Electric power safety area border crossing early warning method, system, equipment and medium

Abstract

The invention discloses an electric power safety area border crossing early warning method, system and device and a medium, and particularly relates to the technical field of safety early warning, and the technical key points are that the method comprises the steps: continuously obtaining the foot position information of a plurality of pedestrians, and constructing the walking track of the pedestrians according to the foot position information of the plurality of pedestrians; judging the continuity of the walking track of the pedestrian, and when the walking track of the pedestrian is continuous, extracting the foot position information of a plurality of pedestrians in the walking track of the pedestrian and performing position judgment on the foot position information and a preset safe area boundary so as to obtain a first position judgment result; when the pedestrian traveling track is discontinuous, predicting and complementing the missing pedestrian traveling track by using a recessive semi-Markov model, and extracting position information of a plurality of pedestrian feet in the complemented pedestrian traveling track and performing position judgment on the position information and a preset safe area boundary to obtain a second position judgment result; and according to the first position determination result or the second position determination result, outputting a boundary-crossing early warning signal to carry out boundary-crossing early warning on the pedestrian.

Description

Technical Field

[0001] The present invention relates to the technical field of safety early warning, and in particular to a method, system, equipment and medium for early warning of crossing a power safety zone. Background Art

[0002] At power production sites, since some high-voltage equipment is dangerous, different power safety zone boundaries are set up according to the degree of danger of the equipment to warn workers if they cross the boundary. Once they cross the boundary, it may bring safety hazards and reduce production efficiency.

[0003] To address this problem, current technical solutions mainly rely on monitoring systems based on image processing to monitor the behavior of staff in real time. When a staff member is detected to have crossed a preset safety boundary, the system will sound an alarm. However, in actual monitoring, the video monitoring system may encounter problems such as video interruption, occlusion, and transmission delay. These problems may cause some trajectory data to be lost, thereby affecting the complete monitoring of personnel behavior. If the data is missing or interrupted, it may not be possible to accurately determine whether a person has crossed the set safety area. At the same time, when a cross-border incident or other security incident occurs, it is necessary to trace the complete trajectory of the person to analyze the course of the incident. Missing data will result in incomplete event tracing, affecting the comprehensive analysis of the incident. Moreover, traditional video monitoring systems may have monitoring blind spots when faced with occlusion and interruption.

[0004] Therefore, the present invention aims to provide a method, system, device and medium for early warning of crossing the boundary of a power safety zone to solve the above-mentioned related problems. Summary of the invention

[0005] The technical problem to be solved by the present invention is that the pedestrian behavior trajectory is incomplete due to the incompleteness of the video image in the prior art, and it is difficult to ensure the accuracy of the cross-border warning. The purpose is to provide a method, system, equipment and medium for cross-border warning in a power safety area. First, the continuity of the pedestrian's trajectory is judged to screen out the discontinuous pedestrian trajectory; then the hidden semi-Markov model is used to predict and complete the missing pedestrian trajectory in the discontinuous pedestrian trajectory; and then the cross-border judgment is made on the completed pedestrian trajectory, thereby solving the related problems in the prior art that the pedestrian behavior trajectory is incomplete due to the incompleteness of the video image, and it is difficult to ensure the accuracy of the cross-border warning.

[0006] The present invention is achieved through the following technical solutions:

[0007] A method for early warning of crossing a power safety zone, the method comprising:

[0008] Continuously obtain the position information of multiple pedestrians' feet, and construct the pedestrian's moving trajectory based on the multiple pedestrians' foot position information;

[0009] The continuity of the pedestrian's trajectory is judged. When the pedestrian's trajectory is continuous, the position information of multiple pedestrians' feet in the pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a first position determination result;

[0010] When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, and the position information of multiple pedestrians' feet in the completed pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a second position determination result;

[0011] According to the first position determination result or the second position determination result, a boundary crossing warning signal is output to warn the pedestrian of the boundary crossing.

[0012] Furthermore, when the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, which is specifically:

[0013] The historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, where the model parameters of the hidden Markov model include hidden state probabilities, hidden state transition probability matrices, observed state parameters, and hidden state duration distributions;

[0014] The pedestrian's trajectory is input into the trained hidden Markov model, and the hidden state and hidden state duration at the missing moment are predicted based on the hidden state transition probability matrix and the hidden state duration distribution.

[0015] Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory.

