Method and system for accurately positioning electric power operating personnel based on three-dimensional space technology

Abstract

The embodiment of the invention provides a method and a system for accurately positioning an electric power operator based on a three-dimensional space technology, and belongs to the technical field of operator positioning. The method comprises the following steps: acquiring a field image of a power grid project; constructing a three-dimensional power grid project real scene model according to the field image of the power grid project; acquiring three-dimensional point cloud data of the power grid project; constructing a three-dimensional remote sensing space according to the three-dimensional point cloud data of the power grid project; performing joint mapping on the three-dimensional power grid project live-action model and the three-dimensional remote sensing space to obtain a three-dimensional point cloud map; a camera in a power grid project is mapped in a three-dimensional point cloud map, and a coordinate conversion relation is constructed, so that a real-time image of the camera can be mapped to the three-dimensional point cloud map in real time, the spatial position of an electric power worker can be determined conveniently, safe operation monitoring of the electric power worker is realized, and the working efficiency of the electric power worker is improved. The working efficiency and effect of the electric power operation personnel are guaranteed, and the operation convenience of the electric power operation personnel is improved.

Description

Technical Field

[0001] This invention relates to the field of worker positioning technology, and more specifically to a method and system for precise positioning of power workers based on three-dimensional spatial technology. Background Technology

[0002] With the continuous development of science and technology, the production technology level of power supply companies is also changing rapidly. The continuous improvement of production technology level has also put forward higher requirements for management level. Operational risk management at the power operation site is an extremely important part of management. In order to achieve real-time early warning and control of typical violations by power operation personnel, such as entering live intervals by mistake, climbing heights without supervision, climbing live circuits on poles by mistake, and personnel staying in the wrong place, it is necessary to accurately locate power operation personnel in real time.

[0003] In existing technologies, the method for locating personnel at power operation sites generally involves having workers wear wearable devices such as chips. However, this method further increases the burden on on-site workers. Wearable devices such as chips have issues with radiation and battery life, and can also affect the work efficiency and effectiveness of workers, increasing the complexity of their work.

[0004] In the process of realizing this invention, the inventors of this application discovered that the above-mentioned solutions in the prior art have the drawbacks of affecting the work efficiency of operators and increasing the complexity of the work. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for precise positioning of power workers based on three-dimensional spatial technology. This method and system for precise positioning of power workers based on three-dimensional spatial technology has the function of not affecting the operation of the workers.

[0006] To achieve the above objectives, embodiments of the present invention provide a method for precise positioning of power workers based on three-dimensional spatial technology, comprising: Acquire on-site images of power grid projects; A three-dimensional real-scene model of the power grid project is constructed based on the on-site images of the power grid project; Obtain the three-dimensional point cloud data of the power grid project; A three-dimensional remote sensing space is constructed based on the three-dimensional point cloud data of the power grid project; A three-dimensional point cloud map is obtained by jointly mapping the three-dimensional power grid engineering real scene model and the three-dimensional remote sensing space. The cameras in the power grid project are mapped in the 3D point cloud map; Acquire real-time images from the camera; The spatial location of the power workers is obtained by matching the real-time images with the 3D point cloud map.

[0007] Optionally, constructing a three-dimensional real-scene model of the power grid project based on the on-site images of the power grid project includes: The feature map of the scene image is obtained by dilated convolution. The feature map is upsampled using bilinear interpolation. The upsampled feature map is then processed using pyramid pooling. A three-dimensional power grid engineering real-world model is constructed based on the feature map after pooling.

[0008] Optionally, constructing a 3D power grid engineering reality model based on the pooled feature map includes data augmentation using techniques such as inversion, rotation, scaling, clipping, translation, and artificially added noise.

[0009] Optionally, the joint mapping based on the three-dimensional power grid engineering reality model and the three-dimensional remote sensing space includes: Label the three-dimensional power grid engineering reality model; Obtain the coordinates of the points in the three-dimensional remote sensing space; The labels of the three-dimensional power grid engineering real-scene model are matched and mapped with the point coordinates in the three-dimensional remote sensing space.

