Evacuation plan generation method, electronic device, and storage medium

Abstract

This disclosure discloses an evacuation plan generation method, electronic device, and storage medium. The method includes: obtaining the original distribution of evacuation objects; the original distribution being the distribution pattern of the original locations of the evacuation objects; updating the current evacuation distribution of the evacuation objects; the current evacuation distribution being the distribution pattern of the evacuation locations of the evacuation objects; the evacuation location being one of N evacuation exits; N being a positive integer; the total number of evacuation objects in the current evacuation distribution being consistent with the total number of evacuation objects in the original distribution; calculating the Wasserstein distance between the current evacuation distribution and the original distribution; and generating an evacuation plan based on the original distribution and the current evacuation distribution in response to the convergence of the Wasserstein distance. The evacuation plan generation method provided by this disclosure can find the evacuation plan with the minimum transportation cost, solving the problem that existing evacuation algorithms are prone to getting trapped in local optima.

Description

Technical Field

[0001] This disclosure generally relates to the field of computer technology. More specifically, this disclosure relates to a method for generating evacuation plans, electronic devices, and storage media. Background Technology

[0002] In public places such as subways, airports, and large performance venues, large crowds gather, and in the event of a sudden emergency, it is necessary to guide the crowd to evacuate and take shelter. For example, in the event of a fire, it is necessary to guide people in large venues to evacuate in an orderly manner through various safety exits. Traditional evacuation guidance usually involves manual intervention, such as commanders issuing action signals or voice signals. This method relies heavily on the commander's subjective experience, resulting in low evacuation efficiency and a high risk of errors.

[0003] To improve the evacuation efficiency of evacuation plans, existing technologies provide a method that generates an evacuation plan by calculating the fastest evacuation path for each individual. However, this method focuses on the path strategy of each individual and often suffers from getting trapped in local optima, resulting in poor accuracy of the evacuation plan.

[0004] Therefore, there is an urgent need to provide a method for generating evacuation plans, so as to automatically generate globally optimal evacuation plans and improve the accuracy of evacuation methods. Summary of the Invention

[0005] In order to at least address one or more of the technical problems mentioned above, this disclosure proposes a method for generating evacuation plans in several aspects.

[0006] In a first aspect, this disclosure provides a method for generating an evacuation plan, comprising: obtaining the original distribution of evacuation objects; the original distribution being the distribution pattern of the original locations of the evacuation objects; updating the current evacuation distribution of the evacuation objects; the current evacuation distribution being the distribution pattern of the evacuation locations of the evacuation objects; the evacuation location being one of N evacuation exits; N being a positive integer; the total number of evacuation objects in the current evacuation distribution being consistent with the total number of evacuation objects in the original distribution; calculating the Wasserstein distance between the current evacuation distribution and the original distribution; and generating an evacuation plan based on the original distribution and the current evacuation distribution in response to the convergence of the Wasserstein distance.

[0007] In some embodiments, after the Wasserstein distance calculation step, the method further includes: in response to the Wasserstein distance not converging, returning to the step of performing the update of the current evacuation distribution of the evacuation objects until the Wasserstein distance converges.

[0008] In some embodiments, the step of updating the current evacuation distribution includes: updating the current evacuation weight of each evacuation exit; wherein the sum of the current evacuation weights of N evacuation exits is 1; calculating the current evacuation quantity of each evacuation exit based on the current evacuation weight of each evacuation exit; and generating the current evacuation distribution based on the current evacuation quantity of each evacuation exit; wherein the current evacuation quantity is the product of the current evacuation weight and the total number of evacuated objects.

[0009] In some embodiments, after the calculation step of the current evacuation quantity, the method further includes: if there is an abnormal evacuation exit, then the current evacuation quantity of the abnormal evacuation exit is rounded and updated, and after the update, the total evacuation quantity of the current evacuation quantity is calculated, and the difference in the number of objects is added to or subtracted from the current evacuation quantity of the preset evacuation exit; wherein, the evacuation quantity of the abnormal evacuation exit is a non-natural number; and the difference in the number of objects is the absolute value of the difference between the total evacuation quantity and the total number of evacuated objects.

