Power line deployment method and device based on three-dimensional model, equipment and medium

Abstract

Embodiments of the present disclosure disclose a power transmission line deployment method and device based on a three-dimensional model, an equipment and a medium. A specific implementation of the method comprises: acquiring a remote sensing atlas and a road network map; generating regional landscape types and weather information according to the remote sensing atlas; selecting a ground feature recognition model; generating a ground feature recognition result using the ground feature recognition model; querying a corresponding power transmission line construction case set; constructing a power transmission line deployment model supporting visual editing for the target region according to the power transmission line construction case set and the ground feature recognition result; and executing power transmission line construction for the target region and anomaly detection in the construction process according to the power transmission line deployment model. The implementation can realize visual display and editing of various virtual lines in the target region by constructing a power transmission line deployment model, intuitively obtain regional content and power transmission line deployment status, and assist in the deployment of power transmission lines, greatly improving the deployment efficiency of power transmission lines.

Description

Technical Field

[0001] Embodiments of this disclosure relate to the field of computer technology, and more specifically to a method, apparatus, electronic device, and computer-readable medium for deploying power transmission lines based on a three-dimensional model. Background Technology

[0002] Currently, with the increasing contradiction between the reverse distribution of energy resources and load centers, the construction of transmission lines faces the dual challenges of land resource constraints and safe operation. How to effectively deploy transmission lines has become one of the main needs. The common approach to transmission line deployment is to use remote sensing images, point cloud layout maps, and plan maps of the target area to set up a corresponding transmission line deployment plan for that area.

[0003] However, when using the above method, the following technical problems often arise: Remote sensing images and point cloud layout maps are greatly affected by regional landscape types and weather, which may lead to regional accuracy issues. In addition, remote sensing images, point cloud layout maps, and plan maps of different formats cannot intuitively show the regional conditions within the target area, and the performance of subsequently deployed power transmission lines cannot be intuitively obtained. The corresponding deployment effect can only be intuitively obtained after the power transmission lines are deployed, which may greatly affect the deployment efficiency of power transmission lines.

[0004] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of this disclosure provide a method, apparatus, electronic device, and computer-readable medium for deploying power transmission lines based on three-dimensional models to address one or more of the technical problems mentioned in the background section above.

[0007] In a first aspect, some embodiments of this disclosure provide a method for deploying power transmission lines based on a three-dimensional model, comprising: acquiring a remote sensing atlas and a road network map corresponding to a target area for power transmission line construction; generating regional landscape type and weather information corresponding to the target area based on the remote sensing atlas; selecting a feature recognition model corresponding to the regional landscape type and the weather information; generating feature recognition results corresponding to the target area using the feature recognition model; querying a power transmission line construction case set corresponding to the regional landscape type and the weather information from a power transmission line case database; constructing a power transmission line deployment model for the target area that supports visual editing based on the power transmission line construction case set and the feature recognition results, wherein the power transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment, and supports the import of multimodal power transmission line construction files; and performing power transmission line construction and anomaly detection during the construction process for the target area based on the power transmission line deployment model.

[0008] Secondly, some embodiments of this disclosure provide a power transmission line deployment device based on a three-dimensional model, including: an acquisition unit configured to acquire a remote sensing atlas and a road network map corresponding to a target area for power transmission line construction; a first generation unit configured to generate regional landscape type and weather information corresponding to the target area based on the remote sensing atlas; a selection unit configured to select a land feature recognition model corresponding to the regional landscape type and the weather information; a second generation unit configured to generate land feature recognition results corresponding to the target area using the land feature recognition model; and a query unit configured to query a power transmission line case database. The system queries the landform types and weather information of the aforementioned regions to obtain a set of power transmission line construction case studies. A construction unit is configured to build a visually editable power transmission line deployment model for the target regions based on the aforementioned power transmission line construction case studies and the aforementioned feature identification results. This model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment. The model also supports the import of multimodal power transmission line construction files. An execution unit is configured to perform power transmission line construction and anomaly detection during the construction process for the target regions based on the aforementioned power transmission line deployment model.

[0009] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0010] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0011] The above-described embodiments of this disclosure have the following beneficial effects: Through the three-dimensional model-based power transmission line deployment method of some embodiments of this disclosure, by constructing a power transmission line deployment model, the visualization and editing of various virtual lines within the target area can be realized, allowing for intuitive acquisition of regional content and power transmission line deployment status, thus assisting in the deployment of power transmission lines and greatly improving deployment efficiency. Specifically, the reason for the low deployment efficiency of related power transmission lines is that remote sensing images and point cloud layout maps are greatly affected by regional landscape types and weather, potentially leading to regional accuracy issues. Furthermore, remote sensing images, point cloud layout maps, and plan maps of different representation forms cannot intuitively represent the regional conditions within the target area, and the performance of subsequently deployed power transmission lines cannot be intuitively obtained. The corresponding deployment effect can only be intuitively obtained after the power transmission lines are deployed, which may significantly affect the deployment efficiency. Based on this, the three-dimensional model-based power transmission line deployment method of some embodiments of this disclosure first acquires a remote sensing atlas corresponding to the target area for constructing the power transmission lines, in order to obtain the regional landscape and weather conditions within the target area from the perspective of remote sensing information. Then, based on the aforementioned remote sensing atlas, the corresponding regional landscape type and weather information for the target area can be accurately generated. Here, generating the regional landscape type and weather information facilitates the selection of a suitable ground feature recognition model based on the remote sensing atlas. Next, the ground feature recognition model corresponding to the aforementioned regional landscape type and weather information is selected to achieve accurate identification of remote sensing content in the remote sensing atlas, ensuring the accuracy of ground feature identification within the target area. Then, a transmission line construction case set corresponding to the aforementioned regional landscape type and weather information is queried from the transmission line case database. Here, querying the transmission line construction case set, from the perspective of construction cases, assists in the subsequent deployment of transmission lines within the target area. Furthermore, based on the aforementioned transmission line construction case set and the aforementioned ground feature identification results, a transmission line deployment model supporting visual editing is constructed for the aforementioned target area. This transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment. The transmission line deployment model also supports the import of multimodal transmission line construction files. Here, by utilizing the results of feature identification and a case study set of power transmission line construction, a precise 3D model of the target area can be constructed. Furthermore, the constructed 3D model supports the creation of virtual power lines and the acquisition of corresponding line information, enabling diverse visualization operations and real-time acquisition of virtual power line deployment within the target area. In addition, importing power transmission line construction files is supported, significantly improving deployment efficiency. Finally, based on the aforementioned power transmission line deployment model, power transmission line construction and anomaly detection during the construction process are performed in the target area.In summary, by constructing a power transmission line deployment model, it is possible to visualize and edit various virtual lines within the target area, intuitively obtain regional content and power transmission line deployment status, thereby assisting in the deployment of power transmission lines and greatly improving the deployment efficiency of power transmission lines. Attached Figure Description

[0012] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0013] Figure 1 This is a flowchart of some embodiments of the power transmission line deployment method based on a three-dimensional model according to the present disclosure; Figure 2 These are schematic diagrams of some embodiments of a power transmission line deployment device based on a three-dimensional model according to the present disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0014] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0015] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0016] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0017] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0018] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0019] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] refer to Figure 1 The diagram illustrates a flow 100 of some embodiments of a three-dimensional model-based power transmission line deployment method according to the present disclosure. This three-dimensional model-based power transmission line deployment method includes the following steps: Step 101: Obtain the remote sensing atlas corresponding to the target area for the construction of power transmission lines.

