CN109670631B - Real-time flow prediction method for macroscopic region - Google Patents
Real-time flow prediction method for macroscopic region Download PDFInfo
- Publication number
- CN109670631B CN109670631B CN201811400258.5A CN201811400258A CN109670631B CN 109670631 B CN109670631 B CN 109670631B CN 201811400258 A CN201811400258 A CN 201811400258A CN 109670631 B CN109670631 B CN 109670631B
- Authority
- CN
- China
- Prior art keywords
- data
- δet
- delta
- area
- time
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 230000014759 maintenance of location Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 2
- 238000007405 data analysis Methods 0.000 abstract 1
- 238000007418 data mining Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a macroscopic region real-time flow prediction method which is based on a macroscopic region real-time pedestrian flow prediction algorithm of a portable mobile terminal unique ID, a collected data timestamp and detection equipment position information, and realizes efficient and accurate prediction of real-time macroscopic region real-time pedestrian flow through data mining and analysis of data collected by a mobile terminal.
Description
Technical Field
The invention belongs to the mobile intelligent Internet technology, and particularly relates to a scenic spot and other people flow detection and evaluation algorithm.
Background
The regional flow data is an important information source of the current smart campus, smart scenic spot and smart city, and can provide decision assistance for regional people flow density evaluation, regional people flow change, flow induction and the like.
With the rapid development of economy and science and technology in China, in order to better manage the areas of daily life and travel of people such as campuses, scenic spots and towns, the convenience and the happiness of daily life and travel of people are improved, and the construction of smart campuses, smart scenic spots and smart cities becomes a new trend of future development.
In the construction process of smart campuses, smart cities and the like, dynamic monitoring, estimation, dynamic detection, prediction and guidance of people flow are indispensable important components.
Taking a scenic spot as an example, a large number of digital monitors are established at present to monitor the flow of people in the scenic spot, for example, the total flow of people in the scenic spot is recorded through an entrance and exit gate; the method comprises the steps that visual screen monitoring is arranged in a key area, and people flow in a current area is identified; and displaying the related information in the scenic spot and the like to the tourists through the digital display screen.
However, although various devices are deployed at present, management requirements cannot be met in terms of people flow detection and prediction, for example, an import-export gate can only record the total reservation in a scenic spot, and cannot record people flow scattered in each scenic spot, and videos and the like can be clearly visualized, even people number can be identified in an image identification mode, full coverage cannot be achieved, global statistics cannot be achieved naturally, and prediction of flow cannot be achieved.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a real-time flow prediction method for a macro area.
In order to solve the technical problems, the invention adopts the following technical scheme: a real-time flow prediction method for a macro area is characterized by comprising the following steps: the system comprises a plurality of detection sub-networks and a remote background server, wherein the detection sub-networks are deployed in a macro area according to physical space, and each detection sub-network comprises a plurality of extension sets and a host computer in data connection with the extension sets; the extension set collects broadcast type data packets sent to the surrounding environment by mobile terminal equipment based on a WIFI protocol randomly within the coverage range of the extension set through a wireless passive sensing mode, screens the data packets with ID information of the mobile terminal equipment for retrieval, marks an extension set label and uploads the extension set label to a host, the host stores the collected data in a unified mode, marks a time label on the collected data, uploads the collected data to a remote background server for storage, and analyzes the data according to the following steps:
step 1: respectively collecting data of a complete time period (T-delta kt) by an extension set, wherein the data comprises an area A and all connecting channels R (g), g is 1,2 and …, dividing the data into k sections by taking a time difference delta T as a period, and each section of complete data is represented as Dk，
Wherein,
indicating the ith extension data of the jth subnet; n represents the number of the sub-networks, and M represents the number of the sub-networks in the jth sub-network;
