Subway platform passenger flow risk prevention and control method and system based on video images

Abstract

The invention provides a subway platform passenger flow risk prevention and control method and system based on a video image, and the method comprises the steps: obtaining a monitoring video of a subway platform, inputting the monitoring video into an improved target detection algorithm YOLOv8, and carrying out the detection of a human head, obtaining real-time passenger flow characteristic parameters based on the extracted video image data set; and obtaining a passenger flow risk value by using a pre-constructed subway platform passenger flow risk model according to the passenger flow characteristic parameters, and performing timely prevention and control on the subway platform passenger flow risk through the passenger flow risk value. On the basis of a subway station video monitoring system and a safety risk level, an initial video image is generated through passenger flow conditions in a monitoring video range, passenger flow detection is carried out by using a polo algorithm with a head detection attention mechanism, the identified number of passenger flows is substituted into a risk model for calculation, a passenger flow risk result changing in real time is obtained, and the passenger flow risk is accurately detected. Therefore, active monitoring and prevention of the passenger flow risk can be realized.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a method and system for preventing and controlling passenger flow risks on a subway platform based on video images. Background Art

[0002] With the gradual improvement of my country's subway network and the increase in residents' travel needs, the number of residents who choose to travel by subway has shown a significant upward trend. Subway stations are important nodes in the urban rail transit network. The station platform, as the only place for passengers to get on and off the train, bears a large amount of passenger flow. Once an accident occurs in the passenger flow distribution area, serious traffic jams will occur, which will greatly affect the passengers' travel experience and the normal operation of the subway. Therefore, the passenger flow risk control work of subway station platforms cannot be ignored. The current station passenger flow risk calculation and evaluation are mainly based on the OD data of the starting and ending points, AFC smart card data, etc., but this type of data cannot reflect the changes in passenger flow risk in real time, and its calculation results are relatively lagging. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for subway platform passenger flow risk prevention and control based on video images, aiming to solve the above-mentioned problems in the prior art.

[0004] The embodiment of the present invention provides a method for preventing and controlling passenger flow risks on subway platforms based on video images, including:

[0005] Obtain surveillance video of the subway platform, input the surveillance video into the improved target detection algorithm YOLOv8 to perform human head detection, and obtain real-time passenger flow feature parameters based on the extracted video image data set; and

[0006] A passenger flow risk value is obtained according to the passenger flow characteristic parameters using a pre-constructed subway platform passenger flow risk model, and the passenger flow risk of the subway platform is timely prevented and controlled using the passenger flow risk value.

[0007] The embodiment of the present invention provides a subway platform passenger flow risk prevention and control system based on video images, including:

[0008] A data module is used to obtain surveillance videos of subway platforms, input the surveillance videos into an improved target detection algorithm YOLOv8 to perform human head detection, and obtain real-time passenger flow feature parameters based on the extracted video image data set; and

[0009] The prevention and control module is used to obtain a passenger flow risk value according to the passenger flow characteristic parameters using a pre-built subway platform passenger flow risk model, and to timely prevent and control the passenger flow risk of the subway platform through the passenger flow risk value.

[0010] An embodiment of the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the above-mentioned subway platform passenger flow risk prevention and control method based on video images are implemented.

[0011] An embodiment of the present invention also provides a computer-readable storage medium, on which a program for implementing information transmission is stored. When the program is executed by a processor, the steps of the above-mentioned subway platform passenger flow risk prevention and control method based on video images are implemented.