[0010] Optionally, matching and mapping the labels of the three-dimensional power grid engineering reality model with the point coordinates in the three-dimensional remote sensing space includes: Construct the Hessian matrix; Based on the Hessian matrix, feature points are extracted from the three-dimensional power grid engineering real scene model and the three-dimensional remote sensing space, respectively. The Harr wavelet features within the circular neighborhood of the feature point are statistically analyzed to determine the main direction of the feature point. Generate the three-dimensional power grid engineering real-scene model and the feature descriptor of the three-dimensional remote sensing space; The feature descriptors of the three-dimensional power grid engineering real-world model and the three-dimensional remote sensing space are matched using Euclidean distance.

[0011] Optionally, mapping the cameras in the power grid project onto the 3D point cloud map includes: Obtain the location coordinates of the cameras in the power grid project; The position coordinates of the 3D point cloud map are mapped based on the position coordinates of the camera. The intrinsic and extrinsic parameters of the camera are obtained through calibration. Map the image coordinates of the images captured by the camera to the camera coordinates; The camera coordinates are mapped to 3D point cloud map coordinates to establish a mapping relationship between the camera-captured images and the 3D point cloud map.

[0012] Optionally, matching the real-time image with the 3D point cloud map to obtain the spatial location of the power workers includes: Extract the personnel features from the real-time image; Based on the mapping relationship, the personnel features in the real-time image are mapped to the three-dimensional point cloud map to obtain the spatial location of the power workers.

[0013] Optionally, the method further includes: Determine whether the spatial location of the electrical worker complies with the location safety regulations; If the location of the electrical worker is determined to be inconsistent with the location safety specifications, an alarm will be issued. Continuous monitoring is conducted once it is determined that the spatial location of the electrical worker complies with the location safety regulations.

[0014] Optionally, the method further includes: Extract the clothing features of the power workers in the real-time images; Determine whether the clothing features of the power workers in the real-time image comply with the safety regulations for clothing. If the wearer's clothing in the real-time image does not conform to the safety standards, an alarm will be issued. If the wear characteristics of the power workers are determined to comply with the safety standards for wearing protective clothing, continuous monitoring shall be conducted.

[0015] On the other hand, the present invention also provides a system for precise positioning of power workers based on three-dimensional spatial technology, comprising: A camera module is installed in the power grid project to acquire images of the power grid project. The point cloud data acquisition module is used to acquire the three-dimensional point cloud data of the power grid project; The controller, connected to the camera module and the point cloud data acquisition module, is used to execute any of the methods described above.

[0016] Through the above technical solution, the method and system for precise positioning of power workers based on three-dimensional spatial technology provided by this invention acquires on-site images of power grid projects and constructs a three-dimensional real-scene model of the power grid project based on these images. Simultaneously, it acquires three-dimensional point cloud data of the power grid project and constructs a three-dimensional remote sensing space. Then, it performs a joint mapping between the three-dimensional real-scene model of the power grid project and the three-dimensional remote sensing space to obtain a three-dimensional point cloud map. By mapping cameras in the power grid project onto the three-dimensional point cloud map and establishing coordinate transformation relationships, real-time images from the cameras can be mapped onto the three-dimensional point cloud map in real time, facilitating the determination of the spatial location of power workers. This enables safe monitoring of power workers' operations, ensuring their work efficiency and effectiveness, and improving the convenience of their work.

[0017] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention; Figure 2 This is a flowchart of a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention, which constructs a three-dimensional power grid engineering scene model; Figure 3 This is a flowchart of a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention, which involves obtaining a three-dimensional point cloud map. Figure 4 This is a flowchart of the matching mapping process in a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention; Figure 5 This is a flowchart illustrating the construction of coordinate transformation relationships in a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention; Figure 6 This is a flowchart illustrating the method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention, which obtains the spatial location of power workers. Figure 7 This is a flowchart of a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention, which includes safety supervision of power workers. Detailed Implementation

[0019] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0020] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0021] Figure 1 This is a flowchart illustrating a method for precise positioning of power workers based on three-dimensional spatial technology according to an embodiment of the present invention. Figure 1 In this context, the method may include: In step S1, on-site images of the power grid project are acquired. These on-site images can include multi-dimensional and multi-angle images, specifically, they can be acquired through methods such as drone photography.

