Risk early warning method, device and electronic equipment for high-altitude operation

Abstract

The application discloses a high-altitude operation risk early warning method and device and electronic equipment, and relates to the field of artificial intelligence. The method comprises the following steps: acquiring a target image of a target user through a camera device, and identifying a target position of the target user in the target image; inputting the target image into a posture recognition model to obtain target posture information of the target user; predicting a target moving position of the target user according to the target position and the target posture information, determining operation risk early warning information of the target user according to the target moving position, and sending the operation risk early warning information to the target user, wherein the operation risk early warning information is used for informing the target user whether there is operation risk. Through the application, the problem that the safety and accuracy of the supervision of high-altitude operation personnel by the artificial supervision mode in the related art are low is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, in particular, to a risk early warning method and device for high-altitude operation and electronic equipment. BACKGROUND

[0002] In the field of power transmission, high-altitude operation is an indispensable part, which covers a series of key tasks such as line erection, equipment maintenance, fault troubleshooting, etc., and is crucial to the stable operation of the power grid. High-altitude operation personnel need to perform delicate operations in the air tens of meters or even hundreds of meters high, which not only tests their professional skills, but also puts high requirements on their safety. However, it is precisely such a working environment that makes high-altitude operation face many potential risks, such as high-altitude falling, electric shock injury, adverse weather influence, etc., each of which may pose a serious threat to the life safety of the operation personnel.

[0003] Traditionally, the safety supervision of high-altitude operation mainly relies on manual methods, but this mode of supervision has obvious defects. First, manual supervision is inefficient and difficult to achieve comprehensive coverage of the wide operation area and long operation process, resulting in that safety hazards are difficult to be discovered and handled in time. Secondly, the supervision effect is easily affected by subjective factors, such as the experience, concentration, judgment of the supervisor, etc., which may affect the accurate assessment and timely response to safety risks. In addition, in the face of emergency situations, the response speed of manual supervision is often not fast enough to provide timely and effective safety protection for high-altitude operation personnel.

[0004] In view of the problem that the safety and accuracy of the manual supervision method for high-altitude operation personnel in the related art are low, no effective solution has been proposed so far. SUMMARY

[0005] The present application provides a risk early warning method and device for high-altitude operation and electronic equipment to solve the problem that the safety and accuracy of the manual supervision method for high-altitude operation personnel in the related art are low.

[0006] According to one aspect of the present application, a risk early warning method for high-altitude operation is provided. The method comprises: acquiring a target image of a target user through a camera device, and identifying a target position of the target user in the target image; inputting the target image into a posture recognition model to obtain target posture information of the target user; predicting a target moving position of the target user according to the target position and the target posture information, and determining operation risk early warning information of the target user according to the target moving position, and sending the operation risk early warning information to the target user, wherein the operation risk early warning information is used to inform the target user whether there is an operation risk.

[0007] Optionally, the obtaining the target image picture of the target user by the camera device comprises: obtaining image pictures captured by a plurality of camera devices to obtain a plurality of initial image pictures, and screening the image pictures containing the target user from the plurality of initial image pictures to obtain a plurality of candidate image pictures; obtaining a resolution level and a size proportion of the target user in each candidate image picture, wherein the resolution level and the resolution size are in a proportional relationship; performing weighted summation on the resolution level and the size proportion of each candidate image picture according to a preset weight to obtain a score of each candidate image picture; and determining the candidate image picture with the highest score as the target image picture.

[0008] Optionally, the identifying the target position of the target user in the target image picture comprises: inputting the target image picture into a position recognition model to obtain the target position of the target user, wherein the target position is determined by boundary point coordinates of a person frame of the target user, and the position recognition model is obtained by training a first sample data set, wherein the first sample data set comprises a plurality of first sample data, each first sample data comprises a historical image picture, a person frame in the historical image picture, and boundary point coordinates of the person frame.

[0009] Optionally, the posture recognition model is obtained by training in the following manner: obtaining a plurality of second sample information, wherein each second sample information comprises a human body image and posture information corresponding to the human body image; and training an initial recognition model according to the plurality of second sample information to obtain the posture recognition model.

[0010] Optionally, the predicting the target moving position of the target user according to the target position and the target posture information comprises: determining a moving acceleration of the target user according to the target posture information; obtaining a plurality of historical moving positions of the target user, and calculating the plurality of historical moving positions, the target position, and the moving acceleration according to a Kalman filtering algorithm to obtain a predicted moving track of the target user; and determining the target moving position of the target user according to the predicted moving track.

[0011] Optionally, the determining the work risk warning information of the target user according to the target moving position comprises: determining whether the target moving position is in a preset area, wherein the preset area is a dangerous area; in the case that the target moving position is in the preset area, determining the work risk warning information as high-risk warning information; in the case that the target moving position is not in the preset area, determining whether the predicted moving track passes through the preset area; in the case that the predicted moving track passes through the preset area, determining the work risk warning information as medium-risk warning information; and in the case that the predicted moving track does not pass through the preset area, determining the work risk warning information as low-risk warning information.

