Deep neural network online integration method for real-time prediction of pedestrian flow

Abstract

The invention discloses a deep neural network online integration method for real-time pedestrian flow prediction, which comprises the following steps: firstly, initializing a pedestrian flow prediction model based on a deep neural network by using offline data, then deploying the model online to predict pedestrian flow, and collecting online data to update the model at the same time; and the model is enabled to quickly adapt to a real-time changing complex scene. According to the multi-resolution updating strategy, the deep neural network is divided into a plurality of sub-modules, and historical data with different resolutions are adopted to train different sub-modules, so that the pedestrian flow prediction capabilities of the sub-modules are diversified; the online integration strategy dynamically selects and combines the sub-module which is most suitable for the current scene according to the change of real-time data distribution, and generates a real-time prediction result. Compared with the prior art, the method has the advantages that the precision and response speed of real-time pedestrian flow prediction are remarkably improved, and the method is particularly suitable for traffic monitoring, public safety management and other scenes needing real-time data processing.

Description

Technical Field

[0001] The present invention relates to an innovative method for real-time analysis and prediction of pedestrian flow data, specifically to a deep neural network online integration method for real-time prediction of pedestrian flow, and to the technical field of intelligent prediction and data analysis. Background Art

[0002] In modern society, real-time prediction of pedestrian flow data plays a key role in many fields. For transportation hubs such as airports, railway stations, and subway stations, accurate pedestrian flow prediction helps optimize operational scheduling, reasonably arrange security inspection channels, allocate transportation resources, improve passenger travel experience, and ensure safe and efficient operations; for commercial places such as shopping malls, supermarkets, and scenic spots, merchants can adjust business hours, number of employees, and product display layout based on pedestrian flow predictions to improve service quality and operating efficiency; in terms of public safety management, real-time monitoring and prediction of pedestrian flow in public places such as squares, parks, and large-scale event sites can help relevant departments formulate safety precautions in advance and prevent accidents such as crowding and trampling.

[0003] Traditional pedestrian flow prediction methods mainly include historical average method, time series analysis method (such as ARIMA model, etc.) and simple machine learning models (such as linear regression, decision tree, etc.). The historical average method and simple time series analysis method can only capture linear laws, and it is difficult to deal with the complex nonlinear relationship of pedestrian flow data caused by special events, holidays, weather changes, etc. When traditional machine learning models process large-scale, high-dimensional pedestrian flow data, the model's expression ability is limited, and it is unable to mine deep-level features. In addition, it lacks online learning and adaptive update capabilities, and it is difficult to adapt to environmental changes caused by the opening of new lines at transportation hubs, promotional activities in commercial places, and temporary activities in public places in real scenarios, causing the prediction accuracy to decrease over time.

[0004] Although deep learning technology has achieved remarkable results in the field of data analysis and prediction, and deep neural networks can automatically extract complex nonlinear features, existing pedestrian flow prediction methods based on deep learning are mostly based on offline training models and cannot adapt to changes in data distribution in real time. When the statistical laws of pedestrian flow data change due to external factors (such as public emergencies, traffic control, seasonal changes, etc.), the performance of offline training models will drop significantly, and retraining and deployment are time-consuming and laborious, which cannot meet the needs of scenarios with high real-time requirements. Therefore, there is an urgent need for a pedestrian flow prediction method that can adapt to environmental changes in real time and update the model online to improve the accuracy and timeliness of the prediction. Summary of the invention

[0005] Purpose of the invention: In view of the problems and shortcomings in the prior art, the present invention provides a deep neural network online integration method for real-time prediction of pedestrian flow, which enables the prediction system to quickly adapt to complex and changeable actual scenarios and achieve high-precision and high-response real-time prediction of pedestrian flow. By designing a multi-resolution update strategy and an online integration strategy, the problem that traditional methods are difficult to cope with dynamic changes in data distribution is solved, the generalization ability and prediction performance of the model in different environments are improved, and reliable decision support is provided for related fields of pedestrian flow management.