step 2: slicing the data and extracting the data D of the local area S to be analyzed and the complete time period (T-delta kt)s'，
Where t is Δ t, DisNumbered asData collected by the extension of i;
and step 3: physical space position mapping, corresponding to the deployed sub-networks, and performing space matching on the deployed sub-networks and the actual area S, wherein each sub-network host is provided with corresponding area corresponding number information and an extension deployment condition list;
and 4, step 4: sequencing the data collected by each extension according to the ID number of the mobile terminal equipment, and establishing a data matrix { Tower (i, s), t };
and 5: classifying the established ID data list according to the times c of occurrence in different data matrixes: let ID denote the ID that appears in only one data matrix Towerc＝1And extracting and analyzing separately, and classifying ID data appearing in more than two data matrixes Tower into IDc＞1；
Step 6: analyzing ID data only appearing in one data matrix Tower in the time period delta t, if the ID data is not found to repeatedly appear in the data matrix Tower in the time period delta t, traversing adjacent subnets around the local area S to find whether the same ID data appears, and if the same ID data does not appear, classifying the ID data as the ID datac＝1Identifying the ID data as an ID if it is present in another subnetc＞1；
And 7: repeating the steps 4-6 until the data processing is finished;
and 8: the data obtained by the processing of the steps 5 and 6 are combined together to form two types of data IDc＝1And IDc＞1；
And step 9: in the time period delta T of the time T-delta et, e-1, 2, … k, the regional flow conservation quantity is VT-Δet， VT-Δet＝DT-Δet{ ID }, the area outflow prediction value W is predicted by the following methodoutAnd area inflow prediction value WinAnd (3) performing prediction evaluation:
(1) first, the ID is addedc＞1The middle data is subjected to data normalization processing and then is added into effective data;
(2) then, the ID is addedc＝1And IDc＞1Middle data through FT-ΔetApproximate fitting function approximation fitting processSetting a bias reference system S in the processT-ΔetAnd with (S)T-Δet-(Vc＞1)T-Δet)/(Vc＝1)T-ΔetAs the feedback judgment condition, when 1-theta is less than or equal to (S)T-Δet-(Vc＞1)T-Δet)/(Vc＝1)T-ΔetCompleting fitting to obtain effective data when the sum of theta is less than or equal to 1 and theta, wherein theta is an allowable error of system setting;
(3) continuing, ID is paired byc＞1And IDc＝1Forming an effective data area with the retention volume of people stream as V through fittingT-Δet：
Wherein x and y are each IDc＞1And IDc＝1Total number of medium valid data;
(4) then, the time range T- Δ et is calculated by the equation (2) to calculate the outflow rate of the person in the local area
And inflow of water
In the formula, delta1And delta2Are respectively T1And T2A very small amount of (c);
(5) finally, according to the area outflow quantity sequence in the time range of delta kt
And regional inflow number series
Generating an approximate function f (x) of regional outflow by adopting a data fitting algorithmoutAnd area inflow approximation function f (x)inCalculating the flow rate of the T + Deltat time region, substituting the time data into f (x)outAnd f (x)inAnd then, the process is finished.
Preferably, the bias reference system ST-ΔetThe video count or the gate count of the inlet and the outlet of the area in the time range of T-delta et is obtained.
Preferably, in step 9, δ1And delta2Is related to the fitting function and is determined by the following method: first, let y be F (·), n be the number of times y meets the system error range, then,
wherein the function F (-) is an approximate fitting function of the ID data, y1,y2,…,ynRespectively corresponding to function F (-) at T1,T2,…,TnValue of time, ymeanIs y1,y2,…,ynAverage value of (d) (-)1And delta2Are respectively T1And T2And is obtained by the above formula delta.
Has the advantages that: the invention provides a macroscopic region real-time pedestrian flow prediction algorithm based on the unique ID of a portable mobile terminal, a collected data timestamp and position information of detection equipment, and the real-time macroscopic region real-time pedestrian flow is efficiently and accurately predicted by mining and analyzing data collected by the mobile terminal.
Drawings
FIG. 1 is a logic flow diagram of a method for predicting regional flow according to the present invention;
fig. 2 is a schematic diagram illustrating a machine learning and self-feedback principle in the region flow prediction method according to the present invention.
Fig. 3 is a schematic diagram of the physical space deployment of the detection system of the present invention.
Fig. 4 is a comparison graph of the regional flow prediction method of the present invention for the real-time people flow prediction data of a certain region.
