CN117952287A - Prediction method and system for number of passengers in terminal building waiting area - Google Patents
Prediction method and system for number of passengers in terminal building waiting area Download PDFInfo
- Publication number
- CN117952287A CN117952287A CN202410358442.7A CN202410358442A CN117952287A CN 117952287 A CN117952287 A CN 117952287A CN 202410358442 A CN202410358442 A CN 202410358442A CN 117952287 A CN117952287 A CN 117952287A
- Authority
- CN
- China
- Prior art keywords
- camera
- points
- camera points
- data
- terminal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000005259 measurement Methods 0.000 claims abstract description 26
- 238000013507 mapping Methods 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 21
- 230000010006 flight Effects 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 5
- 238000010200 validation analysis Methods 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000004913 activation Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000004804 winding Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method and a system for predicting the number of passengers in a terminal area of a terminal building, wherein the method comprises the following steps: presetting a plurality of camera points in a terminal area of a terminal building, and preprocessing image data acquired by the camera points within preset time to obtain the positions of the camera pointsActual measurement result of time; Locating each camera point atActual measurement result of timeInputting into a preset GRU model to obtain a cameraIn the first placePredicted number of people at each camera point; And constructing a two-dimensional space, and mapping the camera points and the view angle information corresponding to the camera points to the two-dimensional space to obtain view angle matrixes corresponding to the camera points. The invention improves the utilization rate of airport resources and the working efficiency of airport service.
Description
Technical Field
The invention relates to the technical field of terminal building terminal management, in particular to a method and a system for predicting the number of passengers in a terminal building terminal area.
Background
The existing terminal waiting area works mainly according to the dynamic development of the resource allocation of the flights, so that the problem of service efficiency reduction caused by local resource shortage is easy to occur, the problem of influence on the normal operation of the flights is further caused, and the airport allocation resources are inconvenient to fully utilize. How to reasonably arrange the resource configuration in the airport terminal and predict potential safety risks and service peaks such as personnel aggregation in advance becomes a main problem faced by the civil aviation airport terminal security departments.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a method and a system for predicting the number of passengers in a terminal area of a terminal building.
The invention provides a method for predicting the number of passengers in a terminal area of a terminal building, which comprises the following steps:
s1, presetting a plurality of camera points in a terminal station terminal area, and preprocessing image data acquired by the camera points in preset time to obtain actual measurement results of the camera points at the time t ；
S2, actually measuring the positions of all the cameras at the time tInputting into a preset GRU model to obtain a camera/>In/>Predicted number of people/>, of camera spots；
S3, constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining ground sight line ranges corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points;
Step S4, obtaining the number of people corresponding to each camera point position according to the preset distance and the ground sight line range corresponding to the camera point positions ；
S5, carrying out global fusion on the number of people corresponding to the camera points one by one to form a passenger number prediction result in the corresponding area。
Preferably, the method for constructing the preset GRU model includes:
Acquiring relevant flight information data based on an airport information system, grouping the flight information data according to the dimensions of flights, routes and boarding gates, and then constructing integral data features of each camera point position in a terminal building by taking every 15 minutes as sampling frequency;
Combining the actual measurement number of each camera at the camera point and the predicted data of the next camera point, and carrying out standardized processing on the combined data to obtain a characteristic data set;
setting related GRU network parameters, and dividing the characteristic data set according to a preset strategy to form a training set, a testing set and a verification data set;
inputting the features in the terminal building and the actual measurement data of the cameras as input features and the actual measurement data of the next camera point position camera as target variables into a GRU network for model training so as to obtain a GRU model;
And adjusting GRU network parameters according to training results until a GRU model obtained after reaching an average absolute percentage error target of less than n percent is obtained, wherein n is a target error rate, and the target error rate is adjusted according to different airport scales, data and physical conditions.
Preferably, the preset strategy is specifically:
The training set was 70% feature data set, the test set was 20% feature data set, and the validation data set was 10% feature data set.