[0010] In some embodiments, the calculation step of the Wasserstein distance includes: calculating the Wasserstein distance according to the following formula: Among them, W p (,v) represents the Wasserstein distance between the original distribution μ(x) and the current evacuation distribution v(y); x represents the original location of the evacuated object; y represents the current evacuation location of the evacuated object; T(x) represents the transport transformation that converts the original distribution μ(x) to the current evacuation distribution v(y); X represents the metric space corresponding to the original distribution μ(x); operator T # satisfy Y represents the metric space corresponding to the current evacuation distribution v(y).

[0011] In some embodiments, the step of generating an evacuation plan includes: determining a transmission transformation T(x) that transforms the original distribution into the current evacuation distribution to obtain an evacuation plan; the transmission transformation T(x) is the path from the original location of each evacuation object to the evacuation location.

[0012] In some embodiments, prior to the current evacuation weight update step, the method further includes: setting an initial evacuation weight for each evacuation exit to obtain the current evacuation distribution for each evacuation exit; wherein the initial evacuation weight is equal to 1 / N.

[0013] In some embodiments, obtaining the original distribution of evacuation objects includes: obtaining a building floor plan and monitoring images collected by monitoring equipment; and generating the original distribution of evacuation objects by combining the positions of the evacuation objects in the monitoring images and the positions of the monitoring equipment on the building floor plan.

[0014] In a second aspect, this disclosure provides an electronic device including: a processor; and a memory storing executable program instructions that, when executed by the processor, cause the electronic device to implement the evacuation scheme generation method as described in any of the first aspects.

[0015] In a third aspect, this disclosure provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by one or more processors, cause the processors to implement the evacuation scheme generation method as described in any of the first aspects.

[0016] Using the evacuation plan generation method provided above, this embodiment of the disclosure regards the Wasserstein distance between the original distribution and the current evacuation distribution as the transportation cost of the evacuation plan. By continuously updating the current evacuation distribution, with the goal of Wasserstein distance convergence and the constraint that the total number of evacuated objects in the evacuation distribution and the original distribution remain consistent, the evacuation plan with the minimum transportation cost is found to guide each individual in the evacuation objects to evacuate from the corresponding evacuation exit. Wasserstein distance not only provides a distance metric but also offers a transformation between distributions. Furthermore, Wasserstein distance reflects the geometric characteristics between probability distributions. Therefore, the transformation process during Wasserstein distance convergence preserves the geometric features of the original distribution, effectively suppressing the concentration of evacuation locations of evacuation objects at a single evacuation exit. This prevents the overload of a particular evacuation exit in pursuit of the fastest individual evacuation path when the repetition of the fastest individual evacuation path is too high. It solves the problem that existing evacuation algorithms are prone to getting trapped in local optima. Compared to evacuation algorithms based on the fastest individual evacuation path, the evacuation scheme obtained in this disclosed embodiment can be regarded as the global optimal solution to a two-level optimization problem (finding the current optimal transmission scheme based on the original distribution and the current evacuation distribution, and finding the minimum transmission cost among the optimal transmission schemes corresponding to all potential evacuation distributions). Attached Figure Description

[0017] The above and other objects, features, and advantages of exemplary embodiments of this disclosure will become readily apparent upon reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this disclosure are illustrated by way of example and not limitation, and like or corresponding reference numerals denote like or corresponding parts, wherein:

[0018] Figure 1 An exemplary flowchart of an evacuation plan generation method according to some embodiments of this disclosure is shown;

[0019] Figure 2 An exemplary flowchart of an evacuation distribution update method according to some embodiments of this disclosure is shown;

[0020] Figure 3 An exemplary flowchart of an evacuation plan generation method according to other embodiments of this disclosure is shown;

[0021] Figure 4 Example diagrams of building floor plans of some embodiments of this disclosure are shown;

[0022] Figure 5 An example diagram of the original distribution of some embodiments of this disclosure is shown;

[0023] Figure 6 An example diagram of the evacuation distribution of some embodiments of this disclosure is shown;

[0024] Figure 7 An exemplary structural block diagram of an electronic device according to an embodiment of this disclosure is shown. Detailed Implementation

[0025] The technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, not all of them. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0026] It should be understood that the terms “comprising” and “including” used in this disclosure and claims indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0027] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure. As used in this disclosure and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this disclosure and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0028] As used in this specification and claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."

