Regional risk early warning method, electronic equipment, storage medium and program product

Abstract

The invention provides a regional risk early warning method, electronic equipment, a storage medium and a program product. Relates to the technical field of power safety management. The method comprises the following steps: performing moving object detection on multiple frames of images continuously acquired by an acquisition assembly to obtain corresponding first pose information of a moving object in each frame of target image containing the moving object; based on the first pose information, mapping the moving object to the three-dimensional twin model of the transformer substation to obtain second pose information of the moving object in the three-dimensional twin model of the transformer substation; according to the second pose information, whether the moving object has a risk behavior in a corresponding target risk area is determined, and the target risk area is determined based on the electric power security database; and if the risk behavior exists, outputting early warning information corresponding to the risk behavior for the moving object. The method and the device are used for improving the early warning accuracy.

Description

Technical Field

[0001] The present application relates to the technical field of power safety management, and in particular to a regional risk early warning method, electronic equipment, storage medium, and program product. Background Art

[0002] In the monitoring and management of substations, real-time and accurate acquisition of the spatial location information of mobile objects can effectively monitor the dynamics of personnel and equipment in the substation, promptly discover potential safety hazards, and ensure the stable operation of the substation.

[0003] In related technologies, images of moving objects are captured by cameras, and risk warnings are issued when the moving objects are determined to be in risk areas. However, there are problems of missed reports and false reports, which interfere with the normal working process of the substation.

[0004] Based on this, there is an urgent need for a regional risk warning solution that can improve the accuracy of warnings. Summary of the Invention

[0005] This application provides a regional risk warning method, electronic equipment, storage medium and program product to achieve the effect of improving the warning accuracy.

[0006] In a first aspect, the present application provides a regional risk early warning method, comprising:

[0007] Performing moving object detection on multiple frames of images continuously acquired by the acquisition component to obtain the first position information corresponding to the moving object in each frame of the target image containing the moving object;

[0008] Based on the first pose information, the mobile object is mapped to the substation 3D twin model to obtain the second pose information of the mobile object in the substation 3D twin model;

[0009] determining, based on the second posture information, whether the mobile object has engaged in risky behavior within a corresponding target risk area, where the target risk area is determined based on a power safety database;

[0010] If there is any risky behavior, the warning information corresponding to the risky behavior is output for the mobile object.

[0011] In one possible implementation, based on the first pose information, the mobile object is mapped to the three-dimensional twin model of the substation to obtain the second pose information of the mobile object in the three-dimensional twin model of the substation, including:

[0012] Obtain the translation matrix and rotation matrix of the acquisition component when acquiring each frame of the target image containing the moving object, where the translation matrix represents the translation position of the acquisition component in the 3D twin model of the substation, and the rotation matrix represents the rotation position of the acquisition component in the 3D twin model of the substation;

[0013] For any frame target image, based on the first pose information corresponding to the target image, the alignment transformation parameters corresponding to the target image are determined according to the translation matrix and rotation matrix corresponding to the target image; according to the alignment transformation parameters corresponding to the target image, the mobile object is mapped to the three-dimensional twin model of the substation to obtain the second pose information of the mobile object in the three-dimensional twin model of the substation.

[0014] In one possible implementation, the second posture information includes an outline of the mobile object, and determining whether the mobile object has engaged in risky behavior within a corresponding target risk area based on the second posture information includes:

[0015] According to the outline, determine the outline coordinates of the mobile object in the substation 3D twin model;

[0016] If there is a contour coordinate within the target risk area corresponding to the moving object, it is determined that the moving object has risky behavior within the corresponding target risk area;

[0017] Accordingly, outputting warning information corresponding to the risk behavior for the moving object includes: outputting first warning information for the moving object, the first warning information indicating warning information for exiting the target risk area.

[0018] In one possible implementation, the second posture information includes the motion and type of the mobile object. Determining whether the mobile object has engaged in risky behavior within the corresponding target risk area based on the second posture information includes:

[0019] Based on the action and object type, the movement trend of the moving object within a preset time period in the future is predicted;

[0020] Determine the predicted coordinates of the moving object in the substation 3D twin model based on the movement trend;

[0021] If the predicted coordinates of the moving object are within the corresponding target risk area within a preset time period in the future, it is determined that the moving object has risky behavior within the corresponding target risk area;

[0022] Accordingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting second warning information for the mobile object, the second warning information indicating that the mobile object is prohibited from entering the target risk area.

[0023] In one possible implementation, the second position information further includes a posture, and the motion trend of the moving object within a preset time period in the future is predicted based on the motion and object type, including:

[0024] According to the action, posture and object type, a motion prediction algorithm is used to predict the action of the mobile object within a preset time period in the future, and the motion trend of the mobile object within the preset time period in the future is obtained.

[0025] In one possible implementation, the second posture information includes the outline, motion, posture, and object type of the mobile object. Determining whether the mobile object has engaged in risky behavior within the corresponding target risk area based on the second posture information includes:

[0026] When the object type of the moving object is a set type, determine whether the moving object has violated any regulations within the corresponding target risk area based on its outline, motion, and posture;

[0027] If the moving object has violated the rules in the corresponding target risk area, it is determined that the moving object has engaged in risky behavior in the corresponding target risk area;

[0028] Accordingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting third warning information for the mobile object, the third warning information instructing to stop performing the illegal behavior in the corresponding target risk area.

[0029] In a second aspect, the present application provides a regional risk early warning device, comprising:

[0030] A detection module is used to detect moving objects on multiple frames of images continuously acquired by the acquisition component, and obtain the first position information corresponding to the moving object in each frame of the target image containing the moving object;

[0031] The processing module is used to map the mobile object to the three-dimensional twin model of the substation based on the first posture information, and obtain the second posture information of the mobile object in the three-dimensional twin model of the substation; according to the second posture information, determine whether the mobile object has risky behavior in the corresponding target risk area, and the target risk area is determined based on the power safety database; if there is risky behavior, output warning information corresponding to the risk behavior for the mobile object.

[0032] In a possible implementation manner, the processing module is further configured to:

[0033] Obtain the translation matrix and rotation matrix of the acquisition component when acquiring each frame of the target image containing the moving object, where the translation matrix represents the translation position of the acquisition component in the 3D twin model of the substation, and the rotation matrix represents the rotation position of the acquisition component in the 3D twin model of the substation;

[0034] For any frame target image, based on the first pose information corresponding to the target image, the alignment transformation parameters corresponding to the target image are determined according to the translation matrix and rotation matrix corresponding to the target image; according to the alignment transformation parameters corresponding to the target image, the mobile object is mapped to the three-dimensional twin model of the substation to obtain the second pose information of the mobile object in the three-dimensional twin model of the substation.

