CN109034460A - Prediction technique, device and system for scenic spot passenger flow congestion level - Google Patents
Prediction technique, device and system for scenic spot passenger flow congestion level Download PDFInfo
- Publication number
- CN109034460A CN109034460A CN201810711677.4A CN201810711677A CN109034460A CN 109034460 A CN109034460 A CN 109034460A CN 201810711677 A CN201810711677 A CN 201810711677A CN 109034460 A CN109034460 A CN 109034460A
- Authority
- CN
- China
- Prior art keywords
- scenic spot
- passenger flow
- congestion level
- flow congestion
- tourist
- 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.)
- Pending
Links
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"
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Abstract
A kind of prediction technique for scenic spot passenger flow congestion level, device and system, method include: to obtain scenic spot map, and carry out grid dividing to scenic spot map；The location information of tourist in scenic spot is obtained, in real time to obtain the volume of the flow of passengers of each grid；Pass through scenic spot trails history passenger flow data training grid passenger flow congestion level model using convolutional neural networks algorithm, to predict scenic spot future time period passenger flow congestion level, scenic spot passenger flow congestion level can be predicted according to scenic spot map and the location information of tourist, to solve the problems, such as not predicting passenger flow congestion level, it is predicted based on congestion level, the generation of scenic spot safety accident can effectively be prevented, and provided personalized service for tourist, the experience of tourist is promoted.
Description
Technical field
The present invention relates to scenic spot traffic prediction technology fields, and in particular to a kind of prediction for scenic spot passenger flow congestion level
Methods, devices and systems.
Background technique
The scenic spot quantity to open for free with the current country is increasing, and it is crowded that tourist all occur in festivals or holidays many scenic spots
The phenomenon that.In order to ensure tourist security, scenic spot density of stream of people situation is grasped in time, becomes particularly important.
In the related technology, tourist's number is usually calculated by statistics presell admission ticket and gate passage amount.But related skill
The problem of art is, since the real-time volume of the flow of passengers to each sight spot in scenic spot inside but can not be obtained accurately, to be easy to cause scenic spot
Interior sight spot tourist is crowded, security risk occurs, so that tourist experience is poor.
Summary of the invention
The application provides a kind of prediction technique for scenic spot passenger flow congestion level, and it is accurate to realize to the scenic spot volume of the flow of passengers
Prediction.
According in a first aspect, provide a kind of prediction technique for scenic spot passenger flow congestion level in a kind of embodiment, including
Following steps: scenic spot map is obtained, and grid dividing is carried out to scenic spot map；The location information of tourist in scenic spot is obtained in real time,
To obtain the volume of the flow of passengers of each grid；Pass through scenic spot trails history passenger flow data training grid visitor using convolutional neural networks algorithm
Congestion level model is flowed, to predict scenic spot future time period passenger flow congestion level.
Further, the real-time location information for obtaining tourist in scenic spot, comprising: obtain and respectively moved in the scenic spot map
The location information of dynamic terminal and information reporting time；First of frequency of occurrence greater than the first preset value moves within a preset time for identification
Dynamic terminal and/or within a preset time second mobile terminal of the frequency of occurrence less than the second preset value；The first movement is whole
The location information and information reporting time of end and the second mobile terminal are rejected, to obtain the location information of tourist in the scenic spot.
Further, described to pass through scenic spot trails history passenger flow data training grid passenger flow using convolutional neural networks algorithm
Congestion level model, further includes: the volume of the flow of passengers in the scenic spot is converted into eigenmatrix；Convolution is carried out to the eigenmatrix
It calculates；Eigenmatrix after convolution is subjected to pond；The eigenmatrix behind the pond is converted by activation primitive；It will
The eigenmatrix converted by activation primitive inputs full articulamentum；According to full articulamentum as a result, being obtained by classifier
The passenger flow congestion level.
Further, the prediction technique for scenic spot passenger flow congestion level further include: by the passenger flow of prediction
Congestion level is sent to the mobile terminal of tourist.
According to second aspect, a kind of computer readable storage medium, including program, described program are provided in a kind of embodiment
It can be executed by processor to realize the prediction technique for scenic spot passenger flow congestion level.
According to the third aspect, a kind of forecasting system for scenic spot passenger flow congestion level is provided in a kind of embodiment, comprising:
Memory, processor and it is stored in the computer program that can be run on the memory and on the processor, the place
When managing the device execution computer program, the prediction technique for scenic spot passenger flow congestion level is realized.
According to fourth aspect, a kind of prediction meanss for scenic spot passenger flow congestion level are provided in a kind of embodiment, comprising:
First obtains module, carries out grid dividing for obtaining scenic spot map, and to scenic spot map；Second obtains module, for real-time
The location information of tourist in scenic spot is obtained, to obtain the volume of the flow of passengers of each grid；Object module, for utilizing convolutional neural networks
Algorithm is by scenic spot trails history passenger flow data training grid passenger flow congestion level model, to scenic spot future time period passenger flow congestion
Degree is predicted.
Further, described second module is obtained, be also used to: obtaining the position letter of each mobile terminal in the scenic spot map
Breath and information reporting time；Identification within a preset time frequency of occurrence be greater than the first preset value first movement terminal and/or
Second mobile terminal of the frequency of occurrence less than the second preset value in preset time；The first movement terminal and second is mobile eventually
The location information at end and information reporting time reject, to obtain the location information of tourist in the scenic spot.
