An intelligent transportation management system, method, electronic device and storage medium

Abstract

The application provides an intelligent transportation management system, method, electronic device and storage medium. In the system, a first camera group is positioned to capture first video data of passengers entering or leaving through an entrance or an exit of a platform; a second camera group is positioned to capture second video data of passengers boarding or alighting in the platform; a data processing center is configured to analyze the first video data and the second video data corresponding to each platform to determine the number of passengers boarding and alighting in the corresponding platform and the number of passengers entering and leaving the platform; and management data is obtained according to the above data, the management data including passenger capacity information of a vehicle, an average number of passengers on the platform, an average waiting time of passengers on the platform and a passenger transfer probability between platforms. Compared with the prior art, the application can use camera data at a limited number of allowed positions to simulate data from virtual cameras with ubiquitous coverage, so as to effectively manage the transportation system.

Description

Technical Field

[0001] This application relates to the field of transportation management technology, specifically to an intelligent transportation management system, method, electronic device, and storage medium. Background Technology

[0002] Managing intelligent transportation systems (such as buses, trains, subways, and infrastructure like stations and transfer hubs) requires administrators to have real-time information on vehicle location and movement, as well as real-time passenger counts and movement on vehicles and platforms. This information can primarily be obtained through cameras or webcams. Ideally, cameras should be deployed in as many locations as possible on vehicles and at stations to image and analyze all areas. However, deploying a large number of cameras is very expensive and may still not cover every location. In particular, the deployment locations of cameras are often limited due to regulations, privacy concerns, and other requirements.

[0003] Therefore, how to deploy a small number of cameras in permitted locations to effectively manage the transportation system is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0004] In view of this, this application provides an intelligent transportation management system, method, electronic device, and storage medium that can use camera data from a limited number of permitted locations to simulate data from ubiquitous virtual cameras for effective management of the transportation system.

[0005] The first aspect of this application provides an intelligent transportation management system, characterized in that it includes: a data processing center and a first camera group and a second camera group installed at each station along the route of the vehicle;

[0006] The first camera group is positioned to capture first video data of passengers entering or leaving through the entrance or exit of the platform;

[0007] The second camera group is positioned to capture second video data of passengers boarding or alighting within the platform;

[0008] The data processing center includes:

[0009] The first processing module is used to analyze the first video data and the second video data corresponding to each platform to determine the number of passengers getting on and off the train at the corresponding platform and the number of passengers entering and exiting the platform.

[0010] The second processing module is used to obtain management data based on the number of passengers getting on and off the train at each platform and the number of passengers entering and exiting the platform. The management data includes: vehicle passenger capacity information, average number of passengers on the platform, average waiting time for passengers on the platform, and passenger transfer probability between platforms.

[0011] In one possible implementation, the first processing module analyzes the first video data and the second video data corresponding to each platform to determine the number of passengers boarding and alighting at the corresponding platform and the number of passengers entering and exiting the platform, including:

[0012] The system analyzes and tracks passengers in the first and second video data based on a deep learning model, and counts passengers who cross specific lines of interest or regions of interest to obtain the number of passengers getting on and off the train within the platform and the number of passengers entering and exiting the platform.

[0013] In one possible implementation, the first processing module analyzes and tracks passengers in the first video data and the second video data based on a deep learning model, including:

[0014] The first processing module is specifically used for:

[0015] Human body detection or head detection is performed on passengers in the first video data and the second video data based on a deep learning model to identify all first passengers in a specified location or area.

[0016] The first passenger is continuously tracked based on the identifier assigned by the re-identification algorithm, and an online algorithm is used to verify whether the first passenger has crossed the line of interest or region of interest.

[0017] In one possible implementation, the first processing module continuously tracks the detected first passenger based on the identifier assigned by the re-identification algorithm, including:

[0018] The first processing module is specifically used for

[0019] Based on time slices or crowd heatmaps, the first detected passenger is continuously tracked according to the identifier assigned by the re-identification algorithm.

[0020] In one possible implementation, the system further includes: passenger signs installed at each station along the vehicle's route;

[0021] The passenger sign is used to receive and display management data sent by the data processing center.

[0022] A second aspect of this application provides an intelligent transportation management method, comprising:

[0023] Acquire the first video data of passengers entering or leaving each platform through the platform's entrance or exit;

[0024] Acquire second video data of passengers boarding or alighting on each platform;

[0025] The first video data and the second video data corresponding to each platform are analyzed to determine the number of passengers getting on and off the train at the corresponding platform and the number of passengers entering and exiting the platform.