[0021] In some embodiments, the implementing entity (e.g., an electronic device) of the above-described 3D model-based power transmission line deployment method can acquire remote sensing atlases corresponding to the target area for power transmission line construction via wired or wireless means. The target area can be the region where power transmission lines are to be constructed and deployed. The remote sensing images in the atlas can be remote sensing images taken of the target area. In practice, each remote sensing image in the atlas can be an image taken by a remote sensing device configured with different resolutions.

[0022] Step 102: Based on the above remote sensing atlas, generate the regional landscape type and weather information corresponding to the above target area.

[0023] In some embodiments, the aforementioned executing entity can generate regional landscape type and weather information corresponding to the target area based on the aforementioned remote sensing atlas. The regional landscape type can be any type of landscape within the target area. In practice, the regional landscape type can be one of the following: mountainous landscape, plains landscape, mixed hilly and mountainous landscape, or river landscape. The weather information can be the weather conditions in the target area during the power transmission line construction process.

[0024] As an example, the aforementioned execution entity can input the highest-resolution remote sensing image from the remote sensing atlas into a regional landscape type recognition model to obtain the regional landscape type. This regional landscape type recognition model can be a pre-trained recognition model used for regional landscape identification. For example, it could be a YOLO model pre-trained based on regional landscape knowledge to support regional landscape type recognition. The execution entity can then query the weather forecast information for the target region to obtain weather information.

[0025] In some optional implementations of certain embodiments, the aforementioned remote sensing atlas includes subsets of remote sensing images at different resolutions. Specifically, the remote sensing atlas may include subsets of remote sensing images at high resolution, subsets of remote sensing images at medium resolution, and subsets of remote sensing images at low resolution.

[0026] Optionally, the aforementioned implementing entity can generate regional landscape type and weather information corresponding to the aforementioned target area based on the aforementioned remote sensing atlas, including the following steps: The first step involves using a subset of low-resolution remote sensing images from the aforementioned remote sensing atlas to generate preliminary regional landscape types and confidence levels for each sub-region corresponding to the target region. Each sub-region can be a portion of the target region. Each sub-region has a corresponding preliminary regional landscape type and confidence level. The preliminary regional landscape type can be the landscape type corresponding to the initially generated sub-region. The confidence level represents the probability that the sub-region's landscape type is the preliminary regional landscape type. In practice, the confidence level can be a value between 0 and 1. A higher confidence level indicates a higher probability that the sub-region's landscape type is the preliminary regional landscape type. There is a one-to-one correspondence between each sub-region and the remote sensing images in the subset of low-resolution remote sensing images.

[0027] It should be noted that the sub-regions can be pre-defined. The remote sensing images in the subset are low-resolution images taken for specific sub-regions. Here, for images with identical content captured in different images within the low-resolution subset, graph processing techniques can be used to remove duplicate content, thereby reducing the computational resources required for subsequent input into the regional landscape type recognition model.

[0028] As an example, the aforementioned implementing entity can first input each remote sensing image from the subset of remote sensing images into a pre-trained regional landscape type recognition model to obtain preliminary regional landscape types and confidence levels. The regional landscape type recognition model can be a recognition model pre-trained based on regional landscape knowledge. For example, the regional landscape type recognition model could be a YOLO model.

[0029] The second step involves merging the aforementioned sub-regions based on the preliminary regional landscape types and confidence levels, resulting in at least one merged sub-region. This merging can be achieved by combining adjacent sub-regions with the same preliminary regional landscape type and confidence levels higher than the target confidence level. The merged sub-region can be the result of merging at least one sub-region.

[0030] As an example, firstly, for the target sub-region within each sub-region, the first step is to identify at least one neighboring sub-region corresponding to the target sub-region. Secondly, for each neighboring sub-region within the at least one neighboring sub-region, it is determined whether the regional landscape type corresponding to the neighboring sub-region is the same as the regional landscape type corresponding to the target sub-region. In response to the determination that they are the same and both have a confidence level higher than the target confidence level, the neighboring sub-region is merged into the target sub-region to obtain a merged region. Thirdly, the merged sub-region and the target sub-region are removed from each of the above sub-regions to obtain the removed sub-regions. Fourthly, in response to the determination that there are no adjacent regions with the same regional landscape type among the removed sub-regions and that the confidence level of each is higher than the target confidence level, the removed sub-regions are identified as at least one merged sub-region.

[0031] The third step involves determining the regional construction probability of each of the at least one merging sub-regions based on a pre-established power transmission line installation scheme in the adjacent areas of the target region, in response to the determination of such scheme. The power transmission line installation scheme can be a construction scheme for power transmission lines already installed in adjacent areas. Adjacent areas can be regions geographically adjacent to the target region. In practice, the power transmission line installation scheme is a scheme that has been finalized and / or is currently under construction. The regional construction probability can be the probability of constructing power transmission lines within the merging sub-region.

[0032] As an example, for each integrated sub-region, the first transmission line construction score is determined based on the corresponding regional landscape type. The second transmission line construction score is determined based on the transmission line layout information in the transmission line installation plan. The third transmission line construction score is determined based on the compensation value within the integrated sub-region. Different regional landscape types have varying degrees of suitability for transmission line construction; regions with higher suitability have higher first transmission line construction scores. Transmission line layout information can be the transmission line layout direction in adjacent areas. In practice, the second transmission line construction score is higher when the integrated sub-region is positioned within the transmission line layout direction of adjacent areas, as this results in lower transmission line laying costs. The higher the compensation value within the integrated sub-region, the lower the corresponding third transmission line construction score. The compensation value can be the total compensation required for transmission line construction within the integrated sub-region. Higher compensation values ​​correspond to lower third transmission line construction scores. In practice, the first, second, and third transmission line construction scores are all values ​​between 0 and 1. Finally, the construction scores of the first, second, and third transmission lines are weighted and summed to obtain the regional construction probability.