[0022] In step S2, a three-dimensional real-scene model of the power grid project is constructed based on the on-site images of the power grid project. After acquiring multi-dimensional, multi-angle on-site images, feature points can be extracted and matched from multiple images to construct the corresponding three-dimensional real-scene model of the power grid project. Specifically, the three-dimensional real-scene model of the power grid project can include equipment and facilities of the power grid project.

[0023] In step S3, three-dimensional point cloud data of the power grid project is acquired. This acquisition can be achieved using methods such as depth cameras and LiDAR.

[0024] In step S4, a three-dimensional remote sensing space is constructed based on the three-dimensional point cloud data of the power grid project. This three-dimensional remote sensing space, constructed from the three-dimensional point cloud data of the power grid project, contains the coordinate information of each point cloud data point.

[0025] In step S5, a joint mapping is performed based on the 3D power grid engineering real-scene model and the 3D remote sensing space to obtain a 3D point cloud map. Specifically, mapping the 3D power grid engineering real-scene model to the 3D remote sensing space yields a 3D point cloud map of the power grid engineering containing point cloud coordinate information.

[0026] In step S6, the cameras in the power grid project are mapped onto a 3D point cloud map. Specifically, the location of each camera at the power grid project site needs to be mapped onto the 3D point cloud map so that the real-time dynamic images captured by the cameras can be displayed on the map. The monitoring locations of each camera at the power grid project site are generally key operational areas, key monitoring areas, etc.

[0027] In step S7, real-time images from the camera are acquired. The camera's shooting pose information is synchronized with the camera pose information mapped in the 3D point cloud map.

[0028] In step S8, the spatial location of the power workers is obtained by matching the real-time image with the 3D point cloud map. Specifically, the features of the personnel in the real-time image are extracted and mapped onto the 3D point cloud map to obtain the spatial location information of the power workers. Based on this spatial location information, safety supervision of the power workers can be carried out.

[0029] In steps S1 to S8, on-site images of the power grid project are first acquired, and a 3D real-world model of the power grid project is constructed based on these images. Simultaneously, 3D point cloud data of the power grid project is acquired, and a 3D remote sensing space is constructed based on this point cloud data. The 3D real-world model of the power grid project and the 3D remote sensing space are then jointly mapped to obtain a 3D point cloud map. Furthermore, multiple cameras used for safety monitoring of power workers in the power grid project are mapped onto the 3D point cloud map. Therefore, for each camera's real-time image, the real-time image can be matched and mapped with the 3D point cloud map to identify power workers and obtain their spatial location, thus facilitating convenient, effective, and reliable monitoring of the safety of power workers.

[0030] Traditional methods for locating personnel at power work sites typically involve wearing wearable devices such as chips on workers. However, this approach increases the burden on on-site personnel, as these devices pose issues such as radiation and battery life problems, and can negatively impact work efficiency and effectiveness, increasing the complexity of the work. In this embodiment of the invention, a 3D point cloud map is obtained by jointly mapping a 3D power grid engineering scene model with 3D remote sensing space. A coordinate transformation relationship is then established between the camera and the 3D point cloud map, enabling effective identification of the spatial location of power workers at the power grid engineering site. This achieves safe monitoring of power workers' operations, ensures their work efficiency and effectiveness, and improves the convenience of their work.

[0031] In this embodiment of the invention, after acquiring on-site images of the power grid project, feature extraction is required from the on-site images to facilitate the construction of a three-dimensional power grid project real-scene model. Specifically, the steps for constructing the three-dimensional power grid project real-scene model can be as follows: Figure 2 As shown. Specifically, in Figure 2 The steps for constructing this 3D power grid engineering reality model may include: In step S20, dilated convolution is used to obtain feature maps of the scene image. Dilated convolution expands the convolution kernel by inserting space between its parts; the increased parameter (dilation rate) indicates how large the kernel should be. The number of parameters in the dilation operation is essentially the same, allowing for the observation of a large receptive field without increasing computational costs. Therefore, this invention only requires stacking multiple dilated convolutions to increase the receptive field of the output unit at low cost, without increasing the kernel size. By adjusting the dilation rate, the receptive field is expanded without changing the resolution, thereby more effectively segmenting and detecting large targets, improving pixel localization accuracy, and obtaining image feature maps with different scale information under different receptive fields.