[0029] The specific embodiments disclosed herein will now be described in detail with reference to the accompanying drawings. Exemplary application scenarios

[0030] In the event of an emergency in a crowded place, an efficient evacuation plan is needed to guide people to evacuate through various exits. This process requires considering not only the location of the exits but also minimizing evacuation time to effectively improve efficiency and reduce casualties. Taking subways as an example, subway platforms have a limited number of exits with fixed locations. In the event of an emergency such as a fire, relying solely on the personal experience of staff for evacuation is both inefficient and prone to errors. Therefore, a highly efficient and reliable evacuation algorithm is urgently needed.

[0031] Existing evacuation algorithms design evacuation plans based on the fastest evacuation path for each individual. Although this path is the optimal evacuation path for each individual, precisely because this algorithm focuses on the individual's path strategy, it is prone to getting trapped in local optima, resulting in poor accuracy of the evacuation plan. Exemplary solution

[0032] In view of this, the present disclosure provides a scheme for generating an evacuation plan, which takes the Wasserstein distance between the original distribution and the current evacuation distribution as the transportation cost of the evacuation plan, and seeks the evacuation distribution that makes the Wasserstein distance converge through continuous iteration, thereby obtaining the evacuation plan with the minimum transportation cost. That is, the global optimal solution of the evacuation problem is obtained by combining distance metric and geometric features.

[0033] Figure 1 An exemplary flowchart of an evacuation plan generation method according to some embodiments of this disclosure is shown.

[0034] like Figure 1 As shown, in step 101, the original distribution of the evacuation objects is obtained.

[0035] In this disclosed embodiment, the original distribution refers to the distribution pattern of the original locations of the evacuation objects. Taking a crowd evacuation scenario in a public place as an example, the crowd is the evacuation object referred to in this disclosed embodiment, and the original location of each person is the first coordinate where they were before they moved. Correspondingly, the original distribution is the distribution information of the location of each person before evacuation.

[0036] In practical applications, the above distribution information can be represented by the mathematical expression μ(x), where x represents the original location of an individual in the population, and μ(x) can be understood as the distribution density function corresponding to the original distribution.

[0037] In step 102, the current evacuation distribution of the evacuation objects is updated.

[0038] In this disclosed embodiment, the current evacuation distribution refers to the distribution pattern of the evacuation locations of the evacuees. Taking the evacuation scenario of a crowd in a public place as an example, the evacuation location can be understood as the location of each person after evacuation, that is, the second coordinate corresponding to each person after movement. It can be understood that the current evacuation distribution is the distribution information of the location of each person after evacuation.

[0039] In this disclosed embodiment, the evacuees need to evacuate via evacuation exits. That is, the evacuation location for each individual is one of N evacuation exits, where N is a positive integer. In this case, the second coordinate of each individual coincides with the coordinate of an evacuation exit. Adapted to a subway evacuation scenario, this corresponds to the evacuation location of each passenger being the location of one of the N subway platform exits.

[0040] It should be noted that during the evacuation process, the evacuated objects only move, that is, their location changes, but the number of evacuated objects remains unchanged. Therefore, it is necessary to set the total number of evacuated objects in the current evacuation distribution to be consistent with the total number of evacuated objects in the original distribution as the update constraint condition for the current evacuation distribution, which is also the constraint condition in the evacuation plan generation process.

[0041] The current evacuation distribution of the evacuation objects obtained in step 102 can be understood as a potential evacuation plan. Each execution of step 102 is a process of proposing a new evacuation plan, and the subsequent steps 103 and 104 verify whether the evacuation plan is the optimal plan.

[0042] Furthermore, before updating the current evacuation distribution of the evacuation objects for the first time, an initial evacuation distribution can be set as the initial value for the update process.

