CN117575684A - Passenger flow volume prediction method and system - Google Patents
Passenger flow volume prediction method and system Download PDFInfo
- Publication number
- CN117575684A CN117575684A CN202410050982.9A CN202410050982A CN117575684A CN 117575684 A CN117575684 A CN 117575684A CN 202410050982 A CN202410050982 A CN 202410050982A CN 117575684 A CN117575684 A CN 117575684A
- Authority
- CN
- China
- Prior art keywords
- time
- passenger flow
- sequence
- fluctuation
- sample point
- 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 38
- 238000009499 grossing Methods 0.000 claims abstract description 70
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 235000019580 granularity Nutrition 0.000 claims description 64
- 238000005070 sampling Methods 0.000 claims description 18
- 230000011218 segmentation Effects 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 3
- 238000013468 resource allocation Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000013486 operation strategy Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011282 treatment Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The invention relates to the technical field of data prediction, in particular to a passenger flow prediction method and a passenger flow prediction system, wherein the method comprises the following steps: acquiring historical passenger flow data of each entrance and exit position; dividing the time sequence sample point sequence according to each preset time granularity to obtain each sample point subsequence; for each sample point of the time sequence sample point sequence, obtaining the fluctuation significance degree of the sample point according to the relation between the data fluctuation conditions of the sub-sequence of the sample point where the sample point is located under the adjacent time granularity; clustering sample points in the fluctuation significance level sequence to obtain clustering clusters; taking the time span of the continuous sample points with the number larger than 1 in each cluster as each time period of the fluctuation saliency sequence; and according to the difference condition of the fluctuation significance degree of the sample points in the time periods, the trend smoothing coefficients of each time period are adjusted to obtain integrally adjusted trend smoothing coefficients, and the Hall reference number smoothing algorithm is adopted to finish passenger flow prediction. The invention improves the accuracy of passenger flow data prediction.
Description
Technical Field
The application relates to the technical field of data prediction, in particular to a passenger flow prediction method and system.
Background
The market passenger flow prediction can provide important references for market operation, help optimize operation strategies and resource allocation, and can reasonably arrange personnel, goods and facilities by analyzing the variation trend and the law of the passenger flow, so that the operation efficiency and the service quality are improved, and the market passenger flow prediction accuracy is an extremely important data analysis purpose in coping with peak periods, sales promotion activities and the like.
Because the acquired market passenger flow is a time sequence data sequence, and the passenger flow data is influenced by various factors such as customer flow, weather and the like, the passenger flow data can cause larger data fluctuation on the historical passenger flow data for passenger flow prediction, and the accurate change rule of the passenger flow is not beneficial to be effectively analyzed. In the process of time sequence prediction by utilizing the Holter index smoothing algorithm, the fluctuation degree of the passenger flow data under different time granularity often shows different remarkable characteristics, and a single trend smoothing coefficient often corresponds to a larger time granularity, so that accurate analysis on the passenger flow change rule is difficult to realize.
Disclosure of Invention
In order to solve the technical problems, the invention provides a passenger flow volume prediction method and a passenger flow volume prediction system, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for predicting a passenger flow volume, including the following steps:
acquiring historical passenger flow data of each entrance and exit position; recording historical passenger flow data as a time sequence sample point sequence;
dividing the time sequence sample point sequence according to each preset time granularity to obtain a sample point sub-sequence in which each sample point is positioned under each time granularity, wherein the sample points are passenger flow data at sampling time; for each sample point of the time sequence sample point sequence, obtaining the fluctuation significance degree of the sample point according to the relation between the data fluctuation conditions of the sub-sequence of the sample point where the sample point is located under the adjacent time granularity; the fluctuation significance of each sample point is formed into a fluctuation significance sequence according to the time sequence; clustering sample points in the fluctuation significance level sequence to obtain clustering clusters;
taking the time span of the continuous sample points with the number larger than 1 in each cluster as each time period of the fluctuation saliency sequence; obtaining the credibility of the trend influence of the time period according to the difference condition of the fluctuation significance of the sample points in the time period; the trend smoothing coefficients of all time periods are adjusted according to the trend influence credibility degree of all time periods to obtain overall adjusted trend smoothing coefficients;
and (5) according to the trend smoothing coefficient after the integral adjustment, adopting a Hall index smoothing algorithm to finish passenger flow prediction.
Preferably, the acquiring historical passenger flow data of each gateway location includes:
inputting video frame images of the monitoring video of each entrance and exit position into a semantic segmentation neural network, outputting a personnel segmentation graph in the video frame images by the semantic segmentation neural network, and counting personnel in the segmentation graph to obtain historical passenger flow data;
the historical passenger flow data are passenger flow data of all sampling moments collected under a preset sampling period, and statistics is carried out only once in the time of a preset sampling interval for the same personnel area.
Preferably, the dividing the sequence of time-sequence sample points according to each preset time granularity to obtain a sub-sequence of sample points where each sample point is located under each time granularity includes:
uniformly dividing the time sequence sample point sequence according to different time granularities to obtain each sample point subsequence under each time granularity;
for each sample point of the time sequence of sample points, a sub-sequence of sample points in which the sub-sequence of sample points is located at each time granularity can be obtained.
Preferably, the obtaining the fluctuation significance of the sample point according to the relation between the data fluctuation conditions of the sub-sequence of the sample point where the sample point is located at the adjacent time granularity includes:
for each time granularity, obtaining a fluctuation change coefficient according to the data fluctuation condition of the sample points in the sample point subsequence where the sample points are located under the time granularity;
and calculating the absolute value of the difference value of the fluctuation change coefficient under any two adjacent time granularities, and taking the normalized value of the sum value of the absolute value of the difference values of the sample points under all time granularities as the fluctuation significance of the sample points.