[0035] In one possible implementation, the second posture information includes a contour of the moving object, and the processing module is further configured to:

[0036] According to the outline, determine the outline coordinates of the mobile object in the substation 3D twin model;

[0037] If there is a contour coordinate within the target risk area corresponding to the moving object, it is determined that the moving object has risky behavior within the corresponding target risk area;

[0038] Accordingly, outputting warning information corresponding to the risk behavior for the moving object includes: outputting first warning information for the moving object, the first warning information indicating warning information for exiting the target risk area.

[0039] In one possible implementation, the second posture information includes the motion and type of the moving object, and the processing module is further configured to:

[0040] Based on the action and object type, the movement trend of the moving object within a preset time period in the future is predicted;

[0041] Determine the predicted coordinates of the moving object in the substation 3D twin model based on the movement trend;

[0042] If the predicted coordinates of the moving object are within the corresponding target risk area within a preset time period in the future, it is determined that the moving object has risky behavior within the corresponding target risk area;

[0043] Accordingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting second warning information for the mobile object, the second warning information indicating that the mobile object is prohibited from entering the target risk area.

[0044] In one possible implementation, the second posture information further includes a posture, and the processing module is further configured to:

[0045] According to the action, posture and object type, a motion prediction algorithm is used to predict the action of the mobile object within a preset time period in the future, and the motion trend of the mobile object within the preset time period in the future is obtained.

[0046] In one possible implementation, the second posture information includes the outline, action, posture, and object type of the moving object, and the processing module is further configured to:

[0047] When the object type of the moving object is a set type, determine whether the moving object has violated any regulations within the corresponding target risk area based on its outline, motion, and posture;

[0048] If the moving object has violated the rules in the corresponding target risk area, it is determined that the moving object has engaged in risky behavior in the corresponding target risk area;

[0049] Accordingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting third warning information for the mobile object, the third warning information instructing to stop performing the illegal behavior in the corresponding target risk area.

[0050] In a third aspect, the present application provides an electronic device, comprising: a memory, a processor;

[0051] Memory stores computer-executable instructions;

[0052] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0053] In a fourth aspect, the present application provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementation methods of the first aspect.

[0054] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementations of the first aspect.

[0055] The regional risk warning method, electronic device, storage medium, and program product provided in this application detect mobile objects through multiple frames of images continuously collected by the acquisition component in the substation, and obtain the first pose information of the mobile object in each frame of the target image. The first pose information collected by the acquisition component can reflect the motion information of the mobile object. Based on the first pose information, combined with the substation three-dimensional twin model, the first pose information is accurately mapped to the substation three-dimensional twin model to obtain second pose information. The second pose information includes the positional relationship and motion relationship of the mobile object relative to the substation three-dimensional twin model. The second pose information can accurately determine the positional relationship of the mobile object relative to the substation three-dimensional twin model. Based on the second pose information, the information of the mobile object in the substation three-dimensional twin model is obtained, and combined with the power safety database, it is determined whether the mobile object is in the corresponding target risk area and whether the mobile object has risky behavior. If the mobile object is detected to have risky behavior, the corresponding risk behavior warning information will be output for the mobile object. The above scheme can automatically determine whether the mobile object has risky behavior and output the corresponding warning information if the mobile object has risky behavior, thereby effectively improving the accuracy of risk warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0057] Figure 1 A schematic diagram of a scenario of a regional risk warning method provided in an embodiment of the present application;

[0058] Figure 2 Schematic diagram of the process of regional risk warning method provided in the embodiment of this application Figure 1 ;

[0059] Figure 3 Schematic diagram of the process of regional risk warning method provided in the embodiment of this application Figure 2 ;

[0060] Figure 4 A schematic diagram of the structure of a regional risk warning device provided in an embodiment of the present application;

[0061] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0062] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0063] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0064] In related technologies, cameras within substations identify moving objects and confirm their coordinates to determine their location. However, this leads to inaccurate positioning of moving objects, making it impossible to accurately determine their exact location within the substation. This leads to missed and false alarms when determining whether a moving object is in a risky area and issuing risk warnings, disrupting the normal working process of the substation.

[0065] The regional risk warning method provided in the embodiment of the present application can track mobile objects in real time through the acquisition component in the substation and accurately map the mobile objects to the substation's three-dimensional twin model. By integrating the mobile objects with the substation's three-dimensional twin model, the position of the mobile objects in the substation can be accurately located. Then, combined with the power safety database, the target risk area of the mobile object in the substation is determined, and whether the mobile object has risky behavior is determined. If the mobile object has risky behavior, warning information corresponding to the risk behavior will be output for the mobile object to achieve risk management and prevention, effectively improve the accuracy of the warning, and ensure the safe operation of the substation.

[0066] Figure 1 This is a schematic diagram of a scenario of a regional risk warning method provided in an embodiment of the present application. Figure 1 As shown, the specific application scenario of this application includes a collection component 11, a substation 12, a mobile object 13, an early warning center 14 and a moving trajectory 15 of the mobile object.

[0067] Multiple acquisition components 11 are deployed within substation 12 for capturing multiple frames of images in real time. These components are connected to an early warning center 14 via a wired or wireless network, transmitting the captured image data to the center in real time. As a mobile object 13 moves within substation 12, it forms a trajectory 15. Simultaneously, the image of the mobile object 13 is captured by the acquisition components 11, which then use intelligent algorithms to detect and obtain the first position information of the mobile object 13.

[0068] The early warning center 14 is internally provided with an electric power safety database, through which information such as the operating status of the equipment area and the permitted access status in the substation 12 can be obtained. The early warning center 14 uses the substation three-dimensional twin model and the automatic calibration algorithm to accurately map the first posture information of the mobile object 13 to the substation three-dimensional twin model, and obtains the second posture information of the mobile object 13. The early warning center 14 monitors the dynamic behavior of the mobile object 13 in the substation in real time through the second posture information, and determines whether the mobile object 13 is in the target risk area by combining the information in the electric power safety database. If the early warning center 14 detects that the mobile object 13 has risky behavior, the early warning center 14 will issue a warning message to notify on-site personnel or take direct measures to eliminate the risk hazards of the mobile object 13.