[0032] In step S21, bilinear interpolation is used to upsample the feature map. Since the feature map obtained in the previous step has undergone convolutional compression, it loses the location information of key features. However, the final step is pixel-by-pixel semantic region segmentation, which is not equivalent to pixel-level segmentation. Therefore, to achieve the final pixel-by-pixel classification, the feature map must be upsampled to restore the lost feature location information. At this point, bilinear interpolation is used to fill in the missing pixels.

[0033] In step S22, the upsampled feature map is processed by pyramid pooling. Spatial pyramid pooling is introduced to address the issue of variable input image data size. Furthermore, because spatial pyramid pooling extracts features from feature spaces of different granularities and then aggregates them, it enriches the feature matching capabilities, improves the robustness of the recognition matching algorithm, and enhances accuracy in object recognition.

[0034] In step S23, a three-dimensional power grid engineering scene model is constructed based on the feature map after pooling. This can include, but is not limited to, methods such as obtaining the depth information of each point in the feature map, constructing a depth map, and fusing multiple depth maps to construct the three-dimensional power grid engineering scene model.

[0035] In steps S20 to S23, dilated convolution is first used to obtain feature maps of the site image, then bilinear interpolation is used to upsample the feature maps, followed by pyramid pooling. Matching and fusion are then performed based on the processed feature maps to obtain a 3D power grid engineering real-world model. Specifically, the 3D power grid engineering real-world model can include different entities, such as personnel, equipment, and facilities.

[0036] In this embodiment of the invention, to further improve the robustness of the 3D power grid engineering scene model recognition and facilitate joint mapping with 3D remote sensing space, the 3D power grid engineering scene model can be further data augmented. Specifically, data augmentation can be performed using methods such as inversion, rotation, scaling, cropping, translation, and artificially adding noise. The 3D power grid engineering scene model can be projected and sliced ​​at different angles and sizes to enrich the data representation.

[0037] In this embodiment of the invention, after obtaining the three-dimensional power grid engineering scene model and the three-dimensional remote sensing space, the two can be jointly mapped. Specifically, the steps for joint mapping can be as follows: Figure 3 As shown. Specifically, in Figure 3 In this context, the joint mapping step may include: In step S50, the 3D power grid engineering reality model is labeled. Specifically, different 3D power grid engineering reality models are labeled and stored according to the category of the physical object.

[0038] In step S51, the coordinates of the points in the three-dimensional remote sensing space are obtained.

[0039] In step S52, the labels of the 3D power grid engineering scene model are matched and mapped with the point coordinates in the 3D remote sensing space. Specifically, for the 3D power grid engineering scene model at the power grid engineering site, its labels can be used to map it to the 3D remote sensing space containing point cloud coordinate information, thereby obtaining a 3D point cloud map with point cloud coordinate information.

[0040] In this embodiment of the invention, the matching and mapping between the three-dimensional power grid engineering real-world model and the three-dimensional remote sensing space can be achieved using SURF feature descriptors. Specifically, the matching and mapping steps can be as follows: Figure 4 As shown. Specifically, in Figure 4 In this context, the matching mapping step may include: In step S520, the Hessian matrix is ​​constructed.

[0041] In step S521, feature points are extracted from the 3D power grid engineering real-world model and the 3D remote sensing space based on the Hessian matrix. The feature point localization process may include comparing each pixel processed by the Hessian matrix with 26 points in the 2D image space and scale space neighborhood to initially locate key points. Then, weakly located key points and incorrectly located key points are filtered out to select the final stable feature points.

[0042] In step S522, the Harr wavelet features within the circular region of the feature points are statistically analyzed to determine the main direction of the feature points.

[0043] In step S523, a 3D power grid engineering real-world model and a 3D remote sensing spatial feature descriptor are generated. Specifically, after determining the main direction of a feature point by statistically analyzing the HAR wavelet features within its circular neighborhood, a feature point descriptor is generated. This involves taking a 4x4 rectangular region around the feature point, but the direction of this rectangular region is along the main direction of the feature point. For each sub-region, HAR wavelet features of 25 pixels in the horizontal and vertical directions are statistically analyzed, resulting in four directions: horizontal value, vertical value, absolute value of the horizontal direction, and the sum of the absolute values ​​of the vertical direction.