[0016] Furthermore, the hidden semi-Markov process is used to segmentally encode the historical pedestrian trajectories to obtain different hidden states, and the initial hidden Markov model is trained using the maximization algorithm to obtain the trained hidden Markov model, which is specifically:

[0017] Obtain historical pedestrian trajectory data, set multiple hidden states according to different travel states in the historical pedestrian travel trajectory, and set initial model parameters of the initial hidden semi-Markov model, the initial model parameters including initial hidden state probability, initial hidden state transition probability matrix, initial observation state parameters and initial hidden state duration distribution;

[0018] The log-likelihood function is used to calculate the log-likelihood and gradient in combination with the initial model parameters to estimate the hidden state distribution corresponding to each observed state. The updating of the initial model parameters is stopped until the log-likelihood is maximized to obtain the updated hidden Markov model.

[0019] Furthermore, based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory, which is specifically:

[0020] Determining the weight matrix of the linear quadratic tracker and initial and terminal conditions of the desired travel trajectory;

[0021] Based on the predicted hidden state at the missing moment, a corresponding moving state model is constructed, and the hidden state duration is used as a time constraint. The expected moving trajectory at the missing moment is determined by a linear quadratic tracker according to the moving state model.

[0022] The expected trajectory is inserted into the corresponding position of the missing pedestrian trajectory to obtain the completed pedestrian trajectory.

[0023] The present invention also provides a power safety zone crossing warning system, which is used in any one of the power safety zone crossing warning methods described above, and the system includes:

[0024] The first module is used to continuously obtain the position information of the feet of multiple pedestrians and construct the pedestrian's moving trajectory according to the position information of the feet of multiple pedestrians;

[0025] The second module is used to determine the continuity of the pedestrian's trajectory. When the pedestrian's trajectory is continuous, the position information of multiple pedestrians' feet in the pedestrian's trajectory is extracted and compared with the preset safety area boundary to perform position determination to obtain a first position determination result.

[0026] When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, and the position information of multiple pedestrians' feet in the completed pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a second position determination result;

[0027] The third module is used to output a boundary crossing warning signal to warn pedestrians of boundary crossing according to the first position determination result or the second position determination result.

[0028] Furthermore, when the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, which is specifically:

[0029] The historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, where the model parameters of the hidden Markov model include hidden state probabilities, hidden state transition probability matrices, observed state parameters, and hidden state duration distributions;

[0030] The pedestrian's trajectory is input into the trained hidden Markov model, and the hidden state and hidden state duration at the missing moment are predicted based on the hidden state transition probability matrix and the hidden state duration distribution.

[0031] Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory.

[0032] Furthermore, the hidden semi-Markov process is used to segmentally encode the historical pedestrian trajectories to obtain different hidden states, and the initial hidden Markov model is trained using the maximization algorithm to obtain the trained hidden Markov model, which is specifically:

[0033] Obtain historical pedestrian trajectory data, set multiple hidden states according to different travel states in the historical pedestrian travel trajectory, and set initial model parameters of the initial hidden semi-Markov model, the initial model parameters including initial hidden state probability, initial hidden state transition probability matrix, initial observation state parameters and initial hidden state duration distribution;

[0034] The log-likelihood function is used to calculate the log-likelihood and gradient in combination with the initial model parameters to estimate the hidden state distribution corresponding to each observed state. The updating of the initial model parameters is stopped until the log-likelihood is maximized to obtain the updated hidden Markov model.

[0035] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0036] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the methods described above are implemented.

[0037] The present invention also provides a computer program product comprising instructions, and when the instructions are executed by a computer device cluster, the computer device cluster executes any of the above methods.

[0038] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0039] In the present invention, the continuity of the pedestrian's trajectory is first judged to screen out discontinuous pedestrian trajectories; then the hidden semi-Markov model is used to predict and complete the missing pedestrian trajectories in the discontinuous pedestrian trajectories; and then the pedestrian's trajectory after completion is judged to be out of bounds, thereby solving the related problems in the prior art that the pedestrian behavior trajectory is incomplete due to the incompleteness of the video image, making it difficult to ensure the accuracy of the out-of-bounds warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0041] Figure 1 This is a method flow chart of a method for warning of crossing the boundary of a power safety area in this embodiment;

[0042] Figure 2 This is a schematic diagram of module connections of a power safety area boundary crossing warning system in this embodiment;