Preferably, step S1 specifically includes:
presetting a plurality of camera point positions in a terminal station waiting area;
acquiring image data of a plurality of camera points in a preset time according to a preset acquisition frequency;
the collected image data of the camera points are extracted one by one according to time, so as to obtain the actual measurement result of each camera point at the time t 。
Preferably, the viewing angle information includes position coordinatesAngle/>Horizontal viewing angle/>Vertical viewing angle/>; The construction of the two-dimensional space and mapping of the camera points and the view angle information corresponding to the camera points to the two-dimensional space to obtain view angle matrixes corresponding to the camera points comprises the following steps:
constructing a two-dimensional space, and distributing position coordinates of a plurality of camera points in a one-to-one correspondence manner Angle ofHorizontal viewing angle/>Vertical viewing angle/>；
Position coordinates allocated in correspondence with single camera pointsAngle/>Horizontal viewing angle/>Vertical viewing angle/>Calculating a view angle matrix/>, corresponding to the point positions of the single camera。
Preferably, the ground sight line range corresponding to the plurality of camera points is obtained according to the plurality of view angle information and the plurality of view angle matrixes corresponding to the plurality of camera points, which specifically includes:
The sight line ranges of four corners of a single camera picture corresponding to the single camera point positions are obtained one by one for the plurality of camera point positions, and the sight line ranges of the four corners specifically comprise: left upper part Upper right/>Lower right/>Lower left/>；
Position coordinates allocated in correspondence with single camera pointsAngle/>Horizontal viewing angle/>Perpendicular viewing angle/>Viewing angle matrix/>Calculating the sight range of four corners;
The view range of the four corners is taken as the view range of the ground of the camera.
Preferably, step S4 specifically includes:
according to the preset distance, the number of the people predicted by the single camera point in the ground sight range of the camera is correspondingly calculated Evenly distributing to obtain the number of people/>, corresponding to each camera point position。
Preferably, step S5 specifically includes:
defining a passenger distribution probability density function, and determining the number of people corresponding to each camera point in the target area The estimated number/>, corresponding to each camera point position, is obtained by being carried into a probability density function；
For a plurality of camera points corresponding to the estimated number of peopleSumming to obtain prediction of the number of passengers in an area/>。
The invention provides a prediction system for the number of passengers in a terminal area of a terminal building, which comprises the following steps:
The data acquisition module is used for preprocessing image data acquired by a plurality of camera points within preset time to obtain each camera point Actual measurement results of time/>；
A data processing module for locating each camera pointActual measurement results of time/>Inputting into a preset GRU model to obtain a camera/>In/>Predicted number of people/>, of camera spots；
The data generation module is used for constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining a ground sight line range corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points;
the data prediction module is used for obtaining the number of people corresponding to each camera point position according to the preset distance and the ground sight ranges corresponding to the camera point positions ；
The region prediction module is used for carrying out global fusion on the number of people corresponding to the camera points one by one to form a passenger number prediction result in the corresponding region。
The invention provides a method and a system for predicting the number of passengers in a terminal area of a terminal building, which are used for acquiring the number of people actually measured by a plurality of camera points in a target area; locating each camera point atActual measurement results of time/>Inputting into a preset GRU model to obtain a camera/>In/>Predicted number of people/>, of camera spots; Constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining ground sight line ranges corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points; obtaining the number of people/>, corresponding to each camera point position, according to the preset distance and the ground sight ranges corresponding to the camera point positions; Global fusion is carried out on the number of people corresponding to the camera points one by one to form a passenger number prediction result/>, in the corresponding area. The method realizes the prediction of the number of passengers in the target area of the terminal building in the preset time, and is convenient for assisting the terminal building manager to make a resource allocation scheduling decision according to the prediction result of the number of passengers, thereby improving the utilization rate of airport resources and the working efficiency of airport service.