[0043] In practical applications, this initial value can be obtained by averaging, for example, by distributing the total number of evacuees evenly across the above N evacuation exits.

[0044] In step 103, the Wasserstein distance between the current evacuation distribution and the original distribution is calculated.

[0045] For example, the Wasserstein distance can be calculated using the following formula:

[0046]

[0047] Among them, W p(,v) represents the Wasserstein distance between the original distribution μ(x) and the current evacuation distribution v(y); x represents the original location of the evacuated object; y represents the current evacuation location of the evacuated object; T(x) represents the transport transformation that converts the original distribution μ(x) to the current evacuation distribution v(y); X represents the metric space corresponding to the original distribution μ(x); operator T # satisfy Y represents the metric space corresponding to the current evacuation distribution v(y).

[0048] T # μ = v, p ≥ 1 can be considered as a constraint condition set on T(x). In this disclosed embodiment, the Wasserstein distance defines the distance from the original distribution to the sparse distribution, and p indicates the norm type of the above distance formula. p = 1 corresponds to the L1 norm, which corresponds to the Manhattan distance between two vectors; p = 2 corresponds to the L2 norm, which corresponds to the Euclidean distance between two vectors; p = +∞ corresponds to the Lp norm, which corresponds to the Chebyshev distance between two vectors. In some embodiments, the value of p in this disclosed embodiment can be 1.

[0049] In step 104, it is determined whether the Wasserstein distance has converged.

[0050] If so, proceed to step 105;

[0051] If not, return to step 102.

[0052] The convergence of the Wasserstein distance is the process by which the current evacuation distribution gradually approaches the optimal evacuation distribution. When the Wasserstein distance converges, it can be considered that the Wasserstein distance has been minimized. Alternatively, in this disclosed embodiment, the Wasserstein distance is used as the loss function in the evacuation scheme generation process, and the loss function is minimized by continuously updating and iterating the current evacuation distribution.

[0053] The Wasserstein distance, also known as the Earth Mover's Distance (EMD), is both a measure of the distance between two distributions and provides insight into how one distribution is transformed into the other. In this disclosed embodiment, the Wasserstein distance is introduced to find a mapping that minimizes transportation costs for two discrete distributions—the original distribution and the evacuation distribution—while keeping the total number of evacuees constant. Furthermore, the Wasserstein distance can preserve the geometric characteristics of the distributions while transforming them.

[0054] To facilitate understanding of the meaning of the Wasserstein distance in this embodiment, the Wasserstein distance will be explained in detail below.

[0055] Given two metric spaces X and Y, typically assumed to be Euclidean spaces, and corresponding distributions μ and v, this embodiment aims to find a transport transformation T: X→Y that transforms random variables following distribution μ into random variables following distribution v, while minimizing the expected transport cost c(,T()). Its mathematical expression is as follows:

[0056]

[0057] Among them, T -1 ()={|∈X,T()∈A}, the constraint condition can be simplified to T # If μ = v, then the operator T # satisfy

[0058] The total cost of optimal transmission is called the Wasserstein distance between μ and v, and its mathematical expression has been described in detail above, so it will not be repeated here.

[0059] It can be understood that the problem to be solved by the embodiments disclosed herein can essentially be regarded as a two-level optimization problem: first, based on the original distribution, calculate the Wasserstein distance between the two in the current evacuation distribution and the corresponding optimal transmission scheme; second, find the current evacuation distribution corresponding to the optimal transmission scheme that minimizes the Wasserstein distance, and regard it as the best evacuation distribution.

[0060] To help those skilled in the art understand the purpose of introducing Wasserstein distance in this disclosure, the following explanation uses a subway evacuation scenario as an example.

[0061] Existing evacuation algorithms in subway evacuation scenarios generate evacuation plans as follows: For each individual, the nearest subway platform exit is determined, and the path from the individual's original location to this determined exit is the individual's fastest evacuation path. Each individual moves along their fastest evacuation path to complete the evacuation. However, in real-world scenarios, there exists a situation where all or most people's original locations are relatively concentrated. When the nearest subway platform exit for all or most people is Exit A, evacuating according to the above plan will lead to a large number of people congesting at Exit A. Although the individual's evacuation path is optimal, the evacuation capacity of Exit A is limited, and considering the time spent queuing, the evacuation efficiency of this plan is actually reduced.