[0012] Optionally, after the target image of the target user is obtained by the camera device, the method further comprises: inputting the target image into the equipment recognition model, determining whether the target user carries the target equipment through the equipment recognition model, and obtaining a recognition result, wherein the equipment recognition model is obtained by training a plurality of third sample data, each third sample data comprises at least one equipment and corresponding label information of each equipment; in a case where the recognition result represents that the target user does not carry the target equipment, determining the work risk warning information as high-risk warning information; in a case where the recognition result represents that the target user carries the target equipment, determining the work risk warning information as low-risk warning information.

[0013] According to another aspect of the present application, a risk warning device for high-altitude work is provided. The device comprises: a first identification unit configured to obtain a target image of a target user through a camera device and identify a target position of the target user in the target image; a second identification unit configured to input the target image into a posture recognition model and obtain target posture information of the target user; a first determination unit configured to predict a target moving position of the target user according to the target position and the target posture information, determine work risk warning information of the target user according to the target moving position, and send the work risk warning information to the target user, wherein the work risk warning information is used to inform the target user whether there is a work risk.

[0014] According to another aspect of the present application, a computer program product is also provided, comprising a computer program, which, when executed by a processor, implements a risk warning method for high-altitude work.

[0015] According to another aspect of the present application, an electronic device is also provided, comprising one or more processors and a memory; the memory stores computer readable instructions, and the processor is configured to run the computer readable instructions, wherein the computer readable instructions perform a risk warning method for high-altitude work when running.

[0016] By the present application, the following steps are adopted: obtaining a target image picture of a target user by a camera device, and identifying a target position of the target user in the target image picture; inputting the target image picture into a posture recognition model to obtain target posture information of the target user; predicting a target moving position of the target user according to the target position and the target posture information, and determining work risk warning information of the target user according to the target moving position, and sending the work risk warning information to the target user, wherein the work risk warning information is used to inform the target user whether there is a work risk. The safety and accuracy of the way of monitoring the high-altitude worker by manual supervision in the related art are low. By identifying the target position of the worker in the image picture, and identifying the target posture of the worker, the moving position of the worker is predicted according to the target position and the target posture, so as to determine the position change of the worker in the future period of time, and according to the predicted moving position, it is judged whether the high-altitude work has a risk, thereby the effect of improving the accuracy and timeliness of risk monitoring is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0017] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application, and the illustrative embodiments of the present application and their description serve the purpose of explaining the present application. The accompanying drawings should not be construed in a limiting manner as to the present application.

[0018] Figure 1 is a flowchart of a work risk warning method of high-altitude work provided according to an embodiment of the present application;

[0019] Figure 2 is a flowchart of a picture selection method provided according to an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of a work risk warning device of high-altitude work provided according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] It should be noted that the embodiments and the features in the embodiments in the present application can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] In order to enable personnel in the technical field to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0025] It should be noted that the high-altitude operation risk early warning method, device and electronic equipment determined by the present disclosure can be used in the field of artificial intelligence, and can also be used in any field other than the field of artificial intelligence. The application field of the high-altitude operation risk early warning method, device and electronic equipment determined by the present disclosure is not limited.

[0026] It should be noted that the information, user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) collected in the present application are all information and data authorized by the user or authorized by all parties. The collection, storage, use, processing, transmission, provision, disclosure and application of related data comply with relevant laws, regulations and standards in relevant regions, take necessary security measures, do not violate public order and good customs, and provide corresponding operation portal for users to choose authorized use or refuse to use. For example, the system and related users or institutions are provided with an interface. Before obtaining the relevant information, the interface needs to send a request to the aforementioned user or institution, and after receiving the consent information feedback from the aforementioned user or institution, the relevant information is obtained.

[0027] For the convenience of description, some nouns or terms related to the embodiments of the present application are described as follows:

[0028] YOLOv5: YOLOv5 (You Only Look Once) is a deep learning model for object detection.

[0029] OpenPose network architecture: a framework for real-time human keypoint detection and full-body pose estimation based on deep learning.

[0030] Heatmap: A heat map used to predict the location of key points on the human body. Each key point has a corresponding peak area on the heatmap. By finding the peak points on the heatmap, the specific location of the key point can be determined.

[0031] Kalman filtering is a highly efficient recursive filter (autoregressive filter) that can estimate the state of a dynamic system from a series of measurements containing statistical noise.

[0032] According to an embodiment of this application, a risk warning method for high-altitude operations is provided.

[0033] Figure 1 This is a flowchart of a risk warning method for high-altitude operations provided according to an embodiment of this application. For example... Figure 1 As shown, the method includes the following steps:

[0034] Step S101: Acquire a target image of the target user using a camera device, and identify the target location of the target user in the target image.

[0035] Specifically, the operators can be those performing high-altitude operations. First, they can use cameras, webcams, or other imaging devices to capture the scene and obtain real-time image images that include the target user. For example, a high-resolution camera can be used to capture video images of the high-altitude operation scene to obtain target image images that include the target user.

[0036] Furthermore, after acquiring the target image, the content of the image can be analyzed using a pre-trained machine learning model or deep learning algorithm to identify the target user. After identifying the target user in the image, it is also necessary to determine the target user's location within the target image. This can be achieved by drawing a bounding box around the target user's body in the target image. The coordinates of the bounding box are typically represented by four values: the x1 and y1 coordinates of the upper left corner and the x2 and y2 coordinates of the lower right corner. These two coordinates define the specific range of the target user's location in the image. Using the coordinates of the target location not only accurately determines the target user's position but also provides precise location information for subsequent analyses (such as behavior analysis and path tracking).