Drawings
FIG. 1 is a schematic diagram of the workflow of a method for predicting the number of passengers in a terminal area of a terminal according to the present invention;
fig. 2 is a schematic diagram of a system frame structure of a system for predicting the number of passengers in a terminal area of a terminal according to the present invention.
Detailed Description
Referring to fig. 1, the method for predicting the number of passengers in a terminal area of a terminal provided by the invention comprises the following steps:
S1, presetting a plurality of camera points in a terminal station terminal area, and preprocessing image data acquired by the camera points within preset time to obtain the positions of the camera points Actual measurement results of time/>。
In this embodiment, step S1 specifically includes: presetting a plurality of camera point positions in a terminal station waiting area; acquiring image data of a plurality of camera points in a preset time according to a preset acquisition frequency; the collected image data of the camera points are extracted one by one according to time, so that the positions of the camera points are obtainedActual measurement results of time/>。
S2, locating each camera pointActual measurement results of time/>Inputting into a preset GRU model to obtain a camera/>In/>Predicted number of people/>, of camera spots。
In this embodiment, the method for constructing the preset GRU model includes:
Acquiring relevant flight information data based on an airport information system, grouping the flight information data according to the dimensions of flights, routes and boarding gates, and then constructing integral data features of each camera point position in a terminal building by taking every 15 minutes as sampling frequency; after the actual measurement number of the camera point position and the predicted data of the next camera point position are combined, each path of camera performs standardization processing on the combined data to obtain a characteristic data set; setting related GRU network parameters, and dividing the characteristic data set according to a preset strategy to form a training set, a testing set and a verification data set; inputting the features in the terminal building and the actual measurement data of the cameras as input features and the actual measurement data of the next camera point position camera as target variables into a GRU network for model training so as to obtain a GRU model; and adjusting GRU network parameters according to training results until a GRU model obtained after reaching an average absolute percentage error target of less than n percent is obtained, wherein n is a target error rate, and the target error rate is adjusted according to different airport scales, data and physical conditions.
In this embodiment, the preset policy specifically includes: the training set was 70% feature data set, the test set was 20% feature data set, and the validation data set was 10% feature data set.
In this embodiment, historical flight information is obtained from an information integration system of an airport, passenger number and boarding information are obtained from an departure information system, security inspection information is obtained from a security inspection information system, all flights with the same static characteristics of departure place, destination, season, holiday, spring delivery summer transportation, departure time period (divided into a time period every 2 hours) and the like are clustered and analyzed to form a characteristic route, and historical data of the advance of boarding, security inspection and boarding of the flights corresponding to the flights of the same characteristic route are fitted to form a route passenger activity distribution characteristic。
Applying airline passenger activity distribution feature functions to each flight in a training datasetObtaining the navigation
Number sequence of check-in and boarding persons passing through per-time advance of class passengers。
Collecting historical data from data sources such as an integrated system, ACDM systems, departure systems, flow systems, security systems, public transportation systems and the like of an airport, correlating weather, public transportation, channel flow control, number of opening of a cabinet table, number of opening of security check openings, whether bridge is leaned, number of flights and departure time offset data to a characteristic route, onehot coding the characteristic route number, weather data and position number, normalizing coding the channel flow control affecting the flights, the number of opening of the cabinet table in a correlation area, the number of opening of the security check openings in the correlation area, number of flights and departure time offset (minutes) to a section of [0,1] to form characteristic data of the characteristic route。
The take-off time of the flight isAdvance of flight activation is/>The flight activation time is/>. Calculating an activated flight set of each boarding gate, wherein the activation condition is/>Wherein/>For sampling time points, the conditional flights are pressed/>Ordering, gate/>At the time point/>Is/>At most 6 flights are calculated for each boarding gate, 0 is deleted and the out-of-flight flights are removed.