Further, the object module, is also used to: the volume of the flow of passengers in the scenic spot is converted to eigenmatrix；To institute
It states eigenmatrix and carries out convolutional calculation；Eigenmatrix after convolution is subjected to pond；It will be behind the pond by activation primitive
Eigenmatrix is converted；The eigenmatrix converted by activation primitive is inputted into full articulamentum；According to full articulamentum
As a result, obtaining the passenger flow congestion level by classifier.
Further, the object module, is also used to: by the passenger flow congestion level of prediction, being sent to the shifting of tourist
Dynamic terminal.
According to the prediction technique for scenic spot passenger flow congestion level of above-described embodiment, by obtaining scenic spot map, and it is right
Scenic spot map carries out grid dividing, obtains the location information of tourist in scenic spot in real time, the volume of the flow of passengers of each grid is obtained, thus sharp
With convolutional neural networks algorithm by scenic spot trails history passenger flow data training grid passenger flow congestion level model, with to scenic spot not
Carry out period passenger flow congestion level to be predicted.The prediction technique of the embodiment of the present invention can be according to scenic spot map and tourist as a result,
Location information scenic spot passenger flow congestion level is predicted, to solve asking of can not being predicted passenger flow congestion level
Topic is predicted based on congestion level, can effectively prevent the generation of scenic spot safety accident, and provide personalized service for tourist,
Promote the experience of tourist.
Detailed description of the invention
Fig. 1 is the flow chart of the prediction technique for scenic spot passenger flow congestion level of the embodiment of the present invention；
Fig. 2 is the scenic spot map that the Summer Palace after grid dividing is carried out using 7 GeoHash values；
Fig. 3 is the volume of the flow of passengers for obtaining in the scenic spot of the Summer Palace tourist in the scenic spot of a period more than a day；
Fig. 4 is the structural schematic diagram of convolution kernel；
Fig. 5 is the volume of the flow of passengers schematic diagram in the past 1 year of grid g1；
Fig. 6 is characterized the demonstration schematic diagram of matrix pool process；
Fig. 7 is the structural schematic diagram of activation primitive；And
Fig. 8 is the block diagram of the prediction meanss for scenic spot passenger flow congestion level of the embodiment of the present invention.
Specific embodiment
Below by specific embodiment combination attached drawing, invention is further described in detail.Wherein different embodiments
Middle similar component uses associated similar element numbers.In the following embodiments, many datail descriptions be in order to
The application is better understood.However, those skilled in the art can recognize without lifting an eyebrow, part of feature
It is dispensed, or can be substituted by other elements, material, method in varied situations.In some cases, this Shen
Please it is relevant it is some operation there is no in the description show or describe, this is the core in order to avoid the application by mistake
More descriptions are flooded, and to those skilled in the art, these relevant operations, which are described in detail, not to be necessary, they
Relevant operation can be completely understood according to the general technology knowledge of description and this field in specification.
It is formed respectively in addition, feature described in this description, operation or feature can combine in any suitable way
Kind embodiment.Meanwhile each step in method description or movement can also can be aobvious and easy according to those skilled in the art institute
The mode carry out sequence exchange or adjustment seen.Therefore, the various sequences in the description and the appended drawings are intended merely to clearly describe a certain
A embodiment is not meant to be necessary sequence, and wherein some sequentially must comply with unless otherwise indicated.
It is herein component institute serialization number itself, such as " first ", " second " etc., is only used for distinguishing described object,
Without any sequence or art-recognized meanings.And " connection ", " connection " described in the application, unless otherwise instructed, include directly and
It is indirectly connected with (connection).
It should be noted that traditional method calculated based on region passenger flow, is usually set using gate machine, infrared ray
The means such as standby, ticket sale system, WIFI probe or video monitoring obtain tourist's data, are then based on history passenger flow data and carry out phase
Close statistical analysis.But the method for the calculating volume of the flow of passengers in the related technology is not only at high cost, but also anti-interference is not strong, Chang Yin
The multifactor influence such as weather, illumination, causes collected data imperfect, so that prediction result accuracy rate is low, availability is not
By force.
Based on this, the embodiment of the present invention, which proposes, a kind of for the prediction technique of scenic spot passenger flow congestion level, device and is
System.
Below with reference to the accompanying drawings the prediction technique and system for scenic spot passenger flow congestion level of the embodiment of the present invention described.
Fig. 1 is the flow chart of the prediction technique for scenic spot passenger flow congestion level of the embodiment of the present invention.As shown in Figure 1,
The prediction technique for scenic spot passenger flow congestion level of the embodiment of the present invention, comprising the following steps:
S101: scenic spot map is obtained, and grid dividing is carried out to scenic spot map.
It should be noted that scenic spot map can be the map for including scenic spot region.Since the global shape at scenic spot is not
Rule, by acquiring the position of mobile terminal, usually it is unable to judge accurately whether mobile terminal is in scenic spot, therefore, needs
Grid dividing is carried out to scenic spot.