[0069] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0070] Figure 2Schematic diagram of the process of regional risk warning method provided in the embodiment of this application Figure 1 .like Figure 2 As shown, the method includes:

[0071] S201: Perform moving object detection on multiple frames of images continuously acquired by an acquisition component to obtain first position information corresponding to the moving object in each frame of target image containing the moving object.

[0072] The acquisition component is a mobile image acquisition device installed within the substation, used to capture multiple frames of real-time images of the substation. For example, the acquisition component is a pan-tilt camera. Moving objects include people, vehicles, and other equipment moving within the substation.

[0073] The acquisition component continuously captures images within the substation at a preset frame rate. The preset frame rate is an acquisition parameter set according to actual risk warning requirements. Exemplarily, the preset frame rate can be 30 frames per second or 60 frames per second. The acquired images are transmitted to the processing unit of the acquisition component, which analyzes each frame of the image through a target detection algorithm to identify moving objects in the image. For the identified moving objects, the acquisition component further extracts the first position information corresponding to the moving object. Optionally, the first position information represents information such as the position and posture of the moving object in the image captured by the acquisition component.

[0074] Exemplarily, the target detection algorithms used include but are not limited to background difference algorithms and optical flow methods. The background difference algorithm detects moving targets by comparing the differences between the pixels of the current frame and the background frame, and is particularly suitable for environments where the background is relatively fixed and the lighting conditions do not change much. The optical flow method determines the motion trajectory of the target by analyzing the movement of pixels in the image sequence, and can provide detailed information on the target movement. It is suitable for scenarios that require accurate analysis of the target motion trajectory. In the embodiments of the present application, different target detection algorithms are used according to the risk warning environment of different substations, or a combination of different target detection algorithms to achieve the best moving object detection effect.

[0075] In one possible implementation, the first pose information also includes the object type of the moving object. The feature information corresponding to the moving object is determined by the moving object in the image. Thereafter, the feature information is matched with the information in the power safety database through database matching to obtain the corresponding object type. The power safety database stores safety information related to the mobile object and the target risk area corresponding to the mobile object. Optionally, mobile objects of different object types have different target risk areas. For example, the object types of staff members and external visitors in the substation are different, and the corresponding target risk areas are also different. For example, external visitors can only enter the visiting area and are not allowed to enter the maintenance area of the power equipment. Therefore, the target risk area for external visitors is the maintenance area. Staff members are not allowed to enter the high-voltage equipment area and the non-working area. Therefore, the risk area for staff members is the high-voltage equipment area and the non-working area. For example, different staff members in the substation have different object types, and the corresponding target risk areas are also different.

[0076] In one possible implementation, when the acquisition component is a pan-tilt camera, the camera's internal parameters are acquired. These parameters include focal length, principal point, distortion parameters, image size, and zoom factor. Focal length, measured in pixels, describes the lens's ability to focus light, determining the image's viewing angle and the size of moving objects within the image. The principal point, also measured in pixels, is the intersection of the optical axis and the image plane. It is typically located at the center of the image, but in practice, this may deviate due to factors such as installation. Correcting the image using the principal point can produce an image that better reflects actual conditions and reduce acquisition errors within the acquisition component. Distortion parameters describe lens shape distortion and are divided into radial and tangential distortion. Radial distortion causes deviations in pixel positions between the center and edges of the image, resulting in barrel or pincushion distortion. Tangential distortion is caused by the lens being non-parallel or non-perpendicular to the image plane. Image size describes the width and height of the image, determining its resolution and number of pixels, and impacting the ability to capture details of moving objects. The scaling factor describes the difference in scale between the x and y directions because the pixels on the image sensor may not be the same size horizontally or vertically. The image captured by the PTZ camera is then corrected based on the internal parameters to ensure more accurate positioning and shape of objects in the image. This ensures that the captured position and posture of moving objects are highly consistent with the actual situation, improving data accuracy. Obtaining the internal parameters of the PTZ camera and performing image correction can significantly improve the accuracy of risk warnings.

[0077] S202. Based on the first pose information, map the mobile object to the three-dimensional twin model of the substation to obtain second pose information of the mobile object in the three-dimensional twin model of the substation.

[0078] The 3D twin model of a substation is a virtual model built based on the physical entity of the substation, which can reflect the entire life cycle of the equipment and environment within the substation. The second pose information is obtained by mapping the first pose information to the 3D twin model of the substation, and is used to represent the position, posture, and movement information of the mobile object in the 3D twin model of the substation. The position and posture information of the mobile object in the physical space is accurately reflected in the virtual 3D twin model through the automatic calibration algorithm, which can determine the precise position of the mobile object in the substation and provide accurate data support for risk warning. The automatic calibration algorithm is an algorithm that uses automated means to perform parameter correction and optimization on the first pose information. The automatic calibration algorithm can map the first pose information to the 3D twin model of the substation.

[0079] For example, a data fusion transformation algorithm is used as the automatic calibration algorithm. This algorithm can remove noise from the first pose information and obtain key features from the first pose information. It also maps the data coordinates in the first pose information to the substation 3D twin model to obtain the second pose information of the mobile object in the substation 3D twin model.

[0080] For example, a visual mapping algorithm is used for automatic calibration. This algorithm maps the first-order pose information of a mobile object from physical space to the model object in the substation's 3D twin model. By matching the first-order pose information with the corresponding position in the 3D twin model, the second-order pose information of the mobile object in the substation's 3D twin model is obtained, enabling precise positioning of the mobile object in the virtual model.

[0081] For example, the automatic calibration algorithm can employ a deep learning-based 3D calibration algorithm. Using deep learning techniques, the mapping relationship between background objects in the real-time video imagery captured by the acquisition component and the static point cloud image of the substation is learned. The background objects can be moving objects or static equipment or devices within the substation. During the test and training phase, each frame of the real-time video imagery captured by the acquisition component and the point cloud data in the static point cloud image of the substation are used as input. At the output layer, alignment transformation parameters between the real-time video imagery and the point cloud data are obtained. Subsequently, the alignment transformation parameters are used to calibrate the first pose information, thereby obtaining the corresponding second pose information.

[0082] S203. Determine, based on the second posture information, whether the mobile object has engaged in risky behavior within a corresponding target risk area, where the target risk area is determined based on a power safety database.