The time span of the historical data of the planned training is divided according to one sampling every 15 minutes, the sample data of the time point is formed by sequencing all boarding gates according to numbers,
Wherein n is the number of boarding gates and m is the number of cameras.Is a boarding gate/>At/>Moment gate feature,/>Is a camera/>At/>Actual measurement number of people at time,/>And/>For sample characteristics,/>Is the target variable.
Collecting all the sampled data to obtain a sample data setWherein/>Is the number of samples of the historical data of the planned training in 15 minute spans, each/>Is at the time point/>For data set/>Segmentation was performed using the first 70% of samples as training samples/>20% Of the samples thereafter served as test samples/>Last 10% sample as validation sample/>。
The sequence length is 24, the actual sample feature quantity (different airport position quantity and different feature quantity) is taken as the input feature quantity, the actual sample quantity is 0.2 as the initial hidden layer size, the camera quantity is the output size, 2 is the layer number, the average absolute percentage error (MAPE) is taken as the loss function to create a GRU training network, and the GRU training network is usedAnd/>Network training, use/>And (3) verifying, namely adjusting network parameters according to the actual training result, and obtaining a prediction model after achieving the target training result with the error rate lower than 5%.
And S3, constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining a ground sight line range corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points.
In the present embodiment, the angle-of-view information includes position coordinatesAngle/>Horizontal viewing angleVertical viewing angle/>; Constructing a two-dimensional space, and mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, wherein the view angle matrixes specifically comprise: constructing a two-dimensional space, and distributing position coordinates/>, which are allocated to a plurality of camera points in a one-to-one correspondence mannerAngle/>Horizontal viewing angleVertical viewing angle/>; Position coordinates/>, which are correspondingly allocated according to single camera point positionsAngle ofHorizontal viewing angle/>Vertical viewing angle/>Calculating a view angle matrix/>, corresponding to the point positions of the single camera。
In the present embodiment, the viewing angle matrixThe calculation process is as follows:
Representing the winding/> Rotated rotation matrix,/>Indicating the rotation angle.
Representing the winding/>Rotated rotation matrix,/>Indicating the rotation angle.
Representing the winding/>Rotated rotation matrix,/>Indicating the rotation angle.
In this embodiment, according to the multiple view angle information and multiple view angle matrixes corresponding to the multiple camera points, a ground sight line range corresponding to the multiple camera points is obtained, which specifically includes: the sight line ranges of four corners of a single camera picture corresponding to the single camera point positions are obtained one by one for the plurality of camera point positions, and the sight line ranges of the four corners specifically comprise: left upper partUpper right/>Lower right/>Lower left/>; Position coordinates/>, which are correspondingly allocated according to single camera pointsAngle ofHorizontal viewing angle/>Perpendicular viewing angle/>Viewing angle matrix/>Calculating the sight range of four corners; the view range of the four corners is taken as the view range of the ground of the camera.
In the present embodiment, the upper leftUpper right/>Lower right/>Lower left/>The specific calculation process is as follows:
for the vertical distance of the camera to the target level.
S4, obtaining the ground sight ranges corresponding to the camera points according to the preset distance and the preset ground sight ranges
To the number of people corresponding to each camera point。
In this embodiment, step S4 specifically includes: according to the preset distance, the number of the people predicted by the single camera point in the ground sight range of the camera is correspondingly calculatedEvenly distributed to obtain each camera point position
Corresponding number of people. In the present embodiment, according to/>The number of people in the view angle is evenly distributed by taking 5 meters as a distance, and the point number of each camera/>The method comprises the following steps:
s5, carrying out global fusion on the number of people corresponding to the camera points one by one to form a passenger number prediction result in the corresponding area 。
In this embodiment, step S5 specifically includes:
defining a passenger distribution probability density function, and determining the number of people corresponding to each camera point in the target area The estimated number/>, corresponding to each camera point position, is obtained by being carried into a probability density function; For a plurality of camera points, the predicted number/>Summing to obtain prediction of the number of passengers in an area/>。
In this embodiment, (1) a passenger distribution probability density function is defined:
Wherein, Is the first Gaussian distribution of mixed weights,/>Is the number of gaussian distributions.