[0026] Management data is obtained based on the number of passengers boarding and alighting at each platform and the number of passengers entering and exiting the platform. The management data includes: vehicle passenger capacity information, average number of passengers on the platform, average waiting time for passengers on the platform, and passenger transfer probability between platforms.

[0027] In one possible implementation, analyzing the first video data and the second video data corresponding to each platform to determine the number of passengers boarding and alighting at the corresponding platform and the number of passengers entering and exiting the platform includes:

[0028] The system analyzes and tracks passengers in the first and second video data based on a deep learning model, and counts passengers who cross specific lines of interest or regions of interest to obtain the number of passengers getting on and off the train within the platform and the number of passengers entering and exiting the platform.

[0029] In one possible implementation, the step of analyzing and tracking passengers in the first video data and the second video data based on a deep learning model includes:

[0030] Human body detection or head detection is performed on passengers in the first video data and the second video data based on a deep learning model to identify all first passengers in a specified location or area.

[0031] The first passenger is continuously tracked based on the identifier assigned by the re-identification algorithm, and an online algorithm is used to verify whether the first passenger has crossed the line of interest or region of interest.

[0032] A third aspect of this application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the second aspect of this application.

[0033] A fourth aspect of this application provides a computer-readable medium having computer-readable instructions stored thereon, the computer-readable instructions being executable by a processor to implement the method described in the second aspect of this application.

[0034] Compared to existing technologies, the intelligent transportation management system, method, electronic device, and storage medium provided in this application include a first camera group positioned to capture first video data of passengers entering or leaving through the entrance or exit of a platform; a second camera group positioned to capture second video data of passengers boarding or alighting within the platform; a data processing center used to analyze the first and second video data corresponding to each platform to determine the number of passengers boarding and alighting within the corresponding platform and the number of passengers entering and exiting the platform; and management data obtained based on the number of passengers boarding and alighting within the platform and the number of passengers entering and exiting the platform, the management data including: vehicle passenger capacity information, average number of passengers on the platform, average waiting time for passengers on the platform, and passenger transfer probability between platforms. Compared to existing technologies, this application can use camera data from a limited number of permitted locations to simulate data from ubiquitous virtual cameras for effective management of the transportation system. Attached Figure Description

[0035] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0036] Figure 1 This demonstrates a typical application scenario for ubiquitous virtual camera technology;

[0037] Figure 2 A schematic diagram of an intelligent transportation management system provided in an embodiment of this application is shown;

[0038] Figure 3 This illustrates a typical deployment scenario for a platform provided in this application;

[0039] Figure 4 This paper illustrates the overall UVC architecture provided in this application and its relationship with AI-based video analytics.

[0040] Figure 5 A flowchart of an intelligent transportation management method provided in an embodiment of this application is shown;

[0041] Figure 6 A schematic diagram of an electronic device provided by some embodiments of this application is shown. Detailed Implementation

[0042] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0043] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application pertains.

[0044] Furthermore, the terms "first" and "second," etc., are used to distinguish different objects, not to describe a specific order. Additionally, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to those processes, methods, products, or apparatuses.

[0045] This application provides an intelligent transportation management system, method, electronic device, and storage medium, which will be described below with reference to the accompanying drawings.

[0046] To facilitate understanding, the principle of the intelligent transportation management system of this application will be introduced first.

[0047] Unlike existing systems that deploy cameras in as many locations as possible on vehicles and at stations, the intelligent transportation management system of this application only requires the deployment of a small number of cameras in permitted locations to acquire video data from these few locations. From the video data captured by a limited number of cameras, this application can obtain the complete set of relevant information needed to manage the transportation system through AI-based video analysis, camera calibration, and mathematical analysis, achieving an effect similar to that provided by a camera system with extensive and comprehensive coverage. The final result is akin to having a ubiquitous virtual camera with advanced video analysis capabilities, which can be termed Ubiquitous Virtual Camera (UVC) technology.

[0048] Typical application scenarios of UVC technology include: Figure 1As shown, UVC provides comprehensive data for intelligent transportation management by processing video footage from cameras in stations and inside trains. Information from UVC can include passenger capacity information for vehicles, average number of passengers on platforms, and average passenger waiting time on platforms, among other things. This information can be used for train timetables and their updates; for monitoring, measuring, and scheduling the utilization of transportation infrastructure; and for providing up-to-date passenger guidance, such as the presentation of station signage. For example, signage can display carriage density and hotspots, and use color coding to identify the relative crowding levels in carriages so that waiting passengers can find and queue for relatively empty carriages.