[0012] The use of the embodiments of the present invention can include the following beneficial effects: the embodiments of the present invention are based on the video surveillance system of the subway station and the safety risk level, generate an initial video image by monitoring the passenger flow within the video range, use the YOLO algorithm with a head detection attention mechanism to detect passenger flow, and bring the identified number of passengers into the risk model for calculation to obtain real-time changing passenger flow risk results, thereby realizing active monitoring and prevention of passenger flow risks, so as to conduct situation analysis and dispatch disposal measures for upcoming risks in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0014] Figure 1 It is a flow chart of a subway platform passenger flow risk prevention and control method based on video images according to an embodiment of the present invention;

[0015] Figure 2 It is a schematic diagram of the structure of the improved yolov8 algorithm of an embodiment of the present invention;

[0016] Figure 3 This is a camera area coverage diagram of the east platform of Line 13 at Xi'erqi Station according to an embodiment of the present invention;

[0017] Figure 4 1 is a schematic diagram of the detection results of the platform passenger flow based on the improved yolov8 according to an embodiment of the present invention (where a is the passengers on platform ①, b is the passengers on platform ②, c is the passengers on platform ③, and d is the passengers on platform ④);

[0018] Figure 5 is a schematic diagram of a change trend of the number of passengers in the platform waiting area from 6:30 to 11:30 according to an embodiment of the present invention;

[0019] Figure 6 This is a schematic diagram of the changing trend of the passenger flow risk on the station platform from 6:30 to 11:30 according to an embodiment of the present invention;

[0020] Figure 7 It is a schematic diagram of a subway platform passenger flow risk prevention and control system based on video images according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0022] Method Embodiment

[0023] According to an embodiment of the present invention, a method for preventing and controlling passenger flow risks on subway platforms based on video images is provided. Figure 1 is a flow chart of a method for preventing and controlling passenger flow risks on a subway platform based on video images according to an embodiment of the present invention. Figure 1 As shown, the subway platform passenger flow risk prevention and control method based on video images according to an embodiment of the present invention specifically includes:

[0024] Step S101, obtaining a surveillance video of a subway platform, inputting the surveillance video into an improved target detection algorithm YOLOv8 to perform human head detection, and obtaining real-time passenger flow feature parameters based on the extracted video image data set;

[0025] The improved target detection algorithm YOLOv8 is a target detection algorithm YOLOv8 that introduces an attention mechanism module including a human head detection network, wherein the target detection algorithm YOLOv8 includes a backbone network Backbone, a neck network Neck and a head network Head; Backbone includes a Focus module and a CSP module, and Neck includes an FPN structure and a PAN structure;

[0026] The passenger flow characteristic parameters include passenger flow congestion degree and passenger flow congestion duration;

[0027] The real-time passenger flow feature parameters obtained based on the extracted video image data set include:

[0028] Passenger flow density and passenger flow speed are divided into thresholds according to relevant standards of subway platforms to obtain a passenger flow density threshold range and a passenger flow speed threshold range, and real-time passenger flow density and passenger flow speed are calculated by formula 1 and formula 2. The real-time passenger flow density and passenger flow speed are normalized by formula 3 and formula 4 according to the passenger flow density threshold range and the passenger flow speed threshold range, and the passenger flow congestion degree is obtained by formula 5 based on the normalized passenger flow density and passenger flow speed, and the passenger flow congestion duration is obtained from the real-time monitoring video;

[0029] ρ={ρ(t)=ρ waiting (t)∪ρ walking (t)|t∈T} Formula 1;

[0030]

[0031] Among them, ρ(t) represents the passenger flow density at time t, T represents the entire evaluation period, and ρ waiting (t) represents the passenger flow density in the platform waiting area at time t, ρ walking (t) represents the passenger flow density in the platform pedestrian zone at time t, v(t) is the passenger flow velocity at time t, and v i (t) represents the walking speed of passenger i at time t, n is the total number of passengers in the area, ρ′(t) represents the normalized passenger flow density index at time t, ρ min represents the minimum value among density thresholds, ρ max represents the maximum value of the density threshold, v′(t) represents the normalized passenger flow speed index at time t, and v min Represents the minimum value among the speed thresholds, v max represents the maximum value of the speed threshold, c(t) represents the congestion level at time t, and c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the running area at time t;