[0083] Risky behaviors are actions by mobile objects within target risk zones that could potentially cause safety incidents, including unauthorized entry into restricted areas and proximity to operating equipment. The target risk zone corresponding to the mobile object is determined based on the mobile object and the power safety database. Optionally, the target risk zone corresponding to the mobile object's type is retrieved from the power safety database. Target risk zones include high-voltage equipment areas and maintenance areas.

[0084] Exemplarily, based on the information in the power safety database, a target risk area within the substation is determined. The second position information of the mobile object is compared with the target risk area to determine whether the mobile object has engaged in risky behavior within the corresponding target risk area. Optionally, the coordinates of the mobile object are determined based on the second position information. Based on the relationship between the coordinates of the mobile object and the target risk area, it is determined whether the mobile object has engaged in risky behavior. Optionally, the behavior and posture of the mobile object are determined based on the second position information. Based on the behavior and posture of the mobile object, it is determined whether the mobile object has engaged in risky behavior within the corresponding target risk area.

[0085] S204: If there is any risky behavior, output warning information corresponding to the risky behavior for the mobile object.

[0086] If a moving object is detected in the corresponding target area and there is risky behavior, an early warning message corresponding to the risky behavior is output for the moving object. The early warning method can be output in a variety of ways to ensure that the moving object can receive the early warning in time. Exemplarily, the early warning information is displayed on the monitoring center screen of the substation, and at the same time, an early warning information text message is sent or an early warning information notification is pushed to the mobile device of the staff in the substation. Exemplarily, an early warning message is issued through the voice broadcasting system in the substation to remind the moving objects on site to pay attention to safety. Optionally, when the moving object is a wild animal, a driving voice is issued through the voice broadcasting system in the substation. At the same time, according to the corresponding type of wild animal, an early warning message to avoid the wild animal or an early warning message to drive away the wild animal is sent to the mobile device of the staff in the substation.

[0087] The regional risk warning method provided by the embodiment of the present application continuously captures multiple frames of images within the substation through an acquisition component, and detects mobile objects in the multiple frames using a target detection algorithm. This method obtains the corresponding first-position information for each frame of the target image containing the mobile object, thereby understanding the mobile object's behavior in real time. Based on the first-position information, the mobile object is accurately mapped to the substation's three-dimensional twin model using a three-dimensional substation model and an automatic calibration algorithm, obtaining the second-position information of the mobile object in the substation's three-dimensional twin model. This second-position information can accurately simulate the mobile object's movements, behaviors, and postures in the substation's three-dimensional twin model, enabling dynamic monitoring and accurate positioning of the mobile object, thereby improving the intuitiveness and accuracy of risk warnings. Based on the second-position information, combined with the corresponding target risk area in the power safety database, it is determined whether the mobile object has risky behavior within the corresponding target risk area. If risky behavior exists, warning information corresponding to the risky behavior is output for the mobile object. By outputting the warning information, relevant personnel can be quickly notified to take measures to avoid or reduce the occurrence of safety accidents, improve the safety management level of the substation, and achieve the effect of improving the accuracy of risk warnings.

[0088] In one possible implementation, based on the first posture information, the mobile object is mapped to the three-dimensional twin model of the substation. After obtaining the second posture information of the mobile object in the three-dimensional twin model of the substation, it also includes: determining the first relative relationship between the mobile object and the equipment in the three-dimensional twin model of the substation based on the second posture information and the three-dimensional twin model of the substation; identifying the second relative relationship between the mobile object and the equipment based on the image of the mobile object; and updating and correcting the second posture information based on the first relative relationship and the second relative relationship.

[0089] First, using the second posture information and the fixed position information of the equipment in the three-dimensional twin model of the substation, the specific position and direction relationship of the mobile object relative to each device can be determined. For example, the first relative relationship can be determined by determining the straight-line distance and angular deviation between the mobile object and a certain device in the three-dimensional twin model of the substation. Furthermore, by processing and analyzing the image containing the mobile object, computer vision technology is used to identify the feature information of the mobile object and the device in the image, where the feature information includes feature points, contours and shapes. For example, the contour of the mobile object and the contour of the device are determined by an edge detection algorithm; based on the contour of the mobile object and the contour of the device, the straight-line distance and angular deviation between the mobile object and a certain device in the substation are determined by a feature matching algorithm to determine the second relative relationship.

[0090] Finally, based on the first and second relative relationships, the second pose information is updated and corrected using an appropriate correction algorithm, such as the least squares method and Kalman filter. If the first and second relative relationships deviate in certain dimensions, the position and attitude parameters in the second pose information are adjusted accordingly based on the magnitude and direction of the deviation.

[0091] For example, if the distance and offset angle between a mobile object and a device are used to represent the relative relationship between the mobile object and a device, then in the first relative relationship, the distance between the mobile object and the device in the substation 3D twin model is determined to be a first distance, the offset angle relationship between the mobile object and the device in the substation 3D twin model is determined to be a first offset angle, and the first distance and the first offset angle are used to describe the first relative relationship between the mobile object and the device in time. Similarly, in the second relative relationship, the distance between the mobile object and the device in the substation is determined to be a second distance, the offset angle relationship between the mobile object and the device in the substation is determined to be a second offset angle, and the second distance and the second offset angle are used to describe the second relative relationship between the mobile object and the device in time. If the first distance is 40 cm, the first offset angle is 25°, the second distance is 30 cm, and the second offset angle is 23°, the target distance between the mobile object and the device is determined to be the average of the first distance and the second distance, and the target offset angle between the mobile object and the device is determined to be the average of the first offset angle and the second offset angle. Based on the position coordinates, target distance, and target offset angle of a device in the substation 3D twin model, the position coordinates of the mobile object are updated using a position determination algorithm to obtain updated second pose information. The position determination algorithm may calculate the target position coordinates of the mobile object based on the coordinates, target distance, and target offset angle of the device, and update the position coordinates of the mobile object according to the target position coordinates.

[0092] Figure 3 Schematic diagram of the process of regional risk warning method provided in the embodiment of this application Figure 2 .like Figure 3 As shown, this embodiment Figure 2 Based on the embodiment, the regional risk early warning method is described in detail, including:

[0093] In one possible implementation, Figure 2 In the embodiment, step S202 further includes:

[0094] S2021. Obtain the translation matrix and rotation matrix of the acquisition component when acquiring each frame of the target image containing the moving object, wherein the translation matrix represents the translation position of the acquisition component in the three-dimensional twin model of the substation, and the rotation matrix represents the rotation position of the acquisition component in the three-dimensional twin model of the substation.

