Crowd trampling and crushing risk assessment method and apparatus, device, and storage medium

Abstract

The present application relates to a crowd trampling and crushing risk assessment method and apparatus, a device, and a storage medium. The method comprises: rasterizing image data of a crowd activity area, and extracting static attribute information of each raster geographic unit, so as to construct a three-dimensional pedestrian road network; using an image analysis technology to extract dynamic attribute information from image monitoring data of the crowd activity area, and mapping the dynamic attribute information to corresponding raster geographic units, so as to construct a crowd movement interaction network; coupling the three-dimensional pedestrian road network with the crowd movement interaction network, using the static attribute information and the dynamic attribute information to construct a trampling and crushing comprehensive risk index, and obtaining a dynamic risk value of each raster geographic unit, so as to generate a crowd trampling and crushing risk graph. The embodiments of the present application fully consider the static attribute information such as road width and slope of the crowd activity area, thus more precisely assessing crowd trampling and crushing risks and providing early warning.

Description

Crowd stampede risk assessment method, device, equipment and storage medium Technical Field

[0001] The present application relates to the field of urban public safety management and risk assessment and prevention, and in particular to a crowd stampede risk assessment method, device, equipment and storage medium. Background Art

[0002] With the rapid development of urbanization and the proliferation of urban complexes, urban built environments have become increasingly complex, urban public spaces have become increasingly crowded, and urban public safety incidents have become frequent. According to data from York University, from 1952 to 2023, at least 260 stampedes occurred, resulting in over 14,506 deaths and over 26,205 injuries. On September 26, 2015, a stampede occurred during the Hajj pilgrimage in Saudi Arabia, killing at least 2,110 people. In October 2022, a large crowd celebrating Halloween in Itaewon, a nightlife district in Seoul, South Korea, killed at least 153 people, most of whom were teenagers and people in their twenties. In the past decade alone, over 100 large-scale stampede incidents have occurred worldwide, resulting in over 10,000 casualties. Therefore, it is crucial to develop a scientific and rational quantitative assessment method and early warning system for stampede risks.

[0003] In existing research, in 2005, Australian scholars RSC Lee and RL Hughes proposed a thought on the problem of crowd stampede accidents and introduced a continuous pedestrian flow model to quantitatively analyze crowd crowding and stampede. D. Helbing proposed the famous "social force" model based on the panic characteristics of crowds [Lee R, Hughes R L. Exploring Trampling and Crushing in a Crowd [J]. Journal of Transportation Engineering, 2005 (8): 131.]; C.M. Henein and T. White used an improved cellular automaton model to simulate and analyze crowd behavior during evacuation based on multi-agent technology, and believed that "individual casualties in the crowd are due to the interaction force between individuals reaching a certain threshold." In China, Wang Chongyang, based on the "crowd pressure" concept proposed by Helbing, quantitatively evaluated the risk of stampede in crowded crowds through real-time speed variance. In addition, many studies have explored stampede accidents from the perspectives of social force models, crowd flow, and emergency linkage. However, existing assessment methods for crowd stampede risks only focus on the analysis of crowd status, ignoring the combined influence of terrain and architectural design factors such as road slope and width, and therefore cannot accurately assess the stampede risk.

[0004] Summary of the Invention

[0005] The present application provides a crowd crowding and trampling risk assessment method, device, equipment and storage medium, which aims to solve at least one of the above-mentioned technical problems in the prior art to a certain extent.