[0037] Step S102: Input the target image into the pose recognition model to obtain the target user's pose information.

[0038] Specifically, after obtaining the target image picture containing the target user, the target image picture can also be input into the posture recognition model, so as to analyze the body posture, joint position and other information of the target user in the target image picture through the posture recognition model, and further obtain the posture information of the target user.

[0039] The posture recognition model is usually constructed based on deep learning technology, and the video image is input into the deep learning model to obtain the posture recognition result. The posture recognition model is trained by a large number of image training sets labeled with human joint position and other related human posture information, so as to learn how to recognize the posture feature information of the human body in the image, and recognize the posture information according to the feature information. The posture recognition model uses multiple convolution layers to extract image features, uses a pooling layer to reduce the size of the feature map and keep important information, and uses a fully connected layer for classification and regression, so as to obtain the position, category and posture information of the human target.

[0040] For example, the posture recognition model can use the OpenPose network architecture and combine human key point detection and posture estimation algorithm to realize accurate posture recognition. The image needs to be input into the OpenPose network, the convolution layer and the pooling layer are used to extract the image features, the heatmap regression is used to predict the position of the human key point, the key point connection algorithm is used to connect the detected human key points into a skeleton, and the posture estimation is performed, so as to complete the operation process of the posture estimation.

[0041] In step S103, the target moving position of the target user is predicted according to the target position and the target posture information, the work risk warning information of the target user is determined according to the target moving position, and the work risk warning information is sent to the target user. The work risk warning information is used to inform the target user whether there is a work risk.

[0042] Specifically, after obtaining the target position and the target posture information of the target user, the possible future moving position of the target user can be inferred according to the target position and the target posture information, so as to determine whether there is a risk in the high-altitude work.

[0043] Wherein, when predicting the moving position, the prediction algorithm can be used to predict the moving path and the possible target moving position of the target user based on the current position and posture of the user. After predicting the target moving position of the target user, the work risk that the target user may face can be evaluated according to this position information, and the corresponding warning information can be generated.

[0044] Further, when evaluating the work risk that the target user may face, the predicted moving position can be compared with predefined risk areas or safety specifications, which can include dangerous work areas, prohibited entry areas, high-risk operation areas, etc. By comparison, the risk that the target user may encounter in future actions can be evaluated, and work risk warning information can be determined, so that the target user can avoid risks in time according to the work risk warning information, thereby achieving the technical effect of improving the safety of high-altitude work. The work risk warning information can be information, light indication, alarm bell sound, etc., which is not limited here.

[0045] The risk warning method for high-altitude work provided by the embodiments of the present application obtains a target image picture of a target user through a camera device, and identifies a target position of the target user in the target image picture; inputs the target image picture into a posture recognition model to obtain target posture information of the target user; predicts a target moving position of the target user according to the target position and the target posture information, and determines work risk warning information of the target user according to the target moving position, and sends the work risk warning information to the target user, wherein the work risk warning information is used to inform the target user whether there is a work risk. The safety and accuracy of the related art method of monitoring the high-altitude work personnel through artificial supervision are low. By identifying the target position of the work personnel in the image picture and identifying the target posture of the work personnel, the moving position of the work personnel is predicted according to the target position and the target posture, so as to determine the position change of the work personnel in a future period of time, and it is judged whether there is a risk in high-altitude work according to the predicted moving position, thereby achieving the effect of improving the accuracy and timeliness of risk monitoring.

[0046] Optionally, Figure 2 The flowchart of the picture selection method provided by the embodiments of the present application is shown in FIG. 1, and the risk warning method for high-altitude work provided by the embodiments of the present application is shown in FIG. 2. Figure 2 As shown in FIG. 2, in the risk warning method for high-altitude work provided by the embodiments of the present application, the target image picture of the target user is obtained through the camera device, including:

[0047] In step S201, a plurality of image pictures shot by a plurality of camera devices are obtained to obtain a plurality of initial image pictures, and image pictures containing a target user are screened from the plurality of initial image pictures to obtain a plurality of candidate image pictures.

[0048] It should be noted that in order to ensure that the model can accurately identify the target position of the target user in the target image picture, it is necessary to ensure the accuracy and clarity of the image content in the target image picture. Therefore, the initial image pictures can be collected from the plurality of camera devices capable of shooting the target user, and the initial image pictures are analyzed and screened to obtain the target image picture.

[0049] Specifically, firstly, it is judged whether the initial image pictures contain the image of the target user. Image recognition technology can be used to find and filter out the pictures containing the target user from the initial image pictures, so as to filter out the initial image pictures in which the target user is too small or there is no target user, and ensure that the image of the target user in the obtained candidate image pictures can be clearly viewed.

[0050] In step S202, the resolution level and the size proportion of the target user in each candidate image picture are obtained, wherein the resolution level and the resolution size are in a proportional relationship.