[0095] In the three-dimensional model of the substation, the position and attitude of the acquisition component are described by the translation matrix and the rotation matrix. The translation matrix consists of translation components in the x, y, and z directions, and can determine the position offset of the acquisition component in the three-dimensional twin model of the substation. The rotation matrix consists of rotation parameters along the x-axis, y-axis, and z-axis, which represent the orientation and attitude of the acquisition component relative to the three-dimensional twin model of the substation. Optionally, by analyzing the image sequence captured by the acquisition component, identifying the feature points in the image, and determining the position changes of the feature points in different images, the translation matrix and rotation matrix of the acquisition component can be derived.

[0096] For example, when the acquisition component is a pan-tilt camera, the camera's internal parameters are obtained, including focal length, principal point, distortion parameters, image size, and zoom factor. Errors in the image due to factors such as lens distortion and installation deviation are corrected using these internal parameters. The target image captured by the pan-tilt camera is then analyzed in conjunction with feature points in the image to derive the camera's translation and rotation matrices.

[0097] S2022. For any frame target image, based on the first pose information corresponding to the target image, and according to the translation matrix and rotation matrix corresponding to the target image, determine the alignment transformation parameters corresponding to the target image; according to the alignment transformation parameters corresponding to the target image, map the mobile object to the three-dimensional twin model of the substation to obtain the second pose information of the mobile object in the three-dimensional twin model of the substation.

[0098] Alignment transformation parameters are used to describe the translation and rotation relationship between coordinate systems and are usually expressed in the form of a homogeneous transformation matrix. The homogeneous transformation matrix contains information about the translation matrix and the rotation matrix.

[0099] For any target image containing mobile object information in any frame, a homogeneous transformation matrix is constructed based on the corresponding translation matrix and rotation matrix of the target image. The homogeneous transformation matrix can transform the mobile object in the image from the camera coordinate system to the coordinate system of the substation 3D twin model. Using the homogeneous transformation matrix, the first pose information of the mobile object in the image is mapped to the substation 3D twin model, thereby obtaining the second pose information of the mobile object in the substation 3D twin model. The second pose information can accurately represent the actual position and posture of the mobile object in the substation 3D twin model.

[0100] For example, from the translation matrix, the translation vector is obtained ,in Indicates the amount of translation in the X-axis direction, Indicates the amount of translation in the y-axis direction, Indicates the amount of translation in the z-axis direction. The rotation matrix represents the rotation around a certain axis. The rotation matrix can be, for example, the rotation matrix around the x-axis , the rotation matrix around the y-axis and the rotation matrix around the z axis Specifically, the rotation matrix The corresponding homogeneous transformation matrix is:

[0101]

[0102] The homogeneous transformation matrix is used as the alignment transformation parameter. According to the alignment transformation parameter H and the first pose information M corresponding to the target image, the second pose information of the mobile object in the 3D twin model of the substation is obtained as N=M×H.

[0103] In one possible implementation, deep learning techniques are used to learn the mapping relationship between mobile objects in the first pose information collected by the acquisition component and the substation's three-dimensional twin model. Background objects can be mobile objects or static equipment or devices within the substation. During the test and training phase, each frame of the real-time video image captured by the acquisition component and the point cloud data from the substation's static point cloud map are used as input. At the output layer, alignment transformation parameters between the real-time video image and the point cloud data are obtained. Subsequently, the alignment transformation parameters are used to calibrate the first pose information, thereby obtaining the corresponding second pose information.

[0104] In one possible implementation, the second posture information includes the outline of the moving object. Figure 2 In the embodiment, step S203 may further include:

[0105] Exemplarily, the contour of the mobile object is obtained by mapping the contour in the first pose information to the three-dimensional twin model of the substation, and / or, the boundary of the mobile object in the three-dimensional twin model of the substation is identified by an edge detection algorithm to obtain the contour of the mobile object.

[0106] S2031. Determine the contour coordinates of the mobile object in the three-dimensional twin model of the substation based on the contour.

[0107] Contour coordinates are the boundary position information of the mobile object in the three-dimensional twin model of the substation, which are used to describe the shape and position of the mobile object. For example, the contour coordinates can be simply determined by the position where the mobile object contacts the background image. When the mobile object stands on the ground, the three-dimensional coordinates corresponding to the coordinates of the position where the mobile object contacts the ground are the contour coordinates corresponding to the mobile object. At the same time, the contour coordinates can also be the coordinates of a series of contours at different positions in the three-dimensional twin model of the substation, which can reflect the changes in the position and contour of the mobile object in the three-dimensional model. For example, when the mobile object is a person, the contour coordinates can be the position coordinates of various parts of the human body in the three-dimensional twin model of the substation, including head coordinates, limb coordinates, and torso coordinates.

[0108] S2032: If there is a contour coordinate within the target risk area corresponding to the moving object, determine that the moving object has risky behavior within the corresponding target risk area.

[0109] After obtaining the outline coordinates of the moving object, compare them with the target risk area corresponding to the moving object. Determine whether the outline coordinates of the moving object overlap with the target risk area. If the outline coordinates of the moving object overlap with the target risk area, it is determined that the moving object has engaged in risky behavior within the corresponding target risk area. For example, if the target risk area is a high-voltage equipment area, and the outline coordinates of the moving person overlap with the coordinate area corresponding to the high-voltage equipment area, it is determined that the moving object has engaged in risky behavior within the high-voltage equipment area.

[0110] For example, when a moving object has multiple outline coordinates, the overlap range can be further limited. The overlap range includes single-point overlap and multi-point overlap. When the overlap range is single-point overlap, if one of the outline coordinates overlaps with the target risk area, the moving object is determined to have engaged in risky behavior within the corresponding target risk area. When the overlap range is N-point overlap, if N of the outline coordinates overlap with the target risk area, the moving object is determined to have engaged in risky behavior within the corresponding target risk area, where N is an integer greater than or equal to 2.

[0111] Accordingly, step S204 further includes: outputting first warning information for the mobile object, where the first warning information indicates warning information for exiting the target risk area.

[0112] When the system detects that the contour coordinates of a moving object enters the target risk area, it will immediately generate a first warning message to instruct the moving object to exit the target risk area.