[0006] In order to solve the above problems, this application provides the following technical solutions:

[0007] A crowd stampede risk assessment method comprising:

[0008] Acquire image data of the crowd activity area, perform rasterization on the image data, and extract static attribute information of each raster geographic unit to construct a three-dimensional pedestrian road network; wherein the static attribute information includes road width and slope information;

[0009] Obtaining image monitoring data of crowd activity areas, extracting dynamic attribute information from the image monitoring data using image analysis technology, and mapping the dynamic attribute information to corresponding raster geographic units to construct a crowd movement interaction network; wherein the dynamic attribute information includes crowd density and movement force field;

[0010] The three-dimensional pedestrian road network is coupled with the crowd movement interaction network, and a comprehensive risk index of crowding and trampling is constructed using the static attribute information and the dynamic attribute information. The dynamic risk value of each grid geographic unit is quantified according to the comprehensive risk index of crowding and trampling, and a crowd crowding and trampling risk map is generated.

[0011] The technical solution adopted by the embodiment of the present application further includes: performing rasterization processing on the image data and extracting static attribute information of each raster geographic unit includes:

[0012] Based on the image data of the crowd activity area, a three-dimensional reconstruction is performed through 3dsMax to obtain a three-dimensional built environment model of the crowd activity area;

[0013] The three-dimensional built environment model is rasterized using the maxscript script tool of 3dsMax to obtain N raster geographic units of a set size, and the road width W and slope H information corresponding to each raster geographic unit are extracted.

[0014] The technical solution adopted in the embodiment of the present application also includes: the construction of the three-dimensional pedestrian road network is specifically as follows:

[0015] Based on the road width W and slope H information of each grid geographic unit, adjacent grid geographic units are connected to obtain edge attribute information, which includes the road width variation coefficient ΔW and the road slope variation coefficient ΔH. The traffic volume coefficient C between each grid geographic unit is calculated based on the road width variation coefficient ΔW and the road slope variation coefficient ΔH. T , to realize the construction of a three-dimensional pedestrian road network; the traffic volume coefficient C T The calculation formula is: C T =ΔW / ΔH

[0016] The technical solution adopted in the embodiment of the present application further includes: extracting dynamic attribute information from the image monitoring data using image analysis technology, and mapping the dynamic attribute information to corresponding grid geographic units to construct a crowd mobile interaction network specifically as follows:

[0017] Extracting crowd density from the image monitoring data using a multi-column convolutional neural network model

[0018] Based on the population density Updating the crowd density attribute of each grid geographic unit and mapping the crowd density attribute to the corresponding grid geographic unit;

[0019] Calculating the crowd density in the image monitoring data of the grid geographic unit at different angles within a similar time period, and correcting the crowd density attribute of each grid geographic unit;

[0020] The velocity field of the crowd movement is extracted from the raster geographic unit using Matlab's PIVLab tool The data is mapped to the corresponding raster geographic unit after spatial coordinate conversion;

[0021] Velocity field based on each raster geographic unit and crowd density The speed variance is calculated by connecting adjacent grid geographic units and combining the crowd density. and the speed variance Mobile force field Update the input edge attribute information to obtain the crowd mobile interaction network.

[0022] The technical solution adopted in the embodiment of the present application further includes: coupling the three-dimensional pedestrian road network with the crowd movement interaction network, using the static attribute information and the dynamic attribute information to construct a comprehensive risk index for crowding and trampling, quantifying the dynamic risk value of each grid geographic unit based on the comprehensive risk index for crowding and trampling, and generating a crowd crowding and trampling risk map specifically as follows:

[0023] The squeezing strength CSL of any grid geographic unit i is calculated using the static attribute information of the three-dimensional pedestrian road network and the dynamic attribute information of the crowd mobile interaction network. i :

[0024] In the above formula, ΔW i , ΔH i are the road width and slope changes of grid geographic unit i, is the crowd movement force field of grid geographic unit i;

[0025] Through the risk attenuation function F j (i) = 2·e -d(i,j) Calculate the impact of the risk of the adjacent grid geographic unit j on the grid geographic unit i, and calculate the dynamic risk value of the grid geographic unit i using the following formula to generate the crowd stampede risk map: R i =∑ j≠i (CSL i ·F j (i)) F j (i) = α·e -β·d(i,j)

[0026] In the above formula, R i represents the dynamic risk value of grid geographic unit i, F j (i) represents the degree of influence of the risk of raster geographic unit j on raster geographic unit i, α and β represent the risk attenuation empirical parameters, and d(i, j) represents the spatial distance between raster geographic units i and j.

[0027] The technical solution adopted in the embodiment of the present application further includes: after generating the crowd stampede risk map, the following steps are also included:

[0028] The dynamic risk values ​​of each grid geographic unit are sorted based on the crowd stampede risk map, and the grid geographic units with dynamic risk values ​​higher than a set threshold are identified as high-risk areas according to the sorting results, and a visual risk warning is performed for the high-risk areas.

[0029] Another technical solution adopted in the embodiment of the present application is: a crowd stampede risk assessment device, comprising:

[0030] A 3D pedestrian road network construction module is used to obtain image data of crowd activity areas, rasterize the image data, and extract static attribute information of each raster geographic unit to construct a 3D pedestrian road network. The static attribute information includes road width and slope information.

[0031] Crowd mobility interaction network construction module: obtains image monitoring data of crowd activity areas, uses image analysis technology to extract dynamic attribute information from the image monitoring data, and maps the dynamic attribute information to corresponding raster geographic units to construct a crowd mobility interaction network; wherein the dynamic attribute information includes crowd density and movement force field;

[0032] Network coupling construction module: used to couple the three-dimensional pedestrian road network with the crowd movement interaction network, use the static attribute information and dynamic attribute information to construct a comprehensive risk index of congestion and trampling, quantify the dynamic risk value of each grid geographic unit according to the comprehensive risk index of congestion and trampling, and generate a crowd congestion and trampling risk map.

[0033] The technical solution adopted in the embodiment of the present application also includes: a visual warning module, which is used to sort the dynamic risk values ​​of each grid geographic unit based on the crowd stampede risk map, and according to the sorting results, the grid geographic units with dynamic risk values ​​higher than the set threshold are regarded as high-risk areas, and visual risk warnings are performed for the high-risk areas.

[0034] Another technical solution adopted by the embodiment of the present application is: a device, the device comprising a processor and a memory coupled to the processor, wherein:

[0035] The memory stores program instructions for implementing the crowd stampede risk assessment method;

[0036] The processor is used to execute the program instructions stored in the memory to control a crowd stampede risk assessment method.

[0037] Another technical solution adopted in the embodiment of the present application is: a storage medium storing program instructions executable by a processor, wherein the program instructions are used to execute the crowd crowding and trampling risk assessment method.

[0038] Compared with the existing technology, the beneficial effects of the embodiments of the present application are: the crowd congestion and trampling risk assessment method, device, equipment and storage medium of the embodiments of the present application construct a three-dimensional pedestrian road network based on static attribute information such as road width and slope in the crowd activity area, and construct a crowd movement interaction network based on dynamic attribute information such as crowd density and mobile force field. By coupling the three-dimensional pedestrian road network and the crowd movement interaction network, a crowd congestion and trampling risk map is generated, and the mutual influence of static attribute information such as road width and slope in the crowd activity area and dynamic attribute information such as crowd density and mobile force field is fully considered. It can more accurately assess the risk of crowd congestion and trampling and issue early warnings, thereby effectively reducing the occurrence of crowd congestion and trampling incidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] FIG1 is a flow chart of a method for assessing crowd stampede risk according to an embodiment of the present application;

[0040] FIG2 is a schematic diagram of a three-dimensional pedestrian road network construction according to an embodiment of the present application;

[0041] FIG3 is a schematic diagram of a crowd mobile interaction network construction according to an embodiment of the present application;

[0042] FIG4 is a schematic diagram of the coupling between a three-dimensional pedestrian road network and a crowd movement interaction network according to an embodiment of the present application;

[0043] FIG5 is a schematic structural diagram of a crowd crowding and trampling risk assessment device according to an embodiment of the present application;

[0044] FIG6 is a schematic diagram of the device structure of an embodiment of the present application;

[0045] FIG7 is a schematic diagram of the structure of a storage medium according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0047] The terms "first," "second," and "third" in this application are used only for descriptive purposes and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of such features. In the description of this application, "multiple" means at least two, for example, two, three, etc., unless otherwise specifically defined. All directional indications in the embodiments of this application (such as up, down, left, right, front, back...) are only used to explain the relative positional relationship, movement, etc. between the components under a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications also change accordingly. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products, or devices.

[0048] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0049] Figure 1 is a flow chart of a method for assessing the risk of a crowd stampede according to an embodiment of the present application. The method for assessing the risk of a crowd stampede according to an embodiment of the present application comprises the following steps:

[0050] S100: Acquire image data of a crowd activity area, and perform rasterization processing on the image data of the crowd activity area to generate a plurality of raster geographic units;

[0051] In this step, image data, such as images or videos, of the crowd activity area is first obtained. Based on this image data, 3D reconstruction is performed using 3dsMax to obtain a 3D built environment model of the crowd activity area. Then, based on the 3D built environment model, the 3D built environment model is rasterized using 3dsMax's MaxScript scripting tool to obtain N 1.5m*1.5m resolution raster geographic units. Specifically, the resolution of the raster geographic units can be set according to the actual application scenario and is not limited in this embodiment.

[0052] S110: extracting static attribute information such as road width and slope of each raster geographic unit, and connecting adjacent raster geographic units to construct a three-dimensional pedestrian road network;

[0053] In this step, based on static attribute information such as road width and slope for each raster geographic cell, adjacent raster geographic cells are connected, and attribute information such as the road width variation coefficient ΔW and the road slope variation coefficient ΔH are calculated. Furthermore, the three-dimensional pedestrian road network is constructed by combining temporal attributes. Specifically, see Figure 2, which is a schematic diagram of the three-dimensional pedestrian road network construction in accordance with an embodiment of the present application. In this embodiment of the present application, the Itaewon stampede on October 29, 2022 is taken as an example. The Metashape tool is used to perform point cloud 3D reconstruction of the image data of the crowd activity area in the Itaewon area, and the road slope change value ΔH is corrected by the perspective matching tool of 3dsMax, and finally a 3D built environment model of the crowd activity area is obtained; based on the 3D built environment model, the 3D built environment model is rasterized by the maxscript script tool of 3dsMax, and the static attribute information such as road width W and slope H corresponding to each raster geographic unit with a resolution of 1.5m*1.5m is extracted. Based on the attribute information such as road width W and slope H of each raster geographic unit, the road width change coefficient ΔW, road slope change coefficient ΔH and other side attribute information are obtained by connecting adjacent raster geographic units, and the traffic volume coefficient C between each raster geographic unit is calculated by the following formula T , to realize the construction of three-dimensional pedestrian road network: C T =ΔW / ΔH

[0054] S120: Obtain image monitoring data of the crowd activity area, use image analysis technology to extract dynamic attribute information such as crowd density and mobile force field from the image monitoring data, and map the extracted dynamic attribute information to corresponding raster geographic units to construct a crowd movement interaction network;

[0055] In this step, the image monitoring data includes but is not limited to drone videos or photos of the crowd activity area and camera surveillance videos. The crowd mobile interaction network is constructed as follows: first, the crowd density in the crowd activity area is extracted from the image monitoring data using point cloud 3D reconstruction and PIV analysis algorithms. and mobile force fields Then, dynamic attribute information such as crowd density and mobile force field is mapped to the corresponding grid geographic units to generate a crowd movement interaction network.

[0056] Specifically, as shown in FIG3 , which is a schematic diagram of constructing a crowd mobile interactive network according to an embodiment of the present application, still taking the Itaewon stampede on October 29, 2022 as an example, the specific construction process includes:

[0057] S121: Obtain image monitoring data of the crowd activity area, and extract crowd density based on the image monitoring data using the MCNN (Multi-column Convolutional Neural Network) model Among them, image monitoring data includes CCTV data and related image data on social media;

[0058] S122: Based on crowd density Update the crowd density attribute of each raster geographic unit and map the crowd density attribute to the corresponding raster geographic unit;

[0059] S123: Calculating the crowd density of the image monitoring data of the grid geographic unit at different angles within a similar time period, and correcting the crowd density attribute of each grid geographic unit;

[0060] S124: Based on the image monitoring data of the crowd activity area, the velocity field of the crowd movement is extracted using Matlab's PIVLab tool The data is mapped to the corresponding raster geographic unit after spatial coordinate conversion;

[0061] S125: Velocity field based on each grid geographic unit Crowd density Dynamic attribute information is obtained by connecting adjacent grid geographic units and combining crowd density to calculate speed variance. And the speed variance Mobile force field Update the input edge attribute information and finally obtain the crowd mobile interaction network.

[0062] S130: Coupling the three-dimensional pedestrian road network with the crowd movement interaction network, constructing a comprehensive stampede risk index using static attribute information such as road width and slope, and dynamic attribute information such as crowd density and movement force field, quantifying the dynamic risk value of each grid geographic unit based on the comprehensive stampede risk index, and generating a stampede risk map;

[0063] In this step, by coupling the three-dimensional pedestrian road network with the crowd movement interaction network, the mutual influence of static attribute information such as road width and slope between each grid geographic unit and dynamic attribute information such as crowd density and movement force field is fully considered. The dynamic risk value of each grid geographic unit is calculated through relevant formulas, thereby generating a crowd crowding and trampling risk map.

[0064] In the above formula, R i represents the dynamic risk value of grid geographic unit i, F j(i) represents the degree of impact of the risk of grid geographic unit j on grid geographic unit i, CSL i is the squeezing intensity of the grid geographic unit i, α and β represent the risk attenuation empirical parameters, d(i,j) represents the spatial distance between grid geographic units i and j, ΔW i , ΔH i are the road width and slope changes of grid geographic unit i, The crowd movement force field for grid geographic unit i.

[0065] Specifically, please refer to Figure 4, which is a schematic diagram of the coupling between the three-dimensional pedestrian road network and the crowd movement interaction network in an embodiment of the present application. The specific coupling process includes:

[0066] S131: Calculate the compression strength CSL of any grid geographic unit i using the static attribute information of the 3D pedestrian road network and the dynamic attribute information of the crowd mobile interaction network i :

[0067] Where ΔW i , ΔH i are the road width and slope changes of grid geographic unit i, The crowd movement force field for grid geographic unit i.

[0068] S132: Through the risk attenuation function F j (i) = 2·e -d(i,j) Calculate the impact of the risk of the adjacent grid geographic unit j on the grid geographic unit i, and finally calculate the dynamic risk value of the grid geographic unit i through the following formula to generate the crowd stampede risk map: R i =∑ j≠i (CSL i ·F j (i)) F j (i) = α·e -β·d(i,j)

[0069] Among them, R i represents the dynamic risk value of grid geographic unit i, F j (i) represents the degree of influence of the risk of raster geographic unit j on raster geographic unit j, α and β represent the risk attenuation empirical parameters, and d(i,j) represents the spatial distance between raster geographic units i and j.

[0070] S133: Couple the crowd movement interaction network with the three-dimensional pedestrian road network, update the squeezing intensity to each grid geographic unit, and combine the risk attenuation function of the three-dimensional pedestrian road network to obtain the final risk value R, thereby generating a crowd crowding and trampling risk map.

[0071] S140: Sort the dynamic risk values ​​of each grid geographic unit based on the crowd stampede risk map, and according to the sorting results, identify the grid geographic unit with a dynamic risk value higher than a set threshold as a high-risk area, and perform a visual risk warning for the high-risk area.

[0072] Based on the above, the crowd stampede risk assessment method of the second embodiment of the present application constructs a three-dimensional pedestrian road network based on static attribute information such as road width and slope in the crowd activity area, and constructs a crowd movement interaction network based on dynamic attribute information such as crowd density and mobile force field. By coupling the three-dimensional pedestrian road network and the crowd movement interaction network, a crowd stampede risk map is generated. The embodiment of the present application fully considers the mutual influence of static attribute information such as road width and slope and dynamic attribute information such as crowd density and mobile force field in the crowd activity area, and can more accurately assess the risk of crowd stampede and issue early warnings, thereby effectively reducing the occurrence of crowd stampede incidents.

[0073] Please refer to FIG5 , which is a schematic diagram of the structure of a crowd stampede risk assessment device according to an embodiment of the present application. The crowd stampede risk assessment device 40 according to an embodiment of the present application comprises:

[0074] 3D pedestrian road network construction module 41: used to obtain image data of crowd activity areas, rasterize the image data, and extract static attribute information of each raster geographic unit to construct a 3D pedestrian road network; wherein the static attribute information includes road width and slope information;

[0075] Crowd movement interaction network construction module 42: obtains image monitoring data of crowd activity areas, extracts dynamic attribute information from the image monitoring data using image analysis technology, and maps the dynamic attribute information to corresponding raster geographic units to construct a crowd movement interaction network; wherein the dynamic attribute information includes crowd density and movement force field;

[0076] A network coupling construction module 43 is configured to couple the three-dimensional pedestrian road network with the crowd movement interaction network, construct a comprehensive risk index for crowding and trampling using the static attribute information and the dynamic attribute information, quantify the dynamic risk value of each grid geographic unit based on the comprehensive risk index for crowding and trampling, and generate a crowd crowding and trampling risk map;

[0077] Visual warning module 44: used to sort the dynamic risk values ​​of each grid geographical unit based on the crowd crowding and trampling risk map, and according to the sorting results, the grid geographical units with dynamic risk values ​​higher than the set threshold are regarded as high-risk areas, and visual risk warnings are performed for the high-risk areas.

[0078] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment section and will not be repeated here.

[0079] The device provided in the embodiment of the present application can be applied in the aforementioned method embodiment. For details, please refer to the description of the aforementioned method embodiment, which will not be repeated here.

[0080] Please refer to Figure 6, which is a schematic diagram of the device structure of an embodiment of the present application. The device 50 includes:

[0081] A memory 51 storing executable program instructions;

[0082] a processor 52 connected to the memory 51;

[0083] The processor 52 is used to call the executable program instructions stored in the memory 51 and execute the following steps: obtaining image data of the crowd activity area, rasterizing the image data, and extracting static attribute information of each raster geographic unit to construct a three-dimensional pedestrian road network; wherein the static attribute information includes road width and slope information; obtaining image monitoring data of the crowd activity area, extracting dynamic attribute information from the image monitoring data using image analysis technology, and mapping the dynamic attribute information to the corresponding raster geographic unit to construct a crowd mobile interaction network; wherein the dynamic attribute information includes crowd density and mobile force field; coupling the three-dimensional pedestrian road network with the crowd mobile interaction network, using the static attribute information and dynamic attribute information to construct a comprehensive risk index of congestion and trampling, quantifying the dynamic risk value of each raster geographic unit according to the comprehensive risk index of congestion and trampling, and generating a crowd congestion and trampling risk map.

[0084] The processor 52 may also be referred to as a CPU (Central Processing Unit). The processor 52 may be an integrated circuit chip having signal processing capabilities. The processor 52 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The general-purpose processor may be a microprocessor or any conventional processor.

[0085] Please refer to Figure 7, which is a schematic diagram of the structure of the storage medium of the embodiment of the present application. The storage medium of the embodiment of the present application stores program instructions 61 that can implement the following steps: obtaining image data of the crowd activity area, rasterizing the image data, and extracting static attribute information of each raster geographic unit to construct a three-dimensional pedestrian road network; wherein the static attribute information includes road width and slope information; obtaining image monitoring data of the crowd activity area, extracting dynamic attribute information from the image monitoring data using image analysis technology, and mapping the dynamic attribute information to the corresponding raster geographic unit to construct a crowd movement interaction network; wherein the dynamic attribute information includes crowd density and mobile force field; coupling the three-dimensional pedestrian road network with the crowd movement interaction network, using the static attribute information and dynamic attribute information to construct a comprehensive risk index of congestion and trampling, quantifying the dynamic risk value of each raster geographic unit according to the comprehensive risk index of congestion and trampling, and generating a crowd congestion and trampling risk map. Among them, the program instructions 61 can be stored in the above-mentioned storage medium in the form of a software product, including a number of instructions for enabling a device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the various implementation methods of the present application. The aforementioned storage medium includes: various media that can store program instructions, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a terminal device such as a computer, a server, a mobile phone, or a tablet. Among them, the server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0087] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units. The above are merely embodiments of the present application and do not limit the scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the scope of patent protection of the present application.

Claims

1. A crowd stampede risk assessment method, characterized in that: include: Acquire image data of the crowd activity area, perform rasterization on the image data, and extract static attribute information of each raster geographic unit to construct a three-dimensional pedestrian road network; wherein the static attribute information includes road width and slope information; Obtain image monitoring data of crowd activity areas, extract dynamic attribute information from the image monitoring data using image analysis technology, and map the dynamic attribute information to corresponding grid geographic units to construct a crowd mobile interaction network; wherein the dynamic attribute information includes crowd density and mobile force field; The three-dimensional pedestrian road network is coupled with the crowd mobile interaction network, and a comprehensive risk index of crowding and trampling is constructed using the static attribute information and the dynamic attribute information. The dynamic risk value of each grid geographic unit is quantified according to the comprehensive risk index of crowding and trampling, and a crowd crowding and trampling risk map is generated.

2. The crowd stampede risk assessment method according to claim 1 is characterized in that: The rasterizing of the image data and extracting static attribute information of each raster geographic unit includes: Based on the image data of the crowd activity area, a three-dimensional reconstruction is performed through 3dsMax to obtain a three-dimensional built environment model of the crowd activity area; The three-dimensional built environment model is rasterized by using the maxscript script tool of 3dsMax to obtain N raster geographic units of a set size, and the road width W and slope H information corresponding to each raster geographic unit are extracted.

3. The crowd stampede risk assessment method according to claim 2 is characterized in that: The construction of the three-dimensional pedestrian road network is specifically as follows: Based on the road width W and slope H information of each grid geographic unit, adjacent grid geographic units are connected to obtain edge attribute information, wherein the edge attribute information includes a road width variation coefficient ΔW and a road slope variation coefficient ΔH. ΔH calculates the traffic coefficient C between each grid geographic unit T , to realize the construction of a three-dimensional pedestrian road network; the traffic volume coefficient C T The calculation formula is: C T =ΔW / ΔH。 4. The crowd stampede risk assessment method according to claim 3 is characterized in that: The method of extracting dynamic attribute information from the image monitoring data by using image analysis technology and mapping the dynamic attribute information to corresponding grid geographic units to construct a crowd mobile interaction network is specifically as follows: Extracting crowd density from the image monitoring data through a multi-column convolutional neural network model Based on the population density Updating the crowd density attribute of each grid geographic unit, and mapping the crowd density attribute to the corresponding grid geographic unit; Calculating the crowd density in the image monitoring data of the grid geographic unit at different angles in a similar time, and correcting the crowd density attribute of each grid geographic unit; The velocity field of crowd movement is extracted from the grid geographic unit using Matlab's PIVLab tool Data is mapped to the corresponding raster geographic unit after spatial coordinate conversion; Velocity field based on each raster geographic unit and crowd density The speed variance is calculated by connecting adjacent grid geographic units and combining the crowd density. and the speed variance Mobile Force Field Update the input edge attribute information to obtain the crowd mobile interaction network.

5. The crowd stampede risk assessment method according to claim 4 is characterized in that: The three-dimensional pedestrian road network is coupled with the crowd mobile interaction network, and the static attribute information and the dynamic attribute information are used to construct a comprehensive risk index of crowding and trampling. The dynamic risk value of each grid geographic unit is quantified according to the comprehensive risk index of crowding and trampling, and the crowd crowding and trampling risk map is generated as follows: The squeezing strength CSL of any grid geographic unit i is calculated by using the static attribute information of the three-dimensional pedestrian road network and the dynamic attribute information of the crowd mobile interaction network. i : In the above formula, ΔW i , ΔH i are the road width and slope changes of grid geographic unit i, The force field for the crowd movement of grid geographic unit i; Through the risk attenuation function F j (i) = 2·e -d(i,j) Calculate the impact of the risk of the adjacent grid geographic unit j on the grid geographic unit i, and calculate the dynamic risk value of the grid geographic unit i through the following formula to generate the crowd stampede risk map: R i =∑ j≠i (CSL i ·F j (i)) F j (i)=a·e -β·d(i,j) In the above formula, R i represents the dynamic risk value of the grid geographic unit i, F j (i) represents the degree of influence of the risk of raster geographic unit j on raster geographic unit i, α and β represent the risk attenuation empirical parameters, and d(i,j) represents the spatial distance between raster geographic units i and j.

6. The crowd stampede risk assessment method according to any one of claims 1 to 5, characterized in that: After generating the crowd stampede risk map, the method further includes: The dynamic risk values ​​of each grid geographic unit are sorted based on the crowd stampede risk map, and the grid geographic units with dynamic risk values ​​higher than a set threshold are identified as high-risk areas according to the sorting results, and a visual risk warning is performed for the high-risk areas.

7. A crowd stampede risk assessment device, characterized in that: include: 3D pedestrian road network construction module: used to obtain image data of crowd activity areas, perform rasterization on the image data, and extract static attribute information of each raster geographic unit to construct a 3D pedestrian road network; wherein the static attribute information includes road width and slope information; Crowd mobile interaction network construction module: obtain image monitoring data of crowd activity areas, extract dynamic attribute information from the image monitoring data using image analysis technology, and map the dynamic attribute information to corresponding grid geographic units to construct a crowd mobile interaction network; wherein the dynamic attribute information includes crowd density and mobile force field; Network coupling construction module: used to couple the three-dimensional pedestrian road network with the crowd mobile interaction network, use the static attribute information and dynamic attribute information to construct a comprehensive risk index of congestion and trampling, quantify the dynamic risk value of each grid geographic unit according to the comprehensive risk index of congestion and trampling, and generate a crowd congestion and trampling risk map.

8. The crowd stampede risk assessment device according to claim 7, characterized in that: It also includes a visual warning module, which is used to sort the dynamic risk values ​​of each grid geographic unit based on the crowd stampede risk map, and according to the sorting results, the grid geographic units with dynamic risk values ​​higher than the set threshold are regarded as high-risk areas, and a visual risk warning is performed for the high-risk areas.

9. A device, characterized in that: The device comprises a processor and a memory coupled to the processor, wherein: The memory stores program instructions for implementing the crowd stampede risk assessment method according to any one of claims 1 to 6; The processor is used to execute the program instructions stored in the memory to control the crowd trampling risk assessment method.

10. A storage medium, characterized in that: The device stores program instructions executable by a processor, wherein the program instructions are used to execute the crowd stampede risk assessment method as described in any one of claims 1 to 6.

Images