[0062] The evacuation plan in this embodiment is generated based on the evacuation distribution, considering the distribution of evacuation locations of evacuees, rather than the evacuation path of a single individual. Wasserstein distance convergence indicates that the geometric features of the current evacuation distribution can maintain the geometric features of the original distribution, thus effectively avoiding the situation where distribution points move towards the same evacuation exit. That is to say, in the evacuation plan generated in this embodiment, the evacuation location of one evacuee needs to take into account the evacuation locations of other evacuees. When multiple evacuation locations are concentrated at one evacuation exit, considering the time consumption caused by congestion, queuing, etc., the fastest evacuation path of an individual will be abandoned, that is, the local optimum will be abandoned, and other evacuation exits will be selected to reduce the overall evacuation time, that is, to ensure the global optimum.

[0063] Furthermore, in practical applications, deep learning can be used to complete steps 102 to 104 above. For example, the model records the current evacuation distribution and its corresponding Wasserstein distance after each update, compares it with historical data, obtains feedback, and guides the next round of update process until the calculated Wasserstein distance tends to stabilize.

[0064] It should be noted that the above description of the Wasserstein distance convergence process is merely an example and does not constitute a limitation of this disclosure.

[0065] In step 105, an evacuation plan is generated based on the original distribution and the current evacuation distribution.

[0066] In step 105, the process of generating the evacuation plan can be understood as the process of determining the transmission transformation T(x), which can transform the original distribution into the current evacuation distribution. The transmission transformation T(x) reflects the path from the original location of each evacuation object to the evacuation location.

[0067] The evacuation plan generation method disclosed in this embodiment essentially seeks a mapping that meets a target condition for two discrete distributions—the original distribution and the evacuation distribution—while satisfying the condition of distribution measure preservation. In this embodiment, the distribution measure preservation condition is that the total number of evacuation objects in the current evacuation distribution is the same as the total number of evacuation objects in the original distribution, and the target condition is to minimize the transportation cost, wherein the Wasserstein distance between the original distribution and the evacuation distribution is used as the transportation cost.

[0068] Since the evacuation location of the evacuation object in this disclosure is one of a finite number of evacuation exits, in some embodiments, updating the evacuation distribution can be refined to adjusting the evacuation weights of the evacuation exits, wherein the evacuation weights are used to determine the number of evacuations corresponding to each evacuation exit.

[0069] For ease of understanding, the evacuation distribution update method disclosed herein is explained below.

[0070] Figure 2 An exemplary flowchart of an evacuation distribution update method according to some embodiments of this disclosure is shown. It can be understood that the evacuation distribution update method is a specific implementation of step 102 described above; therefore, the preceding text, in conjunction with... Figure 1 The described features can be applied similarly here.

[0071] like Figure 2 As shown, in step 201, the current evacuation weight of each evacuation exit is updated.

[0072] Evacuation weight is a variable set to determine the number of evacuees corresponding to each evacuation exit. Specifically, the current evacuation number at each evacuation exit is equal to the product of the current evacuation weight and the total number of evacuees. Here, the current evacuation number at a particular evacuation exit refers to the total number of individuals among those evacuated through that exit under the current evacuation distribution. Taking a subway evacuation scenario as an example, the current evacuation number at a particular evacuation exit refers to the number of people evacuated through that exit.

[0073] Since the total number of evacuees in the current evacuation distribution is the same as the total number of evacuees in the original distribution, the sum of the current evacuation weights of the aforementioned N evacuation exits must be 1. When the evacuation weight of one of the N evacuation exits changes, the evacuation weight of at least another evacuation exit must also be adjusted accordingly to satisfy the distribution measure preservation condition. That is to say, a change in the evacuation path of one individual will affect the evacuation paths of other individuals. The evacuation scheme generation process in this embodiment needs to consider the mutual influence between the evacuation paths of multiple individuals.