[0051] It should be noted that after the plurality of candidate image pictures are filtered out, the resolution level and the size proportion of the target user in each candidate image picture are obtained, so as to select the target image picture from the plurality of candidate image pictures according to the resolution level and the size proportion of the target user. The resolution level can be determined according to a resolution level table. The resolution level table can contain a plurality of resolution intervals. Different resolution intervals correspond to different resolution levels. The higher the resolution, the higher the level.

[0052] Specifically, the resolution of each candidate image picture can be obtained, and the resolution level of each candidate image picture can be determined according to the resolution level table. The approximate outline size of the target user is obtained, and the size proportion of the target user in the candidate image picture is determined, for example, the target user occupies 5% of the area in the candidate image picture, so as to obtain the size proportion of the target user in the candidate image picture.

[0053] In step S203, the resolution level and the size proportion of each candidate image picture are weighted and summed according to a preset weight, to obtain the score of each candidate image picture.

[0054] Further, after the resolution level and the size proportion of each candidate image picture are obtained, the resolution level and the size proportion of each candidate image picture can be weighted and summed according to a preset weight, to obtain the score of each candidate image picture, and then the target image picture can be selected from the candidate image pictures according to the score.

[0055] In step S204, the candidate image picture with the highest score is determined as the target image picture.

[0056] Specifically, after the score of each candidate image picture is obtained, the image with the highest score can be determined as the target image picture, and then the target image picture with the largest and clearest image of the target user can be obtained.

[0057] The embodiment improves the accuracy of recognizing the posture of the target user according to the image information by filtering the images taken by the plurality of camera devices.

[0058] Optionally, in the high-altitude operation risk early warning method provided in the embodiments of the present application, the target position of the target user in the target image picture is identified by inputting the target image picture into a position recognition model to obtain the target position of the target user, wherein the target position is determined by the boundary point coordinates of the person frame of the target user, and the position recognition model is obtained by training a first sample data set, wherein the first sample data set includes a plurality of first sample data, each first sample data includes a historical image picture, a person frame in the historical image picture, and boundary point coordinates of the person frame.

[0059] Specifically, when determining the target position of the target user, the target image picture can be input into a pre-trained position recognition model, and the task of the position recognition model is to identify the specific position of the target user from the input image picture, wherein the target position refers to the specific position of the target user in the image picture, and this position is determined by the boundary point coordinates of the person frame (usually a rectangular frame used to mark and frame the person) of the target user. The boundary point coordinates define the specific position and size of the person frame in the picture, thereby indirectly indicating the position of the target user.

[0060] It should be noted that the position recognition model is obtained by training a large amount of historical data (i.e., a first sample data set), and the position recognition model can automatically extract features from the image picture and predict or identify the position of the target user in the picture based on these features. The first sample data set is composed of historical image pictures, and at least one person frame is marked in each historical image picture, and the boundary point coordinates of each person frame are given. These boundary point coordinates accurately define the position and size of the person frame in the picture, i.e., the position of the target user, so that the first sample data set can be used to train the model to obtain the position recognition model.

[0061] It should be noted that the position recognition model can use the network architecture of YOLOv5, input the image into the position recognition model, use multiple convolution layers to extract image features, use a pooling layer to reduce the size of the feature map while preserving important information, and use a fully connected layer for classification and regression to obtain the position information of the human target.

[0062] For example, the target position can be represented by the upper left corner coordinates and the lower right corner coordinates of the person frame, and the model output result can be: target position: boundary frame upper left corner coordinates (x1=200, y1=300), boundary frame lower right corner coordinates (x2=250, y2=350).

[0063] The embodiments achieve the technical effect of accurately identifying the target position of the target user by training the position recognition model.

[0064] To improve the accuracy of posture recognition, optionally, in the high-altitude operation risk early warning method provided in the embodiments of the present application, the posture recognition model is obtained by training in the following manner: a plurality of second sample information is obtained, wherein each second sample information includes a human body image and posture information corresponding to the human body image; and the initial recognition model is trained according to the plurality of second sample information to obtain the posture recognition model.

[0065] Specifically, in order to ensure that the posture information of the target user is accurately obtained, the posture recognition model needs to be trained. First, a large amount of second sample information needs to be collected as training data, and each second sample information includes two parts: a human body image and posture information corresponding to the image. The human body image can be a real human body photo or video frame from different scenes, different angles, and different postures. The posture information is a detailed description of the human body posture in the image, usually including the positions of the human body key points (such as the head, shoulders, elbows, knees, etc.) and the relative relationships between these key points, as well as the name of the posture.

[0066] Further, after collecting a plurality of second sample information, the initial recognition model can be trained using the collected plurality of second sample information to obtain a posture recognition model that can accurately recognize human body postures. During the training process, the model learns how to extract useful features from the human body image and establishes a mapping relationship between these features and the human body posture information.

[0067] It should be noted that in order to achieve accurate posture recognition, OpenPose can be selected as the network architecture of the model. OpenPose is a deep learning-based pose estimation method that combines human key point detection and pose estimation algorithms to extract the positions of human key points from images or videos in real time and infer the posture of the human body based on the positions. When a video image is input into the OpenPose network, the image is first feature-extracted using convolutional layers and pooling layers. The convolutional layer slides over the image using a convolution kernel and calculates the convolution operation to extract local features in the image. The pooling layer down-samples the feature map output by the convolutional layer to reduce the amount of calculation and extract higher-level features. Further, the positions of the human key points are predicted using a heatmap regression method, and then the key point connection algorithm is used to connect these key points into a skeleton to form the skeleton structure of the human body, so as to estimate the posture based on the skeleton structure and infer the overall posture of the human body.