The probability density function for each gaussian distribution is:
(2) Calculating posterior probability:
(3) Updating each Gaussian distribution parameter:
a. updating the mixing weight parameters:
b. updating the mean value:
c. Updating the covariance matrix:
d. According to the actual conditions of airport scale, camera coverage and the like, comprehensively considering, defining weight parameter threshold iterative steps b and c until:
Wherein the method comprises the steps of
Stopping iteration to obtain a probability distribution function;
e. Dividing an airport plane into five longitudinal and transverse grains, bringing each camera point in a target area into a probability density function, and obtaining the estimated number of people at the point: ；
the number of people predicted in the area is:
Referring to fig. 2, the system for predicting the number of passengers in a terminal area of a terminal according to the present invention includes:
The data acquisition module is used for preprocessing image data acquired by a plurality of camera points within preset time to obtain each camera point Actual measurement results of time/>；
A data processing module for locating each camera pointActual measurement results of time/>Inputting into a preset GRU model to obtain a camera/>In/>Predicted number of people/>, of camera spots；
The data generation module is used for constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining a ground sight line range corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points;
the data prediction module is used for obtaining the number of people corresponding to each camera point position according to the preset distance and the ground sight ranges corresponding to the camera point positions ；
The region prediction module is used for carrying out global fusion on the number of people corresponding to the camera points one by one to form a passenger number prediction result in the corresponding region。
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art, who is within the scope of the present invention, should make equivalent substitutions or modifications according to the technical scheme of the present invention and the inventive concept thereof, and should be covered by the scope of the present invention.
Claims (9)
1. The method for predicting the number of passengers in the terminal area of the terminal building is characterized by comprising the following steps of:
s1, presetting a plurality of camera points in a terminal station terminal area, and preprocessing image data acquired by the camera points in preset time to obtain actual measurement results of the camera points at the time t ；
S2, actually measuring the positions of all the cameras at the time tInputting into a preset GRU model to obtain a cameraIn/>Predicted number of people/>, of camera spots；
S3, constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining ground sight line ranges corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points;
Step S4, obtaining the number of people corresponding to each camera point position according to the preset distance and the ground sight line range corresponding to the camera point positions ；
S5, carrying out global fusion on the number of people corresponding to the camera points one by one to form a passenger number prediction result in the corresponding area。
2. The method for predicting the number of passengers in a terminal area of a terminal as set forth in claim 1, wherein the method for constructing the preset GRU model comprises:
Acquiring relevant flight information data based on an airport information system, grouping the flight information data according to the dimensions of flights, routes and boarding gates, and then constructing integral data features of each camera point position in a terminal building by taking every 15 minutes as sampling frequency;
Combining the actual measurement number of each camera at the camera point and the predicted data of the next camera point, and carrying out standardized processing on the combined data to obtain a characteristic data set;
setting related GRU network parameters, and dividing the characteristic data set according to a preset strategy to form a training set, a testing set and a verification data set;
inputting the features in the terminal building and the actual measurement data of the cameras as input features and the actual measurement data of the next camera point position camera as target variables into a GRU network for model training so as to obtain a GRU model;
And adjusting GRU network parameters according to training results until a GRU model obtained after reaching an average absolute percentage error target of less than n percent is obtained, wherein n is a target error rate, and the target error rate is adjusted according to different airport scales, data and physical conditions.
3. The method for predicting the number of passengers in a terminal area of a terminal according to claim 2, wherein the preset strategy is specifically:
The training set was 70% feature data set, the test set was 20% feature data set, and the validation data set was 10% feature data set.