[0043] Figure 3 It is a structural schematic diagram of a computer device in this embodiment. DETAILED DESCRIPTION

[0044] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0045] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0046] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0047] Example 1

[0048] See also Figure 1 As shown, this embodiment provides a method for warning of crossing the boundary of a power safety area, the method comprising:

[0049] S1: continuously obtain multiple pedestrian foot position information, and construct pedestrian movement trajectories based on the multiple pedestrian foot position information;

[0050] It should be noted that, in this embodiment, first, a monitoring image of the power generation area is obtained by a high-definition camera, and at the same time, a scene calibration is performed on the monitoring area, and a mapping relationship between image coordinates and actual coordinates is established through known reference objects; in the monitoring process, since the foot position can provide reliable trajectory position information, the foot position information of the person is detected through an image processing algorithm, so as to construct a pedestrian travel trajectory based on the continuously acquired foot position information of multiple pedestrians.

[0051] S2: judging the continuity of the pedestrian's trajectory. When the pedestrian's trajectory is continuous, extracting multiple pedestrian foot position information in the pedestrian's trajectory and performing position determination with the preset safety zone boundary to obtain a first position determination result; when the pedestrian's trajectory is discontinuous, using the hidden semi-Markov model to predict and complete the missing pedestrian's trajectory to obtain a completed pedestrian's trajectory, extracting multiple pedestrian foot position information in the completed pedestrian's trajectory and performing position determination with the preset safety zone boundary to obtain a second position determination result;

[0052] It should be noted that, in this embodiment, it is necessary to judge the continuity of the collected pedestrian's trajectory to detect whether there are missing parts in the pedestrian's trajectory that cause the trajectory to be discontinuous; in this embodiment, the real-time clock is used to judge whether the video frame interruption exceeds the set threshold, so as to judge whether the trajectory is continuous; at the same time, the hidden semi-Markov model is used to predict the subsequent trajectory based on the pedestrian's foot position information.

[0053] S3: Outputting a boundary crossing warning signal according to the first position determination result or the second position determination result to warn the pedestrian of the boundary crossing.

[0054] It should be noted that, in this embodiment, both the first position determination result and the second position determination result are two situations: crossing the boundary and not crossing the boundary. When it is determined that the pedestrian has crossed the boundary, a crossing warning is issued to the pedestrian.

[0055] Specifically, in this embodiment, the continuity of the pedestrian's trajectory is first judged to screen out discontinuous pedestrian trajectories; then the hidden semi-Markov model is used to predict and complete the missing pedestrian trajectories in the discontinuous pedestrian trajectories; and then the pedestrian's trajectory after completion is judged to be out of bounds, thereby solving the problem in the prior art that the pedestrian behavior trajectory is incomplete due to the incompleteness of the video image, making it difficult to ensure the accuracy of the out-of-bounds warning.

[0056] As a possible implementation method, when the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, specifically:

[0057] The historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model (HSMM model) is trained using a maximization algorithm to obtain a trained hidden Markov model, where the model parameters of the hidden Markov model include hidden state probability, hidden state transition probability matrix, observation state parameters and hidden state duration distribution;

[0058] It should be noted that, in this embodiment, it is of great significance to use the hidden semi-Markov process to segmentally encode the historical pedestrian trajectories to obtain different hidden states. By using the historical pedestrian trajectories and adopting the expectation maximization algorithm (EM) to estimate the model parameters of the HSMM model, it can provide strong support for the subsequent future state prediction; at the same time, the hidden state probability a i,j Represents the probability of transferring from hidden state i to hidden state j, which reflects the possibility of transition of the trajectory between different states; hidden state transition probability matrix.

[0059] The pedestrian's trajectory is input into the trained hidden Markov model, and the hidden state and hidden state duration at the missing moment are predicted based on the hidden state transition probability matrix and the hidden state duration distribution.

[0060] Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory.

[0061] As a possible implementation method, the historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, specifically:

[0062] Obtain historical pedestrian trajectory data, set multiple hidden states according to different travel states in the historical pedestrian travel trajectory, and set initial model parameters of the initial hidden semi-Markov model, the initial model parameters including initial hidden state probability, initial hidden state transition probability matrix, initial observation state parameters and initial hidden state duration distribution;

[0063] The log-likelihood function is used to calculate the log-likelihood and gradient in combination with the initial model parameters to estimate the hidden state distribution corresponding to each observed state. The updating of the initial model parameters is stopped until the log-likelihood is maximized to obtain the updated hidden Markov model.