[0068] In order to ensure the accuracy of the risk prediction, optionally, in the high-altitude operation risk early warning method provided in the embodiments of the present application, the target movement position of the target user is predicted according to the target position and the target posture information, which comprises: determining the movement acceleration of the target user according to the target posture information; obtaining a plurality of historical movement positions of the target user, and calculating the plurality of historical movement positions, the target position and the movement acceleration according to the Kalman filtering algorithm to obtain the predicted movement trajectory of the target user; and determining the target movement position of the target user according to the predicted movement trajectory.

[0069] Specifically, after obtaining the target posture information, the target posture information can be analyzed to obtain the movement acceleration of the target user. In this step, the target posture information needs to be analyzed to obtain posture information such as speed, direction, inclination angle and other parameters. These parameters can directly or indirectly reflect the motion state of the user, so that by analyzing the posture information, the current movement acceleration of the target user can be calculated, and then the predicted movement trajectory of the target user can be predicted according to the movement acceleration.

[0070] For example, the posture information can be recognized by a posture parameter recognition model, that is, the skeleton information, key point information and current posture and other related contents in the target posture information are input into the posture parameter recognition model, and the posture information is obtained through the posture parameter recognition model.

[0071] Further, in order to more accurately predict the future movement trajectory of the target user, a plurality of historical movement positions of the target user in the past period of time also need to be obtained, so that the plurality of historical movement positions, the target position and the movement acceleration can be calculated according to the Kalman filtering algorithm to obtain the predicted movement trajectory of the target user. The Kalman filtering algorithm predicts the next position based on the plurality of historical movement positions and the movement acceleration, and then updates the plurality of historical movement positions and the movement acceleration in combination with the new position, so that the predicted movement trajectory of the target user is obtained by continuously iterating the above prediction process.

[0072] Further, after the above iterative calculation is completed, a predicted movement trajectory of the target user based on historical data, current state and future trend prediction can be obtained. The predicted movement trajectory not only contains every position point that the user may pass through, but also reflects the speed and direction change of the user movement, so that the target movement position of the target user at a future time point can be further determined according to the predicted movement trajectory, and then whether the high-altitude operation performed by the target user has an operation risk can be determined according to the target movement position.

[0073] Optionally, in the high-altitude operation risk warning method provided in the embodiments of the present application, the operation risk warning information of the target user is determined according to the target moving position, including: judging whether the target moving position is in a preset area, wherein the preset area is a dangerous area; in the case that the target moving position is in the preset area, determining the operation risk warning information as high-risk warning information; in the case that the target moving position is not in the preset area, judging whether the predicted moving track passes through the preset area; in the case that the predicted moving track passes through the preset area, determining the operation risk warning information as medium-risk warning information; in the case that the predicted moving track does not pass through the preset area, determining the operation risk warning information as low-risk warning information.

[0074] Specifically, after obtaining the target moving position of the target user, it can be determined whether the target moving position is in a preset area, wherein the preset area can be a preset dangerous area, and the user can be determined to have a high operation risk when located in the dangerous area.

[0075] Further, in the case that the target moving position is in the preset area, the operation risk warning information is determined as high-risk warning information, and the target user is timely notified to suspend operation and adjust the moving track, and feedback is performed through a red light at the target user end. In the case that the target moving position is not in the preset area, it is necessary to determine whether the target user will pass through the preset area according to the predicted moving track, and in the case that the predicted moving track passes through the preset area, the operation risk warning information is determined as medium-risk warning information, and the target user is timely notified to pay attention to the change of the moving track, and feedback is performed through a yellow light at the target user end. In the case that the predicted moving track does not pass through the preset area, it is represented that the target user will not pass through the preset area during operation, and at this time, the operation risk warning information can be determined as low-risk warning information, and feedback is performed through a green light at the target user end, so that the target user can continue operation, thereby achieving the technical effects of ensuring operation safety and timeliness of risk prompt.

[0076] In order to improve the accuracy of risk identification and improve the safety of the target user's work, optionally, in the high-altitude work risk early warning method provided by the embodiment of the application, after the target image picture of the target user is obtained through the camera equipment, the method further comprises: inputting the target image picture into the equipment identification model, determining whether the target user carries the target equipment through the equipment identification model, and obtaining an identification result, wherein the equipment identification model is obtained by training a plurality of third sample data, each third sample data comprises at least one equipment and corresponding label information of each equipment; in the case that the identification result represents that the target user does not carry the target wearing device, the work risk early warning information is determined as high-risk early warning information; in the case that the identification result represents that the target user carries the target wearing device, the work risk early warning information is determined as low-risk early warning information.

[0077] It should be noted that when judging the high-altitude work risk, it is also necessary to determine whether the target user correctly wears the safety wearing device, such as a safety belt / rope, a safety helmet, and an insulating garment, so as to determine whether the work is at risk according to the equipment condition of the safety wearing device.