4. The method of claim 1, wherein step S1 specifically comprises:
presetting a plurality of camera point positions in a terminal station waiting area;
acquiring image data of a plurality of camera points in a preset time according to a preset acquisition frequency;
the collected image data of the camera points are extracted one by one according to time, so as to obtain the actual measurement result of each camera point at the time t 。
5. The method of claim 1, wherein the view information includes location coordinatesAngle/>Horizontal viewing angle/>Vertical viewing angle/>; The construction of the two-dimensional space and mapping of the camera points and the view angle information corresponding to the camera points to the two-dimensional space to obtain view angle matrixes corresponding to the camera points comprises the following steps:
constructing a two-dimensional space, and distributing position coordinates of a plurality of camera points in a one-to-one correspondence manner Angle ofHorizontal viewing angle/>Vertical viewing angle/>；
Position coordinates allocated in correspondence with single camera pointsAngle/>Horizontal viewing angleVertical viewing angle/>Calculating a view angle matrix/>, corresponding to the point positions of the single camera。
6. The method for predicting the number of passengers in a terminal area of a terminal according to claim 5, wherein the obtaining a ground sight line range corresponding to a plurality of camera points according to a plurality of view angle information and a plurality of view angle matrixes corresponding to a plurality of camera points specifically includes:
The sight line ranges of four corners of a single camera picture corresponding to the single camera point positions are obtained one by one for the plurality of camera point positions, and the sight line ranges of the four corners specifically comprise: left upper part Upper right/>Lower right/>Lower left/>；
Position coordinates allocated in correspondence with single camera pointsAngle/>Horizontal viewing anglePerpendicular viewing angle/>Viewing angle matrix/>Calculating the sight range of four corners;
The view range of the four corners is taken as the view range of the ground of the camera.
7. The method of claim 6, wherein step S4 specifically comprises:
according to the preset distance, the number of the people predicted by the single camera point in the ground sight range of the camera is correspondingly calculated Evenly distributing to obtain the number of people/>, corresponding to each camera point position。
8. The method of claim 7, wherein step S5 specifically comprises:
defining a passenger distribution probability density function, and determining the number of people corresponding to each camera point in the target area The estimated number/>, corresponding to each camera point position, is obtained by being carried into a probability density function；
For a plurality of camera points corresponding to the estimated number of peopleSumming to obtain prediction of the number of passengers in an area/>。
9. A system for predicting the number of passengers in a terminal area, comprising:
The data acquisition module is used for preprocessing image data acquired by a plurality of camera points within preset time to obtain actual measurement results of the camera points at the time t ；
The data processing module is used for processing the actual measurement result of each camera point at the time tInputting into a preset GRU model to obtain a camera/>In/>Predicted number of people/>, of camera spots；
The data generation module is used for constructing a two-dimensional space, mapping a plurality of camera points and a plurality of view angle information corresponding to the camera points to the two-dimensional space to obtain a plurality of view angle matrixes corresponding to the camera points, and obtaining a ground sight line range corresponding to the camera points according to the view angle information and the view angle matrixes corresponding to the camera points;
the data prediction module is used for obtaining the number of people corresponding to each camera point position according to the preset distance and the ground sight ranges corresponding to the camera point positions ；
The region prediction module is used for carrying out global fusion on the number of people corresponding to the camera points one by one to form a passenger number prediction result in the corresponding region。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410358442.7A CN117952287B (en) | 2024-03-27 | 2024-03-27 | Prediction method and system for number of passengers in terminal building waiting area |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410358442.7A CN117952287B (en) | 2024-03-27 | 2024-03-27 | Prediction method and system for number of passengers in terminal building waiting area |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117952287A true CN117952287A (en) | 2024-04-30 |