[0064] As a possible implementation method, based on the predicted hidden state and hidden state duration at the missing moment, a linear quadratic tracker is used to synthesize the expected trajectory, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory, which is specifically:

[0065] Determining the weight matrix of the linear quadratic tracker and initial and terminal conditions of the desired travel trajectory;

[0066] It should be noted that, in this embodiment, the weight matrices Q and R in the linear quadratic tracker (LQT) are first defined, where Q∈R n×n Used to punish the error of the moving state deviating from the expected trajectory; R∈R m×m It is used to penalize the size of the control input; at the same time, it clarifies the initial and terminal conditions of the expected trajectory, such as the starting position and target position of the pedestrian, so as to provide boundary constraints for the LQT calculation.

[0067] Based on the predicted hidden state at the missing moment, a corresponding moving state model is constructed, and the hidden state duration is used as a time constraint. The expected moving trajectory at the missing moment is determined by a linear quadratic tracker according to the moving state model.

[0068] It should be noted that, in this embodiment, a corresponding travel state model is constructed for the predicted hidden state at the missing moment, and the travel state model includes the pedestrian's position coordinates, speed, acceleration and other travel parameters, and the model parameters of the corresponding travel state model are determined according to different travel states; the predicted hidden state duration is then used as a time constraint condition to be incorporated into the LQT trajectory planning process, and the expected travel state at each missing moment in the missing time period is determined according to the hidden state duration and the travel state model, and the expected travel trajectory at the missing moment is determined; by solving the optimal control problem of LQR, that is, minimizing the quadratic cost function Among them, x d (t) represents the desired trajectory; the optimal control input sequence u is obtained within the duration of the hidden state * (t),u * (t) = -K(t)(x(t) - x d (t))+u ff (t).

[0069] It should also be noted that in this embodiment, the feedback control term -K(t)(x(t)-x d (t)), used to calculate the current state error x(t)-x d (t) to adjust the control input so that the system can track the desired trajectory x as much as possible. d (t), the feedback gain matrix K(t) is obtained by solving the Riccati equation obtained, among which is a positive definite symmetric matrix, which is the solution of the Riccati differential equation with terminal conditions , which is the cost weight of the system at the end point. Solving the Riccati equation is a relatively complex mathematical process, which usually requires the use of numerical methods for calculation. Once the solution is obtained The feedback gain matrix K(t) can be calculated through a specific formula, thereby determining the specific form of the feedback control term and achieving precise control of the system.

[0070] The expected trajectory is inserted into the corresponding position of the missing pedestrian trajectory to obtain the completed pedestrian trajectory.

[0071] It should be noted that, in this embodiment, the core of the continuous-time linear quadratic tracker (LQT) is to determine the appropriate control input, whose function is to intervene in the system state at each moment, just like manipulating the key parameters or variables of the system, with the aim of guiding the system state as close as possible to the smooth desired trajectory, while minimizing a specific quadratic cost function to achieve minimum control of the state error.

[0072] Specifically, in this embodiment, the continuity of the pedestrian's trajectory is first judged to screen out discontinuous pedestrian trajectories; then the hidden semi-Markov model is used to predict and complete the missing pedestrian trajectories in the discontinuous pedestrian trajectories; and then the pedestrian's trajectory after completion is judged to be out of bounds, thereby solving the problem in the prior art that the pedestrian behavior trajectory is incomplete due to the incompleteness of the video image, making it difficult to ensure the accuracy of the out-of-bounds warning.

[0073] Example 2

[0074] See also Figure 2 As shown, the present invention also provides a power safety zone crossing warning system, which is used in any one of the power safety zone crossing warning methods described above, and the system includes:

[0075] The first module 100 is used to continuously obtain the position information of the feet of multiple pedestrians and construct the pedestrian's moving trajectory according to the position information of the feet of multiple pedestrians;

[0076] The second module 200 is used to determine the continuity of the pedestrian's trajectory. When the pedestrian's trajectory is continuous, the position information of multiple pedestrians' feet in the pedestrian's trajectory is extracted and compared with the preset safety area boundary to perform position determination to obtain a first position determination result;

[0077] When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, and the position information of multiple pedestrians' feet in the completed pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a second position determination result;

[0078] The third module 300 is used to output a boundary crossing warning signal to warn pedestrians of boundary crossing according to the first position determination result or the second position determination result.