Wherein, scenic spot map can be subjected to grid dividing by GeoHash (perceptual knowledge) algorithm.GeoHash is a kind of
Mature address coding method, it can be the longitude and latitude data encoding of two-dimensional space at a character string, that is, understands the earth
For a two-dimensional surface, plane recurrence is resolved into smaller sub-block, each sub-block possesses identical within the scope of certain longitude and latitude
Coding.
That is, in the location information for the mobile terminal for getting tourist, can by the longitude and latitude in location information into
Row judgement, determines longitude and latitude range locating for the location information, and carries out unified volume according to the longitude and latitude range locating for it to it
Code, so that it is determined that grid locating for the mobile terminal of tourist, the i.e. location information of tourist.
Further, it is contemplated that be easy the presence of drift by the longitude and latitude report point information that the mobile terminal of tourist acquires tourist
It moves, the location information that the tourist at scenic spot edge reports can float to outside scenic spot, therefore, in embodiments of the present invention, using 7
The GeoHash value of length, that is, when carrying out grid dividing using GeoHash, along scenic spot edge to extending out 7 GeoHash
The grid of value, error are about 76 meters.
For example, Fig. 2 is the scenic spot map that the Summer Palace after grid dividing is carried out using 7 GeoHash values.Wherein,
Grid after division is 200.
S102: the location information of tourist in scenic spot is obtained, in real time to obtain the volume of the flow of passengers of each grid.
According to one embodiment of present invention, the location information of tourist in scenic spot is obtained in real time, comprising: is obtained each in scenic spot
The location information of mobile terminal and information reporting time；Frequency of occurrence is greater than the first of the first preset value within a preset time for identification
The second mobile terminal of mobile terminal and/or within a preset time frequency of occurrence less than the second preset value；By first movement terminal
It is rejected with the location information of the second mobile terminal and information reporting time, to obtain the location information of tourist in scenic spot.
It should be understood that position built-in in mobile terminal can be passed through when user uses mobile terminal in scenic spot
Obtain current location information and information reporting time that module obtains mobile terminal.But comprehensively obtain the position of mobile terminal
Confidence breath and information reporting time, noise data can be got, for example, the location information of scenic spot management person, to live in scenic spot attached
The location information of the location information of close resident and the people accidentally passed by, it is therefore, accurate and effective for data, it is also necessary to will
Above-mentioned noise data removal.
Specifically, obtain scenic spot map in each mobile terminal location information and after the information reporting time, identification exist
Preset time (for example, in three months) interior frequency of occurrence is greater than the first movement terminal of the first preset value (for example, 10 times), determines
First movement terminal is scenic spot staff or periphery civic mobile terminal, corresponding location information and information reporting time
It should be rejected, and in preset time (for example, in three months) interior frequency of occurrence less than the second preset value (for example, 2 times)
The second mobile terminal, determine that the second mobile terminal is the mobile terminal of the people accidentally passed by, corresponding location information and letter
Calling time on breath should be rejected.
It should be understood that being occurred getting in effective scenic spot after the location information of tourist according to tourist in grid
Quantity, can be obtained the volume of the flow of passengers of each grid.For example, when grid g1 gets the location information of 5 tourists, it is determined that
The volume of the flow of passengers of grid g1 is 5.Specifically, the volume of the flow of passengers should be corresponding with the time, that is, when grid g1 gets 5 trips in the T period
When the location information of visitor, it is determined that the volume of the flow of passengers of the grid g1 in the T period is 5.
By taking the grid chart in the above-mentioned Summer Palace as an example, the grid based on Fig. 2, Fig. 3 is a more than a day in acquisition Summer Palace scenic spot
The volume of the flow of passengers of tourist in the scenic spot of period.
S103: pass through scenic spot trails history passenger flow data training grid passenger flow congestion level using convolutional neural networks algorithm
Model, to predict scenic spot future time period passenger flow congestion level.
According to one embodiment of present invention, it is instructed using convolutional neural networks algorithm by scenic spot trails history passenger flow data
Practice grid passenger flow congestion level model, comprising:
S201: the volume of the flow of passengers in scenic spot is converted into eigenmatrix.
It should be noted that the volume of the flow of passengers converted is the scenic spot of each period to be predicted daily in the past in 1 year
The interior volume of the flow of passengers.It should be understood that when only need to the specific T period carry out passenger flow congestion level prediction when, can also only by
In the scenic spot, the volume of the flow of passengers of the T period in past 1 year is converted.
Wherein, the past 1 year was using current date as 360 days of starting point.
Wherein, eigenmatrix is the matrix of 12*30.That is, need to pass by 1 year in scenic spot, i.e., 360 days,
The volume of the flow of passengers of tourist is converted to the matrix of 12*30.
It should also be noted that, in embodiments of the present invention, just for a period of a grid to be once prediction
Object specifically when the passenger flow situation of predicted grid g1 T tomorrow period, is then needed 360 days before from today, grid
The practical volume of the flow of passengers of the g1 in the T period is converted to eigenmatrix.
S202: convolutional calculation is carried out to eigenmatrix.
Wherein, as shown in Figure 4, the convolution kernel progress convolutional calculation of 3*3 can be used.