[0006] Technical solution: A deep neural network online integration method for real-time pedestrian flow prediction, including a multi-resolution update strategy and an online integration strategy. First, a pedestrian flow prediction model based on a deep neural network is initialized using offline data, and then the model is deployed to predict pedestrian flow, while collecting online data to update the model.

[0007] The multi-resolution update strategy refers to dividing the deep neural network into multiple sub-modules, using historical data of different resolutions to train different sub-modules, so that the pedestrian flow prediction capabilities of the sub-modules are diverse.

[0008] The online integration strategy refers to dynamically selecting and combining sub-modules that are most suitable for the current scenario according to changes in real-time data distribution to generate real-time prediction results.

[0009] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned deep neural network online integration method for real-time prediction of pedestrian flow are implemented.

[0010] A computer-readable storage medium stores a computer program for executing the deep neural network online integration method for real-time pedestrian flow prediction as described above.

[0011] Beneficial effects: The real-time pedestrian flow prediction method of the present invention has significant advantages in terms of accuracy and response speed. Through the collaboration of online integration strategy and multi-resolution update strategy, the model can quickly adapt to the dynamic changes of pedestrian flow data in complex scenarios, effectively capture nonlinear characteristics and long-term trends, provide more accurate prediction results, and reduce the computational complexity of model updates. In traffic monitoring scenarios, it can timely predict the peak and trough of passenger flow at transportation hubs, help traffic management departments to reasonably arrange transportation capacity, optimize diversion plans, reduce passenger congestion waiting time, and improve traffic operation efficiency; in the field of public safety management, it can provide early warning of crowded public places, help relevant departments formulate emergency plans, strengthen safety precautions, and reduce accident risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 An overall workflow diagram for real-time prediction of pedestrian flow data according to an embodiment of the present invention;

[0013] Figure 2 A detailed workflow diagram of offline initialization for real-time prediction of pedestrian flow data according to an embodiment of the present invention;

[0014] Figure 3 A detailed workflow diagram for online updating of real-time prediction of pedestrian flow data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The present invention is further explained below in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, various equivalent forms of modifications to the present invention by those skilled in the art all fall within the scope defined by the claims attached to this application.

[0016] The deep neural network online integration method for real-time pedestrian flow prediction is used to complete the online pedestrian flow prediction task, including a multi-resolution update strategy and an online integration strategy. First, a pedestrian flow prediction model based on a deep neural network is initialized using offline data, and then the model is deployed to predict pedestrian flow, while collecting online data to update the model. The structures of the deep neural network models that can be selected include MLP, ResNet, UNet, LSTM, Transformer and RWKV or a multi-modal structure composed of multiple structures.

[0017] The multi-resolution update strategy refers to dividing the deep neural network into multiple sub-modules, using historical data of different resolutions to train different sub-modules, so that the pedestrian flow prediction capabilities of the sub-modules are diverse.

[0018] The online integration strategy refers to dynamically selecting and combining the sub-modules that best adapt to the current scenario based on changes in real-time data distribution to generate real-time prediction results.

[0019] Use offline data to initialize a pedestrian flow prediction model based on a deep neural network. The specific steps are as follows:

[0020] Step 101: Collect offline data sets for pedestrian flow prediction. in, is a d-dimensional real number space, is x i The set of all possible values ​​of x i It is a feature used to predict pedestrian flow. It has d dimensions, each of which is a real value. The data has m 0 y i ∈{1,…,K},y iis x i There are K types of corresponding real pedestrian flows.

[0021] Step 102: Using offline dataset S 0 Initialize pedestrian flow prediction model h 0 (x) = h(x; Φ 0 ,W 0 )=

[0022] w(φ(x;φ 0 );W 0 );Model Map the features of d-dimensional real numbers to the confidence of K-dimensional real numbers. The category predicted by the pedestrian flow prediction model is the category corresponding to the dimension with the highest confidence. If there are multiple dimensions with the highest confidence, one of them is selected; each model consists of two parts. It is a deep neural network model that maps the features of d-dimensional real numbers to the Hilbert space representation of d′-dimensional real numbers, where is the deep neural network parameter, with a total of n real number parameters; is a linear classification model that maps the Hilbert space representation of d′-dimensional real numbers to the confidence of K-dimensional real numbers, where It is the linear classification model parameter, which is a d′×K real matrix.