[0032] Step S102, obtaining a passenger flow risk value according to the passenger flow characteristic parameters by using a pre-built subway platform passenger flow risk model, and timely preventing and controlling the passenger flow risk of the subway platform by using the passenger flow risk value, specifically includes:

[0033] Constructing a subway platform passenger flow risk model as shown in Formula 6, and obtaining a passenger flow risk value through Formula 7 according to the passenger flow congestion degree and the passenger flow congestion duration and based on the subway platform passenger flow risk model;

[0034] Σ=(ρ,V,C,D,R) Formula 6;

[0035] R(t)=[c walking (t)·d walking(t)]+[c waiting (t)·d waiting (t)] Formula 7;

[0036] Where ρ represents the passenger flow density, V represents the passenger flow speed, C represents the degree of passenger flow congestion, D represents the duration of passenger flow congestion, R represents the passenger flow risk value, R(t) represents the passenger flow risk value at time t, and d walking (t) represents the duration of passenger congestion in the travel area at time t, d waiting (t) represents the duration of passenger congestion in the waiting area at time t, c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the walking area at time t.

[0037] The above technical solution of the embodiment of the present invention is described in detail below in combination with the specific situation of the subway platform passenger flow risk prevention and control method based on video images in the embodiment of the present invention.

[0038] The embodiment of the present invention combines the existing video surveillance system in the station, uses real-time video image data to obtain corresponding passenger flow parameters, and brings the passenger flow parameters into a specific risk model to obtain real-time passenger flow risks, and assists station managers in making prevention and control decisions based on changes in risks.

[0039] The embodiment of the present invention aims to provide a subway platform passenger flow risk control method based on video images, which is used to solve the problems existing in the prior art. Specifically, it includes the following steps:

[0040] Step 1: Passenger flow characteristic parameters based on crowding degree and crowding duration are proposed, and the passenger flow characteristic parameters are divided into thresholds according to risk concepts;

[0041] Among them, the passenger flow congestion degree is used to indicate the passenger flow congestion state in the area. Subway platforms can be divided into two types according to their regional characteristics: platform waiting area and platform running area. Passengers in the waiting area are mainly stationary waiting for the train, so the embodiment of the present invention only uses passenger flow density to indicate the congestion degree; passengers in the running area are mainly dynamic, so the embodiment of the present invention uses the ratio of passenger flow density to passenger flow speed to indicate the congestion degree.

[0042] The calculation method of passenger flow density ρ(t) is:

[0043]

[0044] Where Q1(t) represents the number of people gathered on the subway platform at time t, obtained through video image data; S eff It represents the effective standing area, which is calculated by subtracting the structural area from the total area; S total is the total platform area, Sstru The area occupied by the building structure.

[0045] The calculation method of passenger flow velocity v(t) is:

[0046]

[0047] In the formula, v i (t) represents the walking speed of passenger i, n is the total number of passengers in the area; v i The calculation method of (t) is: l i represents the walking distance of passenger i; Δt i is the time passenger i takes to walk.

[0048] Congestion duration is a description of the congestion situation in the time dimension. When the congestion level is high and the duration is long, the passenger flow risk will increase accordingly.

[0049] At the same time, considering that the model needs to be normalized according to the passenger flow density and speed, the first step is to divide and describe the passenger flow density and passenger flow speed of the waiting area and the running area according to the relevant standards of the "Metro Design Specifications" and the "Urban Rail Transit Safety Service 1000-Point Evaluation Management Measures". The results are shown in Table 1.