[0074] Furthermore, before updating the evacuation weights for the first time, an initial evacuation weight can be set for each evacuation exit as the initial value for the update process. In some embodiments, the initial evacuation weight can be set to 1 / N.

[0075] In step 202, the current number of evacuees at each evacuation exit is calculated based on the current evacuation weight of each evacuation exit.

[0076] Multiply the current evacuation weight by the total number of evacuees to obtain the current evacuation quantity.

[0077] Given the application scenario of this embodiment, the individual in the evacuation object is the smallest indivisible unit, and its corresponding value is 1. Therefore, the number of individuals evacuated from each evacuation exit should be a positive integer, and correspondingly, the current evacuation quantity should also be a positive integer.

[0078] Furthermore, after step 202, if it is found that the current evacuation quantity is not a natural number, it can be rounded down to match the application scenario.

[0079] Specifically, this involves: after calculating the current evacuation quantity, if an abnormal evacuation exit is found, the current evacuation quantity for the abnormal exit is rounded down, and then the rounded current evacuation quantity is used as the updated current evacuation quantity. After the update, the total evacuation quantity is calculated. Then, the absolute value of the difference between the total evacuation quantity and the total number of evacuees is calculated to obtain the object number difference. This object number difference is added to or subtracted from the current evacuation quantity of the preset evacuation exits. The evacuation quantity for abnormal evacuation exits is a non-natural number. The purpose of adding or subtracting the object number difference in this process is to maintain consistency in the total number of evacuees.

[0080] For example, assuming the total number of evacuees is 100, and there are four evacuation exits: A, B, C, and D, the initial evacuation weights for these four exits can be set to 0.25, 0.25, 0.25, and 0.25 respectively. After one evacuation weight update, assuming the current evacuation weights change to 0.225, 0.275, 0.25, and 0.25, the current evacuation quantity at exit A changes to 22.5, and the current evacuation quantity at exit B changes to 27.5. However, in reality, there is no 0.5 people; the evacuation quantity at each exit should be a positive integer. In this case, the current evacuation quantities at exits A and B can be rounded down. Rounding methods include, but are not limited to, rounding up, rounding down, and rounding to the nearest whole number.

[0081] Assuming rounding is used, the current evacuation numbers at evacuation exits A, B, C, and D are 23, 28, 25, and 25 respectively. At this point, the total number of evacuees is 101, which is 1 more than the total number of evacuees in the original distribution. This difference of 1 can be subtracted from the current evacuation numbers at the preset evacuation exits. Assuming the preset evacuation exit is A, the final evacuation numbers are 22, 28, 25, and 25 respectively.

[0082] It should be noted that the above-mentioned preset evacuation exit is a pre-set evacuation exit, which can be changed according to the actual situation.

[0083] In step 203, the current evacuation distribution is generated based on the current number of evacuees at each evacuation exit.

[0084] After determining the current number of evacuees at each evacuation exit, a corresponding number of individuals are allocated to each evacuation exit based on that current number of evacuees.

[0085] In some embodiments, assuming the current evacuation capacity of an evacuation exit is 20, the 20 people closest to it can be selected to evacuate via that evacuation exit.

[0086] It should be noted that in practical applications, there are other allocation methods that assign a corresponding number of individuals to each evacuation exit, which will not be elaborated here.

[0087] In practical applications, the evacuation plan is generated by a processor connected to a monitoring device. This monitoring device is used to collect image data for the processor to process, thereby obtaining the distribution information of the evacuation targets.

[0088] Figure 3 An exemplary flowchart of an evacuation plan generation method according to other embodiments of this disclosure is shown.

[0089] like Figure 3 As shown, in step 301, the building floor plan and the monitoring images collected by the monitoring equipment are obtained.

[0090] Figure 4 Example diagrams of building floor plans of some embodiments of this disclosure are shown, such as Figure 4 As shown, the building floor plan includes information on the location of evacuation exits. Furthermore, the building floor plan also includes information on the location of monitoring equipment.