[0033] The fourth step is to obtain, based on the obtained construction probability of at least one region, a first remote sensing image subset at high resolution or a second remote sensing image subset at medium resolution corresponding to each of the above at least one fused sub-regions.

[0034] As an example, for each fused sub-region, in response to the region construction probability corresponding to the aforementioned fused sub-region being higher than a first probability, a first subset of remote sensing images at high resolution corresponding to the aforementioned fused sub-region is obtained. In response to determining that the region construction probability is higher than a second probability but lower than the first probability, a second subset of remote sensing images at medium resolution corresponding to the aforementioned fused sub-region is obtained. The first probability is higher than the second probability.

[0035] The fifth step involves generating the aforementioned regional landscape types and weather information based on at least one subset of first remote sensing images, at least one subset of second remote sensing images, and the aforementioned preliminary regional landscape types.

[0036] As an example, firstly, for each first remote sensing image in at least one subset of first remote sensing images, a first regional landscape type identification model is used to determine the corresponding regional landscape type identification result for the first remote sensing image. Then, for each second remote sensing image in at least one subset of second remote sensing images, a second regional landscape type identification model is used to determine the corresponding regional landscape type identification result for the second remote sensing image. Next, the preliminary regional landscape types in each subset, corresponding to the regions identified by the respective regional landscape type identification results, are replaced with the respective regional landscape type identification results to obtain the replaced regional landscape types, which are then used as the regional landscape types.

[0037] As another example, the aforementioned implementing entity can leverage a broader external weather knowledge base to obtain precise weather identification for each region corresponding to at least one subset of first and at least one subset of second remote sensing images, thereby obtaining at least one set of first weather information. Then, for at least one other region besides those corresponding to at least one subset of first and at least one subset of second remote sensing images, it utilizes a basic external weather knowledge base to obtain at least one set of second weather information for that other region. Finally, the at least one set of first weather information and at least one set of second weather information are aggregated to obtain the overall weather information.

[0038] Optionally, the aforementioned executing entity can generate the aforementioned regional landscape type and weather information based on at least one subset of first remote sensing images, at least one subset of second remote sensing images, and the aforementioned preliminary regional landscape types, including the following steps: The first step is to determine the first candidate regional landscape type corresponding to each of the above preliminary regional landscape types, based on the aforementioned preliminary regional landscape types. The first candidate regional landscape type can be any candidate regional landscape type.

[0039] As an example, the aforementioned implementing entity can find the corresponding preliminary regional landscape type for each fusion sub-region from each preliminary regional landscape type, and use it as the first candidate regional landscape type.

[0040] The second step is to select target fusion sub-regions from the above at least one fusion sub-regions that contain corresponding high-resolution or medium-resolution remote sensing image subsets, thereby obtaining at least one target fusion sub-region.

[0041] Third, for each target fusion sub-region, perform the following first generation step: Sub-step 1: Based on the first or second remote sensing image subset corresponding to the target fusion sub-region, generate the second candidate region landscape type corresponding to the target fusion sub-region.

[0042] As an example, a subset of remote sensing images corresponding to the target fusion sub-region is designated as the first remote sensing image subset. Each first remote sensing image in the first remote sensing image subset is input into a first regional landscape type recognition model to generate a first candidate regional landscape subtype, resulting in a first candidate regional landscape subtype subset. The various first candidate regional landscape subtypes in the first candidate regional landscape subtype subset are then summarized to obtain a second candidate regional landscape type. Similarly, a subset of remote sensing images corresponding to the target fusion sub-region is designated as the second remote sensing image subset. Each second remote sensing image in the second remote sensing image subset is input into a second regional landscape type recognition model to generate a second candidate regional landscape subtype, resulting in a second candidate regional landscape subtype subset. The various second candidate regional landscape subtypes in the second candidate regional landscape subtype subset are then summarized to obtain a second candidate regional landscape type. Note that the model parameters and model structure of the first regional landscape type recognition model are more complex than those of the second regional landscape type recognition model. For example, the model structure of the image extraction model included in the first regional landscape type recognition model is the same as the model structure of the image extraction model included in the second regional landscape type recognition model. Furthermore, the model structure of the image extraction model included in the first regional landscape type recognition model has more model layers than the model structure of the image extraction model included in the second regional landscape type recognition model.

[0043] Sub-step 2: Replace the first candidate region landscape type corresponding to the above target fusion sub-region with the above second candidate region landscape type.

[0044] The fourth step is to generate the aforementioned regional landscape types based on at least one first candidate regional landscape type obtained.

[0045] As an example, the aforementioned implementing entity can summarize at least one first candidate regional landscape type to obtain the regional landscape type.

[0046] The fifth step involves revising the initial weather forecast for the target region based on the aforementioned regional landform types, thereby obtaining weather information. The initial weather forecast can be a prediction of the target region's weather over a future time period based on a weather-related knowledge base.

[0047] As an example, firstly, the aforementioned entity can query the initial forecast weather for the target region from a weather-related knowledge base. Then, it inputs the regional wind type and the initial forecast weather into a multimodal large language model to obtain weather information.

[0048] Optionally, the aforementioned execution entity may, based on the obtained at least one region construction probability, acquire a first subset of high-resolution remote sensing images or a second subset of medium-resolution remote sensing images corresponding to each of the at least one fused sub-regions, including the following steps: The first step involves obtaining a first subset of remote sensing images corresponding to the aforementioned fused sub-region, in response to the determination that the probability of construction in the corresponding area falls within a first probability interval. The first probability interval can be a probability interval set for high resolution. Each first probability value in the first probability interval is higher than each second probability value in a second probability interval. Each first probability value in the first probability interval can be set based on historical experience. That is, when the probability of construction in the corresponding area of ​​the fused sub-region is determined to be high, a corresponding high-resolution subset of remote sensing images is obtained for subsequent generation of regional landscape type and weather information.

[0049] The second step involves obtaining a second subset of remote sensing images corresponding to the aforementioned fused sub-region, in response to the determination that the probability of construction in the corresponding area falls within the second probability interval. The values ​​corresponding to the first probability region are higher than those corresponding to the second probability interval. The second probability interval can be a probability interval set for medium resolution. Each second probability value within the second probability interval is greater than each third probability value within the third probability interval. Each second probability value within the second probability interval can be set based on historical experience. That is, when the probability of construction in the corresponding area of ​​the fused sub-region is determined to be relatively high, a corresponding subset of medium-resolution remote sensing images is obtained for subsequent generation of regional landscape type and weather information.