[0079] As a possible implementation method, when the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, specifically:

[0080] The historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, where the model parameters of the hidden Markov model include hidden state probabilities, hidden state transition probability matrices, observed state parameters, and hidden state duration distributions;

[0081] The pedestrian's trajectory is input into the trained hidden Markov model, and the hidden state and hidden state duration at the missing moment are predicted based on the hidden state transition probability matrix and the hidden state duration distribution.

[0082] Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory.

[0083] As a possible implementation method, the historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, specifically:

[0084] Obtain historical pedestrian trajectory data, set multiple hidden states according to different travel states in the historical pedestrian travel trajectory, and set initial model parameters of the initial hidden semi-Markov model, the initial model parameters including initial hidden state probability, initial hidden state transition probability matrix, initial observation state parameters and initial hidden state duration distribution;

[0085] The log-likelihood function is used to calculate the log-likelihood and gradient in combination with the initial model parameters to estimate the hidden state distribution corresponding to each observed state. The updating of the initial model parameters is stopped until the log-likelihood is maximized to obtain the updated hidden Markov model.

[0086] It should be noted that the modules in the system of Example 2 correspond to the steps in the method of Example 1. The steps in the method of Example 1 have been described in detail in Example 1. In this Example 2, the contents of the modules in the system will not be described in detail.

[0087] Example 3

[0088] See also Figure 3 As shown, this embodiment further provides a computer device, including a system memory 1005 and a processor 1001, wherein the system memory 1005 stores a computer program, and the processor 1001 implements the steps of any of the above methods when executing the computer program.

[0089] It should be noted that the processor 1001 is used to execute the steps in the above method embodiments according to the instructions in the program code. Alternatively, the processor 1001 implements the functions of each module / unit in the above system / device embodiments when executing the computer program.

[0090] Specifically, in this embodiment, the computer program may be divided into one or more modules / units, one or more modules / units are stored in the system memory 1005, and are executed by the processor 1001 to complete the present application. One or more modules / units may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0091] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor 1001 and a system memory 1005. Those skilled in the art will appreciate that this does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include an input / output device 1003, a network access device 1002, a bus 1006, etc.

[0092] The processor 1001 may be a central processing unit (CPU), or other general-purpose processors 1001, digital signal processors 1001 (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 1001 may be a microprocessor or any conventional processor, etc.

[0093] The system memory 1005 may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The system memory 1005 may also be a storage device 1004 of the terminal device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the terminal device. Further, the system memory 1005 may also include both the internal storage unit of the terminal device and the storage device 1004. The system memory 1005 is used to store computer programs and other programs and data required by the terminal device. The system memory 1005 may also be used to temporarily store data that has been output or is to be output.

[0094] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0095] Example 4

[0096] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0097] Among them, the computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, system or device, or any combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk. Random Access Memory (RAM), Read-Only Memory (ROM), Erasable Programmable Read Only Memory (EPROM), registers, hard disks, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above, or any other form of computer-readable storage medium known in the art.

[0098] An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In an embodiment of the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system, or device.

[0099] Example 5

[0100] This embodiment also provides a computer program product including instructions. When the instructions are executed by a computer device cluster, the computer device cluster executes the method described in Embodiment 1.

[0101] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for early warning of crossing the boundary of a power safety area, characterized in that: Methods include: Continuously obtain the position information of multiple pedestrians' feet, and construct the pedestrian's moving trajectory based on the multiple pedestrians' foot position information; The continuity of the pedestrian's trajectory is judged. When the pedestrian's trajectory is continuous, the position information of multiple pedestrians' feet in the pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a first position determination result; When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, and the position information of multiple pedestrians' feet in the completed pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a second position determination result; According to the first position determination result or the second position determination result, a boundary crossing warning signal is output to warn the pedestrian of the boundary crossing.

2. The method for early warning of electric power safety zone crossing boundary according to claim 1, characterized in that: When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, which is specifically: The historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, where the model parameters of the hidden Markov model include hidden state probabilities, hidden state transition probability matrices, observed state parameters, and hidden state duration distributions; The pedestrian's trajectory is input into the trained hidden Markov model, and the hidden state and hidden state duration at the missing moment are predicted based on the hidden state transition probability matrix and the hidden state duration distribution. Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory.