[0023] Step 103: Use gradient descent technology to ensure that the pedestrian flow prediction model h initialized in step 102 0 For the offline dataset S 0 and the loss function e converges; where the loss function It is used to measure the output h of the pedestrian flow prediction model. 0 (x i ) and the actual pedestrian flow y i The difference between them is then used to train and update Φ and W; convergence refers to the training process Φ 0 , Refers to the training process Φ 0 and W 0 Cannot initialize the prediction model h 0 In the offline dataset S 0 The average loss on the dataset is significantly reduced. Gradient descent techniques include SGD and Adam. The available loss functions include Logistic loss function, Sigmoid loss function, negative log-likelihood, and KL Divergence.

[0024] Multi-resolution update strategy, the specific steps are:

[0025] Step 201: Divide the deep neural network into N submodules, and use mask M for the jth submodule jIndicates; mask M j ∈

[0026] {0,1} n is an n-dimensional array consisting only of 0s and 1s. If M j The i-th M j [i] is 1, indicating that the j-th submodule contains the i-th real parameter of the deep neural network, otherwise it does not contain it; for each i∈{1,…,n}, ∑ j∈{1,…,N} M j [i] = 1, which means that the L submodules should be mutually exclusive and cover all parameters of the deep neural network; appropriate division of submodules It is necessary to properly select the value of N, take N∈{4,5,6}, and properly allocate each submodule M j The number of parameters covered, for each j∈{1,…,N}, This means that the number of parameters covered by each submodule is roughly the same.

[0027] Step 202: Using historical data of different resolutions to train different submodules refers to using continuous interval I j =

[0028] [s j ,t-1] historical data training submodule j, interval I j The starting point is s j , interval I j The termination endpoint is (t-1), where t is the number of times the model predicts and collects online data after the pedestrian flow prediction model is deployed online; s j The larger the value is, the shorter the interval of historical data used by the jth submodule is and the higher the resolution is; appropriately distribute historical data The resolution of refers to the resolution of the historical data used by different submodules to be differentiated as much as possible. When t ≥ N, for each j ∈ {1, ..., N}, or It means that the length of the interval to which the historical data used by the j-th sub-module belongs is an exponential function of j or a linear function of j.

[0029] Step 203: Diversify the pedestrian flow prediction capabilities of the submodules, which refers to the appropriate division of the submodules in step 201. and the appropriate resolution of the historical data in step 202

[0030] The online integration strategy includes the following specific steps:

[0031] Step 301: For t = {1, ..., ∞}, at the tth online deployment, according to the existing deep neural network parameters Φ t-1 and the linear classification model parameters W t-1 , deploy model h t-1 (x) = h(x; φ t-1 ,W t-1 ) is used to predict the pedestrian flow in the future, denoted as The corresponding pedestrian flow, and then collect the real pedestrian flow data to form an online dataset Based on S t By updating the existing Φ t-1 and W t-1 To obtain a Φ that is more suitable for the current scene t and W t , and used for the next online deployment.

[0032] Step 302: During the tth online deployment, according to S t Constructing deep neural network optimization objectives that do not rely on linear classification models Refers to L t It can measure whether the deep neural network is suitable for the scenario of the tth online deployment based on the change of real-time data distribution without relying on the linear classification model, and the sampling disturbance coefficient r t ~U([0,1]) refers to uniformly sampling a real number r from the interval [0,1] t , solve the gradient of the deep neural network after perturbation Among them, λ is a real parameter used to eliminate the dependence on the linear classification model; Δ t-1 is the update amount of the deep neural network generated by the online integration strategy during the t-1th online deployment. When t=1, Δ t-1 Each element of is zero, when t>1, Δ t-1 The specific calculation method of is as described in step 305.

[0033] Step 303: During the t-th online deployment, for each j∈{1,…,N}, calculate the submodule M j Update volume Refers to the interval I j =[s j ,t-1] on the deep neural network and the historical update amount of the submodule to calculate the update amount of the submodule when it is deployed online for the tth time; where Δ τ,j is the jth submodule M j The update amount during the τth online deployment, τ = s j When τ,j Each element of is zero, τ>s jWhen τ,j Calculate recursively according to step 303.

[0034] Step 304: During the t-th online deployment, for each j∈{1,…,N}, calculate the j-th submodule M j The weight of the adaptability of the scenario for the tth online deployment p t,j The larger the value, the jth submodule M j The more suitable it is for the scenario of the tth online deployment; t,j In the calculation formula, The function ω(R,C) is defined as follows,

[0035] The function ψ(R,C) is defined as follows: when R and C are both zero, ψ(R,C) = 1; when R and C are not both zero,

[0036] Step 305: Calculate the update amount of the deep neural network during the t-th online deployment And used to update the deep neural network Φ t =Φ t-1 +Δ t , refers to the dynamic selection and combination of submodules that best adapt to the current scenario, and the integration of the deep neural network parameters Φ that adapt to the scenario of the tth online deployment t ; Among them, Δ t,j ⊙M j Indicates Δ t,j and M j Multiply each element at the same position of ;

[0037] Step 306: When the tth online deployment is performed, the linear classification model W is updated t , refers to the use of gradient descent technology to ensure that W t For the online dataset S t And the loss function l and the updated deep neural network Φ t Convergence; where convergence refers to the training process Gradient descent techniques include SGD and Adam.

[0038] Take the real-time prediction of pedestrian flow in popular scenic spots as an example. Traditional pedestrian flow prediction methods, such as historical averaging method, time series analysis method and simple machine learning models, are difficult to cope with complex nonlinear relationships and environmental changes, and it is difficult to predict changes in pedestrian flow caused by emergencies, and they lack online learning and adaptive update capabilities. Existing pedestrian flow prediction methods based on deep learning are mostly based on offline training models, and their prediction accuracy decreases over time, and updating the model requires a lot of computing power. The present invention collaborates with multi-resolution update strategies and online integration strategies to enable the model to quickly adapt to the dynamic changes of pedestrian flow data in complex scenarios with less computing power, thereby improving prediction accuracy and response speed.

[0039] This embodiment needs to use offline data to initialize a pedestrian flow prediction model based on a deep neural network. The steps are as follows: Figure 2 As shown. Collect offline pedestrian flow data set S 0 The data sources include remote sensing information, monitoring information, navigation software information, social media hot spots and other related information, such as table type, visual type, and natural language type data. Select the model structure, for example, choose the architecture of the deep neural network as 4-layer MLP, the Hilbert representation space dimension d′ as 1024, and the loss function l as KL Divergence. Use the gradient descent technology, such as Adam, to initialize the model h 0 , ensuring that the model parameter Φ 0 ,W 0 convergence.

[0040] This embodiment also requires the deep neural network to be divided into blocks and the resolution of each block to be assigned before entering the online stage, such as Figure 1 For example, the 4-layer MLP is divided into N = 4 blocks, and each layer of the 4-layer MLP becomes a submodule. When the t-th online deployment is performed, the j-th submodule M j Use the interval I to which the historical data belongs j =[s j ,t-1], where That is, the length of the interval is roughly a linear function of j.

[0041] This embodiment uses an online integration strategy to update the deep neural network model during the tth online deployment. The steps are as follows: Figure 3 As shown. At this point, the t-1th online deployment has been completed, and the model h is obtained. t-1 and its parameter Φ t-1 and W t-1 Model h t-1 Given D t The prediction on the current pedestrian flow is the prediction of the current pedestrian flow. Collect the online pedestrian flow dataset S t, that is, wait until the current real pedestrian flow is observed. Based on the current online pedestrian flow dataset S t Construct a deep neural network that does not rely on linear classification models to optimize the target L t (Φ;λ), the historical update amount Δ of the deep neural network generated by loading the online ensemble strategy t-1 , sampling disturbance coefficient r t , calculate the current gradient g of the deep neural network after the disturbance t For each submodule j∈{1,…,N}, load the jth submodule M j Use the historical gradient of the deep neural network after the perturbation in the interval to which the historical data belongs and the jth submodule M j Historical update volume Calculate the jth submodule M j The current update amount Δ t,j and its weight p t,j . The current update amount of the deep neural network is obtained by integration Δ t , update the current parameters of the deep neural network Φ t . Use gradient descent techniques, such as SGD, to update the linear classification model to ensure that its parameters W t Convergence. In this way, the model h is obtained. t and its parameter φ t and W t , the tth online deployment can be ended and the t+1th online deployment can be started.

[0042] Obviously, those skilled in the art should understand that the various steps of the real-time prediction of pedestrian flow in popular scenic spots of the above-mentioned embodiment of the present invention can be implemented by a general computing device, which can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices. Optionally, they can be implemented with program codes executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. In this way, the embodiment of the present invention is not limited to any specific combination of hardware and software.

Claims

1. A deep neural network online integration method for real-time pedestrian flow prediction, characterized in that: It includes multi-resolution update strategy and online integration strategy. First, a pedestrian flow prediction model based on deep neural network is initialized using offline data. Then, the model is deployed to predict pedestrian flow and online data is collected to update the model. The multi-resolution update strategy refers to dividing the deep neural network into multiple sub-modules and using historical data of different resolutions to train different sub-modules; The online integration strategy refers to dynamically selecting and combining sub-modules that are most suitable for the current scenario according to changes in real-time data distribution to generate real-time prediction results.

2. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 1 is characterized in that: The offline data is used to initialize a pedestrian flow prediction model based on a deep neural network. The specific steps are as follows: Step 101: Collect offline data sets for pedestrian flow prediction. in, is a d-dimensional real number space, is x i The set of all possible values ​​of x i It is a feature used to predict pedestrian flow. It has d dimensions, each of which is a real value. There are m0 data in total. i ∈{1,...,K},y i is x i There are K types of corresponding real pedestrian flows; Step 102: Initialize the pedestrian flow prediction model h0(x)=h(x;Φ0,W0)=w(φ(x;Φ0);W0) using the offline data set S0; model Map the features of d-dimensional real numbers to the confidence of K-dimensional real numbers. The category predicted by the pedestrian flow prediction model is the category corresponding to the dimension with the highest confidence. If there are multiple dimensions with the highest confidence, one of them is selected; each model consists of two parts. It is a deep neural network model that maps the features of d-dimensional real numbers to the Hilbert space representation of d′-dimensional real numbers, where is the deep neural network parameter, with a total of n real number parameters; is a linear classification model that maps the Hilbert space representation of d′-dimensional real numbers to the confidence of K-dimensional real numbers, where is the linear classification model parameter, which is a d′×K real matrix; Step 103: Use the gradient descent technique to ensure that the pedestrian flow prediction model h0 initialized in step 102 converges to the offline data set S0 and the loss function l; where the loss function It is used to measure the output h0(x i ) and the actual pedestrian flow y i The difference between them is then used to train and update Φ and W.

3. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 1 is characterized in that: The multi-resolution update strategy comprises the following specific steps: Step 201: Divide the deep neural network into N submodules, and use mask M for the jth submodule j Indicates; mask M j ∈{0, 1} n is an n-dimensional array consisting only of 0s and 1s. If M j The i-th M j [i] is 1, indicating that the j-th submodule contains the i-th real parameter of the deep neural network, otherwise it does not contain it; for each i∈{1,...,n}, ∑ j∈{1，...，N} M j [i] = 1, which means that the L submodules should be mutually exclusive and cover all parameters of the deep neural network; dividing submodules It is necessary to select the value of N, take N∈{4, 5, 6}, and assign each submodule M j The number of parameters covered, for each j∈{1,...,N}, Step 202: Using historical data of different resolutions to train different submodules refers to using continuous interval I j =[s j , t-1] historical data training submodule j, interval I j The starting point is s j , interval I j The termination endpoint is (t-1), where t is the number of times the model predicts and collects online data after the pedestrian flow prediction model is deployed online; s j The larger the value is, the shorter the interval of historical data used by the jth submodule is and the higher the resolution is; appropriately distribute historical data The resolution of refers to the resolution of the historical data used by different submodules to be differentiated as much as possible. When t ≥ N, for each j ∈ {1, ..., N}, or It means that the length of the interval to which the historical data used by the j-th submodule belongs is an exponential function of j or a linear function of j; Step 203: Diversify the pedestrian flow prediction capabilities of the submodules, which refers to the appropriate division of the submodules in step 201. and the appropriate resolution of the historical data in step 202 4. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 1 is characterized in that: The online integration strategy includes the following specific steps: Step 301: For t = {1, ..., ∞}, at the tth online deployment, according to the existing deep neural network parameters Φ t-1 and the linear classification model parameters W t-1 , deploy model h t-1 (x) = h(x; Φ t-1 , W t-1 ) is used to predict the pedestrian flow in the future, denoted as The corresponding pedestrian flow, and then collect the real pedestrian flow data to form an online dataset Based on S t By updating the existing Φ t-1 and W t-1 To obtain a Φ that is more suitable for the current scene t and W t , and used for the next online deployment; Step 302: During the tth online deployment, according to S t Constructing deep neural network optimization objectives that do not rely on linear classification models Refers to L t It can measure whether the deep neural network is suitable for the scenario of the tth online deployment based on the change of real-time data distribution without relying on the linear classification model, and the sampling disturbance coefficient r t ~U([0,1]) refers to uniformly sampling a real number r from the interval [0,1] t , solve the gradient of the deep neural network after perturbation Among them, λ is a real parameter used to eliminate the dependence on the linear classification model; Δ t-1 is the update amount of the deep neural network generated by the online integration strategy during the t-1th online deployment. When t=1, Δ t-1 Each element of is zero, when t>1, Δ t-1 The specific calculation method of is as described in step 305; Step 303: During the t-th online deployment, for each j∈{1, ..., N}, the submodule M is calculated. j Update volume Refers to the interval I j =[s j ,t-1] on the deep neural network and the historical update amount of the submodule to calculate the update amount of the submodule when it is deployed online for the tth time; where Δ τ,j is the jth submodule M j The update amount during the τth online deployment, τ = s j When τ,j Each element of is zero, τ>s j When τ,j Recursively calculate according to step 303; Step 304: During the t-th online deployment, for each j∈{1, ..., N}, calculate the j-th submodule M j The weight of the adaptability of the scenario for the tth online deployment p t,j The larger the value, the jth submodule M j The more suitable it is for the scenario of the tth online deployment; t,j In the calculation formula, The function ω(R, C) is defined as follows, The function ψ(R, C) is defined as follows: when R and C are both zero, ψ(R, C) = 1; when R and C are not both zero, Step 305: Calculate the update amount of the deep neural network during the t-th online deployment And used to update the deep neural network Φ t =Φ t-1 +Δ t , refers to the dynamic selection and combination of submodules that best adapt to the current scenario, and the integration of the deep neural network parameters Φ that adapt to the scenario of the tth online deployment t ; Among them, Δ t,j ⊙M j Indicates Δ t,j and M j Multiply each element at the same position of ; Step 306: When the tth online deployment is performed, the linear classification model W is updated t , refers to the use of gradient descent technology to ensure that W t For the online dataset S t And the loss function l and the updated deep neural network Φ t Convergence; where convergence refers to the training process 5. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 2 is characterized in that: The structures of the deep neural network models available for selection in step 102 include MLP, ResNet, UNet, LSTM, Transformer and RWKV, or a multi-modal structure formed by a combination of multiple structures.

6. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 2 is characterized in that: The optional loss functions l in step 103 include Logistic loss function, Sigmoid loss function, negative log-likelihood and KL Divergence.

7. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 2 is characterized in that: The gradient descent techniques in step 103 include SGD and Adam.

8. The deep neural network online integration method for real-time pedestrian flow prediction according to claim 4 is characterized in that: The gradient descent techniques in step 306 include SGD and Adam.

9. A computer device, characterized in that: The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the deep neural network online integration method for real-time prediction of pedestrian flow as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program for executing the deep neural network online integration method for real-time pedestrian flow prediction as described in any one of claims 1-8.

Images