Detailed Description
The invention will be further elucidated with reference to the following description of an embodiment in conjunction with the accompanying drawing. It is to be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
As shown in fig. 1, it is a flow chart of the regional flow prediction algorithm of the present invention, and specifically, it does not inject:
step 1: collecting complete data of a complete time period (T-delta kt), including an area A and all connecting channels R (g) (g 1,2,.), dividing the data segment into k segments by taking a time difference delta T as a period, and expressing each segment of complete data as
Wherein
Indicating the ith extension data of the jth subnet;
step 2: slicing the data, and extracting and analyzing the complete data of the local area S and the time period (T-delta kt)
(time interval t ═ Δ t);
and step 3: physical space position mapping, corresponding to the deployed sub-networks, and performing space matching on the deployed sub-networks and the actual area S, wherein each sub-network host is provided with corresponding area corresponding number information and an extension deployment condition list;
and 4, step 4: sequencing the collected data according to the ID number of the collected data, and establishing a data matrix { Tower (i, S), t };
and 5: classifying the established ID data list according to the times c of appearance in different Tower, and individually extracting the IDs appearing in only one Tower for analysisGrouping ID data occurring in more than two Tower into a class of IDsc＞1；
Step 6: analyzing the ID only appearing in one Tower in the time period delta t, traversing the sub-networks around the area S if the ID is not found to repeatedly appear in the Tower in the time period delta t, searching whether the same ID number appears, and classifying the data into a class ID if the ID does not appearc＝1If the data appears in other sub-networks, marking the data, and attributing the data to the data ID in the step 5c＞1A list;
and 7: repeating the steps 4, 5 and 6 until the data processing is finished;
and 8: merging the effective data obtained by processing in the steps 5 and 6, and dividing the merged data into two types of IDsc＞1And IDc＝1；
And step 9: at time T- Δ et (e ═ 1, 2.. k), during the Δ T time period, the regional flow reserve is VT-Δet＝DT-Δet{ ID }, the real-time regional flow retention is evaluated using the following method, as shown in the following network:
from the figure, IDc＞1_x/IDc＞1Y denotes valid ID data, where IDc＞1The data in _xare normalized and directly added to the effective holding quantity, IDc＞1The data in _ y needs to pass through FT-ΔetAfter approximation by the approximate fitting function, effective inventory can be counted, and the offset reference system S represents effective reference units, such as video counting, inlet and outlet gate counting and the like. Accumulating the ID data and fitting a function F by approximationT-Δet(. o) fitting under the feedback judgment condition
The threshold range of the ratio can be adjusted according to the detection precision requirement, and the detection precision can be improved
Time (theta is the system setting allowable error)
And the data fitting work is completed to form effective data.
Step 10: calculating the time T-delta et, within the time range delta T, the regional flow: regional outflow
And area inflow
Let y be FT-ΔetN is the number of times y is within the system error range, then,
step 11: according to the area outflow quantity sequence in the time range of delta kt
And regional inflow number series
Generating an approximate function f (x) of regional outflow by adopting a data fitting algorithmoutAnd area inflow approximation function f (x)in。
Step 12: calculating the flow quantity of the T + delta T time region, substituting the time data into f (x)outAnd f (x)inAnd (4) finishing.
Step 13: and ending the flow.
As shown in fig. 4, the acquired data of a scene area of Nanjing is compared and analyzed by applying the regional flow prediction method of the present invention, and the data is substantially consistent with the comparative data of the reference system data.