[0044] In step S524, Euclidean distance is used to match the feature descriptors of the 3D power grid engineering scene model and the 3D remote sensing space. Euclidean distance can accurately and effectively match the feature descriptors of the two, thereby achieving the matching of the 3D power grid engineering scene model and the 3D remote sensing space.

[0045] In steps S520 to S524, feature point sets are first extracted from the 3D power grid engineering scene model and the 3D remote sensing space using the Hessian matrix, and then corresponding feature descriptor sets are obtained based on these feature point sets. Finally, the feature descriptors of the 3D power grid engineering scene model and the 3D remote sensing space are matched using Euclidean distance to achieve matching mapping.

[0046] In this embodiment of the invention, after constructing the three-dimensional point cloud map, it is also necessary to map each camera in the power grid project onto the three-dimensional point cloud map and establish coordinate transformation relationships. The specific mapping and coordinate transformation steps can be as follows: Figure 5 As shown. Specifically, in Figure 5 In addition, the method may also include: In step S60, the location coordinates of the cameras in the power grid project are obtained. The location coordinates of the cameras in the power grid project can be, but are not limited to, relative positions, such as a specific location of a particular device. In step S61, the position coordinates of the camera are mapped to the position coordinates of the 3D point cloud map. Specifically, if a point is proportionally located on the 3D point cloud map at a specific location of a device, the position coordinates of the camera in the 3D point cloud map can be obtained.

[0047] In step S62, the intrinsic and extrinsic parameters of the camera are calibrated. The intrinsic parameters of the camera can be calibrated using the OpenCV function `calibrateCamera`. Specifically, the intrinsic parameter matrix can be obtained as shown in formula (1). (1) in, This is the intrinsic parameter matrix. , Focal length , These are the coordinates of the optical center.

[0048] The extrinsic parameter matrix includes a rotation matrix and a translation vector. The translation vector is the coordinate position of the camera in the point cloud map. The rotation matrix can be obtained through sensor fusion or visual SLAM.

[0049] In step S63, the image coordinates of the captured image are mapped to camera coordinates. This may include converting the image coordinates of the captured image to camera coordinates according to formula (2). (2) in, , , , , , The coordinates of the camera. , These are the pixel coordinates of a pixel in the image. This represents the depth value of a pixel in the image.

[0050] In step S64, the camera coordinates are mapped to the 3D point cloud map coordinates to establish a mapping relationship between the camera-captured images and the 3D point cloud map. Specifically, the camera coordinates are converted to the 3D point cloud map coordinates according to formula (3). (3) in, For the coordinates of the 3D point cloud map, For rotation matrix, It is a translation vector. These are the camera coordinates.

[0051] In steps S60 to S64, the intrinsic and extrinsic parameters of the camera are first calibrated and acquired. Then, the image coordinates of the camera are transformed to camera coordinates based on the intrinsic and extrinsic parameters. Next, the camera coordinates are transformed to 3D point cloud map coordinates, thereby realizing the coordinate transformation and mapping between the camera image and the 3D point cloud map, which facilitates the real-time identification and monitoring of the location of power workers in power grid engineering sites.

[0052] In this embodiment of the invention, after acquiring real-time images of the power grid engineering site, it is necessary to determine the spatial location of the power workers based on these real-time images. The specific determination steps can be as follows: Figure 6 As shown, specifically, in Figure 6 In this process, the steps for obtaining the spatial location of the power worker may include: In step S80, the human features of the real-time image are extracted. The extraction and recognition of human features in the real-time image can be achieved using convolutional neural networks, a technique known to those skilled in the art.

[0053] In step S81, the personnel features in the real-time image are mapped to a 3D point cloud map according to the mapping relationship to obtain the spatial location of the power workers. Specifically, after obtaining the features of the power workers in the real-time image, they are mapped to pixel coordinates, and these pixel coordinates are transformed into coordinates on the 3D point cloud map, thereby determining the spatial location of the power workers. Specifically, the spatial location of the power workers can also be accurately determined by identifying and obtaining the corresponding pixels of the head and feet of the power workers in the real-time image.