[0050] Third, in response to determining that the construction probability of the area corresponding to the aforementioned fused sub-region falls within the third probability interval, information is generated indicating that the aforementioned fused sub-region will not acquire the first or second remote sensing image subset. The value corresponding to the third probability interval is lower than the value corresponding to the second probability interval. Each third probability value in the third probability interval can be set based on historical experience. Since the current construction probability of the area is within the third probability interval, it indicates that the fused sub-region will not deploy transmission lines. Therefore, it is not necessary to acquire the corresponding remote sensing image subset.

[0051] Step 103: Select the land feature recognition model corresponding to the above-mentioned regional landscape type and weather information.

[0052] In some embodiments, the executing entity may select a feature recognition model corresponding to the aforementioned regional landscape type and weather information. The feature recognition model may be a neural network model that identifies geographic and physical information within a region. Here, geographic information can represent the transportation network structure information of the target region. For example, geographic information may include: cities, towns, stations, ports, highways, urban roads, and railways. In practice, different regional landscape types and weather information have applicable feature recognition models. Here, by selecting a feature recognition model corresponding to the regional landscape type and weather information, accurate identification of geographic information and physical objects within the target region can be achieved.

[0053] As an example, the aforementioned implementing entity can select the corresponding feature recognition model based on regional landscape type and weather information by querying the association table. The association table represents the mapping relationship between regional landscape type, weather information, and feature recognition model. Each mapping relationship in the association table can be extracted from existing transmission line cases.

[0054] In some optional implementations of certain embodiments, the execution entity may select a land feature recognition model corresponding to the aforementioned regional landscape type and weather information, including the following steps: The first step is to obtain at least one regional construction probability corresponding to the aforementioned target region. Each regional construction probability has a corresponding fusion sub-region.

[0055] The second step involves dividing the target region into regional groups based on at least one of the aforementioned regional development probabilities. Each regional group has a corresponding regional development probability interval. There is a one-to-one correspondence between the regional groups within each regional group and the regional development probability intervals within each regional development probability interval.

[0056] As an example, for each region construction probability in at least one region construction probability, the sub-regions corresponding to the above region construction probabilities are divided into corresponding region groups according to the region construction probability intervals corresponding to the above region construction probabilities, thus obtaining a region group set.

[0057] Third, for each region group, perform the following third generation step: Sub-step 1: Based on the probability intervals corresponding to the aforementioned region groups, determine the corresponding model magnitude and image resolution. The model magnitude can be the magnitude of the land cover recognition model. In practice, the model magnitude can be measured based on the number of model parameters, model structure, and computational complexity. The image resolution can be the resolution of the input image.

[0058] Sub-step 2 involves generating an initial set of ground feature recognition models for the aforementioned region group, based on the model scale and image resolution described above. These initial models can be selected and determined for subsequent ground feature recognition. The models in the initial set can have different structures, all supporting the recognition of ground feature content at the specified image resolution. In practice, the initial set of ground feature recognition models may include: various recognition models based on convolutional neural networks and their variants, YOLO-based detection models, and visual language models (VLMs).

[0059] As an example, the aforementioned execution entity can query the model mapping table to obtain the initial land cover recognition model set corresponding to the model size and image resolution. The model mapping table represents the mapping relationship between model size, processable image resolution, and model identifier.

[0060] Sub-step 3 involves selecting at least one initial ground feature recognition model from the aforementioned set of initial ground feature recognition models, corresponding to the aforementioned regional landscape type and weather information. Each initial ground feature recognition model excels at different scenarios for ground feature recognition. The recognition accuracy of different initial ground feature recognition models varies depending on the landform type and the image under different weather conditions. Each initial ground feature recognition model in the at least one initial ground feature recognition model can be a recognition model for images with deleted regional landscape types and weather information.

[0061] As an example, the aforementioned implementing entity can filter at least one initial land feature recognition model corresponding to the aforementioned regional landform type and weather information from the model description data. The model description data can describe the various landform types and weather information that all initial land feature recognition models are adept at recognizing.

[0062] Sub-step 4: Based on the model usage frequency corresponding to the initial feature identification model, select the target feature identification model from at least one initial feature identification model. The model usage frequency can be the model scheduling frequency.

[0063] As an example, the aforementioned implementing entity can select the initial feature identification model with the highest usage frequency from at least one initial feature identification model as the target feature identification model.

[0064] The fourth step is to determine the obtained target feature recognition model set as the above-mentioned feature recognition model.

[0065] Step 104: Using the above-mentioned feature recognition model, generate the feature recognition results corresponding to the target area.

[0066] In some embodiments, the executing entity may utilize the aforementioned feature recognition model to generate feature recognition results corresponding to the target area. These feature recognition results may include both geographic recognition results and physical object recognition results for the target area. The geographic recognition results may describe the geographic conditions within the target area. The physical object recognition results may describe the category, location, and size of various physical objects within the target area.

[0067] As an example, the aforementioned implementing entity can input the overall regional image corresponding to the target region into the land feature recognition model to obtain the land feature recognition result.

[0068] In addressing the technical problems mentioned in the background section, and considering the application scenario—specifically, high-humidity areas—the following technical challenges arise: These high-humidity areas are often basins or valleys with enclosed terrain, hindering moisture dissipation and resulting in persistent fog and mist. While the air may not be inherently polluted, the naturally low atmospheric transparency leads to lower accuracy in feature identification, resulting in a lack of detailed information for subsequent 3D model construction. To meet the specific requirements of this application scenario—specifically, the steps for feature identification in high-humidity areas—we have decided to adopt the following solution: In some optional implementations of certain embodiments, the target area is a high-humidity region, and the input to the land cover recognition model is the feature information corresponding to the remote sensing image. The high-humidity region can be a basin or valley with high humidity. Due to the enclosed terrain, water vapor does not easily dissipate, resulting in persistent cloud cover and fog. Although the air may not be polluted, atmospheric transparency is naturally low. The feature information corresponding to the remote sensing image can be a feature vector primarily composed of the semantic content of the remote sensing image's corresponding features.

[0069] Optionally, the aforementioned implementing entity may utilize the aforementioned feature recognition model to generate feature recognition results corresponding to the aforementioned target area, including the following steps: The first step is to acquire the infrared image set corresponding to the target area. There is a one-to-one correspondence between the infrared images in the aforementioned infrared image set and the remote sensing images in the aforementioned remote sensing image set. The infrared images in the infrared image set are individual infrared images taken specifically for the target area.

[0070] The second step involves performing the following overlay process for each infrared image: Sub-step 1: Determine whether the area of ​​the sub-region corresponding to the aforementioned infrared image is greater than the target area. Here, the sub-region corresponding to the infrared image can be any area covered by the image in the infrared image. The area can be the size of the region corresponding to the sub-region. The target area can be a threshold used to measure whether the area corresponding to the sub-region is too large. The target area can be set based on historical experience.