[0113] For example, when the moving object is a staff member, a first warning message is sent to the staff member's mobile terminal, with the content "You have entered the dangerous area, please exit immediately", and at the same time, the voice broadcast system in the substation sends a voice corresponding to the first warning message.

[0114] For example, when the moving object is an electric power equipment that can be operated by the substation control center, a first warning message is sent to the substation control center, which reads "The equipment has entered the dangerous area, please exit immediately". At the same time, the substation control center can also be requested to move the moving object out of the target risk area.

[0115] In one possible implementation, the second posture information includes the motion of the moving object and the object type. Figure 2 In the embodiment, step S203 may further include:

[0116] Exemplarily, the action in the first pose information is mapped into the three-dimensional twin model of the substation to obtain the action of the mobile object, and / or the action of the mobile object in the three-dimensional twin model of the substation is detected by a motion detection algorithm to obtain the action of the mobile object.

[0117] S2033: Predict the movement trend of the moving object within a preset time period in the future based on the movement and object type.

[0118] Motion trends refer to the direction, speed, and behavior patterns of a moving object within a preset future duration. By analyzing the history and current motion of a moving object, its future behavior trends can be predicted. The future preset duration is a set time range for predicting motion trends, which can be set according to actual operational requirements. The unit of the future preset duration can be milliseconds, seconds, minutes, or hours. Optionally, the future preset duration can be 10 milliseconds, 20 milliseconds, 35 milliseconds, or 40 milliseconds; the future preset duration can also be 15 seconds, 25 seconds, 30 seconds, or 45 seconds; the future preset duration can also be 1 minute, 1.5 minutes, 5 minutes, or 8 minutes; the future preset duration can also be 0.5 hours, 1 hour, 1.8 hours, or 2 hours.

[0119] Based on the corresponding motion and object type of the moving object, the motion trend of the moving object within a preset time period is predicted. For example, if the corresponding object type of the moving object is a person, the moving object's walking direction and speed within the preset time period are predicted to obtain the motion trend of the moving object within the preset time period. For example, if the corresponding object type of the moving object is a car, the moving object's route within the preset time period is predicted to obtain the motion trend of the moving object within the preset time period.

[0120] S2034. Determine the predicted coordinates of the mobile object in the three-dimensional twin model of the substation based on the movement trend.

[0121] Based on the movement trends of the moving object, the corresponding contour information is obtained. Subsequently, based on this contour information, the predicted coordinates of the moving object in the substation's 3D twin model are further determined. This knowledge of the moving object's predicted coordinates provides an important basis for subsequent risk warnings. The predicted coordinates are the coordinates of the location that the moving object is likely to reach within a preset timeframe in the future.

[0122] Optionally, the predicted coordinates include the different coordinates of the moving object at multiple time points within a preset future duration, reflecting the moving object's position changes at different future time points, and can provide more comprehensive information about the moving object's motion trajectory. For example, if the preset future duration is 30 seconds, the predicted position coordinates of the moving object are at 5 seconds, 10 seconds, 15 seconds, 20 seconds, 25 seconds, and 30 seconds. Using the coordinates at different time points, the moving object's motion trajectory can be mapped, allowing for a more accurate determination of whether the moving object has entered the target risk area.

[0123] S2035: If the predicted coordinates of the moving object are within the corresponding target risk area within a preset time period in the future, determine that the moving object has risky behavior within the corresponding target risk area.

[0124] After obtaining the predicted coordinates of the moving object, compare them with the target risk area corresponding to the moving object. Determine whether the predicted coordinates of the moving object overlap with the target risk area. If the predicted coordinates of the moving object overlap with the target risk area, it is determined that the moving object has engaged in risky behavior within the corresponding target risk area.

[0125] Optionally, the predicted coordinates include different coordinates of the mobile object at multiple time points within a preset future duration. If any of the predicted coordinates overlaps with the target risk area, it is determined that the mobile object has engaged in risky behavior within the corresponding target risk area. For example, assuming the preset future duration is 30 seconds, the predicted coordinates of the mobile object at the 5th second are (8, 8, 23); at the 10th second, (12, 12, 12); at the 15th second, (18, 18, 16); at the 20th second, (22, 22, 2); at the 25th second, (24, 24, 4); and at the 30th second, (26, 26, 4). The target risk area corresponding to the mobile object is a rectangular region with coordinates ranging from (10, 10, 0) to (20, 20, 10). By determining whether each predicted coordinate is within the target risk area, it is determined whether the mobile object has engaged in risky behavior within the corresponding target risk area. At the 10th and 15th seconds, the moving object's coordinates (12, 12, 12) and (18, 18, 16) were both within the target risk zone. Therefore, the moving object exhibited risky behavior within the corresponding target risk zone. By analyzing the moving object's motion trends, early risk warnings can be issued, improving their accuracy and enhancing substation safety management.

[0126] Accordingly, step S204 further includes: outputting a second warning message for the mobile object, where the second warning message indicates that the mobile object is prohibited from entering the target risk area.

[0127] If the system predicts that a moving object will enter a target risk zone within a preset timeframe, it generates a second warning message, explicitly prohibiting the moving object from entering the target risk zone. For example, if the system predicts that a moving object will enter the target risk zone within the next 30 seconds, the system will issue a warning message: "Warning: You are about to enter a dangerous area. Please stop and exit immediately." By predicting the movement trends of moving objects within a preset timeframe, potential risk behaviors can be detected in advance, prompting timely warnings and improving the accuracy and reliability of risk warnings.

[0128] In one possible implementation, the second posture information includes the outline, action, posture, and object type of the moving object. Figure 2 In the embodiment, step S203 may further include:

[0129] Optionally, the posture of the mobile object can be obtained by mapping the posture in the first posture information to the three-dimensional twin model of the substation; or the posture of the mobile object in the three-dimensional twin model of the substation can be identified through a posture recognition algorithm to obtain the posture of the mobile object.

[0130] S2036: When the object type of the moving object is a set type, determine whether the moving object has any violation behavior in the corresponding target risk area based on the outline, action, and posture.

[0131] A set type is a predefined, specific object type. In substation risk warnings, set types include "Staff" and "Vehicle." A violation is defined as any activity performed by a mobile object within a target risk area that does not comply with safety regulations. Based on the set type, the target risk area and the corresponding violation within that target risk area are determined for the mobile object. The mobile object's movements and posture are used to determine whether it is engaging in a violation; its outline is used to determine whether it is within the target risk area.