[0054] In steps S80 to S81, features of the power workers in the real-time image are first extracted to determine their pixel coordinates. These pixel coordinates are then converted into a 3D point cloud map, enabling quick and easy determination of the power workers' spatial location.

[0055] In this embodiment of the invention, after identifying power workers in real-time images, their spatial location and attire can be diagnosed to achieve safety supervision of power workers. Specific diagnostic steps can be as follows: Figure 7 As shown. Specifically, in Figure 7 In addition, the method may also include: In step S9, it is determined whether the spatial position of the power worker complies with the position safety regulations.

[0056] In step S10, an alarm is issued if the spatial location of the power worker does not comply with the location safety regulations. Specifically, if the power worker's location does not comply with the location safety regulations, such as mistakenly entering a certain area or climbing a pole incorrectly, a safety alarm will be issued to alert the power worker and other supervisory personnel to prevent safety accidents.

[0057] In step S11, continuous monitoring is performed if the spatial location of the power worker is determined to comply with the location safety regulations.

[0058] In step S12, the clothing features of the power workers in the real-time image are extracted. These clothing features can be extracted from the real-time image, specifically using methods known to those skilled in the art, such as convolutional neural networks.

[0059] In step S13, it is determined whether the clothing features of the power workers in the real-time image comply with the safety regulations for clothing.

[0060] In step S14, an alarm is issued if it is determined that the clothing characteristics of the power workers in the real-time image do not comply with the safety regulations. Specifically, if the clothing characteristics of the power workers do not comply with the safety regulations, such as not wearing a safety helmet or safety harness, a safety alarm will be issued to alert the power workers and other supervisory personnel to prevent safety accidents from occurring.

[0061] In step S15, continuous monitoring is conducted if it is determined that the electrical worker's attire characteristics comply with the attire safety regulations.

[0062] In steps S9 to S15, after obtaining the spatial location of the power workers, their location is compared with the spatial location specified in the location safety regulations to determine whether the power workers have violated regulations by entering certain areas, thus achieving safety supervision of the power workers' locations. Simultaneously, the clothing features of the power workers are extracted from the real-time images and identified and diagnosed to determine whether their attire meets safety requirements. If not, an alarm is issued to achieve safety supervision of the power workers' clothing. By employing a method that diagnoses both the spatial location and clothing of power workers, the safety and efficiency of their work are effectively ensured.

[0063] On the other hand, the present invention also provides a system for precise positioning of power workers based on three-dimensional spatial technology. Specifically, the system may include a camera module, a point cloud data acquisition module, and a controller. Specifically, the camera module may include, but is not limited to, a camera, and the point cloud data acquisition module may include, but is not limited to, a depth camera, a lidar, etc.

[0064] The camera module is installed in the power grid project to acquire images of the project, while the point cloud data acquisition module acquires 3D point cloud data of the power grid project. The controller is connected to both the camera module and the point cloud data acquisition module to execute any of the methods described above.

[0065] Through the above technical solution, the method and system for precise positioning of power workers based on three-dimensional spatial technology provided by this invention acquires on-site images of power grid projects and constructs a three-dimensional real-scene model of the power grid project based on these images. Simultaneously, it acquires three-dimensional point cloud data of the power grid project and constructs a three-dimensional remote sensing space. Then, it performs a joint mapping between the three-dimensional real-scene model of the power grid project and the three-dimensional remote sensing space to obtain a three-dimensional point cloud map. By mapping cameras in the power grid project onto the three-dimensional point cloud map and establishing coordinate transformation relationships, real-time images from the cameras can be mapped onto the three-dimensional point cloud map in real time, facilitating the determination of the spatial location of power workers. This enables safe monitoring of power workers' operations, ensuring their work efficiency and effectiveness, and improving the convenience of their work.