[0049] The intelligent transportation management system provided in the embodiments of this application will be described in detail below.

[0050] Please refer to Figure 2 It shows a schematic diagram of an intelligent transportation management system provided in an embodiment of this application, such as... Figure 2 As shown in the embodiment of this application, the intelligent transportation management system includes: a data processing center 100 and a first camera group 210 and a second camera group 220 installed at each station along the vehicle's route. The first camera group 210 and the second camera group 220 include one or more cameras, and the number of cameras in each group is determined according to the actual situation.

[0051] The first camera group 210 is positioned to capture initial video data of passengers entering or leaving the platform through entrances or exits. For example, cameras can be positioned at entrances and exits of the platform, with the number of cameras determined by the number of entrances and exits of the platform.

[0052] The second camera group 220 is positioned to capture secondary video data of passengers boarding or alighting on the platform. For example, cameras can be placed on the platform directly opposite the carriage doors. The number of cameras can be determined based on the number of carriage doors on the entire train and the coverage area of ​​the cameras. For example, if 5 cameras can cover 10 carriage doors, then the number of cameras is 5.

[0053] Data processing center 100 includes a first processing module 110 and a second processing module 120. Each processing module has a data processing function, specifically:

[0054] The first processing module 110 is used to analyze the first video data and the second video data corresponding to each platform to determine the number of passengers getting on and off the train at the corresponding platform and the number of passengers entering and exiting the platform.

[0055] The second processing module 120 is used to obtain management data based on the number of passengers getting on and off the train at each platform and the number of passengers entering and exiting the platform. The management data includes: vehicle passenger capacity information, average number of passengers on the platform, average waiting time of passengers on the platform and passenger transfer probability between platforms.

[0056] In actual deployment, a limited number of cameras can generally be deployed only in designated locations, for example:

[0057] Location 1: Deploy cameras at each platform entrance / exit to count the number of passengers entering and leaving the platform.

[0058] Location 2: Deploy cameras on the platform side facing the train tracks to keep track of the number of people getting on and off the train.

[0059] A typical deployment scenario is as follows Figure 3 As shown in the diagram, the platform includes two entrances / exits (A and B in the diagram), a passenger waiting area, and two tracks. Passengers can enter or leave the passenger waiting area from the entrances / exits, or board or alight from a stopping train and enter the passenger waiting area. Cameras (which can be commercially available standard surveillance cameras) are deployed on the platform and installed in specific locations to capture video data. Here, two sets of cameras are deployed at the platform entrances / exits to keep track of the number of passengers entering and exiting the platform; another two sets of cameras are deployed on either side of the passenger waiting area facing the train tracks to keep track of the number of passengers boarding and alighting from trains.

[0060] Next, we will introduce what information the intelligent transportation management system needs, and then discuss how to obtain this information based on the first video data and the second video data.

[0061] It is worth mentioning that some stations have multiple platforms that converge together. In such cases, they can be separated into independent platforms with their own tracks.

[0062] A station contains multiple platforms P (k) k = 0, 1, ..., K. At platform P (k) There are two tracks, X (k) and Y (k) They are distributed on both sides of the platform. For clarity, uppercase letters represent unknown quantities, and lowercase letters represent known quantities. The quantities may be related to time t: t0, t1, ..., t i ... or associated with the time interval [t, t+Δt].

[0063] The following information is required for system administration and operation:

[0064] 1. At a certain time interval [t] i,t i+1 The number of passengers entering and exiting the station is represented by E. in (t i ,t i+1 ) and E out (t i ,t i+1 );

[0065] 2. At platform P (k) The information within k = 1, ..., K is as follows:

[0066] 2.a) At a certain time interval [t] i ,t i+1 The number of passengers entering and exiting the platform is expressed as follows: and This does not include passengers getting on and off the bus;

[0067] 2.b) The number of passengers waiting on the platform at time t, t:Z (k) (t);

[0068] 2.c) The passenger's waiting time at time t, t:W (k) (t);

[0069] 3. When train L (i) Arrive at platform P at time t (k) The number of passengers boarding and alighting is calculated at time t+Δt when the train departs. This application does not assume that there are cameras on the train to monitor the number of passengers. Instead, this application monitors the changes in the number of passengers on each platform as the train passes by and estimates the number of passengers on each train.