[0132] S2037: If the mobile object has violated the regulations in the corresponding target risk area, determine that the mobile object has engaged in risky behavior in the corresponding target risk area.

[0133] When the mobile object has the corresponding violation and is in the corresponding target risk area, it is determined that the mobile object has the violation in the corresponding target risk area. That is, only when the mobile object has the corresponding violation and is in the corresponding target risk area, it is determined that the mobile object has the violation.

[0134] For example, when the object type of the mobile object is "staff," if the staff member is within the target risk area and has committed the violation of "not wearing safety equipment," the mobile object is determined to have committed a violation within the corresponding target risk area, and the mobile object is further determined to have committed a risky behavior within the corresponding target risk area. For example, when the object type of the mobile object is "vehicle," if the vehicle is within the target risk area within the substation and has committed the violation of "speeding," the mobile object is determined to have committed a violation within the corresponding target risk area, and the mobile object is further determined to have committed a risky behavior within the corresponding target risk area.

[0135] Accordingly, step S204 further includes: outputting a third warning message for the mobile object, where the third warning message instructs to stop performing illegal actions in the corresponding target risk area.

[0136] For mobile objects, a third warning message is generated, explicitly instructing the mobile object to stop the illegal behavior in the corresponding target risk area. For example, if the mobile object commits the illegal behavior of "not wearing safety equipment" in the target risk area "high-voltage equipment area", the output third warning message content is "Warning: You are not wearing safety equipment, please wear safety equipment immediately or exit the high-voltage equipment area." For example, if the mobile object commits the illegal behavior of "speeding" in the target risk area "inspection and maintenance area", the output third warning message content is "Warning: You have exceeded the time limit, please slow down immediately or exit the inspection and maintenance area."

[0137] In one possible implementation, the second posture information also includes posture, and the movement trend of the mobile object in step S2033 within the future preset time period is determined in the following manner: based on the movement, posture and object type, a movement prediction algorithm is used to predict the movement of the mobile object within the future preset time period to obtain the movement trend of the mobile object within the future preset time period.

[0138] Based on the object's motion, posture, and object type, the object's object type is encoded to ensure that data from different object types is correctly processed. Next, an appropriate motion prediction algorithm is selected and integrated through multimodal fusion technology. This leverages the data from different modalities to predict motion trends and derive the object's motion trends over a predetermined timeframe. The predicted motion trends can optionally be output as visual charts or descriptive language.

[0139] Exemplarily, the motion prediction algorithm can be an artificial intelligence model prediction algorithm. The motion segmentation algorithm in the artificial intelligence model prediction algorithm is used as the motion prediction algorithm. The motion segmentation algorithm is an artificial intelligence model based on Transformer, which can decompose the motion of the mobile object into multiple small blocks according to the input motion, posture and object type. By predicting the motion of each small block, the motion trend of the mobile object within a preset time period in the future can be accurately predicted. Exemplarily, the motion prediction algorithm can also be obtained by combining a kinematic model and a posture dynamics model. By considering the physical characteristics of the mobile object, such as the range of motion of the biological joints and the dynamic constraints of the object, the motion of the mobile object within a preset time period in the future can be more accurately predicted. Exemplarily, in human motion prediction, a kinematic model can be used to describe the motion laws of human joints, and combined with a posture dynamics model to analyze the changes in the strength and energy of the human body during the motion process, thereby achieving accurate prediction of the motion trend of the mobile object within a preset time period in the future.

[0140] The regional risk warning method provided in the embodiments of the present application provides risk warnings for mobile objects based on their second posture information. By identifying multiple risk behaviors and outputting corresponding risk warning information, it enables real-time monitoring, risk prediction, and early warning of mobile objects within the substation, ensuring the safety of personnel and equipment. By predicting the movement trends of mobile objects, early warnings can be achieved, and countermeasures can be formulated in advance, improving the accuracy and effectiveness of risk warnings. This results in an improvement in the safety management level of substations.

[0141] Figure 4 This is a schematic diagram of the structure of the regional risk warning device provided in the embodiment of the present application. Figure 4 As shown, the regional risk warning device 40 provided in this embodiment includes:

[0142] The detection module 401 is used to detect moving objects on multiple frames of images continuously captured by the capture component, and obtain the first position information corresponding to the moving object in each frame of the target image containing the moving object;

[0143] Processing module 402 is used to map the mobile object to the three-dimensional twin model of the substation based on the first posture information, and obtain the second posture information of the mobile object in the three-dimensional twin model of the substation; according to the second posture information, determine whether the mobile object has risky behavior in the corresponding target risk area, and the target risk area is determined based on the power safety database; if there is risky behavior, output warning information corresponding to the risk behavior for the mobile object.

[0144] In a possible implementation, the processing module 402 is further configured to:

[0145] Obtain the translation matrix and rotation matrix of the acquisition component when acquiring each frame of the target image containing the moving object, where the translation matrix represents the translation position of the acquisition component in the 3D twin model of the substation, and the rotation matrix represents the rotation position of the acquisition component in the 3D twin model of the substation;

[0146] For any frame target image, based on the first pose information corresponding to the target image, the alignment transformation parameters corresponding to the target image are determined according to the translation matrix and rotation matrix corresponding to the target image; according to the alignment transformation parameters corresponding to the target image, the mobile object is mapped to the three-dimensional twin model of the substation to obtain the second pose information of the mobile object in the three-dimensional twin model of the substation.

[0147] In one possible implementation, the second posture information includes a contour of the moving object, and the processing module 402 is further configured to:

[0148] According to the outline, determine the outline coordinates of the mobile object in the substation 3D twin model;

[0149] If there is a contour coordinate within the target risk area corresponding to the moving object, it is determined that the moving object has risky behavior within the corresponding target risk area;

[0150] Accordingly, outputting warning information corresponding to the risk behavior for the moving object includes: outputting first warning information for the moving object, the first warning information indicating warning information for exiting the target risk area.

[0151] In one possible implementation, the second posture information includes the motion and type of the moving object. The processing module 402 is further configured to:

[0152] Based on the action and object type, the movement trend of the moving object within a preset time period in the future is predicted;

[0153] Determine the predicted coordinates of the moving object in the substation 3D twin model based on the movement trend;

[0154] If the predicted coordinates of the moving object are within the corresponding target risk area within a preset time period in the future, it is determined that the moving object has risky behavior within the corresponding target risk area;

[0155] Accordingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting second warning information for the mobile object, the second warning information indicating that the mobile object is prohibited from entering the target risk area.