3. The method for early warning of electric power safety zone crossing boundary according to claim 2 is characterized in that: The hidden semi-Markov process is used to segmentally encode the historical pedestrian trajectories to obtain different hidden states, and the initial hidden Markov model is trained using the maximization algorithm to obtain the trained hidden Markov model, which is specifically: Obtain historical pedestrian trajectory data, set multiple hidden states according to different travel states in the historical pedestrian travel trajectory, and set initial model parameters of the initial hidden semi-Markov model, the initial model parameters including initial hidden state probability, initial hidden state transition probability matrix, initial observation state parameters and initial hidden state duration distribution; The log-likelihood function is used to calculate the log-likelihood and gradient in combination with the initial model parameters to estimate the hidden state distribution corresponding to each observed state. The updating of the initial model parameters is stopped until the log-likelihood is maximized to obtain the updated hidden Markov model.

4. The method for early warning of electric power safety zone crossing boundary according to claim 2 is characterized in that: Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory, which is specifically: Determining the weight matrix of the linear quadratic tracker and initial and terminal conditions of the desired travel trajectory; Based on the predicted hidden state at the missing moment, a corresponding moving state model is constructed, and the hidden state duration is used as a time constraint. The expected moving trajectory at the missing moment is determined by a linear quadratic tracker according to the moving state model. The expected trajectory is inserted into the corresponding position of the missing pedestrian trajectory to obtain the completed pedestrian trajectory.

5. An electric power safety zone boundary crossing warning system, characterized in that: The system is used in a power safety zone boundary crossing warning method as described in any one of claims 1 to 4, and the system comprises: The first module is used to continuously obtain the position information of the feet of multiple pedestrians and construct the pedestrian's moving trajectory according to the position information of the feet of multiple pedestrians; The second module is used to determine the continuity of the pedestrian's trajectory. When the pedestrian's trajectory is continuous, the position information of multiple pedestrians' feet in the pedestrian's trajectory is extracted and compared with the preset safety area boundary to perform position determination to obtain a first position determination result. When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, and the position information of multiple pedestrians' feet in the completed pedestrian's trajectory is extracted to perform position determination with the preset safety area boundary to obtain a second position determination result; The third module is used to output a boundary crossing warning signal to warn pedestrians of boundary crossing according to the first position determination result or the second position determination result.

6. The electric power safety zone crossing warning system according to claim 5 is characterized in that: When the pedestrian's trajectory is discontinuous, the hidden semi-Markov model is used to predict and complete the missing pedestrian's trajectory to obtain the completed pedestrian's trajectory, which is specifically: The historical pedestrian trajectories are segmented and encoded using a hidden semi-Markov process to obtain different hidden states, and the initial hidden Markov model is trained using a maximization algorithm to obtain a trained hidden Markov model, where the model parameters of the hidden Markov model include hidden state probabilities, hidden state transition probability matrices, observed state parameters, and hidden state duration distributions; The pedestrian's trajectory is input into the trained hidden Markov model, and the hidden state and hidden state duration at the missing moment are predicted based on the hidden state transition probability matrix and the hidden state duration distribution. Based on the predicted hidden state and hidden state duration at the missing moment, the expected trajectory is synthesized using a linear quadratic tracker, and the missing pedestrian trajectory is completed using the expected trajectory to obtain the completed pedestrian trajectory.

7. The electric power safety zone boundary crossing warning system according to claim 6 is characterized in that: The hidden semi-Markov process is used to segmentally encode the historical pedestrian trajectories to obtain different hidden states, and the initial hidden Markov model is trained using the maximization algorithm to obtain the trained hidden Markov model, which is specifically: Obtain historical pedestrian trajectory data, set multiple hidden states according to different travel states in the historical pedestrian travel trajectory, and set initial model parameters of the initial hidden semi-Markov model, the initial model parameters including initial hidden state probability, initial hidden state transition probability matrix, initial observation state parameters and initial hidden state duration distribution; The log-likelihood function is used to calculate the log-likelihood and gradient in combination with the initial model parameters to estimate the hidden state distribution corresponding to each observed state. The updating of the initial model parameters is stopped until the log-likelihood is maximized to obtain the updated hidden Markov model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

10. A computer program product comprising instructions, characterized in that When the instructions are executed by a computer device cluster, the computer device cluster executes the method according to any one of claims 1 to 4.

Images