[0070] 3.a) During the time interval [t, t + Δt], from the platform track X (k) / Y (k) The number of passengers boarding from the side is expressed as follows:

[0071] 3.b) During the time interval [t, t+Δt], on track X of the platform... (k) / Y (k) The number of passengers alighting from the side is expressed as follows:

[0072] 3.c) The number of passengers on the train that passes through the platform at time t is expressed as: L (i) (t);

[0073] 4. From platform P (k) Transfer to platform P (r) The number of passengers is represented as: C k,r ,k,r=0,1,...,K.

[0074] From the above four types of information, we can deduce the following information, such as the average number of passengers on the platform, the average waiting time for passengers on the platform, and the probability of passenger transfer between platforms.

[0075] Clearly, the information collected by the cameras is still very limited and cannot provide all the information required by the intelligent transportation management system listed above. To fill these significant gaps, two challenging tasks need to be completed: Task 1, deriving the information listed above as items 1 and 2 from the first and second video data; and Task 2, calculating all the other information listed above from the results obtained in Task 1.

[0076] Task 1: Obtain the required number of passengers from the first and second video data using AI-based video analysis technology, corresponding to the function of the first processing module 110. Of course, ensuring accurate counting is not easy.

[0077] Task 2: Based on the results of Task 1, perform camera calibration and mathematical analysis to obtain all other necessary information listed, corresponding to the function of the second processing module 120.

[0078] All the details here are transparent to the user. From the user's perspective, it's as if there's a continuous virtual camera that can cover any part of the train and station—a ubiquitous virtual camera (UVC).

[0079] In some implementations, the first processing module 110 analyzes and tracks passengers in the first video data and the second video data based on a deep learning model, and counts passengers who cross specific lines of interest or regions of interest to obtain the number of passengers getting on and off the train within the platform and the number of passengers entering and exiting the platform.

[0080] Specifically, the first processing module 110 analyzes and tracks passengers in the first video data and the second video data based on a deep learning model, including: performing human body detection or head detection on passengers in the first video data and the second video data based on a deep learning model to identify all first passengers in a specified location or area; continuously tracking the detected first passengers according to the identifier assigned by the re-identification algorithm; and using an online algorithm to verify whether the first passengers have crossed the line of interest or region of interest.

[0081] To improve accuracy, the first processing module 110 continuously tracks the detected first passenger based on the identifier assigned by the re-identification algorithm, specifically including: continuously tracking the detected first passenger based on the identifier assigned by the re-identification algorithm according to the time slice or crowd heat map.

[0082] Specifically, the AI-based video analysis process of the first processing module 110 in this application is as follows:

[0083] Cameras can be deployed at locations 1 and 2 on the platform to collect video data. The video collected by these cameras is used for AI-based video analysis to determine the number of passengers.

[0084] To calculate passenger numbers—the number of people entering / leaving the platform or boarding / alighting—this application uses computer vision and deep neural networks (DNNs) to track passengers crossing lines of interest (LOIs) or regions of interest (ROIs). This process consists of two stages:

[0085] Phase 1: Perform human or head detection based on a deep learning (DL) model (e.g., Faster RCNN, Mask RCNN, SSD, or YOLO) to identify all passengers in a specified location or area.

[0086] Phase 2: Based on the identifier IDs assigned by the re-identification algorithm and the distances between objects in the video, passengers detected in Phase 1 are continuously tracked so that the online algorithm can verify whether passengers have crossed the LOI or ROI. This allows for real-time acquisition of passenger numbers entering / leaving the platform or boarding / alighting trains. To improve accuracy, other supplementary methods can be used, including time slicing and crowd heatmaps.

[0087] For an overview of the UVC architecture and its relationship to AI-based video analytics, please refer to [link / reference]. Figure 4 UVC can be provided and deployed in the following ways: 1) a dedicated server at the customer's site; 2) a computing center with shared servers; or 3) computing resources rented from the cloud.

[0088] Next is the mathematical analysis for Task 2, which is that after obtaining the number of passengers from the video analysis, the second processing module 120 derives and calculates all the other listed information.

[0089] Specifically, the analysis process of the second processing module 120 in this application is as follows:

[0090] First, information that can be directly obtained from the results of Task 1, or that can be directly calculated, is derived. For clarity, this application lists the corresponding sequence number in the above description before each piece of information.

[0091] For a station, in the time interval (t) i ,t i+1 The number of passengers entering and exiting the station were E in (t i ,t i+1 ) and E out (t i,t i+1 These two pieces of information can be automatically obtained from the entry and exit turnstiles. Therefore,

[0092] 1:E in (t i ,t i+1 ):=e in (t i ,t i+1 ) and E out (t i ,t i+1 ):=e out (t i ,t i+1 ).

[0093] For entering and exiting platform P (k) The number of passengers can be obtained directly through UVC's Task 1. That is,

[0094] 2.a:

[0095] A train passes through platform P (k) The arrival time is t i The departure time is t i +Δt i The arrival time of the next train is t. i+1 Satisfying Δt i <t i+1 -t i When a train travels at a time interval (t... i ,t i +Δt i ) Passing through orbit X (k) At that time, the number of passengers getting on and off the bus is also obtained directly through UVC's Task 1. That is,

[0096] 3.a:

[0097] 3.b:

[0098] Therefore, the change in the number of passengers on the train is as follows:

[0099]

[0100] Now, let's estimate the number of passengers on the platform. Assume this information cannot be obtained and tracked via cameras. In reality, although this information can be derived to some extent using techniques such as object heatmaps and crowd counting, these techniques are difficult to execute due to occlusion and low resolution, often resulting in inaccurate results.

[0101] At platform P (k)Above, time interval (t) i ,t i+1 Within the station, changes in passenger numbers depend on the following counts: the number of passengers boarding and alighting, and the number of passengers entering and exiting the platform. That is,

[0102]

[0103] Therefore, at time t n Platform P (k) The number of passengers on board is:

[0104] 2.b:

[0105] Among them, Z (k) (t0) = 0, that is, at time t0, when the station just opened, the number of passengers on the platform is 0;

[0106] Now, let's deduce the number of passengers on the train as it passes through a series of platforms at different stations. It is assumed that the train's cameras will not be used to record the number of passengers. Instead, this application obtains this information via UVC using the following method.

[0107] Assume train L (i) At time t0, equipped A number of passengers leave the origin station platform k0 together, and arrive at times t1, t2, ..., t3 respectively. n The route passes through a series of stations, including platforms k1, k2, ..., k. n According to equation (1), at time t before getting on and off the bus... n The number of passengers on the train is:

[0108] 3.c:

[0109] in,

[0110] When a train passes the other side of track Y (k) The same approach can be used to handle such situations.

[0111] In equation (3), we obtain the result at any time t. i Platform P (k) Number of passengers on board z (k) (t i ), i = 0, 1, ... Therefore, in the time interval [t u ,t v Within the station, the average number of passengers on the platform is:

[0112] 2.b'

[0113] For integrals like 2.b', there are several methods for numerical calculation, such as the Gaussian summation integral method. Note that a sufficiently high sampling rate is required to ensure accuracy. More importantly, sampling needs to be performed at equal time intervals—regardless of train arrival times, because passengers are constantly entering and exiting the platform between the arrival times of two trains.

[0114] Similarly, in the time interval [t] u ,t v Inside, besides passengers disembarking, those entering platform P... (k) The average number of passengers is:

[0115] 1'

[0116] And in the time interval [t] u ,t v Inside, besides passengers boarding the train, those leaving platform P... (k) The average number of passengers is:

[0117] 1'

[0118] And in the time interval [t] u ,t v Inside, from platform P (k) The average number of passengers on both sides of the track (if time t) v (Passengers boarding at this time are not included) are:

[0119] 3.a'

[0120] Where Δt i ≤t i+1 -t i ,i=0,1,.... When only one train is in the time interval (t i ,t i +t i When passing through the platform, or or

[0121] Please note that the above information will only change when the train arrives. Furthermore, this application does not use integrals to calculate the average quantity.

[0122] Similarly, in the time interval [t] u ,t v Inside, from platform P (k) The average number of passengers getting off the train on both sides of the track is:

[0123] 3.b'

[0124] Where, Δt i ≤t i+1 -t i ,i=0,1,.... When only one train is in the time interval (t i ,t i +Δt i When passing through the platform, or or

[0125] Now, let's examine a time interval (t) u ,t v Within v≥u+1, at platform P (k) The average waiting time of passengers on the platform. An obvious approach is to track each passenger as they enter the platform and record their waiting time. However, this is impractical in real-world implementation because tracking each individual passenger in a crowded situation is essentially infeasible. The method in this application differs from this; the algorithm relies on information from UVC, specifically at time t... i ,i=0,1,...,n. The number of passengers on the platform, please refer to equation (3).

[0126] At time t u , t u+1 Platform P (k) The number of passengers on board are z (k) ( u ),z (k) ( u+1 Thus, at time t u+1 On average The passengers waited for t u+1 -t u The time; and their total waiting time was At time t u+1 The number of passengers on the platform is And at the time when the next train arrives... u+2 The previous total waiting time was However, among them Passengers from time t u And so the wait began. The new passenger number was... Assume a passenger waits for at most two trains before boarding, and consider the time t... i i = u, u+1, ..., v, repeating the above process, then at time t u to t v The total number of people waiting on the platform is:

[0127]

[0128] Their total waiting time is The average waiting time for each passenger is:

[0129] 2.c'

[0130] in,

[0131] The goal is to estimate the number of passengers transferring between platforms within a station. This information could also be obtained by identifying and tracking each passenger. However, this is impractical for large numbers of passengers. Therefore, this application uses an approximate algorithm to calculate transfer statistics, which only requires the information from Task 1 of UVC.

[0132] For simplicity, the set of all station entrances is represented as an abstract platform P. (0) The number of passengers passing through this platform can be obtained from the ticket gates. In this way, the number of passengers entering and exiting the station can always be obtained as E. in :=e in E out :=e out When the entrance is used as platform P (0) hour,

[0133] Suppose a passenger starts from platform P (k) Arrive at platform P (r) The probability of transferring is P. k,r ,k,r=0,1,...,K,and Therefore, from platform P (k) Transfer to platform P (r) The number of passengers is Among them, C i,i =P i,i =0, i=0,1,...,K. Note P k,k =0,k=0,1,...,K.

[0134] For the abstract platform P at the station entrance (0) ,

[0135]

[0136]

[0137] For platform P (k) K = 1, 2, ..., K

[0138]

[0139] that is,

[0140]

[0141] The following is a calculation of the number of passengers transferring between platforms.

[0142]

[0143] The key here is to estimate the probability distribution P. k,r ,k,r=0,1,...,K. However, these probability distributions are often difficult to obtain, and they are also dynamically changing. This application will estimate these probability distributions using an approximate algorithm. For values ​​that change over time, this application introduces a time parameter, thus obtaining:

[0144]

[0145]

[0146]

[0147] Substituting equation (4) into equation (5), we get:

[0148]

[0149] Now, the problem is simplified to solving equation (6), where P k,r =0,r,k =0,1,...,K. This is a subset of (K+1). 2 A 2(K+1) linear equation with 3 variables and a time parameter. Many techniques exist for solving such problems, including: 1) maximum likelihood estimation; 2) time series analysis; 3) machine learning (ML) and deep neural networks (DNN); and 4) linear programming. This application describes an approximation method.

[0150] In equation (6), the number of variables is greater than the number of equations. However, this application has different time points t. i Data is collected for i = 1, 2, ..., n, resulting in (K+1)(n+1) equations. This application sets (n+1)(K+1) = (K+1) / (n+1). 2 Thus, we can obtain n = K. That is, in order to estimate the passenger transfer flow between K platforms, we need to collect K times the number of passengers entering / leaving at each platform. In other words, the collection equation (6) at time t... i Given data where i = 1, 2, ..., K, we ultimately obtain (K+1). 2 A linear equation and variables are used to solve for the probability distribution P. k,r ,k,r=0,1,...,K.

[0151] Note that the sampling time here is not necessarily the arrival / departure time of the vehicle. This process can be repeated periodically to refine and update the probability distribution that changes over time.

[0152] The second processing module 120 can obtain the passenger capacity information of the vehicle, the average number of passengers on the platform, the average waiting time of passengers on the platform, and the passenger transfer probability between platforms, etc., through the above basic calculations, estimations, and approximate calculations, as well as the management data required by the intelligent transportation management system.

[0153] In some embodiments, the intelligent transportation management system provided in this application may further include: passenger signs installed at each station along the vehicle's route; these passenger signs are used to receive and display management data sent by the data processing center 100. This management data can be used for train timetables and their updates; for monitoring, measuring, and scheduling the utilization of transportation infrastructure; and for providing up-to-date passenger guidance, such as the presentation of station signs. For example, the signs may display the density and hotspots of the carriages and apply color coding to identify the relative crowding of passengers in the carriages, so that passengers waiting to board can find and queue for relatively empty carriages.

[0154] In the above embodiments, an intelligent transportation management system is provided. Correspondingly, this application also provides an intelligent transportation management method. The intelligent transportation management method provided in this application can be applied to the above-mentioned intelligent transportation management system. This intelligent transportation management method can be implemented through software, hardware, or a combination of both. For example, the intelligent transportation management method may include integrated or separate functional modules or units to perform the corresponding steps in the above methods. Please refer to... Figure 5 The diagram illustrates a flowchart of an intelligent transportation management method provided by some embodiments of this application. Since the method embodiments are substantially similar to the system embodiments, the description is relatively simple; relevant details can be found in the descriptions of the system embodiments. The method embodiments described below are merely illustrative.

[0155] like Figure 5 As shown, the intelligent transportation management method may include:

[0156] S101. Obtain the first video data of passengers entering or leaving through the entrance or exit of each platform.

[0157] S102. Obtain the second video data of passengers boarding or alighting on each platform.

[0158] S103. Analyze the first video data and the second video data corresponding to each platform to determine the number of passengers getting on and off the train at the corresponding platform and the number of passengers entering and exiting the platform.

[0159] S104. Management data is obtained based on the number of passengers getting on and off the train at each platform and the number of passengers entering and exiting the platform. The management data includes: vehicle passenger capacity information, average number of passengers on the platform, average waiting time for passengers on the platform, and passenger transfer probability between platforms.

[0160] In some embodiments, step S103 specifically includes:

[0161] The system analyzes and tracks passengers in the first and second video data based on a deep learning model, and counts passengers who cross specific lines of interest or regions of interest to obtain the number of passengers getting on and off the train within the platform and the number of passengers entering and exiting the platform.

[0162] In some embodiments, the above steps are based on deep learning models to analyze and track passengers in the first video data and the second video data, specifically including:

[0163] Human body detection or head detection is performed on passengers in the first video data and the second video data based on a deep learning model to identify all first passengers in a specified location or area.

[0164] The first passenger is continuously tracked based on the identifier assigned by the re-identification algorithm, and an online algorithm is used to verify whether the first passenger has crossed the line of interest or region of interest.

[0165] In some embodiments, the above steps of continuously tracking the detected first passenger based on the identifier assigned by the re-identification algorithm specifically include: continuously tracking the detected first passenger based on the identifier assigned by the re-identification algorithm according to a time slice or a crowd heatmap.

[0166] The intelligent transportation management method provided in this application is based on the same inventive concept and has the same beneficial effects as the intelligent transportation management system provided in the foregoing embodiments of this application.

[0167] This application also provides an electronic device corresponding to the intelligent transportation management method provided in the foregoing embodiments. The electronic device may be an electronic device for a client, such as a mobile phone, laptop computer, tablet computer, desktop computer, etc., to execute the above-mentioned intelligent transportation management method.

[0168] Please refer to Figure 6 This illustrates a schematic diagram of an electronic device provided by some embodiments of this application. For example... Figure 6As shown, the electronic device 20 includes: a processor 200, a memory 201, a bus 202, and a communication interface 203. The processor 200, the communication interface 203, and the memory 201 are connected via the bus 202. The memory 201 stores a computer program that can run on the processor 200. When the processor 200 runs the computer program, it executes the intelligent transportation management method provided in any of the foregoing embodiments of this application.

[0169] The memory 201 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 203 (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.

[0170] Bus 202 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory 201 is used to store programs. After receiving an execution instruction, the processor 200 executes the program. The intelligent transportation management method disclosed in any of the foregoing embodiments of this application can be applied to the processor 200, or implemented by the processor 200.

[0171] The processor 200 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 200 or by instructions in software form. The processor 200 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 201. The processor 200 reads the information in memory 201 and, in conjunction with its hardware, completes the steps of the above method.

[0172] The electronic device provided in this application embodiment and the intelligent transportation management method provided in this application embodiment are based on the same inventive concept and have the same beneficial effects as the methods they adopt, operate or implement.

[0173] This application also provides a computer-readable medium corresponding to the intelligent transportation management method provided in the foregoing embodiments, which stores a computer program (i.e., a program product) thereon. When the computer program is run by a processor, it executes the intelligent transportation management method provided in any of the foregoing embodiments.

[0174] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0175] The computer-readable storage medium provided in the above embodiments of this application and the intelligent transportation management method provided in the embodiments of this application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application stored therein.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application.