[0156] In a possible implementation, the second position information further includes a posture, and the processing module 402 is further configured to:

[0157] According to the action, posture and object type, a motion prediction algorithm is used to predict the action of the mobile object within a preset time period in the future, and the motion trend of the mobile object within the preset time period in the future is obtained.

[0158] In one possible implementation, the second posture information includes the outline, action, posture, and object type of the moving object. The processing module 402 is further configured to:

[0159] When the object type of the moving object is a set type, determine whether the moving object has violated any regulations within the corresponding target risk area based on its outline, motion, and posture;

[0160] If the moving object has violated the rules in the corresponding target risk area, it is determined that the moving object has engaged in risky behavior in the corresponding target risk area;

[0161] Accordingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting third warning information for the mobile object, the third warning information instructing to stop performing the illegal behavior in the corresponding target risk area.

[0162] The regional risk warning device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and are not described in detail in this embodiment.

[0163] Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the electronic device 50 further includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected via a bus 504.

[0164] In a specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that the at least one processor 501 performs the above method.

[0165] The specific implementation process of the processor 501 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0166] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules in the processor.

[0167] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0168] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be categorized as address buses, data buses, and control buses. For ease of presentation, the buses in the drawings of the embodiments of this application are not limited to just one bus or just one type of bus.

[0169] An embodiment of the present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0170] An embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed, any of the above methods is implemented.

[0171] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0172] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in a device as discrete components.

[0173] The division of units is merely a logical functional division; actual implementations may employ alternative divisions, such as combining or integrating multiple units or components into another system, or omitting or disabling certain features. Furthermore, any direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units, either through an interface, electrical, mechanical, or other means.

[0174] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0175] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0176] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0177] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0178] Finally, it should be noted that those skilled in the art will readily identify other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. The present invention is not limited to the precise structure described above and illustrated in the accompanying drawings, and various modifications and variations may be made without departing from the scope thereof. The scope of the present invention is limited solely by the appended claims.

Claims

1. A regional risk early warning method, characterized in that: include: Performing moving object detection on multiple frames of images continuously acquired by the acquisition component to obtain first position information corresponding to the moving object in each frame of the target image containing the moving object; Based on the first posture information, mapping the mobile object to the three-dimensional twin model of the substation to obtain second posture information of the mobile object in the three-dimensional twin model of the substation; determining, based on the second posture information, whether the mobile object has engaged in risky behavior within a corresponding target risk area, the target risk area being determined based on a power safety database; If there is any risky behavior, warning information corresponding to the risky behavior is output for the mobile object.

2. The method according to claim 1, characterized in that The mapping of the mobile object to the three-dimensional twin model of the substation based on the first posture information to obtain second posture information of the mobile object in the three-dimensional twin model of the substation includes: Obtaining a translation matrix and a rotation matrix of the acquisition component when acquiring each frame of the target image containing the moving object, wherein the translation matrix represents the translation position of the acquisition component in the three-dimensional twin model of the substation, and the rotation matrix represents the rotation position of the acquisition component in the three-dimensional twin model of the substation; For any frame target image, based on the first pose information corresponding to the target image, the alignment transformation parameters corresponding to the target image are determined according to the translation matrix and rotation matrix corresponding to the target image; according to the alignment transformation parameters corresponding to the target image, the mobile object is mapped to the three-dimensional twin model of the substation to obtain the second pose information of the mobile object in the three-dimensional twin model of the substation.

3. The method according to claim 1 or 2, characterized in that in, The second posture information includes an outline of the mobile object, and determining whether the mobile object has a risky behavior in the corresponding target risk area according to the second posture information includes: Determining the contour coordinates of the mobile object in the three-dimensional twin model of the substation according to the contour; If there is a contour coordinate within the target risk area corresponding to the mobile object, it is determined that the mobile object has a risky behavior within the corresponding target risk area; Correspondingly, outputting warning information corresponding to the risk behavior for the moving object includes: outputting first warning information for the moving object, wherein the first warning information indicates warning information for exiting the target risk area.

4. The method according to claim 1 or 2, characterized in that in, The second posture information includes the action and object type of the mobile object, and determining whether the mobile object has a risky behavior in the corresponding target risk area according to the second posture information includes: Predicting a movement trend of the moving object within a preset time period in the future based on the movement and the object type; Determining predicted coordinates of the mobile object in the three-dimensional twin model of the substation based on the movement trend; If the predicted coordinates of the moving object are within the corresponding target risk area within a preset time period in the future, it is determined that the moving object has risky behavior within the corresponding target risk area; Correspondingly, outputting warning information corresponding to the risky behavior for the mobile object includes: outputting second warning information for the mobile object, wherein the second warning information indicates that entry into the target risk area is prohibited.

5. The method according to claim 4, characterized in that The second posture information also includes a posture, and the prediction of the movement trend of the mobile object within a future preset time period based on the movement and the object type includes: According to the action, the posture and the object type, a motion prediction algorithm is used to predict the action of the mobile object within the future preset time period to obtain the motion trend of the mobile object within the future preset time period.

6. The method according to claim 1 or 2, characterized in that The second posture information includes an outline, a motion, a posture, and an object type of the mobile object. Determining whether the mobile object has a risky behavior in a corresponding target risk area based on the second posture information includes: When the object type of the moving object is a set type, determining whether the moving object has violated any regulations within the corresponding target risk area based on the outline, the action, and the posture; If the mobile object has violated the regulations in the corresponding target risk area, determining that the mobile object has engaged in risky behavior in the corresponding target risk area; Correspondingly, outputting warning information corresponding to the risk behavior for the mobile object includes: outputting third warning information for the mobile object, the third warning information instructing to stop executing the illegal behavior in the corresponding target risk area.

7. A regional risk early warning device, characterized in that: include: A detection module is used to detect a moving object on multiple frames of images continuously acquired by the acquisition component, and obtain the first position information corresponding to the moving object in each frame of the target image containing the moving object; a processing module, configured to map the mobile object to a three-dimensional twin model of a substation based on the first posture information, and obtain second posture information of the mobile object in the three-dimensional twin model of the substation; determining, based on the second posture information, whether the mobile object has engaged in risky behavior within a corresponding target risk area, the target risk area being determined based on a power safety database; If there is any risky behavior, warning information corresponding to the risky behavior is output for the mobile object.

8. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 6 when the computer program is executed.

Images