For example, by n1, n2 ..., n360 be grid g1 go over 1 year in the volume of the flow of passengers for, as shown in figure 5, will
N1, n2 ..., n360 fills to generating eigenmatrix in the matrix of 12*30.Then to n1, n2 ..., each of n360, into
Row convolutional calculation:
N1 '=1*0+0*0+1*0+0*0+1*0+0*0+1*0+0*0+1*n1；
N2 '=1*0+0*0+1*0+0*0+1*0+0*0+1*0+0*n1+1*n2；
N3 '=1*0+0*0+1*0+0*0+1*0+0*0+1*n1+0*n2+1*n3；
N4 '=1*0+0*0+1*0+0*0+1*0+0*n1+1*0+0*0+1*n4；
……
N9 '=1*n1+0*n2+1*n3+0*n4+1*n5+0*n6+1*n7+0*n8+1*n9；
……
Wherein, n1 ', n2 ', n3 ', n4 ', n9 ' are the characteristic value in the eigenmatrix after convolution.
S203: the eigenmatrix after convolution is subjected to pond.
It should be noted that the purpose of pond layer is compressed to the eigenmatrix after convolution, on the one hand make convolution
Eigenmatrix afterwards becomes smaller, and is on the other hand to further extract main feature to simplify network query function complexity.
In this law embodiment, pondization operation is carried out using max Pooling, as shown in fig. 6, with the filtering of a 2*2
Device sums again divided by 4 to the region of 2*2 each in eigenmatrix, to obtain main feature value, can will roll up as a result,
Eigenmatrix after product is compressed into original 1/4, realizes the purpose for reducing the feature vector of convolutional layer output, while can also change
It is apt to as a result, result is made to be less prone to over-fitting.Wherein, Fig. 6 is only the demonstration of eigenmatrix pond process, data value and this reality
It is unrelated to apply example.
S204: the eigenmatrix of Chi Huahou is converted by activation primitive.
Wherein, ReLU function can be used to convert the eigenmatrix of Chi Huahou.Wherein, ReLU function can for f (x)=
max(0,x)。
It should be noted that ReLU function is as shown in Figure 7, wherein x is the characteristic value of Chi Huahou, and as x > 0, gradient is permanent
It is 1, no gradient dissipation issues, convergence is fastly；As x < 0, the output of this layer is 0, and 0 neuron is more after the completion of training, sparse
Property it is bigger, the feature extracted is more representative, generalization ability it is stronger to get arrive same effect, really work
Neuron it is fewer, the Generalization Capability of network is better, increases the sparsity of network, and operand is smaller.
Specifically, each characteristic value in the eigenmatrix of Chi Huahou is converted, when characteristic value x is less than 0, feature
The numerical value value of value x is changed to 0, i.e. x '=0, and when characteristic value x is greater than 0, then the numerical value of characteristic value x is constant, i.e. x '=x.
S205: the eigenmatrix converted by activation primitive is inputted into full articulamentum.
Wherein, following formula can be used in full articulamentum:
npre=w*x '+b
Wherein, w is weight, and b is bigoted amount, and x ' is the characteristic value converted by activation primitive, npreFor the defeated of full articulamentum
Out.
Wherein, w is randomly assigned when model training starts for each x ' one weight of corresponding distribution and bigoted amount
With the initial value of b, wherein w need to meet normal distribution, and standard deviation 0.1, default maximum 1, minimum value is -1, and mean value is
0；The initial value of b is 0.1.
S206: according to full articulamentum as a result, obtaining passenger flow congestion level by classifier.
It should be noted that classifier can classify to the training result of full articulamentum.
In embodiments of the present invention, the output of classifier is passenger flow congestion level, wherein passenger flow congestion level can be divided into 4
A grade: heavy congestion, congestion, unobstructed and comfortable.
Specifically, the classifier of the embodiment of the present invention can be used following formula and classify:
Wherein, NAlwaysFor the at most open ended passenger flow total amount in scenic spot, npreFor the output of full articulamentum, m is the net of scenic spot division
Lattice sum, for example, the grid number at Summer Palace scenic spot is 200.
Further, when getting y≤0.4, classifier exports result " comfortable "；As 0.4 < y≤0.6, classifier
Exporting result is " unobstructed "；As 0.6 < y≤0.8, the result of classifier output is " congestion "；As y > 0.8, model output
For " heavy congestion ".
In the present invention is real-time, as the numerical value that classifier calculated obtains is bigger, passenger flow congestion level is more serious, works as classification
The result of device be " comfortable " when, that is, indicate predict grid g1 it is less in the period scenic spot T passenger flow, tourist can arbitrarily arrange the time into
Row visit；When the result of classifier is " unobstructed ", that is, indicates that the grid g1 of prediction is slightly more than in T period scenic spot passenger flow and " relax
It is suitable ", but do not influence to pass through；When the result of classifier be " congestion " when, that is, indicate prediction grid g1 the period scenic spot T passenger flow very
It is more, it can not be successfully passage；When the result of classifier is " heavy congestion ", that is, indicate the grid g1 of prediction in the period scenic spot T visitor
Flow very more, easy initiation safety accident.
Further, it is based on identical method, also each period to each grid can carry out passenger flow congestion level
Prediction, so as to obtain the passenger flow congestion prediction result as unit of day.
As a result, after scenic spot passenger flow congestion level is effectively predicted, the administrative department at scenic spot can be made to tie according to prediction
Fruit is deployed troops on garrison duty in advance, is carried out the stream of people at the period of " congestion " and " heavy congestion " and sight spot and the work such as is dredged, so as to effectively pre-
Safety accident in anti-scenic spot occurs, and while guaranteeing that scenic spot is safe, promotes the experience of tourist.
According to one embodiment of present invention, also the passenger flow congestion level of prediction can be sent to the mobile terminal of tourist,
It to make tourist in tourism process, selectively avoids going to the sight spot in sight spot heavy congestion, to realize sight of avoiding the peak hour
The purpose of reward reduces the generation of safety accident in scenic spot, promotes the experience of user.
Further, to meet the growing individual demand of tourist, for scenic spot, there is bigger development machine
Meeting.In today of Internet technology prosperity, the technologies such as wifi covering that internet is booked tickets, core area is wireless all are promoting to travel
The upgrading of service.
Based on this, the passenger flow congestion level of prediction can be also sent to joint third party's mobile network data service provider, such as
Aurora big data, drawn a portrait in conjunction with the individual and group of tourist in scenic spot according to these information providing area passenger flow analysings, point
It analyses on tourist's line, Behavior preference under line, is supported to improve its commerce services for scenic spot and provide reliable data, so that scenic spot energy
It enough provides personalized service for tourist, for example, pushing the reservation date after it reserves admission ticket for the tourist for not liking congestion for it
The passenger flow forecast situation of phase can select the specific travel time according to the passenger flow situation of prediction, to promote tourist's
Experience；For another example, the mostly young visitor that certain scenic spot is received for a long time is analyzed, then scenic spot management person can be according to young visitor
Consumption preferences, provided for tourist landscape it is ornamental except, matched, convenient, comfortable, characteristic lodging environment is provided,
With meet night entertainment requirements etc., specifically, can literary trip's performing art be " night is economical " of representative, landscape business block, literature and art
Or the Recreational places such as folk custom performance, light section, scenic spot periphery or internal hot spring, health carried, so that tourist can enjoy
By night life abundant, thus so enchanted by the scenery as to forget to return.Prediction as a result, based on passenger flow congestion level, can rationally excavate visitor behavior
Preference is that the service of tourist's design personalized attracts tourists to preferably promote the consumption experience of tourist, keeps tourist here.
In conclusion the prediction technique for scenic spot passenger flow congestion level of above-described embodiment, by obtaining scenic spot map,
And grid dividing is carried out to scenic spot map, the location information of tourist in scenic spot is obtained in real time, obtains the volume of the flow of passengers of each grid, from
And convolutional neural networks algorithm is utilized by scenic spot trails history passenger flow data training grid passenger flow congestion level model, to scape
Area's future time period passenger flow congestion level is predicted.As a result, the prediction technique of the embodiment of the present invention can according to scenic spot map and
The location information of tourist predicts scenic spot passenger flow congestion level, so that passenger flow congestion level can not be predicted by solving
The problem of, it is predicted based on congestion level, can effectively prevent the generation of scenic spot safety accident, and provide personalized clothes for tourist
Business, promotes the experience of tourist.
The embodiment of the present invention also proposed a kind of non-transitory readable storage medium storing program for executing, including program, which can be located
Device is managed to execute to realize the prediction technique for being previously described for scenic spot passenger flow congestion level.
Non-transitory readable storage medium storing program for executing according to an embodiment of the present invention can be believed according to the position of scenic spot map and tourist
Breath predicts scenic spot passenger flow congestion level, to solve the problems, such as not predicting passenger flow congestion level, is based on
Congestion level prediction, can effectively prevent the generation of scenic spot safety accident, promote the experience of tourist.
The embodiment of the present invention also proposed a kind of forecasting system for scenic spot passenger flow congestion level, including memory, place
The computer program managing device and storage on a memory and can running on a processor, when processor executes computer program,
Realize the prediction technique for being previously described for scenic spot passenger flow congestion level.
Forecasting system according to an embodiment of the present invention for scenic spot passenger flow congestion level, can be according to scenic spot map and trip
The location information of visitor predicts scenic spot passenger flow congestion level, to solve and can not be predicted passenger flow congestion level
Problem is predicted based on congestion level, can effectively prevent the generation of scenic spot safety accident, promote the experience of tourist.
Corresponding for the prediction technique of scenic spot passenger flow congestion level, of the invention one is provided with above-mentioned several embodiments
Kind embodiment additionally provides the prediction meanss for scenic spot passenger flow congestion level, is used for scenic spot due to provided in an embodiment of the present invention
The prediction technique phase for scenic spot passenger flow congestion level that the prediction meanss of passenger flow congestion level are provided with above-mentioned several embodiments
It is corresponding, therefore be also applied in the embodiment for the prediction technique for being previously used for scenic spot passenger flow congestion level provided in this embodiment
For the prediction meanss of scenic spot passenger flow congestion level, no longer describe in the present embodiment.
Fig. 8 is the block diagram of the prediction meanss for scenic spot passenger flow congestion level of the embodiment of the present invention.Such as Fig. 8 institute
Show, the prediction meanss 100 for scenic spot passenger flow congestion level of the embodiment of the present invention, comprising: the first acquisition module 10, second are obtained
Modulus block 20 and object module 30.
Wherein, the first acquisition module 10 carries out grid dividing for obtaining scenic spot map, and to scenic spot map；Second obtains
Module 20 for obtaining the location information of tourist in scenic spot in real time, to obtain the volume of the flow of passengers of each grid；Object module 30 is used for
Using convolutional neural networks algorithm by scenic spot trails history passenger flow data training grid passenger flow congestion level model, to scenic spot
Future time period passenger flow congestion level is predicted.
Further, the second acquisition module 20 is also used to: obtaining the location information and letter of each mobile terminal in the map of scenic spot
It calls time on breath；Frequency of occurrence is greater than the first movement terminal of the first preset value and/or when default within a preset time for identification
Second mobile terminal of the interior frequency of occurrence less than the second preset value；The position of first movement terminal and the second mobile terminal is believed
Breath and information reporting time reject, to obtain the location information of tourist in scenic spot.
Further, object module 30 are also used to: the volume of the flow of passengers in scenic spot is converted to eigenmatrix；To eigenmatrix
Carry out convolutional calculation；Eigenmatrix after convolution is subjected to pond；The eigenmatrix of Chi Huahou is turned by activation primitive
It changes；The eigenmatrix converted by activation primitive is inputted into full articulamentum；According to full articulamentum as a result, being obtained by classifier
Passenger flow congestion level.
Further, object module 30 are also used to: by the passenger flow congestion level of prediction, being sent to the mobile terminal of tourist.
It will be understood by those skilled in the art that all or part of function of various methods can pass through in above embodiment
The mode of hardware is realized, can also be realized by way of computer program.When function all or part of in above embodiment
When being realized by way of computer program, which be can be stored in a computer readable storage medium, and storage medium can
To include: read-only memory, random access memory, disk, CD, hard disk etc., it is above-mentioned to realize which is executed by computer
Function.For example, program is stored in the memory of equipment, when executing program in memory by processor, can be realized
State all or part of function.In addition, when function all or part of in above embodiment is realized by way of computer program
When, which also can store in storage mediums such as server, another computer, disk, CD, flash disk or mobile hard disks
In, through downloading or copying and saving into the memory of local device, or version updating is carried out to the system of local device, when logical
When crossing the program in processor execution memory, all or part of function in above embodiment can be realized.
Use above specific case is illustrated the present invention, is merely used to help understand the present invention, not to limit
The system present invention.For those skilled in the art, according to the thought of the present invention, can also make several simple
It deduces, deform or replaces.
Claims (10)
1. a kind of prediction technique for scenic spot passenger flow congestion level, which comprises the following steps:
Scenic spot map is obtained, and grid dividing is carried out to scenic spot map；
The location information of tourist in scenic spot is obtained, in real time to obtain the volume of the flow of passengers of each grid；
Using convolutional neural networks algorithm by scenic spot trails history passenger flow data training grid passenger flow congestion level model, with right
Scenic spot future time period passenger flow congestion level is predicted.
2. the prediction technique according to claim 1 for scenic spot passenger flow congestion level, which is characterized in that described to obtain in real time
It finds a view the location information of tourist in area, comprising:
Obtain the location information of each mobile terminal and information reporting time in the scenic spot map；
Frequency of occurrence is greater than the first movement terminal of the first preset value and/or occurs within a preset time within a preset time for identification
Second mobile terminal of the number less than the second preset value；
The location information and information reporting time of the first movement terminal and the second mobile terminal are rejected, to obtain the scape
The location information of tourist in area.
3. the prediction technique according to claim 1 for scenic spot passenger flow congestion level, which is characterized in that described to utilize volume
Product neural network algorithm passes through scenic spot trails history passenger flow data training grid passenger flow congestion level model, further includes:
The volume of the flow of passengers in the scenic spot is converted into eigenmatrix；
Convolutional calculation is carried out to the eigenmatrix；
Eigenmatrix after convolution is subjected to pond；
The eigenmatrix behind the pond is converted by activation primitive；
The eigenmatrix converted by activation primitive is inputted into full articulamentum；
According to full articulamentum as a result, obtaining the passenger flow congestion level by classifier.
4. the prediction technique according to claim 1 for scenic spot passenger flow congestion level, which is characterized in that further include:
By the passenger flow congestion level of prediction, it is sent to the mobile terminal of tourist.
5. a kind of computer readable storage medium, feature exist, including program, described program can be executed by processor to realize
Such as the prediction technique of any of claims 1-4 for scenic spot passenger flow congestion level.
6. a kind of forecasting system for scenic spot passenger flow congestion level characterized by comprising memory, processor and deposit
The computer program that can be run on the memory and on the processor is stored up, the processor executes the computer journey
When sequence, such as the prediction technique of any of claims 1-4 for scenic spot passenger flow congestion level is realized.
7. a kind of prediction meanss for scenic spot passenger flow congestion level characterized by comprising
First obtains module, carries out grid dividing for obtaining scenic spot map, and to scenic spot map；
Second obtains module, for obtaining the location information of tourist in scenic spot in real time, to obtain the volume of the flow of passengers of each grid；
Object module, for passing through the training grid passenger flow congestion of scenic spot trails history passenger flow data using convolutional neural networks algorithm
Degree Model, to predict scenic spot future time period passenger flow congestion level.
8. the prediction meanss according to claim 7 for scenic spot passenger flow congestion level, which is characterized in that described second obtains
Modulus block, is also used to:
Obtain the location information of each mobile terminal and information reporting time in the scenic spot map；
Frequency of occurrence is greater than the first movement terminal of the first preset value and/or occurs within a preset time within a preset time for identification
Second mobile terminal of the number less than the second preset value；
The location information and information reporting time of the first movement terminal and the second mobile terminal are rejected, to obtain the scape
The location information of tourist in area.
9. the prediction meanss according to claim 7 for scenic spot passenger flow congestion level, which is characterized in that the result mould
Block is also used to:
The volume of the flow of passengers in the scenic spot is converted into eigenmatrix；
Convolutional calculation is carried out to the eigenmatrix；
Eigenmatrix after convolution is subjected to pond；
The eigenmatrix behind the pond is converted by activation primitive；
The eigenmatrix converted by activation primitive is inputted into full articulamentum；
According to full articulamentum as a result, obtaining the passenger flow congestion level by classifier.
10. the prediction meanss according to claim 7 for scenic spot passenger flow congestion level, which is characterized in that the result
Module is also used to:
By the passenger flow congestion level of prediction, it is sent to the mobile terminal of tourist.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810711677.4A CN109034460A (en) | 2018-07-03 | 2018-07-03 | Prediction technique, device and system for scenic spot passenger flow congestion level |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810711677.4A CN109034460A (en) | 2018-07-03 | 2018-07-03 | Prediction technique, device and system for scenic spot passenger flow congestion level |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109034460A true CN109034460A (en) | 2018-12-18 |
Family
ID=65521328
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810711677.4A Pending CN109034460A (en) | 2018-07-03 | 2018-07-03 | Prediction technique, device and system for scenic spot passenger flow congestion level |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109034460A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858388A (en) * | 2019-01-09 | 2019-06-07 | 武汉中联智诚科技有限公司 | A kind of intelligent tourism management system |
| CN109946724A (en) * | 2019-03-29 | 2019-06-28 | 江苏小牛电动科技有限公司 | A kind of GPS static drift modification method and its device based on GEOHASH algorithm |
| CN110147904A (en) * | 2019-04-23 | 2019-08-20 | 深圳先进技术研究院 | A kind of urban agglomerations event prediction and localization method and device |
| CN110544115A (en) * | 2019-08-16 | 2019-12-06 | 北京慧辰资道资讯股份有限公司 | Method and device for analyzing characteristics of tourists from scenic spot tourism big data |
| CN110827174A (en) * | 2019-09-27 | 2020-02-21 | 恒大智慧科技有限公司 | Method and device for allocating resources in scenic spot and storage medium |
| CN111372196A (en) * | 2020-02-20 | 2020-07-03 | 杭州海康威视系统技术有限公司 | Data processing method and device, electronic equipment and machine-readable storage medium |
| CN111489131A (en) * | 2019-01-25 | 2020-08-04 | 北京搜狗科技发展有限公司 | Information recommendation method and device |
| CN112884267A (en) * | 2019-11-29 | 2021-06-01 | 重庆小雨点小额贷款有限公司 | Travel scheme planning method and device and readable storage medium |
| JP2021196796A (en) * | 2020-06-12 | 2021-12-27 | ヤフー株式会社 | Information processing device, information processing method, and information processing program |
| CN113947758A (en) * | 2021-12-16 | 2022-01-18 | 北京凯泰铭科技文化发展有限公司 | Big data method and system of sponge system based on scenic spot chessboard division |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160140424A1 (en) * | 2014-11-13 | 2016-05-19 | Nec Laboratories America, Inc. | Object-centric Fine-grained Image Classification |
| CN107103758A (en) * | 2017-06-08 | 2017-08-29 | 厦门大学 | A kind of city area-traffic method for predicting based on deep learning |
| CN107529651A (en) * | 2017-08-18 | 2018-01-02 | 北京航空航天大学 | A kind of urban transportation passenger flow forecasting and equipment based on deep learning |
| CN107833161A (en) * | 2017-10-10 | 2018-03-23 | 东南大学 | A kind of tourist communications management system based on big data |
| CN107967532A (en) * | 2017-10-30 | 2018-04-27 | 厦门大学 | The Forecast of Urban Traffic Flow Forecasting Methodology of integration region vigor |
| CN108053080A (en) * | 2017-12-30 | 2018-05-18 | 中国移动通信集团江苏有限公司 | Zone user quantity statistics value Forecasting Methodology, device, equipment and medium |
-
2018
- 2018-07-03 CN CN201810711677.4A patent/CN109034460A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160140424A1 (en) * | 2014-11-13 | 2016-05-19 | Nec Laboratories America, Inc. | Object-centric Fine-grained Image Classification |
| CN107103758A (en) * | 2017-06-08 | 2017-08-29 | 厦门大学 | A kind of city area-traffic method for predicting based on deep learning |
| CN107529651A (en) * | 2017-08-18 | 2018-01-02 | 北京航空航天大学 | A kind of urban transportation passenger flow forecasting and equipment based on deep learning |
| CN107833161A (en) * | 2017-10-10 | 2018-03-23 | 东南大学 | A kind of tourist communications management system based on big data |
| CN107967532A (en) * | 2017-10-30 | 2018-04-27 | 厦门大学 | The Forecast of Urban Traffic Flow Forecasting Methodology of integration region vigor |
| CN108053080A (en) * | 2017-12-30 | 2018-05-18 | 中国移动通信集团江苏有限公司 | Zone user quantity statistics value Forecasting Methodology, device, equipment and medium |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858388A (en) * | 2019-01-09 | 2019-06-07 | 武汉中联智诚科技有限公司 | A kind of intelligent tourism management system |
| CN111489131A (en) * | 2019-01-25 | 2020-08-04 | 北京搜狗科技发展有限公司 | Information recommendation method and device |
| CN109946724A (en) * | 2019-03-29 | 2019-06-28 | 江苏小牛电动科技有限公司 | A kind of GPS static drift modification method and its device based on GEOHASH algorithm |
| CN109946724B (en) * | 2019-03-29 | 2024-01-16 | 江苏小牛电动科技有限公司 | GPS static drift correction method and device based on GEOHOASH algorithm |
| CN110147904A (en) * | 2019-04-23 | 2019-08-20 | 深圳先进技术研究院 | A kind of urban agglomerations event prediction and localization method and device |
| CN110147904B (en) * | 2019-04-23 | 2021-06-18 | 深圳先进技术研究院 | Urban gathering event prediction and positioning method and device |
| CN110544115A (en) * | 2019-08-16 | 2019-12-06 | 北京慧辰资道资讯股份有限公司 | Method and device for analyzing characteristics of tourists from scenic spot tourism big data |
| CN110827174A (en) * | 2019-09-27 | 2020-02-21 | 恒大智慧科技有限公司 | Method and device for allocating resources in scenic spot and storage medium |
| CN112884267A (en) * | 2019-11-29 | 2021-06-01 | 重庆小雨点小额贷款有限公司 | Travel scheme planning method and device and readable storage medium |
| CN111372196A (en) * | 2020-02-20 | 2020-07-03 | 杭州海康威视系统技术有限公司 | Data processing method and device, electronic equipment and machine-readable storage medium |
| JP2021196796A (en) * | 2020-06-12 | 2021-12-27 | ヤフー株式会社 | Information processing device, information processing method, and information processing program |
| JP7012781B2 (en) | 2020-06-12 | 2022-02-14 | ヤフー株式会社 | Information processing equipment, information processing methods and information processing programs |
| CN113947758A (en) * | 2021-12-16 | 2022-01-18 | 北京凯泰铭科技文化发展有限公司 | Big data method and system of sponge system based on scenic spot chessboard division |
| CN113947758B (en) * | 2021-12-16 | 2022-04-29 | 北京凯泰铭科技文化发展有限公司 | Big data method and system of sponge system based on scenic spot chessboard division |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109034460A (en) | Prediction technique, device and system for scenic spot passenger flow congestion level | |
| US10862788B2 (en) | Telecommunications network traffic metrics evaluation and prediction | |
| Latinopoulos | Using a spatial hedonic analysis to evaluate the effect of sea view on hotel prices | |
| US20150292894A1 (en) | Travel route | |
| US11030635B2 (en) | Method and server for providing a set of price estimates, such as air fare price estimates | |
| Kantakumar et al. | SUSM: a scenario-based urban growth simulation model using remote sensing data | |
| Zheng | Urban computing | |
| Pun‐Cheng | Distance decay | |
| JP6473158B2 (en) | Method and server for providing a set of quoted prices, eg airfare price quotes | |
| Szeto et al. | A time-dependent logit-based taxi customer-search model | |
| US11489749B2 (en) | Mobile telecommunications network capacity simulation, prediction and planning | |
| US11341109B2 (en) | Method and system for detecting and using locations of electronic devices of users in a specific space to analyze social relationships between the users | |
| Chava et al. | Gentrification in new-build and old-build transit-oriented developments: The case of Bengaluru | |
| US11909616B2 (en) | Mobile telecommunications network capacity simulation, prediction and planning | |
| Viverit et al. | Application of machine learning to cluster hotel booking curves for hotel demand forecasting | |
| Jiang et al. | Pedestrian volume prediction with high spatiotemporal granularity in urban areas by the enhanced learning model | |
| Liu et al. | An analysis on the spatiotemporal behavior of inbound tourists in Jiaodong Peninsula based on Flickr geotagged photos | |
| Vanegas Sr | Co-integration and error correction estimation to forecast tourism in El Salvador | |
| KR102359902B1 (en) | Crossroads LOS Prediction Method Based on Big Data and AI, and Storage Medium Having the Same | |
| Molloy | Development of a destination choice model for Ontario | |
| Zhang et al. | Evaluating the socioeconomic value of urban underground space in Kunming, China, Using the entropy method and exponential smoothing prediction | |
| Chhetri et al. | Modelling spatial fragmentation of the Brisbane housing market | |
| Liu et al. | Uncertainty-Aware Probabilistic Travel Time Prediction for On-Demand Ride-Hailing at DiDi | |
| Ning et al. | Predicting citywide passenger demand via reinforcement learning from spatio-temporal dynamics | |
| Williams et al. | Utilizing an urban-regional model (MOLAND) for testing the planning and provision of wastewater treatment capacity in the Dublin Region 2006–2026 |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181218 |