Claims

1. An intelligent transportation management system, characterized in that, include: The data processing center, as well as the first and second camera groups installed at each station along the vehicle's route; The first camera group is positioned to capture first video data of passengers entering and leaving through the entrance or exit of the platform; The second camera group is positioned to capture second video data of passengers boarding and alighting within the platform; The data processing center includes: The first processing module is used to analyze the first video data and the second video data corresponding to each platform to determine the number of passengers getting on and off the train at the corresponding platform and the number of passengers entering and exiting the platform. The second processing module is used to obtain management data based on the number of passengers getting on and off the vehicle within the platform and the number of passengers entering and exiting the platform. The management data includes: vehicle passenger capacity information, average number of passengers on the platform, average waiting time of passengers on the platform, and passenger transfer probability between platforms. The passenger transfer probability between the platforms is calculated using the following equation: ； in, Indicates a passenger from the platform platform The probability of a transfer, i.e., the passenger transfer probability; Indicates time t from platform The number of departing passengers, Indicates entering the platform at time t. The number of passengers.

2. The intelligent transportation management system according to claim 1, characterized in that, The first processing module analyzes the first video data and the second video data corresponding to each platform to determine the number of passengers boarding and alighting at the corresponding platform and the number of passengers entering and exiting the platform, including: The system analyzes and tracks passengers in the first and second video data based on a deep learning model, and counts passengers who cross lines of interest or regions of interest to obtain the number of passengers getting on and off the train at the platform and the number of passengers entering and exiting the platform.

3. The intelligent transportation management system according to claim 2, characterized in that, The first processing module analyzes and tracks passengers in the first video data and the second video data based on a deep learning model, including: The first processing module is specifically used for: Human body detection or head detection is performed on passengers in the first video data and the second video data based on a deep learning model to identify all first passengers in a specified location or area. The first passenger is continuously tracked based on the identifier assigned by the re-identification algorithm, and an online algorithm is used to verify whether the first passenger has crossed the line of interest or region of interest.

4. The intelligent transportation management system according to claim 3, characterized in that, The first processing module continuously tracks the detected first passenger based on the identifier assigned by the re-identification algorithm, including: The first processing module is specifically used for Based on time slices or crowd heatmaps, the first detected passenger is continuously tracked according to the identifier assigned by the re-identification algorithm.

5. The intelligent transportation management system according to claim 1, characterized in that, The system also includes passenger signs installed at each station along the route the vehicle travels; The passenger sign is used to receive and display management data sent by the data processing center.

6. An intelligent transportation management method, characterized in that, include: Acquire the first video data of passengers entering and leaving each platform through the platform's entrance or exit; Acquire second video data of passengers boarding and alighting at each platform; The first video data and the second video data corresponding to each platform are analyzed to determine the number of passengers getting on and off the train at the corresponding platform and the number of passengers entering and exiting the platform. Management data is obtained based on the number of passengers getting on and off at each platform and the number of passengers entering and exiting the platform. The management data includes: vehicle passenger capacity information, average number of passengers on the platform, average waiting time for passengers on the platform, and passenger transfer probability between platforms. The passenger transfer probability between the platforms is calculated using the following equation: ； in, Indicates a passenger from the platform platform The probability of a transfer, i.e., the passenger transfer probability; Indicates time t from platform The number of departing passengers, Indicates entering the platform at time t. The number of passengers.

7. The intelligent transportation management method according to claim 6, characterized in that, The analysis of the first video data and the second video data corresponding to each platform to determine the number of passengers boarding and alighting at the corresponding platform and the number of passengers entering and exiting the platform includes: The system analyzes and tracks passengers in the first and second video data based on a deep learning model, and counts passengers who cross lines of interest or regions of interest to obtain the number of passengers getting on and off the train at the platform and the number of passengers entering and exiting the platform.

8. The intelligent transportation management method according to claim 7, characterized in that, The method of analyzing and tracking passengers in the first video data and the second video data based on a deep learning model includes: Human body detection or head detection is performed on passengers in the first video data and the second video data based on a deep learning model to identify all first passengers in a specified location or area. The first passenger is continuously tracked based on the identifier assigned by the re-identification algorithm, and an online algorithm is used to verify whether the first passenger has crossed the line of interest or region of interest.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method as described in any one of claims 6 to 8.

10. A computer-readable medium, characterized in that, It stores computer-readable instructions that can be executed by a processor to implement the method as described in any one of claims 6 to 8.

Images