[0071] Sub-step 2: In response to determining that it is greater than, the infrared image is divided into blocks to obtain the first infrared sub-image block set.

[0072] As an example, the aforementioned execution entity can perform uniform block processing on the infrared image to obtain a first set of infrared sub-image blocks.

[0073] Sub-step 3 involves extracting infrared features from each first infrared sub-image block to obtain first infrared sub-feature information. This first infrared sub-feature information can be in vector form, representing the semantic content of the image features corresponding to the first infrared image block.

[0074] As an example, the aforementioned execution entity can utilize convolutional layers to extract infrared features from the aforementioned first infrared image block to obtain first infrared sub-feature information.

[0075] Sub-step 4: Based on the block processing method corresponding to the first infrared sub-image block set, the infrared image is further divided into blocks to obtain a second infrared sub-image block set. The second infrared sub-image block set includes at least one second infrared sub-image block located at a corresponding edge region between two first infrared sub-image blocks. The block processing method can be how the first infrared sub-image blocks are generated. For example, it can be a uniform block division method. The further block division can be a block division method different from the block processing method, mainly to avoid having too many identical image blocks between the first and second infrared sub-image block sets. The edge region can be the junction area where two corresponding first infrared sub-image blocks meet.

[0076] Sub-step 5 involves extracting infrared features from each second infrared sub-image block to obtain second infrared sub-feature information. Details will not be elaborated further.

[0077] Sub-step 6 involves splicing the obtained first infrared sub-feature information set to obtain first spliced ​​feature information, and splicing the obtained second infrared sub-feature information set to obtain second spliced ​​feature information. The splicing method for the feature information can be a combination of feature information.

[0078] Sub-step 7 involves combining the first and second splicing feature information to obtain combined feature information. This feature information combination can be achieved by averaging the feature information. Specifically, the combined feature information can be obtained by averaging the content of corresponding pixels in the first and second splicing feature information.

[0079] Sub-step 8 involves extracting the remote sensing feature information corresponding to the remote sensing image corresponding to the aforementioned infrared image. This remote sensing feature information characterizes the semantic content of the image features corresponding to the remote sensing image. The specific implementation method will not be elaborated further.

[0080] Sub-step 9 involves fusing the combined feature information with the remote sensing feature information to obtain fused feature information. Here, feature information fusion can be feature information overlay.

[0081] The third step is to generate the corresponding land feature recognition results for the target area based on the obtained fused feature information set and the above-mentioned land feature recognition model.

[0082] As an example, the aforementioned implementing entity can input the fused feature information set into the land cover recognition model to obtain the land cover recognition result.

[0083] The aforementioned "step one to step three" is another inventive aspect of this disclosure, solving the problem of insufficient accuracy in feature recognition models in high-humidity areas. By acquiring the infrared image set corresponding to the target area and overlaying the feature information corresponding to the infrared images onto the corresponding remote sensing images, the feature recognition model can consider richer feature information during the recognition process, thereby achieving accurate generation of feature recognition results in high-humidity areas.

[0084] Step 105: Query the transmission line construction case set corresponding to the landscape type and weather information of the above-mentioned areas from the transmission line case database.

[0085] In some embodiments, the aforementioned implementing entity can query a set of power transmission line construction cases corresponding to the aforementioned regional landscape type and weather information from a power transmission line case database. The power transmission line case database stores construction cases of existing power transmission lines in various regions. The similarity between the regional landscape type and weather information corresponding to the power transmission line construction cases in the case database and the regional landscape type and weather information corresponding to the target region is higher than the target similarity.

[0086] As an example, the aforementioned implementing entity can query the transmission line case database for transmission line construction cases with the same landscape type and weather information as the aforementioned region, thereby obtaining a transmission line construction case set.

[0087] Step 106: Based on the above case study of power transmission line construction and the above ground feature identification results, construct a power transmission line deployment model that supports visual editing for the above target area.

[0088] In some embodiments, the aforementioned execution entity can construct a visually editable power transmission line deployment model for the aforementioned target area based on the aforementioned power transmission line construction case set and the aforementioned feature identification results. The power transmission line deployment model can be a digital twin model of the power transmission line construction in the target area. The power transmission line deployment model achieves a one-to-one reconstruction of the landscape and feature identification results of the target area, obtaining a 3D model of the target area. The aforementioned power transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment. The aforementioned power transmission line deployment model supports the import of multimodal power transmission line construction files. Virtual line construction can be the construction of a virtual power transmission line. In practice, line construction controls can be selected on the digital platform page corresponding to the power transmission line deployment model to draw lines based on the manipulated object, thereby realizing the construction of a virtual power transmission line. Real-time display of virtual line construction costs can display the estimated construction cost after constructing the virtual power transmission line. That is, the amount consumed in constructing the virtual power transmission line. Virtual line construction equipment can be the construction equipment required for constructing the virtual line. Anomaly detection equipment can be equipment used to perform real-time anomaly detection of the line during the construction process of the virtual line. It should be noted that the number and type of construction equipment and anomaly detection equipment required vary greatly depending on the specific virtual transmission line. For example, construction equipment may include, but is not limited to, the following: surveying instruments, earthmoving and concrete equipment, drilling machinery, jacks and winches, traction and fixing devices, and fastening tools. Anomaly detection equipment may include, but is not limited to, the following: intelligent visual monitoring devices, lightning optical observation instruments, abnormal discharge detection equipment, and galloping monitoring terminals. Multimodal transmission line construction documents can be construction concept documents for the transmission line construction in the target area under various modes. For example, multimodal transmission line construction documents can be transmission line construction documents in image mode or transmission line construction documents in video mode.

[0089] In some optional implementations of certain embodiments, the content of the aforementioned multimodal transmission line construction documents is displayed within the aforementioned transmission line deployment model through the following steps: The first step, in response to determining that the aforementioned power transmission line construction file is a file in image mode, is to perform image block marking on the aforementioned power transmission line construction file according to at least one regional construction probability corresponding to the aforementioned target region, thereby obtaining image block marking information. The image block marking information can be power transmission line construction files marked with regional construction probabilities. That is, the regional construction probabilities are marked at the corresponding regional positions in the file to obtain the image block marking information.