[0066] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0067] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0068] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0069] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0070] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0071] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0072] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0073] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0074] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for precise positioning of power workers based on three-dimensional spatial technology, characterized in that, include: Acquire on-site images of power grid projects; A three-dimensional real-scene model of the power grid project is constructed based on the on-site images of the power grid project; Obtain the three-dimensional point cloud data of the power grid project; A three-dimensional remote sensing space is constructed based on the three-dimensional point cloud data of the power grid project; A three-dimensional point cloud map is obtained by jointly mapping the three-dimensional power grid engineering real scene model and the three-dimensional remote sensing space. The cameras in the power grid project are mapped in the 3D point cloud map; Acquire real-time images from the camera; The spatial location of the power workers is obtained by matching the real-time images with the 3D point cloud map.

2. The method according to claim 1, characterized in that, Constructing a 3D real-world model of the power grid project based on on-site images includes: The feature map of the scene image is obtained by dilated convolution. The feature map is upsampled using bilinear interpolation. The upsampled feature map is then processed using pyramid pooling. A three-dimensional power grid engineering real-world model is constructed based on the feature map after pooling.

3. The method according to claim 2, characterized in that, The construction of a 3D power grid engineering real-world model based on the feature map after pooling includes data augmentation using techniques such as inversion, rotation, scaling, clipping, translation, and artificially added noise.

4. The method according to claim 1, characterized in that, The joint mapping based on the three-dimensional power grid engineering real-scene model and the three-dimensional remote sensing space includes: Label the three-dimensional power grid engineering reality model; Obtain the coordinates of the points in the three-dimensional remote sensing space; The labels of the three-dimensional power grid engineering real-scene model are matched and mapped with the point coordinates in the three-dimensional remote sensing space.

5. The method according to claim 4, characterized in that, Matching and mapping the labels of the three-dimensional power grid engineering reality model with the point coordinates in the three-dimensional remote sensing space includes: Construct the Hessian matrix; Based on the Hessian matrix, feature points are extracted from the three-dimensional power grid engineering real scene model and the three-dimensional remote sensing space, respectively. The Harr wavelet features within the circular neighborhood of the feature point are statistically analyzed to determine the main direction of the feature point. Generate the three-dimensional power grid engineering real-scene model and the feature descriptor of the three-dimensional remote sensing space; The feature descriptors of the three-dimensional power grid engineering real-world model and the three-dimensional remote sensing space are matched using Euclidean distance.

6. The method according to claim 1, characterized in that, Mapping the cameras in the power grid project onto the 3D point cloud map includes: Obtain the location coordinates of the cameras in the power grid project; The position coordinates of the 3D point cloud map are mapped based on the position coordinates of the camera. The intrinsic and extrinsic parameters of the camera are obtained through calibration. Map the image coordinates of the images captured by the camera to the camera coordinates; The camera coordinates are mapped to 3D point cloud map coordinates to establish a mapping relationship between the camera-captured images and the 3D point cloud map.

7. The method according to claim 6, characterized in that, The spatial location of power workers is obtained by matching the real-time image with the 3D point cloud map, including: Extract the personnel features from the real-time image; Based on the mapping relationship, the personnel features in the real-time image are mapped to the three-dimensional point cloud map to obtain the spatial location of the power workers.

8. The method according to claim 1, characterized in that, The method further includes: Determine whether the spatial location of the electrical worker complies with the location safety regulations; If the location of the electrical worker is determined to be inconsistent with the location safety specifications, an alarm will be issued. Continuous monitoring is conducted once it is determined that the spatial location of the electrical worker complies with the location safety regulations.

9. The method according to claim 1, characterized in that, The method further includes: Extract the clothing features of the power workers in the real-time images; Determine whether the clothing features of the power workers in the real-time image comply with the safety regulations for clothing. If the wearer's clothing in the real-time image does not conform to the safety standards, an alarm will be issued. If the wear characteristics of the power workers are determined to comply with the safety standards for wearing protective clothing, continuous monitoring shall be conducted.

10. A system for precise positioning of power workers based on three-dimensional spatial technology, characterized in that, include: A camera module is installed in the power grid project to acquire images of the power grid project. The point cloud data acquisition module is used to acquire the three-dimensional point cloud data of the power grid project; The controller, connected to the camera module and the point cloud data acquisition module, is used to perform the method as described in any one of claims 1-9.

Images