| CN117952287B CN117952287B (en) | 2024-06-18 |
Family
ID=90794960
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410358442.7A Active CN117952287B (en) | 2024-03-27 | 2024-03-27 | Prediction method and system for number of passengers in terminal building waiting area |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117952287B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114092866A (en) * | 2020-07-30 | 2022-02-25 | 上海虹桥枢纽交通中心建设发展有限公司 | Method for predicting space-time distribution of airport passenger flow |
| CN114387265A (en) * | 2022-01-19 | 2022-04-22 | 中国民航大学 | Anchor-frame-free detection and tracking unified method based on attention module addition |
| EP4125058A1 (en) * | 2021-07-29 | 2023-02-01 | Siemens Mobility GmbH | Method and assembly for calibration of a simulator of a flow of people |
| CN115983487A (en) * | 2023-01-10 | 2023-04-18 | 大连理工大学 | Airdrome station building passenger space-time distribution prediction method based on chi-square distribution |
-
2024
- 2024-03-27 CN CN202410358442.7A patent/CN117952287B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114092866A (en) * | 2020-07-30 | 2022-02-25 | 上海虹桥枢纽交通中心建设发展有限公司 | Method for predicting space-time distribution of airport passenger flow |
| EP4125058A1 (en) * | 2021-07-29 | 2023-02-01 | Siemens Mobility GmbH | Method and assembly for calibration of a simulator of a flow of people |
| CN114387265A (en) * | 2022-01-19 | 2022-04-22 | 中国民航大学 | Anchor-frame-free detection and tracking unified method based on attention module addition |
| CN115983487A (en) * | 2023-01-10 | 2023-04-18 | 大连理工大学 | Airdrome station building passenger space-time distribution prediction method based on chi-square distribution |
Non-Patent Citations (1)
| Title |
|---|
| 闫江南;杨帅;崔育帅;: "基于视频分析的公交客流统计系统的研究", 软件, no. 06, 15 June 2018 (2018-06-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117952287B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jacquillat et al. | A roadmap toward airport demand and capacity management | |
| Abed et al. | An econometric analysis of international air travel demand in Saudi Arabia | |
| CN112368660A (en) | Unmanned aerial vehicle risk assessment | |
| US6580998B2 (en) | System and method for estimating aircraft flight delay | |
| CN107403325B (en) | Method and device for evaluating reliability of air ticket order | |
| Sidiropoulos et al. | Robust identification of air traffic flow patterns in Metroplex terminal areas under demand uncertainty | |
| CN105528647B (en) | Airport traffic demand possibility prediction method based on big data analysis | |
| CN109460871A (en) | Airport passenger amount prediction technique based on the identification of typical day | |
| Andreatta et al. | A set of approximate and compatible models for airport strategic planning on airside and on landside | |
| CN110166945B (en) | Method, device, medium and equipment for dynamically predicting passenger flow of terminal building | |
| CN114048944A (en) | Estimation method for people to be evacuated and houses to be damaged under rainstorm induced geological disaster | |
| CN114239325B (en) | Airport check-in consignment counter configuration planning method, device, equipment and storage medium | |
| Waltert et al. | Ratio-based design hour determination for airport passenger terminal facilities | |
| Setiawan et al. | Analysis of demand potential and need for passenger terminal facilities at Cikembar Sukabumi Airport | |
| Mott et al. | Estimation of aircraft operations at airports using nontraditional statistical approaches | |
| CN117952287B (en) | Prediction method and system for number of passengers in terminal building waiting area | |
| Fontes et al. | A deep learning approach for predicting bus passenger demand based on weather conditions | |
| Sidiropoulos et al. | Identifying significant traffic flow patterns in multi-airport systems terminal manoeuvring areas under uncertainty | |
| WO2022252850A1 (en) | Seasonal classification method and apparatus for flights, and machine-readable medium | |
| Bauman III et al. | Applied Meteorology Unit-Operational Contributions to Spaceport Canaveral | |
| Nagy et al. | Revealing Influencing Factors of Check-in Time | |
| Cai et al. | A scenario-based optimization approach to robust estimation of airport apron capacity | |
| Doshi et al. | Application of simulation models in airport facility design | |
| CN109146113B (en) | Riding demand prediction method and system | |
| Li et al. | A Least‐Square Model to Estimate Historical Percentages of Itinerant General Aviation Operations by Aircraft Types and Flight Rules at an Airport |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |