, this embodiment of the invention calculates the base station distance between the preset existing network base station and the preset planned base station based on the coverage radius of the existing network base station and the preset coverage radius, so as to determine the range of the planned base station through the base station distance, and remove dense buildings located within the range of the planned base station to obtain a number of candidate buildings. Specifically, to avoid overlapping coverage between newly planned base stations and existing base stations, which could lead to resource waste and potential signal interference, this embodiment of the invention excludes buildings located within the sum of the coverage radius R of the existing base stations and the coverage radius r of the newly planned base stations (base station distance) (the planned base station range). This ensures that the newly planned base stations maintain a certain distance from the existing base stations, thereby optimizing the overall network layout. Correspondingly, the planned base station range refers to the area defined by a pre-defined planned base station as the center and the base station distance between the pre-defined existing network base stations and the pre-defined planned base station as the radius.In this embodiment of the invention, the planned base station range D = the coverage radius R of the existing network base station + the coverage radius r of the planned base station.

[0123] It is easy to understand that the embodiments of the present invention obtain rural building cluster areas through clustering algorithms, and combine them with the base station coverage radius and the number of neighboring buildings. Through building elimination and site selection shrinkage, planned sites are generated, realizing automatic and accurate identification of high-value sites in rural areas. This reduces the cost of manual planning while improving planning efficiency.

[0124] The following section provides a detailed introduction and explanation of the solutions in this embodiment of the invention, using a specific rural base station planning scenario as an example:

[0125] For example, such as Figure 8 As shown, this embodiment of the invention first trains a preset training model using a remote sensing image dataset to obtain an image segmentation and recognition model, i.e., a preset building recognition model. Then, this embodiment acquires remote sensing images of rural areas to be identified, and uses the preset building recognition model to perform image segmentation and recognition processing on these images to obtain rural building vectors, i.e., preset building data. Further, this embodiment uses the DBSCAN clustering algorithm to perform cluster analysis on the rural building vectors to obtain rural residential areas, i.e., preset concentrated area data. Finally, this embodiment performs site planning based on existing network base station information and rural concentrated residential areas to generate the desired base station planning site.

[0126] It is readily understood that this invention, through the fusion of remote sensing imagery and AI technology, extracts the outlines of rural buildings and applies them to rural base station planning. Specifically, this invention acquires high-resolution remote sensing imagery based on geographic location, performs data annotation and image processing, and trains a remote sensing imagery building recognition model (i.e., a preset building recognition model) using image segmentation and recognition techniques in deep learning. Then, based on model inference, rural buildings are extracted, integrated, and output as a rural building layer. Subsequently, a clustering algorithm from machine learning is used to identify concentrated rural residential areas, eliminating buildings covered by existing network base stations and scattered buildings within these areas. Finally, site selection is narrowed down based on base station coverage radius, selecting sites on the buildings with the most surrounding buildings, thus completing the base station planning for rural areas. This invention is suitable for base station planning in rural areas lacking coverage. By integrating GIS and AI image processing technologies, it intelligently supplements rural building data, providing a foundation for rural value assessment. Simultaneously, the site generation method is highly efficient and accurate, significantly saving on wireless network planning costs.

[0127] Please see Figure 9 This application also provides a rural base station planning system that can implement the above-mentioned rural base station planning method. The system includes:

[0128] The first module 410 is used to acquire remote sensing images of rural areas to be identified.

[0129] The second module 420 is used to identify and analyze the rural remote sensing images to be identified using a preset building recognition model, and obtain preset building data.

[0130] The third module 430 is used to perform cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data.

[0131] The fourth module 440 is used to plan site locations based on preset centralized area data and existing network base station information, and generate the desired base station planning site.

[0132] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0133] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described rural base station planning method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0134] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0135] Please see Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0136] The processor 510 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0137] The memory 520 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 520 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 520 and called by the processor 510 to execute the rural base station planning method of the embodiments of this application.

[0138] The input / output interface 530 is used to implement information input and output;

[0139] The communication interface 540 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0140] Bus 550 transmits information between various components of the device (e.g., processor 510, memory 520, input / output interface 530, and communication interface 540);

[0141] The processor 510, memory 520, input / output interface 530 and communication interface 540 are connected to each other within the device via bus 550.

[0142] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described rural base station planning method.

[0143] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0144] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0145] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0146] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0147] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0148] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0149] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0150] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0151] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0152] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0153] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0154] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0155] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for planning rural base stations, characterized in that, The method includes the following steps: Acquire remote sensing images of rural areas to be identified; The rural remote sensing image to be identified is analyzed by a preset building recognition model to obtain preset building data. The preset building data is clustered using a preset clustering algorithm to obtain preset concentrated area data; Based on the preset centralized area data and existing network base station information, site planning is performed to generate the desired base station planning site; The step of identifying and analyzing the rural remote sensing image to be identified using a preset building recognition model to obtain preset building data includes: The rural remote sensing image to be identified is subjected to a second preprocessing to obtain image blocks to be identified; wherein, the second preprocessing includes image segmentation and image normalization; The image block to be identified is input into the preset building recognition model for feature extraction, generating a feature vector; Based on the feature vector, a mask prediction is performed to obtain the image segmentation prediction mask; The image segmentation prediction mask is mapped and transformed to obtain a preset mask raster image; The preset mask raster image is converted into a preset geographic information system data format to obtain the preset building data; The existing network base station information includes data on the coverage area of ​​the existing network base stations; The step of planning site locations based on the preset centralized area data and existing network base station information to generate desired base station planning sites includes: Acquire preset operating parameter data to calculate the coverage area data of the existing network base station; wherein, the preset operating parameter data includes base station identifier, base station coordinate data, base station azimuth angle, base station frequency band, and base station transmit power; Based on the existing network base station coverage area data, the preset centralized area data is filtered to obtain several candidate buildings; The candidate building with the largest number of neighboring buildings among the candidate buildings is selected as the target building, and the desired base station planning site is generated based on the target building; The step of filtering buildings in the preset centralized area data based on the existing network base station coverage area data to obtain several candidate buildings includes: The coverage radius of the existing network base station is determined based on the preset operating parameter data; Based on the coverage radius of the existing network base stations, the covered buildings in the preset centralized area data are removed to obtain the uncovered buildings; The number of buildings within a preset coverage radius of the uncovered buildings is counted to obtain the number of neighboring buildings; wherein, the preset coverage radius includes the expected coverage radius of the preset planned base station; The uncovered buildings whose number of neighboring buildings is less than a preset threshold are removed to obtain dense buildings; The base station distance between the preset existing base station and the preset planned base station is calculated based on the coverage radius of the existing base station and the preset coverage radius. The range of the planned base station is determined by the base station distance, and the dense buildings located within the range of the planned base station are eliminated to obtain a number of candidate buildings.

2. The method according to claim 1, characterized in that, Before performing the identification and analysis of the rural remote sensing image to be identified using a preset building identification model to obtain preset building data, the method further includes: A remote sensing image dataset is acquired, and the remote sensing image dataset is subjected to a first preprocessing step to obtain a model training dataset; wherein, the first preprocessing step includes resizing, image normalization, and data augmentation. Construct a preset training model; wherein, the preset training model includes a visual converter model; The model training dataset is input into the visual converter model to train the model using a backpropagation algorithm and a preset loss function, thereby obtaining the preset building recognition model.

3. The method according to claim 1, characterized in that, The step of performing cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data includes: Construct preset clustering parameters; wherein, the preset clustering parameters include preset neighborhood radius and neighborhood sample threshold; Based on the preset clustering parameters, a density-based spatial clustering algorithm is used to identify the regions where buildings are clustered in the preset building data, thereby obtaining the preset concentrated region data.

4. The method according to claim 3, characterized in that, The step of identifying areas where buildings cluster in the preset building data using a density-based spatial clustering algorithm based on the preset clustering parameters to obtain the preset clustered area data includes: Detect the first unmarked building in the preset building data to determine whether the number of second buildings contained in the first building within the preset neighborhood radius is less than the neighborhood sample threshold; When it is determined that the number of times the first building contains the second building within the preset neighborhood radius is greater than the neighborhood sample threshold, a preset residential area is constructed, and the second building is added to the preset candidate set; Determine whether the number of unmarked third buildings in the preset candidate set that contain fourth buildings within the preset neighborhood radius is greater than the neighborhood sample threshold. When it is determined that the number of fourth buildings contained within the preset neighborhood radius of the third building is greater than the neighborhood sample threshold, the fourth building is added to the preset candidate set. When it is determined that the third building is not marked in the residential area, the third building is added to the preset residential area, and the process returns to the step of determining whether the number of fourth buildings contained in the preset neighborhood radius of the unmarked third building in the preset candidate set is greater than the neighborhood sample threshold, until the preset candidate set is traversed. When it is determined that the preset candidate set has been traversed, the step of returning to detect the first unmarked building in the preset building data, and determining whether the number of second buildings contained in the first building within the preset neighborhood radius is less than the neighborhood sample threshold, continues until all the first buildings in the preset building data are marked, and the preset concentrated area data is obtained.

5. A rural base station planning system, characterized in that, The system includes: The first module is used to acquire remote sensing images of rural areas to be identified; The second module is used to identify and analyze the rural remote sensing image to be identified using a preset building identification model to obtain preset building data. The third module is used to perform cluster analysis on the preset building data using a preset clustering algorithm to obtain preset concentrated area data; The fourth module is used to plan site locations based on the preset centralized area data and existing network base station information, and generate the desired base station planning site. The step of identifying and analyzing the rural remote sensing image to be identified using a preset building recognition model to obtain preset building data includes: The rural remote sensing image to be identified is subjected to a second preprocessing to obtain image blocks to be identified; wherein, the second preprocessing includes image segmentation and image normalization; The image block to be identified is input into the preset building recognition model for feature extraction, generating a feature vector; Based on the feature vector, a mask prediction is performed to obtain the image segmentation prediction mask; The image segmentation prediction mask is mapped and transformed to obtain a preset mask raster image; The preset mask raster image is converted into a preset geographic information system data format to obtain the preset building data; The existing network base station information includes data on the coverage area of ​​the existing network base stations; The step of planning site locations based on the preset centralized area data and existing network base station information to generate desired base station planning sites includes: Acquire preset operating parameter data to calculate the coverage area data of the existing network base station; wherein, the preset operating parameter data includes base station identifier, base station coordinate data, base station azimuth angle, base station frequency band, and base station transmit power; Based on the existing network base station coverage area data, the preset centralized area data is filtered to obtain several candidate buildings; The candidate building with the largest number of neighboring buildings among the candidate buildings is selected as the target building, and the desired base station planning site is generated based on the target building; The step of filtering buildings in the preset centralized area data based on the existing network base station coverage area data to obtain several candidate buildings includes: The coverage radius of the existing network base station is determined based on the preset operating parameter data; Based on the coverage radius of the existing network base stations, the covered buildings in the preset centralized area data are removed to obtain the uncovered buildings; The number of buildings within a preset coverage radius of the uncovered buildings is counted to obtain the number of neighboring buildings; wherein, the preset coverage radius includes the expected coverage radius of the preset planned base station; The uncovered buildings whose number of neighboring buildings is less than a preset threshold are removed to obtain dense buildings; The base station distance between the preset existing base station and the preset planned base station is calculated based on the coverage radius of the existing base station and the preset coverage radius. The range of the planned base station is determined by the base station distance, and the dense buildings located within the range of the planned base station are eliminated to obtain a number of candidate buildings.

6. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method as described in any one of claims 1-4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 4.

Images