[0078] Specifically, whether the target user carries the target equipment can be determined through the equipment identification model, wherein the equipment identification model can be based on the network architecture of YOLOv5, and features of the safety protection device are extracted and classified, the target image picture is input into the YOLOv5 network, image features are extracted using convolutional layers and pooling layers, multi-scale feature fusion is performed in combination with an attention mechanism and a feature pyramid network, and detection and identification are performed using a fully connected layer to obtain an identification result of the safety protection device, so as to determine whether the target user carries the target equipment according to the identification result.

[0079] Further, in the case that the identification result represents that the target user does not carry the target wearing device, the work risk early warning information is determined as high-risk early warning information, that is, in the case that the target user does not carry the target wearing device, the work needs to be stopped immediately, therefore, the work risk early warning information needs to be determined as high-risk early warning information; in the case that the identification result represents that the target user carries the target wearing device, it represents that the wearing device is normally worn, at this time, the work risk early warning information is determined as low-risk early warning information, and the work personnel can normally perform high-altitude work, thereby achieving the technical effect of accurately identifying the risk of high-altitude work.

[0080] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0081] The embodiment of the present application further provides a high-altitude operation risk early warning device. It should be noted that the high-altitude operation risk early warning device of the embodiment of the present application can be used to execute the high-altitude operation risk early warning method provided by the embodiment of the present application. The high-altitude operation risk early warning device provided by the embodiment of the present application is introduced as follows.

[0082] Figure 3 is a schematic diagram of the high-altitude operation risk early warning device provided by the embodiment of the present application. As shown in the figure, Figure 3 the device comprises a first identification unit 31, a second identification unit 32, and a first determination unit 33.

[0083] The first identification unit 31 is configured to acquire a target image of a target user through a camera device and identify a target position of the target user in the target image.

[0084] The second identification unit 32 is configured to input the target image into a posture recognition model to obtain target posture information of the target user.

[0085] The first determination unit 33 is configured to predict a target moving position of the target user according to the target position and the target posture information, determine operation risk early warning information of the target user according to the target moving position, and send the operation risk early warning information to the target user, wherein the operation risk early warning information is used to inform the target user whether there is an operation risk.

[0086] The high-altitude operation risk early warning device provided by the embodiment of the present application is configured to acquire a target image of a target user through a camera device by the first identification unit 31 and identify a target position of the target user in the target image; input the target image into a posture recognition model by the second identification unit 32 to obtain target posture information of the target user; and predict a target moving position of the target user according to the target position and the target posture information by the first determination unit 33, determine operation risk early warning information of the target user according to the target moving position, and send the operation risk early warning information to the target user, wherein the operation risk early warning information is used to inform the target user whether there is an operation risk. The safety and accuracy of the high-altitude operation personnel supervised by the manual supervision method in the related art are low. The target position of the operation personnel in the image is identified, the target posture of the operation personnel is identified, the moving position of the operation personnel is predicted according to the target position and the target posture, the position change of the operation personnel in a future period of time is determined, and it is judged whether there is a risk in the high-altitude operation according to the predicted moving position, thereby achieving the effect of improving the accuracy and timeliness of risk monitoring.

[0087] Optionally, in the high-altitude operation risk early warning device provided by the embodiment of the application, the first identification unit 31 comprises: a first acquisition module, configured to acquire image pictures captured by a plurality of camera devices, to obtain a plurality of initial image pictures, and to screen image pictures containing the target user from the plurality of initial image pictures to obtain a plurality of candidate image pictures; a second acquisition module, configured to acquire a resolution level of each candidate image picture and a size proportion of the target user in the candidate image picture, wherein the resolution level and the resolution size are in a proportional relationship; a first calculation module, configured to weight and sum the resolution level and the size proportion of each candidate image picture according to a preset weight to obtain a score of each candidate image picture; and a first determination module, configured to determine the candidate image picture with the highest score as the target image picture.

[0088] Optionally, in the high-altitude operation risk early warning device provided by the embodiment of the application, the first identification unit 31 comprises: an identification module, configured to input the target image picture into a position identification model to obtain a target position of the target user, wherein the target position is determined by boundary point coordinates of a person frame of the target user, and the position identification model is obtained by training a first sample data set, wherein the first sample data set comprises a plurality of first sample data, each first sample data comprises a historical image picture, a person frame in the historical image picture, and boundary point coordinates of the person frame.

[0089] Optionally, in the high-altitude operation risk early warning device provided by the embodiment of the application, the posture identification model is obtained by training in the following manner: an acquisition unit, configured to acquire a plurality of second sample information, wherein each second sample information comprises a human body image and posture information corresponding to the human body image; and a training unit, configured to train an initial identification model according to the plurality of second sample information to obtain the posture identification model.

[0090] Optionally, in the high-altitude operation risk early warning device provided by the embodiment of the application, the first determination unit 33 comprises: a second determination module, configured to determine a moving acceleration of the target user according to the target posture information; a second calculation module, configured to acquire a plurality of historical moving positions of the target user, and to calculate the plurality of historical moving positions, the target position, and the moving acceleration according to a Kalman filtering algorithm to obtain a predicted moving track of the target user; and a third determination module, configured to determine a target moving position of the target user according to the predicted moving track.