Claims (3)
1. A real-time flow prediction method for a macro area is characterized by comprising the following steps: the system comprises a plurality of detection sub-networks and a remote background server, wherein the detection sub-networks are deployed in a macro area according to physical space, and each detection sub-network comprises a plurality of extension sets and a host computer in data connection with the extension sets; the extension set collects broadcast type data packets sent to the surrounding environment by mobile terminal equipment based on a WIFI protocol randomly within the coverage range of the extension set through a wireless passive sensing mode, screens the data packets with ID information of the mobile terminal equipment for retrieval, marks an extension set label and uploads the extension set label to a host, the host stores the collected data in a unified mode, marks a time label on the collected data, uploads the collected data to a remote background server for storage, and analyzes the data according to the following steps:
step 1: respectively collecting data of a complete time period (T-delta kt) by an extension set, wherein the data comprises an area A and all connecting channels R (g), g is 1,2 and …, dividing the data into k sections by taking a time difference delta T as a period, and each section of complete data is represented as Dk，
Wherein,
indicating the ith extension data of the jth subnet; n represents the number of the sub-networks, and M represents the number of the sub-networks in the jth sub-network;
step 2: slicing the data and extracting data D 'of a local area S to be analyzed and a complete time period (T-delta kt)'s，
Where t is Δ t, DisThe data collected by the extension set with the number i under the s sub-network;
and step 3: physical space position mapping, corresponding to the deployed sub-networks, and performing space matching on the deployed sub-networks and the actual area S, wherein each sub-network host is provided with corresponding area corresponding number information and an extension deployment condition list;
and 4, step 4: sequencing the data collected by each extension according to the ID number of the mobile terminal equipment, and establishing a data matrix { Tower (i, s), t };
and 5: classifying the established ID data list according to the times c of occurrence in different data matrixes: let ID denote the ID that appears in only one data matrix Towerc＝1And extracting and analyzing separately, and classifying ID data appearing in more than two data matrixes Tower into IDc＞1；
Step 6: analyzing ID data only appearing in one data matrix Tower in the time period delta t, if the ID data is not found to repeatedly appear in the data matrix Tower in the time period delta t, traversing adjacent subnets around the local area S to find whether the same ID data appears, and if the same ID data does not appear, classifying the ID data as the ID datac＝1Identifying the ID data as an ID if it is present in another subnetc＞1；
And 7: repeating the steps 4-6 until the data processing is finished;
and 8: the data obtained by the processing of the steps 5 and 6 are combined together to form two types of data IDc＝1And IDc＞1；
And step 9: at T-Deltaet, e equals to 1,2, … k, the regional flow reserve is VT-Δet，VT-Δet＝DT-Δet{ ID }, the area outflow prediction value W is predicted by the following methodoutAnd area inflow prediction value WinAnd (3) performing prediction evaluation:
(1) first, the ID is addedc＞1The middle data is subjected to data normalization processing and then is added into effective data;
(2) then, the ID is addedc＝1And IDc＞1Middle data through FT-ΔetApproximate fitting process of approximate fitting function, setting bias reference system S in the fitting processT-ΔetAnd with (S)T-Δet-(Vc＞1)T-Δet)/(Vc＝1)T-ΔetAs the feedback judgment condition, when 1-theta is less than or equal to (S)T-Δet-(Vc＞1)T-Δet)/(Vc＝1)T-ΔetCompleting fitting to obtain effective data when the sum of theta is less than or equal to 1 and theta, wherein theta is an allowable error of system setting;
(3) continuing, ID is paired byc＞1And IDc＝1Forming an effective data area with the retention volume of people stream as V through fittingT-Δet：
Wherein x and y are each IDc＞1And IDc＝1Total number of medium valid data;
(4) then, the time range T- Δ et is calculated by the equation (2) to calculate the outflow rate of the person in the local area
And inflow of water
In the formula, delta1And delta2Are respectively T1And T2A very small amount of (c);
(5) finally, according to the area outflow quantity sequence in the time range of delta kt
And regional inflow number series
Generating an approximate function f (x) of regional outflow by adopting a data fitting algorithmoutAnd area inflow approximation function f (x)inCalculating the flow rate of the T + Deltat time region, substituting the time data into f (x)outAnd f (x)inAnd then, the process is finished.
2. The macro-region real-time flow prediction method of claim 1, wherein: said biased reference frame ST-ΔetThe video count or the gate count of the inlet and the outlet of the area in the time range of T-delta et is obtained.