[0050] Table 1 Results of threshold division of passenger flow characteristic parameters in the platform waiting area and platform walking area

[0051]

[0052]

[0053] Step 2: A passenger flow risk model for subway platforms was constructed, and the risk index and risk level were calculated using real-time passenger flow characteristic parameters;

[0054] The passenger flow risk model of a station platform is defined as a five-tuple Σ=(ρ,V,C,D,R), where:

[0055] ρ={ρ(t)=ρ waiting (t)∪ρ walking (t)|t∈T} (3);

[0056] Where ρ(t) is the passenger flow density at time t, T is the entire evaluation period, which includes the passenger flow density in the platform waiting area and the platform walking area, and ρ waiting (t) represents the passenger flow density in the platform waiting area, ρ walking (t) represents the passenger flow density in the platform pedestrian zone;

[0057]

[0058] Where v(t) is the passenger flow velocity at time t;

[0059] C = f(ρ, V) (5);

[0060] Where C is the degree of congestion at time t, and f() represents the function of C with respect to ρ and V;

[0061] D = {d(t)|t∈T} (6);

[0062] Where d(t) is the congestion duration at time t;

[0063] R = g(C,D) (7);

[0064] Where R is the passenger flow risk at time t, and g() represents the function of R with respect to C and D.

[0065] The obtained real-time data ρ(t) and v(t) are normalized according to the thresholds in Table 1:

[0066]

[0067] Where ρ′(t) is the normalized passenger flow density index, ρ min is the minimum value among the density thresholds, ρ max is the maximum value among the density thresholds; v′(t) is the normalized passenger flow speed index, v min is the minimum value among the speed thresholds, v max is the maximum value among the speed thresholds.

[0068] The calculation method of the congestion degree c(t) at time t is:

[0069]

[0070] In the formula, c waiting (t) is the congestion level of the waiting area at time t, c walking (t) is the degree of congestion in the walking area at time t. The calculation method of the passenger flow risk value R(t) at time t is:

[0071] R(t)=[c walking (t)·d walking (t)]+[c waiting (t)·d waiting (t)] (11);

[0072] Where, d walking (t) is the duration of passenger congestion in the travel area at time t, d waiting (t) is the congestion duration of the passenger flow in the waiting area at time t.

[0073] Step 3: A head detection algorithm based on the improved yolov8 algorithm is proposed, which can improve the detection effect of dense crowds in subway stations.

[0074] That is, introduce the attention mechanism with human head detection network into the yolov8 algorithm, such as Figure 2 The backbone network model uses the Focus structure and CSP (Cross Stage Partial) network to enhance the learning ability of the model, making the model lightweight while ensuring accuracy; the neck network model uses the FPN (Feature Pyramid Network) and PAN (Path Aggregation Network) structures to perform multi-scale feature fusion on the image, which can improve the detection ability of larger and smaller targets.

[0075] The embodiment of the present invention selects the video data of the east platform of Line 13 of Xi'erqi Station of Beijing Subway from 06:30 to 11:30 on February 13, 2023 for example verification. First, the video surveillance area of ​​the east platform of Line 13 of Xi'erqi Station is divided. The division results are as follows: Figure 3 As shown;

[0076] Considering that the changes between each frame of the video are extremely small, the embodiment of the present invention extracts an image every 2 frames, which means that one second of video will generate 12 images. In the process of video copying, the complete video is edited into several 10-minute videos. Finally, the algorithm generates a data set of 86,400 images from 120 videos. The results of some algorithm implementations are shown below. Figure 4 As shown;

[0077] The passenger number change curve of the platform waiting area extracted based on video data is as follows: Figure 5 As shown, it can be used to calculate passenger flow density and congestion. As can be seen from the figure, the number of passengers will change greatly in a short period of time. This is because of the arrival of subway vehicles: there will be a large number of passengers in the waiting area before the vehicle arrives, and there will be almost no passengers in the pedestrian area at this time; after the subway arrives, the passengers in the vehicle get off or transfer, causing the passenger flow in the pedestrian area to increase instantly, while the number of passengers waiting for the subway will drop sharply because of getting on the train.

[0078] Figure 6 The passenger flow risk trend from 06:30 to 11:30 is shown. Two risk curves are drawn here based on the actual congestion duration and the predicted time. The actual congestion time is obtained through video images, and the calculation method of the predicted time d(t) is:

[0079]

[0080] In the formula, Q1(t) represents the number of people gathered on the subway platform at time t, Q2(t) is the number of people on the subway train that is about to arrive, A1 is the escalator capacity (unit: person / (m·min)); A2 is the capacity of the stairs (unit: person / (m·min)); N is the number of escalators; B is the total width of the stairs. By calculating d(t), the duration of congestion can be estimated based on the number of passengers on the platform and the number of passengers on the subway, and it can also be used to predict passenger flow risks. The "×" curve is derived from the actual time of the station video image, and the "·" curve is derived from the duration calculated by d(t). There is a large deviation between the two curves in the period from 07:45 to 10:15.

[0081] The calculation result of the congestion duration d(t) is affected by the number of passengers in the waiting area and the number of subway passengers. The more passengers there are, the longer the congestion time will be. When passengers queue up in an orderly manner to get on and off the train, the congestion duration will not be longer. Despite the large number of passengers, the passengers maintained a stable speed and were able to leave the platform in a short time, which made the actual value of the congestion duration smaller than the predicted value. Managers can set corresponding risk thresholds according to the specific situation of the station, so as to start corresponding passenger flow control measures in real time when the curve fluctuates and reaches the corresponding risk level.

[0082] System Example

[0083] According to an embodiment of the present invention, a subway platform passenger flow risk prevention and control system based on video images is provided. Figure 7 Schematic diagram of a subway platform passenger flow risk prevention and control system based on video images according to an embodiment of the present invention. Figure 7 As shown, the subway platform passenger flow risk prevention and control system based on video images according to an embodiment of the present invention specifically includes:

[0084] The data module 70 is used to obtain the surveillance video of the subway platform, input the surveillance video into the improved target detection algorithm YOLOv8 to perform human head detection, and obtain real-time passenger flow feature parameters based on the extracted video image data set, which is specifically used for:

[0085] Passenger flow density and passenger flow speed are divided into thresholds according to relevant standards of subway platforms to obtain a passenger flow density threshold range and a passenger flow speed threshold range, and real-time passenger flow density and passenger flow speed are calculated by formula 1 and formula 2. The real-time passenger flow density and passenger flow speed are normalized by formula 3 and formula 4 according to the passenger flow density threshold range and the passenger flow speed threshold range, and the passenger flow congestion degree is obtained by formula 5 based on the normalized passenger flow density and passenger flow speed, and the passenger flow congestion duration is obtained from the real-time monitoring video;

[0086] ρ={ρ(t)=ρ waiting (t)∪ρwalking (t)|t∈T} Formula 1;

[0087]

[0088] Among them, ρ(t) represents the passenger flow density at time t, T represents the entire evaluation period, and ρ waiting (t) represents the passenger flow density in the platform waiting area at time t, ρ walking (t) represents the passenger flow density in the platform pedestrian zone at time t, v(t) is the passenger flow velocity at time t, and v i (t) represents the walking speed of passenger i at time t, n is the total number of passengers in the area, ρ′(t) represents the normalized passenger flow density index at time t, ρ min represents the minimum value among density thresholds, ρ max represents the maximum value of the density threshold, v′(t) represents the normalized passenger flow speed index at time t, and v min Represents the minimum value among the speed thresholds, v max represents the maximum value of the speed threshold, c(t) represents the congestion level at time t, and c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the running area at time t;

[0089] The prevention and control module 72 is used to obtain a passenger flow risk value based on the passenger flow characteristic parameters using a pre-built subway platform passenger flow risk model, and to timely prevent and control the passenger flow risk of the subway platform through the passenger flow risk value, specifically for:

[0090] Constructing a subway platform passenger flow risk model as shown in Formula 6, and obtaining a passenger flow risk value through Formula 7 according to the passenger flow congestion degree and the passenger flow congestion duration and based on the subway platform passenger flow risk model;

[0091] Σ=(ρ,V,C,D,R) Formula 6;

[0092] R(t)=[c walking (t)·d walking (t)]+[c waiting (t)·d waiting (t)] Formula 7;

[0093] Where ρ represents the passenger flow density, V represents the passenger flow speed, C represents the degree of passenger flow congestion, D represents the duration of passenger flow congestion, R represents the passenger flow risk value, R(t) represents the passenger flow risk value at time t, and d walking (t) represents the duration of passenger congestion in the travel area at time t, d waiting (t) represents the duration of passenger congestion in the waiting area at time t, c waiting(t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the walking area at time t.

[0094] The embodiment of the present invention is a system embodiment corresponding to the above-mentioned method embodiment. The specific operations of each module can be understood by referring to the description of the method embodiment, which will not be repeated here.

[0095] Device Example 1

[0096] An embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps described in the method embodiment when executed by the processor.

[0097] Device Example 2

[0098] An embodiment of the present invention provides a computer-readable storage medium, on which a program for implementing information transmission is stored. When the program is executed by a processor, the steps described in the method embodiment are implemented.

[0099] The computer-readable storage medium in this embodiment includes, but is not limited to, ROM, RAM, magnetic disk or optical disk, etc.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for preventing and controlling passenger flow risks on subway platforms based on video images, characterized in that include: Obtain surveillance video of the subway platform, input the surveillance video into the improved target detection algorithm YOLOv8 to perform human head detection, and obtain real-time passenger flow feature parameters based on the extracted video image data set; as well as A passenger flow risk value is obtained according to the passenger flow characteristic parameters using a pre-constructed subway platform passenger flow risk model, and the passenger flow risk of the subway platform is timely prevented and controlled using the passenger flow risk value.

2. The method according to claim 1, characterized in that The improved target detection algorithm YOLOv8 is a target detection algorithm YOLOv8 that introduces an attention mechanism module including a human head detection network, wherein the target detection algorithm YOLOv8 includes a backbone network Backbone, a neck network Neck and a head network Head; Backbone includes a Focus module and a CSP module, and Neck includes an FPN structure and a PAN structure.

3. The method according to claim 1, characterized in that The passenger flow characteristic parameters include passenger flow congestion degree and passenger flow congestion duration.

4. The method according to claim 3, characterized in that The real-time passenger flow feature parameters obtained based on the extracted video image data set include: Passenger flow density and passenger flow speed are divided into thresholds according to relevant standards of subway platforms to obtain a passenger flow density threshold range and a passenger flow speed threshold range, and real-time passenger flow density and passenger flow speed are calculated by formula 1 and formula 2. The real-time passenger flow density and passenger flow speed are normalized by formula 3 and formula 4 according to the passenger flow density threshold range and the passenger flow speed threshold range, and the passenger flow congestion degree is obtained by formula 5 based on the normalized passenger flow density and passenger flow speed, and the passenger flow congestion duration is obtained from the real-time monitoring video; ρ = {ρ(t) = ρ waiting (t) ∪ ρ walking (t) | t ∈ T} Equation 1; Among them, ρ(t) represents the passenger flow density at time t, T represents the entire evaluation period, and ρ waiting (t) represents the passenger flow density in the platform waiting area at time t, ρ walking (t) represents the passenger flow density in the platform pedestrian zone at time t, v(t) is the passenger flow velocity at time t, and v i (t) represents the walking speed of passenger i at time t, n is the total number of passengers in the area, ρ′(t) represents the normalized passenger flow density index at time t, ρ min represents the minimum value among density thresholds, ρ max represents the maximum value of the density threshold, v′(t) represents the normalized passenger flow speed index at time t, and v min Represents the minimum value among the speed thresholds, v max represents the maximum value of the speed threshold, c(t) represents the congestion level at time t, and c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the walking area at time t.

5. The method according to claim 4, characterized in that The passenger flow risk value is obtained by using the pre-built subway platform passenger flow risk model according to the passenger flow characteristic parameters, specifically including: Constructing a subway platform passenger flow risk model as shown in Formula 6, and obtaining a passenger flow risk value through Formula 7 according to the passenger flow congestion degree and the passenger flow congestion duration and based on the subway platform passenger flow risk model; Σ=(ρ,V,C,D,R) Formula 6; R(t) = [c walking (t)·d walking (t)] + [c waiting (t)·d waiting (t)] Formula 7; Where ρ represents the passenger flow density, V represents the passenger flow speed, C represents the degree of passenger flow congestion, D represents the duration of passenger flow congestion, R represents the passenger flow risk value, R(t) represents the passenger flow risk value at time t, and d walking (t) represents the duration of passenger congestion in the travel area at time t, d waiting (t) represents the duration of passenger congestion in the waiting area at time t, c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the walking area at time t.

6. A subway platform passenger flow risk prevention and control system based on video images, characterized in that include: A data module is used to obtain surveillance videos of subway platforms, input the surveillance videos into the improved target detection algorithm YOLOv8 to perform human head detection, and obtain real-time passenger flow feature parameters based on the extracted video image data set; as well as The prevention and control module is used to obtain a passenger flow risk value according to the passenger flow characteristic parameters using a pre-built subway platform passenger flow risk model, and to timely prevent and control the passenger flow risk of the subway platform through the passenger flow risk value.

7. The system according to claim 6, characterized in that The data module is specifically used for: Passenger flow density and passenger flow speed are divided into thresholds according to relevant standards of subway platforms to obtain a passenger flow density threshold range and a passenger flow speed threshold range, and real-time passenger flow density and passenger flow speed are calculated by formula 1 and formula 2. The real-time passenger flow density and passenger flow speed are normalized by formula 3 and formula 4 according to the passenger flow density threshold range and the passenger flow speed threshold range, and the passenger flow congestion degree is obtained by formula 5 based on the normalized passenger flow density and passenger flow speed, and the passenger flow congestion duration is obtained from the real-time monitoring video; ρ ={ρ(t)=ρ waiting (t)∪ρ walking (t)|t∈T} Formula 1; Among them, ρ(t) represents the passenger flow density at time t, T represents the entire evaluation period, and ρ waiting (t) represents the passenger flow density in the platform waiting area at time t, ρ walking (t) represents the passenger flow density in the platform pedestrian zone at time t, v(t) is the passenger flow velocity at time t, and v i (t) represents the walking speed of passenger i at time t, n is the total number of passengers in the area, ρ′(t) represents the normalized passenger flow density index at time t, ρ min represents the minimum value among density thresholds, ρ max represents the maximum value of the density threshold, v′(t) represents the normalized passenger flow speed index at time t, and v min Represents the minimum value among the speed thresholds, v max represents the maximum value of the speed threshold, c(t) represents the congestion level at time t, and c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the walking area at time t.

8. The system according to claim 7, characterized in that The prevention and control module is specifically used for: Constructing a subway platform passenger flow risk model as shown in Formula 6, and obtaining a passenger flow risk value through Formula 7 according to the passenger flow congestion degree and the passenger flow congestion duration and based on the subway platform passenger flow risk model; Σ=(ρ,V,C,D,R) Formula 6; R(t) = [c walking (t)·d walking (t)] + [c waiting (t)·d waiting (t)] Formula 7; Where ρ represents the passenger flow density, V represents the passenger flow speed, C represents the degree of passenger flow congestion, D represents the duration of passenger flow congestion, R represents the passenger flow risk value, R(t) represents the passenger flow risk value at time t, and d walking (t) represents the duration of passenger congestion in the travel area at time t, d waiting (t) represents the duration of passenger congestion in the waiting area at time t, c waiting (t) represents the degree of congestion in the waiting area at time t, c walking (t) represents the congestion level of the walking area at time t.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the subway platform passenger flow risk prevention and control method based on video images as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by the processor, the steps of the subway platform passenger flow risk prevention and control method based on video images as described in any one of claims 1 to 5 are implemented.

Images