[0091] Optionally, in the high-altitude operation risk warning device provided by the embodiment of the application, the first determination unit 33 comprises: a first judgment module, configured to judge whether the target moving position is in a preset area, wherein the preset area is a dangerous area; a fourth determination module, configured to determine the operation risk warning information as high-risk warning information in the case that the target moving position is in the preset area; a second judgment module, configured to judge whether the predicted moving track passes through the preset area in the case that the target moving position is not in the preset area; a fifth determination module, configured to determine the operation risk warning information as medium-risk warning information in the case that the predicted moving track passes through the preset area; and a sixth determination module, configured to determine the operation risk warning information as low-risk warning information in the case that the predicted moving track does not pass through the preset area.

[0092] Optionally, in the high-altitude operation risk warning device provided by the embodiment of the application, after the target image picture of the target user is acquired by the camera device, the device further comprises: a third identification unit, configured to input the target image picture into an equipment identification model to determine whether the target user carries target equipment through the equipment identification model, and obtain an identification result, wherein the equipment identification model is obtained by training a plurality of third sample data, each third sample data comprises at least one equipment and corresponding label information of each equipment; a second determination unit, configured to determine the operation risk warning information as high-risk warning information in the case that the identification result represents that the target user does not carry the target equipment; and a third determination unit, configured to determine the operation risk warning information as low-risk warning information in the case that the identification result represents that the target user carries the target equipment.

[0093] The high-altitude operation risk warning device comprises a processor and a memory, the first identification unit 31, the second identification unit 32, the first determination unit 33 and the like are stored in the memory as program units, and the corresponding functions are realized by the processor executing the program units stored in the memory.

[0094] The processor comprises a core, and the core calls the corresponding program units from the memory. The core can be one or more, and the safety and accuracy of the high-altitude operation personnel supervised by the artificial supervision in the related art are improved by adjusting the core parameters.

[0095] The memory can include a non-persistent memory in a computer readable medium, a random access memory (RAM) and / or a non-volatile memory such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory comprises at least one memory chip.

[0096] The embodiment of the application provides a computer readable storage medium, which stores a program, and the program is executed by a processor to realize the high-altitude operation risk warning method.

[0097] The embodiment of the present application provides a processor used for running a program, wherein the processor executes the risk early warning method of the aerial work when the program is running.

[0098] Figure 4 is a schematic diagram of an electronic device provided by the embodiment of the present application, as Figure 4 shown, the embodiment of the present application provides an electronic device, the electronic device 40 comprises a processor, a memory, and a program stored in the memory and capable of running on the processor, and the processor implements the steps of the risk early warning method of the aerial work when executing the program. The device in the present application can be a server, a PC, a PAD, a mobile phone, and the like.

[0099] The present application further provides a computer program product, when executed on a data processing device, is suitable for executing the program of the steps of the risk early warning method of the aerial work.

[0100] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of being implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0101] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device implemented in the flowcharts and / or block diagrams. Figure 1 The function of one flow or multiple flows and / or blocks Figure 1 The function of one block or multiple blocks.

[0102] These computer program instructions can also be stored in a computer readable storage medium capable of guiding the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The function of one flow or multiple flows and / or blocks Figure 1 The function of one block or multiple blocks.

[0103] These computer program instructions can also be loaded into computer or other programmable data processing devices to cause a series of operational steps to be performed on the computer or other programmable devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable devices provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 Figure 1

[0104] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0105] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) about which the processor can execute instructions. The memory can also include non-volatile memory, such as read only memory (ROM), electrically programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, or other memory technologies, CD-ROM, digital versatile disc (DVD), or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information for access by a computing device. In no case does the medium include a transitory signal.

[0106] Computer readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology for storage of information such as computer readable instructions, data structures, program modules 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 programmable read only memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information that can be accessed by a computing device. According to the definition herein, computer readable media does not include transitory media such as modulated data signals and carrier waves.

[0107] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to encompass a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not include only those elements recited, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without further limitation, an element preceded by "comprises a" does not, without more constraints, preclude the existence of additional identical elements in the process, method, article or apparatus that includes the element.

[0108] ​​The above merely provides an example of the present application, and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall fall into the scope of claims of the present application.

Claims

1. A risk early warning method for aerial work, characterized in that, The method comprises the following steps: obtaining a target image of a target user through a camera device, and identifying a target position of the target user in the target image; inputting the target image into a posture recognition model to obtain target posture information of the target user; predicting a target moving position of the target user according to the target position and the target posture information, and determining work risk warning information of the target user according to the target moving position, and sending the work risk warning information to the target user, wherein the work risk warning information is used to inform the target user whether there is a work risk; obtaining a target image of a target user through a camera device comprises the following steps: obtaining image pictures taken by a plurality of camera devices to obtain a plurality of initial image pictures, and screening image pictures containing the target user in the plurality of initial image pictures to obtain a plurality of candidate image pictures; obtaining a resolution level of each candidate image picture and a size proportion of the target user in the candidate image picture, wherein the resolution level and the resolution size are in a proportional relationship; according to a preset weight, the resolution level and the size proportion of each candidate image picture are weighted and summed to obtain a score of each candidate image picture; the candidate image picture with the highest score is determined as the target image picture; predicting a target moving position of the target user according to the target position and the target posture information comprises the following steps: determining a moving acceleration of the target user according to the target posture information; obtaining a plurality of historical moving positions of the target user, and calculating the plurality of historical moving positions, the target position and the moving acceleration according to a Kalman filtering algorithm to obtain a predicted moving track of the target user; determining a target moving position of the target user according to the predicted moving track; determining work risk warning information of the target user according to the target moving position comprises the following steps: judging whether the target moving position is in a preset area, wherein the preset area is a dangerous area; in the case that the target moving position is in the preset area, the work risk warning information is determined as high-risk warning information; in the case that the target moving position is not in the preset area, judging whether the predicted moving track passes through the preset area; in the case that the predicted moving track passes through the preset area, the work risk warning information is determined as medium-risk warning information; in the case that the predicted moving track does not pass through the preset area, the work risk warning information is determined as low-risk warning information; After obtaining the target image picture of the target user through the camera device, the method further comprises: inputting the target image picture into an equipment identification model, determining whether the target user carries target equipment through the equipment identification model, and obtaining an identification result, wherein the equipment identification model is obtained by training a plurality of third sample data, each third sample data comprises at least one equipment and corresponding label information of each equipment; in the case that the identification result represents that the target user does not carry the target equipment, determining the work risk warning information as high-risk warning information; in the case that the identification result represents that the target user carries the target equipment, determining the work risk warning information as low-risk warning information.

2. The method of claim 1, wherein, The target position of the target user in the target image picture is identified by: inputting the target image picture into a position identification model to obtain the target position of the target user, wherein the target position is determined by the boundary point coordinates of the person frame of the target user, and the position identification model is obtained by training a first sample data set, wherein the first sample data set comprises a plurality of first sample data, each first sample data comprises a historical image picture and a person frame in the historical image picture and boundary point coordinates of the person frame.

3. The method of claim 1, wherein, The posture identification model is obtained by training in the following way: obtain a plurality of second sample information, wherein each second sample information comprises a human body image and corresponding posture information of the human body image; train an initial identification model according to the plurality of second sample information to obtain the posture identification model.

4. A risk warning device for aerial work, characterized in that, Comprise: The first identification unit is configured to obtain a target image picture of a target user through a camera device and identify a target position of the target user in the target image picture. The second identification unit is configured to input the target image picture into a posture identification model to obtain target posture information of the target user. The first determination unit is configured to predict a target moving position of the target user according to the target position and the target posture information, determine work risk warning information of the target user according to the target moving position, and send the work risk warning information to the target user, wherein the work risk warning information is used to inform the target user whether there is a work risk. The first identification unit comprises: a first acquisition module configured to obtain a plurality of image pictures captured by a plurality of camera devices to obtain a plurality of initial image pictures, and select image pictures containing a target user from the plurality of initial image pictures to obtain a plurality of candidate image pictures; a second acquisition module configured to obtain a resolution level and a size ratio of the target user in each candidate image picture, wherein the resolution level and the resolution size are in a proportional relationship; a first calculation module configured to weight and sum the resolution level and the size ratio of each candidate image picture according to a preset weight to obtain a score of each candidate image picture; and a first determination module configured to determine the candidate image picture with the highest score as the target image picture. The first determination unit comprises: a second determination module configured to determine the moving acceleration of the target user according to the target attitude information; a second calculation module configured to obtain a plurality of historical moving positions of the target user, and calculate the plurality of historical moving positions, the target position and the moving acceleration according to a Kalman filtering algorithm to obtain a predicted moving track of the target user; and a third determination module configured to determine the target moving position of the target user according to the predicted moving track. The first determination unit comprises: a first judgment module configured to judge whether the target moving position is in a preset area, wherein the preset area is a dangerous area; a fourth determination module configured to determine the work risk warning information as high-risk warning information in the case that the target moving position is in the preset area; a second judgment module configured to judge whether the predicted moving track passes through the preset area in the case that the target moving position is not in the preset area; a fifth determination module configured to determine the work risk warning information as medium-risk warning information in the case that the predicted moving track passes through the preset area; and a sixth determination module configured to determine the work risk warning information as low-risk warning information in the case that the predicted moving track does not pass through the preset area. The device further comprises: a third identification unit configured to input the target image picture into an equipment identification model to determine whether the target user carries the target equipment through the equipment identification model to obtain an identification result, wherein the equipment identification model is obtained by training a plurality of third sample data, each third sample data comprises at least one equipment and corresponding label information of each equipment; a second determination unit configured to determine the work risk warning information as high-risk warning information in the case that the identification result represents that the target user does not carry the target equipment; and a third determination unit configured to determine the work risk warning information as low-risk warning information in the case that the identification result represents that the target user carries the target equipment.

5. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the risk warning method of high-altitude work of any one of claims 1 to 3.

6. An electronic device, comprising: The device comprises one or more processors and a memory, and the memory is configured to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the risk warning method of high-altitude work of any one of claims 1 to 3.

Images