3. The macro-region real-time flow prediction method of claim 1, wherein: in step 9, δ1And delta2Is related to the fitting function and is determined by the following method: first, let y be F (·), n be the number of times y meets the system error range, then,
wherein the function F (-) is an approximate fitting function of the ID data, y1,y2,…,ynRespectively corresponding to function F (-) at T1,T2,…,TnOf time of dayValue, ymeanIs y1,y2,…,ynAverage value of (d) (-)1And delta2Are respectively T1And T2And is obtained by the above formula delta.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811400258.5A CN109670631B (en) | 2018-11-22 | 2018-11-22 | Real-time flow prediction method for macroscopic region |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811400258.5A CN109670631B (en) | 2018-11-22 | 2018-11-22 | Real-time flow prediction method for macroscopic region |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109670631A CN109670631A (en) | 2019-04-23 |
| CN109670631B true CN109670631B (en) | 2021-09-17 |
Family
ID=66142147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811400258.5A Active CN109670631B (en) | 2018-11-22 | 2018-11-22 | Real-time flow prediction method for macroscopic region |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109670631B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751369B (en) * | 2019-09-19 | 2024-02-27 | 金陵科技学院 | Inter-region urban flow space flow situation measurement method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9860123B2 (en) * | 2014-04-11 | 2018-01-02 | International Business Machines Corporation | Role and proximity-based management of networks |
| CN104573859A (en) * | 2014-12-31 | 2015-04-29 | 合肥城市云数据中心有限公司 | Human traffic prediction method based on Wifi positioning and cloud data processing technology |
| CN107844848B (en) * | 2016-09-20 | 2020-12-29 | 中国移动通信集团湖北有限公司 | Regional pedestrian flow prediction method and system |
| CN107564283B (en) * | 2017-08-24 | 2018-10-26 | 南京茶非氪信息科技有限公司 | A kind of macroscopical wagon flow flow detection algorithm based on WIFI signal |
-
2018
- 2018-11-22 CN CN201811400258.5A patent/CN109670631B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109670631A (en) | 2019-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cai et al. | Deep learning-based video system for accurate and real-time parking measurement | |
| KR102215041B1 (en) | Method and system for tracking an object in a defined area | |
| WO2018180588A1 (en) | Facial image matching system and facial image search system | |
| CN106295788A (en) | The statistical method of the volume of the flow of passengers and device | |
| GB2569752A (en) | Method for detecting human traffic in public place by using Wi-Fi probe | |
| CN111160243A (en) | Passenger flow volume statistical method and related product | |
| CN112465866B (en) | Multi-target track acquisition method, device, system and storage medium | |
| CN110874583A (en) | Passenger flow statistics method and device, storage medium and electronic equipment | |
| CN111325954B (en) | Personnel loss early warning method, device, system and server | |
| CN111814510B (en) | Method and device for detecting legacy host | |
| CN108810808A (en) | A kind of region passenger flow saturation computation method based on mobile phone signaling data | |
| US20170161564A1 (en) | Queue Monitoring Based on Imprint Profiles | |
| CN108710827A (en) | A kind of micro- police service inspection in community and information automatic analysis system and method | |
| Li et al. | A top–bottom clustering algorithm based on crowd trajectories for small group classification | |
| CN109670631B (en) | Real-time flow prediction method for macroscopic region | |
| CN109359689B (en) | Data identification method and device | |
| WO2022126668A1 (en) | Method for pedestrian identification in public places and human flow statistics system | |
| CN109511109B (en) | Real-time pedestrian flow reserve prediction method for macroscopic region | |
| CN111210634B (en) | Intelligent traffic information processing method and device, intelligent traffic system and server | |
| CN109362034B (en) | Macroscopic region connecting channel heat degree prediction method | |
| CN109657701B (en) | Real-time pedestrian flow retention amount detection method for macroscopic region | |
| CN109657700B (en) | Macroscopic region communicating channel heat degree detection method | |
| CN109525981B (en) | Real-time flow detection method for macroscopic region | |
| Das et al. | Cam-stitch: Trajectory cavity stitching method for stereo vision cameras in a public building | |
| Gupta et al. | A comparative study of crowd counting and profiling through visual and non-visual sensors |
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 | ||
| GR01 | Patent grant |