[0091] In practical applications, the monitoring angle of surveillance equipment is often not entirely overhead, and there are often multiple surveillance devices. Therefore, point cloud data can be generated from surveillance images at multiple locations and angles, and angle conversion can be performed through matrix transformation to obtain an overhead-view image of the distribution of evacuation objects, which is then matched with the overhead-view building floor plan. The resulting image of the distribution of evacuation objects can reflect the distribution pattern of the evacuation objects.

[0092] In step 302, the original distribution of evacuation objects is generated by combining the positions of the evacuation objects in the monitoring images and the positions of the monitoring equipment on the building floor plan.

[0093] Figure 5 An example diagram of the original distribution of some embodiments of this disclosure is shown. Figure 5 The dots in the diagram represent objects to be evacuated.

[0094] In step 303, the current evacuation distribution of the evacuation objects is updated.

[0095] In step 304, the Wasserstein distance between the current evacuation distribution and the original distribution is calculated.

[0096] In this embodiment, the contents of steps 303 and 304 are the same as those of steps 102 and 103 in the previous embodiment, and will not be described again here.

[0097] In step 305, in response to the convergence of the Wasserstein distance, an evacuation plan is generated based on the original distribution and the current evacuation distribution.

[0098] Figure 6 An example diagram of evacuation distribution according to some embodiments of this disclosure is shown. Assuming... Figure 6 The Wasserstein distance between the corresponding sparse distribution and the original distribution is minimized. Figure 6 The dashed lines in the diagram represent the optimal evacuation path for each individual in the evacuation plan.

[0099] Corresponding to the aforementioned functional embodiments, this disclosure also provides a method such as... Figure 7 The electronic device shown. Figure 7 An exemplary structural block diagram of an electronic device according to an embodiment of this disclosure is shown.

[0100] Figure 7 The illustrated electronic device 700 includes: a processor 710; and a memory 720 storing executable program instructions that, when executed by the processor 710, cause the electronic device to perform any of the methods described above.

[0101] exist Figure 7 In the electronic device 700, only the components relevant to this embodiment are shown. Therefore, it will be apparent to those skilled in the art that the electronic device 700 may also include components related to... Figure 7 The following are common constituent elements with different constituent elements.

[0102] Processor 710 can control the operation of electronic device 700. For example, processor 710 controls the operation of electronic device 700 by executing a program stored in memory 720 on electronic device 700. Processor 710 can be implemented by a central processing unit (CPU), application processor (AP), artificial intelligence processor chip (IPU), etc., provided in electronic device 700. However, this disclosure is not limited thereto. In this embodiment, processor 710 can be implemented in any suitable manner. For example, processor 710 can take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc.

[0103] The memory 720 can be used to store various data and instructions processed in the electronic device 700. For example, the memory 720 can store processed data and data to be processed in the electronic device 700. The memory 720 can store datasets that have been processed or are to be processed by the processor 710. In addition, the memory 720 can store applications, drivers, etc., to be driven by the electronic device 700. For example, the memory 720 can store various programs related to distance calculations, parameter iterations, etc., to be executed by the processor 710. The memory 720 can be DRAM, but this disclosure is not limited thereto. The memory 720 can include at least one of volatile memory or non-volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, phase-change RAM (PRAM), magnetic RAM (MRAM), resistive RAM (RRAM), ferroelectric RAM (FRAM), etc. Volatile memory may include dynamic RAM (DRAM), static RAM (SRAM), synchronous DRAM (SDRAM), PRAM, MRAM, RRAM, ferroelectric RAM (FeRAM), etc. In embodiments, memory 720 may include at least one of hard disk drive (HDD), solid-state drive (SSD), high-density flash memory (CF), secure digital card (SD), micro-secure digital card (Micro-SD), mini-secure digital card (Mini-SD), extreme digital card (xD), cache, or memory stick.

[0104] In summary, the specific functions implemented by the memory 720 and processor 710 of the electronic device 700 provided in this specification can be explained in comparison with the aforementioned embodiments in this specification, and can achieve the technical effects of the aforementioned embodiments. Therefore, they will not be repeated here.

[0105] Alternatively, this disclosure may be implemented as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) storing computer program instructions (or computer program, or computer instruction code) that, when executed by a processor of an electronic device (or electronic device, server, etc.), cause the processor to perform some or all of the steps of the methods described above according to this disclosure.

[0106] While numerous embodiments of this disclosure have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and intent of this disclosure. It should be understood that various alternatives to the embodiments of this disclosure described herein may be employed in the practice of this disclosure. The appended claims are intended to define the scope of this disclosure and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. A method of evacuation plan generation, characterized by, include: Obtain the original distribution of the evacuation objects; the original distribution is the distribution pattern of the original locations of the evacuation objects; Update the current evacuation distribution of the evacuation objects; The current evacuation distribution is the distribution pattern of the evacuation locations of the evacuation objects; the evacuation location is one of N evacuation exits; N is a positive integer; the total number of evacuation objects in the current evacuation distribution is the same as the total number of evacuation objects in the original distribution; Calculate the Wasserstein distance between the current evacuation distribution and the original distribution; In response to the convergence of the Wasserstein distance, an evacuation plan is generated based on the original distribution and the current evacuation distribution. The steps for calculating the Wasserstein distance include: The Wasserstein distance is calculated using the following formula: ； wherein, denotes the Wasserstein distance between the original distribution and the current evacuation distribution ; denotes the original location of the evacuee; denotes the current evacuation location of the evacuee; denotes the transport transformation that converts the original distribution to the current evacuation distribution ; denotes the original distribution and the corresponding metric space; the operator satisfies ; denotes the current evacuation distribution and the corresponding metric space.

2. The evacuation plan generation method according to claim 1, characterized in that, Following the calculation of the Wasserstein distance, the following steps are also included: In response to the Wasserstein distance not converging, return to the step of updating the current evacuation distribution of the evacuation objects until the Wasserstein distance converges.

3. The evacuation plan generation method according to claim 1 or 2, characterized in that, The current evacuation distribution update steps include: Update the current evacuation weight of each evacuation exit; wherein the sum of the current evacuation weights of the N evacuation exits is 1; The current number of evacuees at each evacuation exit is calculated based on the current evacuation weight of each exit. The current evacuation distribution is generated based on the current number of evacuees at each evacuation exit; The current evacuation quantity is the product of the current evacuation weight and the total number of evacuation targets.

4. The evacuation plan generation method according to claim 3, characterized in that, Following the calculation of the current evacuation numbers, the following steps are also included: If there is an abnormal evacuation exit, the current evacuation quantity of the abnormal evacuation exit is rounded and updated. After the update, the total evacuation quantity of the current evacuation quantity is calculated, and the difference in the number of objects is added or subtracted from the current evacuation quantity of the preset evacuation exit. Wherein, the number of evacuees at the abnormal evacuation exits is a non-natural number; the difference in the number of objects is the absolute value of the difference between the sum of the evacuation numbers and the total number of evacuation objects.

5. The evacuation plan generation method according to claim 1, characterized in that, The steps for generating the evacuation plan include: determining a transport transform that transforms the original distribution to a current evacuation distribution , resulting in an evacuation plan; the transport transform is a path for each evacuation object from its original location to its evacuation location.

6. The evacuation plan generation method according to claim 3, wherein Also includes: Prior to the current evacuation weight update step, the following also applies: An initial evacuation weight is set for each evacuation exit to obtain the current evacuation distribution for each evacuation exit; where the initial evacuation weight is equal to 1 / N.

7. The evacuation plan generation method according to claim 1, characterized in that, Obtaining the original distribution of the evacuated objects includes: Acquire building floor plans and surveillance images captured by monitoring equipment; By combining the positions of the evacuated objects in the monitoring image and the positions of the monitoring equipment on the building floor plan, the original distribution of the evacuated objects is generated.

8. An electronic device, comprising: include: processor; as well as A memory storing executable program instructions that, when executed by the processor, cause the electronic device to implement the evacuation scheme generation method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, It stores computer program instructions that, when executed by one or more processors, cause the processors to implement the evacuation scheme generation method as described in any one of claims 1-7.

Images