[0090] The second step is to construct a three-dimensional model based on the image patch labeling information using a pre-trained neural field model. The construction accuracy of the transmission line in different model regions of the three-dimensional model is different. The higher the construction probability of the corresponding region, the higher the construction accuracy of the transmission line. The neural field model here can be a neural radiance field (NeRF), which can be learned and represented by a neural network to represent the geometry and appearance of the entire three-dimensional scene. It should be noted that the input of the neural field model is a file in the image modality. The file in the image modality can be a set of two-dimensional photos taken from multiple different angles around the target area. These photos need to cover all the key parts of the object, and the camera pose (position and orientation) needs to be recorded or can be calculated. The output is a three-dimensional model. The training method of the neural field model is as follows: (1) Construct an MLP neural network as the initial neural radiance field model. (2) For each pixel in the training image, according to its corresponding camera pose, project a ray into the scene and randomly sample a set of 3D points on the ray. (3) Input the coordinates of these sampled points and the corresponding view direction into the MLP network to obtain the predicted color and density of each point. (4) Based on the physical volume rendering equation, integrate the color and density of the sampling point along the light rays to calculate the predicted color of the pixel. Loss calculation and optimization: Compare the predicted color with the true color of the pixel in the training image to calculate the loss. Then update the weights of the MLP network through the backpropagation algorithm to minimize this loss. This process will be iterated until the model can accurately predict the pixel color under all training views.

[0091] The third step is to input the model parameters corresponding to the above-mentioned 3D model into the above-mentioned power transmission line deployment model in order to display the model content corresponding to the above-mentioned 3D model.

[0092] Fourth, in response to determining that the aforementioned power transmission line construction file is a video mode file, the target region is divided into regions based on at least one regional construction probability corresponding to the target region, resulting in a region set. Regions with construction probabilities within the same interval are grouped into the same region set.

[0093] Fifth, for each region group, perform the following second generation step: Sub-step 1: Based on the probability intervals corresponding to the aforementioned region groups, determine the corresponding keyframe number intervals. The keyframe number interval refers to the range of keyframes required for building the 3D model for each region group. In other words, if the number of key images corresponding to a region group falls within the keyframe number interval, the corresponding 3D model can be built.

[0094] Sub-step 2 involves extracting a set of key video frames for the aforementioned region group from the transmission line construction documents, based on the aforementioned key frame number range. The number of video frames corresponding to the key video frame set must fall within the aforementioned key frame number range. The extraction method will not be elaborated further here. Each key video frame in the key video frame set can be a key image captured for the region group. The number of key video frames corresponding to each region must not be less than the set number.

[0095] Sub-step 3: Based on the aforementioned key video frame set, and using the aforementioned neural field model, construct a set of 3D sub-models for the aforementioned region groups. There is a one-to-one correspondence between the regions in the region group and the 3D sub-models in the 3D sub-model set. The 3D sub-models can represent the 3D architecture of the regions.

[0096] As an example, firstly, for each region in the region group, the aforementioned execution entity can filter out the corresponding key video frame group from the key video frame set. Then, based on the key video frame group, a corresponding 3D sub-model is generated using a neural field model.

[0097] The sixth step is to input the model parameters corresponding to the obtained three-dimensional sub-model set into the above-mentioned power transmission line deployment model to display the model content corresponding to the above-mentioned three-dimensional model.

[0098] Step 107: Based on the above power transmission line deployment model, perform power transmission line construction and anomaly detection during the construction process for the target area.

[0099] In some embodiments, the aforementioned execution entity may perform power transmission line construction and anomaly detection during the construction process in the aforementioned target area based on the aforementioned power transmission line deployment model.

[0100] In addressing the technical problems mentioned in the background section, and considering the specific application scenario—characterizing the terrain features of a target area using a 3D model—the following technical issues arise: the 3D model is often a fixed model corresponding to the target area, failing to achieve sufficient rendering accuracy and real-time updates of the rendered content. This means that if power transmission lines are deployed in areas with low rendering accuracy, a complete 3D model of the target area may need to be regenerated. To address the specific requirements of this application scenario—avoiding the fixation of the 3D model corresponding to the target area—we have decided to adopt the following solution: In some optional implementations of certain embodiments, after step 107, the steps further include: The first step involves determining the rendering enhancement level corresponding to the height rendering control in response to clicking on the target sub-region of the aforementioned power transmission line deployment model and selecting the height rendering control corresponding to the target sub-region. The target sub-region can be the region clicked within the target area. The height rendering control is a control that performs height model rendering on the target sub-region. That is, by clicking the height rendering control, the 3D model corresponding to the target sub-region can be rendered to a higher degree. After clicking the target sub-region, an editing page corresponding to the target sub-region will pop up. The editing page displays various types of rendering controls, including rendering controls with different rendering levels. By selecting the corresponding rendering level control, the model accuracy and content of the 3D sub-model corresponding to the target sub-region can be adjusted. In practice, rendering controls for adjusting the rendering level can include: height rendering control, medium rendering control, and low rendering control. The rendering enhancement level can be the degree of adjustment to the rendering level of the 3D sub-model corresponding to the target sub-region. For example, the rendering enhancement level can include: -1, -2, +1, +2.

[0101] As an example, the rendering level corresponding to the above-mentioned height-rendered control can be improved by querying the mapping relationship.

[0102] The second step is to set the probability interval corresponding to the target sub-region based on the above rendering enhancement level, as the target probability interval.

[0103] As an example, first, determine the probability boost size corresponding to the rendering boost level. Then, determine the original probability interval corresponding to the target sub-region. Finally, add the probability boost size to the original corresponding probability interval to obtain the target probability interval.

[0104] The third step involves re-determining the regional landscape type, weather information, and feature recognition model for the target area based on the aforementioned target probability interval. This process yields updated regional landscape types, updated weather information, and updated feature recognition models. The updated regional landscape type can be the result of updating the regional landscape type corresponding to the target area. The updated weather information can be the result of updating the weather information. The updated feature recognition model can be the re-selected feature recognition model for the target probability interval. The specific implementation details are not elaborated here.

[0105] The fourth step is to generate the corresponding sub-results of the target sub-region based on the updated feature recognition model.

[0106] As an example, firstly, a subset of remote sensing images corresponding to the target sub-region is obtained. Then, the subset of remote sensing images is input into the updated land cover recognition model described above to obtain the land cover recognition sub-results.

[0107] Fifth, based on the above feature identification sub-results, update the above feature identification results to obtain updated feature identification results. The updated feature identification results can be the updated results obtained after updating the feature identification results.

[0108] As an example, the aforementioned executing entity can replace the sub-results corresponding to the target area in the feature identification results with the aforementioned feature identification sub-results to obtain updated feature identification results.

[0109] Step 6: Perform identification result connection processing on the connecting regions in the updated feature identification results to obtain the processed feature identification results. The connecting regions can be the various connecting areas between the target sub-region and its adjacent sub-regions. The identification result connection processing can unify the feature identification results in each connecting region to avoid inconsistencies. Specifically, the highest confidence level of the feature identification results can be used to achieve uniformity in the connecting regions.

[0110] The sixth step involves querying the updated transmission line construction case set from the transmission line case database, matching the updated regional landscape type with the processed feature identification results. Further details are omitted.

[0111] Step 7: Based on the updated transmission line construction case set, update the sub-models corresponding to the target sub-regions in the transmission line deployment model to obtain the updated transmission line deployment model. Details will not be elaborated further.

[0112] The aforementioned "Steps 1-7" constitute another inventive aspect of this disclosure. By setting different levels of rendering controls, this disclosure can achieve varying degrees of rendering enhancement for corresponding sub-regions, thereby updating the regional landscape type, weather information, and feature recognition model of the corresponding sub-regions. Based on this, by combining corresponding updated power transmission line construction cases, precise updates to the sub-model can be achieved, enabling dynamic updates of the three-dimensional details of the corresponding sub-regions to effectively consider whether power transmission lines should be deployed within that sub-region.

[0113] The above-described embodiments of this disclosure have the following beneficial effects: Through the three-dimensional model-based power transmission line deployment method of some embodiments of this disclosure, by constructing a power transmission line deployment model, the visualization and editing of various virtual lines within the target area can be realized, allowing for intuitive acquisition of regional content and power transmission line deployment status, thus assisting in the deployment of power transmission lines and greatly improving deployment efficiency. Specifically, the reason for the low deployment efficiency of related power transmission lines is that remote sensing images and point cloud layout maps are greatly affected by regional landscape types and weather, potentially leading to regional accuracy issues. Furthermore, remote sensing images, point cloud layout maps, and plan maps of different representation forms cannot intuitively represent the regional conditions within the target area, and the performance of subsequently deployed power transmission lines cannot be intuitively obtained. The corresponding deployment effect can only be intuitively obtained after the power transmission lines are deployed, which may significantly affect the deployment efficiency. Based on this, the three-dimensional model-based power transmission line deployment method of some embodiments of this disclosure first acquires a remote sensing atlas corresponding to the target area for constructing the power transmission lines, in order to obtain the regional landscape and weather conditions within the target area from the perspective of remote sensing information. Then, based on the aforementioned remote sensing atlas, the corresponding regional landscape type and weather information for the target area can be accurately generated. Here, generating the regional landscape type and weather information facilitates the selection of a suitable ground feature recognition model based on the remote sensing atlas. Next, the ground feature recognition model corresponding to the aforementioned regional landscape type and weather information is selected to achieve accurate identification of remote sensing content in the remote sensing atlas, ensuring the accuracy of ground feature identification within the target area. Then, a transmission line construction case set corresponding to the aforementioned regional landscape type and weather information is queried from the transmission line case database. Here, querying the transmission line construction case set, from the perspective of construction cases, assists in the subsequent deployment of transmission lines within the target area. Furthermore, based on the aforementioned transmission line construction case set and the aforementioned ground feature identification results, a transmission line deployment model supporting visual editing is constructed for the aforementioned target area. This transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment. The transmission line deployment model also supports the import of multimodal transmission line construction files. Here, by utilizing the results of feature identification and a case study set of power transmission line construction, a precise 3D model of the target area can be constructed. Furthermore, the constructed 3D model supports the creation of virtual power lines and the acquisition of corresponding line information, enabling diverse visualization operations and real-time acquisition of virtual power line deployment within the target area. In addition, importing power transmission line construction files is supported, significantly improving deployment efficiency. Finally, based on the aforementioned power transmission line deployment model, power transmission line construction and anomaly detection during the construction process are performed in the target area.In summary, by constructing a power transmission line deployment model, it is possible to visualize and edit various virtual lines within the target area, intuitively obtain regional content and power transmission line deployment status, thereby assisting in the deployment of power transmission lines and greatly improving the deployment efficiency of power transmission lines.

[0114] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a power transmission line deployment device based on a three-dimensional model. These device embodiments are similar to... Figure 1 Corresponding to the method embodiments shown, this three-dimensional model-based power transmission line deployment device can be specifically applied to various electronic devices.

[0115] like Figure 2 As shown, a power transmission line deployment device 200 based on a 3D model includes: an acquisition unit 201, a first generation unit 202, a selection unit 203, a second generation unit 204, a query unit 205, a construction unit 206, and an execution unit 207. The acquisition unit 201 is configured to acquire a remote sensing atlas corresponding to the target area for power transmission line construction; the first generation unit 202 is configured to generate regional landscape type and weather information corresponding to the target area based on the remote sensing atlas; the selection unit 203 is configured to select a land feature recognition model corresponding to the regional landscape type and weather information; the second generation unit 204 is configured to generate land feature recognition results corresponding to the target area using the land feature recognition model; and the query unit 205 is configured to query the regional landscape type and weather information from a power transmission line case database. The system includes a set of power transmission line construction case studies; a construction unit 206, configured to construct a power transmission line deployment model for the target area that supports visual editing based on the power transmission line construction case studies and the ground feature identification results. The power transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment. The power transmission line deployment model also supports the import of multimodal power transmission line construction files. An execution unit 207, configured to perform power transmission line construction and anomaly detection during the construction process for the target area based on the power transmission line deployment model.

[0116] It is understandable that the units described in the three-dimensional model-based power transmission line deployment device 200 are similar to those in the reference model. Figure 1 The steps in the described method correspond accordingly. Therefore, the operations, features, and beneficial effects described above for the method also apply to the three-dimensional model-based power transmission line deployment device 200 and the units contained therein, and will not be repeated here.

[0117] The following is for reference. Figure 3It shows a schematic diagram of the structure of an electronic device (e.g., an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0118] like Figure 3 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0119] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0120] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0121] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0122] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0123] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire a remote sensing atlas corresponding to the target area for power transmission line construction; generate regional landscape type and weather information corresponding to the target area based on the remote sensing atlas; select a feature recognition model corresponding to the regional landscape type and weather information; generate feature recognition results corresponding to the target area using the feature recognition model; query a power transmission line construction case set corresponding to the regional landscape type and weather information from a power transmission line case database; construct a power transmission line deployment model for the target area that supports visual editing based on the power transmission line construction case set and the feature recognition results, wherein the power transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment, and anomaly detection equipment, and supports the import of multimodal power transmission line construction files; and perform power transmission line construction and anomaly detection during the construction process for the target area based on the power transmission line deployment model.

[0124] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0125] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0126] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an acquisition unit, a first generation unit, a selection unit, a second generation unit, a query unit, a construction unit, and an execution unit. The names of these units do not necessarily limit the specific unit; for example, the acquisition unit may also be described as "a unit that acquires remote sensing atlases corresponding to target areas for power transmission line construction."

[0127] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0128] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for deploying power transmission lines based on a three-dimensional model, comprising: Obtain remote sensing atlases corresponding to the target areas for the construction of power transmission lines; Based on the remote sensing atlas, generate the regional landscape type and weather information corresponding to the target area; Select the land feature recognition model corresponding to the regional landscape type and the weather information; Using the aforementioned feature recognition model, feature recognition results corresponding to the target area are generated; Search the power transmission line case database for the set of power transmission line construction cases corresponding to the landscape type and weather information of the region; Based on the transmission line construction case set and the ground feature identification results, a transmission line deployment model supporting visual editing is constructed for the target area. The transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment and anomaly detection equipment. The transmission line deployment model supports the import of multimodal transmission line construction files. Based on the power transmission line deployment model, perform power transmission line construction and anomaly detection during the construction process for the target area.

2. The method according to claim 1, wherein, The remote sensing atlas includes: subsets of remote sensing images at different resolutions; and The step of generating regional landscape type and weather information corresponding to the target area based on the remote sensing atlas includes: Using a subset of low-resolution remote sensing images from the remote sensing atlas, preliminary regional landscape types and confidence levels are generated for each sub-region corresponding to the target region. Based on the preliminary regional landscape types and the confidence levels, the sub-regions are merged to obtain at least one merged sub-region; In response to determining that a pre-set power transmission line configuration scheme exists in the neighboring area corresponding to the target area, the regional construction probability corresponding to each fusion sub-region in the at least one fusion sub-region is determined according to the power transmission line configuration scheme; Based on the obtained construction probability of at least one region, obtain a first remote sensing image subset at high resolution or a second remote sensing image subset at medium resolution corresponding to each fusion sub-region in the at least one fusion sub-region; Based on at least one subset of first remote sensing images, at least one subset of second remote sensing images, and the respective preliminary regional landscape types, the regional landscape type and the weather information are generated.

3. The method according to claim 2, wherein, The step of generating the regional landscape type and the weather information based on at least one subset of first remote sensing images, at least one subset of second remote sensing images, and the various preliminary regional landscape types includes: Based on the preliminary regional landscape types, determine the first candidate regional landscape type corresponding to each of the at least one fusion sub-regions; From the at least one fused sub-region, select the target fused sub-region that contains a corresponding high-resolution remote sensing image subset or a medium-resolution remote sensing image subset, to obtain at least one target fused sub-region; For each target fusion sub-region, perform the following first generation step: Based on the first or second remote sensing image subset corresponding to the target fused sub-region, generate the second candidate region landscape type corresponding to the target fused sub-region; Replace the first candidate region landscape type corresponding to the target fusion sub-region with the second candidate region landscape type; The region landscape type is generated based on at least one first candidate region landscape type obtained; Based on the landscape type of the region, the initial weather forecast for the target region is corrected to obtain weather information.

4. The method according to claim 2, wherein, The step of obtaining a first subset of high-resolution remote sensing images or a second subset of medium-resolution remote sensing images corresponding to each of the at least one fused sub-regions based on the obtained construction probability of at least one region includes: In response to determining that the construction probability of the region corresponding to the fused sub-region is in the first probability interval, the first remote sensing image subset corresponding to the fused sub-region is obtained; In response to determining that the construction probability of the region corresponding to the fused sub-region is in the second probability interval, a second remote sensing image subset corresponding to the fused sub-region is obtained, wherein the value corresponding to the first probability region is higher than the value corresponding to the second probability interval. In response to determining that the construction probability of the region corresponding to the fused sub-region is in the third probability interval, information is generated to indicate that the fused sub-region does not acquire the first remote sensing image subset or the second remote sensing image subset.

5. The method according to claim 1, wherein, The contents of the multimodal transmission line construction documents are displayed within the transmission line deployment model through the following steps: In response to determining that the power transmission line construction file is a file in image mode, the power transmission line construction file is marked with image blocks according to the construction probability of at least one region corresponding to the target region, and image block marking information is obtained; Using a pre-trained neural field model, a three-dimensional model is constructed for the image patch labeling information. The construction accuracy of the power transmission line varies in different model regions of the three-dimensional model. The higher the construction probability of the model region, the higher the construction accuracy of the corresponding power transmission line. The model parameters corresponding to the 3D model are input into the power transmission line deployment model to display the model content corresponding to the 3D model; In response to determining that the transmission line construction file is a file in video mode, the target region is divided into regions according to the construction probability of at least one region corresponding to the target region, and a region set is obtained; For each region group, perform the following second generation step: Based on the probability intervals corresponding to the region groups, determine the corresponding keyframe number intervals; Based on the key frame number range, a set of key video frames for the region group is extracted from the power transmission line construction document, wherein the number of video frames corresponding to the key video frame set is within the key frame number range; Based on the set of key video frames, a set of three-dimensional sub-models for the set of regions is constructed using the neural field model. The obtained three-dimensional sub-model set and its corresponding model parameters are input into the power transmission line deployment model to display the model content corresponding to the three-dimensional model.

6. The method according to claim 1, wherein, The step of selecting the land cover recognition model corresponding to the regional landscape type and the weather information includes: Obtain the construction probability of at least one area corresponding to the target area; Based on the construction probability of at least one region, the target region is divided into regions to obtain a region set; For each region group, perform the following third generation step: Based on the probability intervals corresponding to the region groups, determine the corresponding model magnitude and image resolution; Based on the model size and the image resolution, an initial set of ground feature recognition models for the region group is generated; At least one initial land feature recognition model corresponding to the regional landscape type and the weather information is selected from the initial land feature recognition model set; Based on the frequency of use of the model corresponding to the initial feature identification model, a target feature identification model is selected from the at least one initial feature identification model; The obtained target feature recognition model set is determined as the feature recognition model.

7. A power transmission line deployment device based on a three-dimensional model, comprising: The acquisition unit is configured to acquire remote sensing atlases and road network maps corresponding to the target areas for the construction of power transmission lines; The first generation unit is configured to generate the regional landscape type and weather information corresponding to the target region based on the remote sensing atlas; The selection unit is configured to select the land feature recognition model corresponding to the regional landscape type and the weather information; The second generation unit is configured to use the land feature recognition model to generate land feature recognition results corresponding to the target area; The query unit is configured to query the transmission line construction case set corresponding to the regional landscape type and the weather information from the transmission line case database; The construction unit is configured to construct a visually editable power transmission line deployment model for the target area based on the power transmission line construction case set and the ground feature identification results. The power transmission line deployment model supports virtual line construction, real-time display of virtual line construction costs, virtual line construction equipment and anomaly detection equipment, and supports the import of multimodal power transmission line construction files. The execution unit is configured to perform anomaly detection during the construction of power transmission lines in the target area, based on the power transmission line deployment model.

8. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

9. A computer-readable medium having a computer program stored thereon, wherein, When the program is executed by the processor, it implements the method as described in any one of claims 1-6.

Images