Preferably, the obtaining the fluctuation change coefficient according to the data fluctuation condition of the sample point sub-sequence where the sample point is located under the time granularity includes:
marking each sample point in a sub-sequence of the sample points where the sample points are located at the time granularity as a first sample point, calculating the sum of absolute values of differences of all first sample points except the sample points in the sub-sequence of the sample points where the sample points are located at the time granularity and passenger flow data values of the sample points, and taking the sum as a fluctuation change coefficient.
Preferably, the clustering the sample points in the fluctuation significance level sequence to obtain clusters includes:
the clustering distance is the fluctuation significance degree of each sample point, a Gaussian mixture model is adopted in the clustering process, and the sample points in each cluster obtained by the clustering result are ordered according to the time sequence.
Preferably, the obtaining the credibility of the trend influence of the time period according to the difference condition of the remarkable fluctuation degree of the sample points in the time period includes:
acquiring the mean value of the fluctuation significance degree of all sample points in a time period as a first mean value;
acquiring a fluctuation significant degree mean value of all time periods in the fluctuation significant degree sequence as a second mean value;
calculating the absolute value of the difference between the first mean value and the second mean value, and calculating the ratio of the absolute value of the difference to the second mean value;
the inverse number of the ratio is taken as an index of an exponential function based on a natural constant, and the calculation result of the exponential function is taken as the trend of the time period to influence the credibility.
Preferably, the adjusting the trend smoothing coefficient of each time period according to the reliability degree of the trend influence of each time period to obtain the overall adjusted trend smoothing coefficient includes:
calculating the mean value of the credibility of the trend influence in all time periods;
for each time period, calculating the ratio of the credibility of the trend influence of the time period to the average value of the credibility of the trend influence, and calculating the product of the ratio and the trend smoothing coefficient of the time period;
taking the average value of the products of all the time periods as the trend smoothing coefficient after overall adjustment.
Preferably, the predicting the passenger flow volume according to the trend smoothing coefficient after the overall adjustment by adopting a holter index smoothing algorithm includes:
according to the trend smoothing coefficient after overall adjustment, the trend smoothing coefficient is used as a trend smoothing coefficient in a Hall index smoothing algorithm, and a passenger flow prediction model is constructed;
the passenger flow volume data of the previous time period of the passenger flow volume to be predicted is used as the input of the passenger flow volume prediction model, and is output as the passenger flow volume prediction data of the next time period.
In a second aspect, an embodiment of the present invention further provides a passenger flow volume prediction system, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of any one of the methods described above when executing the computer program.
The invention has at least the following beneficial effects:
according to the invention, the fluctuation significance degree of each sample point is obtained by analyzing the difference between the historical passenger flow data on different time granularities, the sensitivity degree of the sample point on the values on the different time granularities is reflected, and the classification of the sample point types is realized to a certain extent;
meanwhile, as the influence degree of the conventional passenger flow and the variable passenger flow on each sample point is different, the influence degree on the acquisition of the trend smoothing parameters is different, the sample points are clustered according to the fluctuation significance, each time period obtained by dividing the sample points of the same type is realized, the influence of historical passenger flow data on the trend smoothing coefficient in the time period is quantized according to the fluctuation significance difference in the time period, the adjustment of the trend smoothing coefficient of the Hall specific number smoothing algorithm is realized, the prediction accuracy of the Hall specific number smoothing algorithm on the real-time passenger flow data is improved, and the subsequent management strategy optimization, resource allocation and other treatments are completed.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a passenger flow volume prediction method provided by the invention;
fig. 2 is a prediction flow chart of the passenger flow volume prediction data.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to specific embodiments, structures, features and effects of a passenger flow volume prediction method and system according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a passenger flow prediction method and a system provided by the invention with reference to the accompanying drawings.
The invention provides a passenger flow volume prediction method and a passenger flow volume prediction system.
Specifically, referring to fig. 1, the following passenger flow volume prediction method is provided, and the method includes the following steps:
step S001, obtaining a plurality of historical passenger flow data of the gateway positions of the mall.
According to the embodiment, the machine vision device is arranged at the position of the store entrance and exit, namely, the monitoring video of the store entrance and exit is obtained through the camera, semantic segmentation and identification are carried out on the video frame image, and the real-time passenger flow in the store is recorded. The sampling period of the passenger flow is 30D, the sampling interval is 5 min/time, the sampling period of the passenger flow is 30 days, the semantic segmentation neural network adopts an Encoder-Decoder structure, and the specific training content is as follows:
1) The data set is a video frame image of the acquired monitoring video of the store entrance.
2) Labels are of two types, personnel area and background area. The method is pixel-level classification, namely, corresponding labels need to be marked on all pixels in the image. Pixels belonging to the personnel area are labeled with a value of 1, pixels belonging to the background are labeled with a value of 0.
3) The loss function used by the network is a cross entropy loss function.
The semantic segmentation neural network is a known technology, and this embodiment is not described in detail. The method comprises the steps of identifying and dividing personnel targets entering and exiting a market in a video frame image, counting the same personnel area in a similar video frame image only once in a sampling interval, obtaining real-time passenger flow corresponding to a plurality of sampling time periods, and obtaining passenger flow data of the market in the past for a long time as historical passenger flow data by adopting the same processing mode, wherein the historical passenger flow data are time sequence data sequences closely related to the sampling time.
And step S002, according to the obtained historical passenger flow data, combining the data fluctuation characteristics under different time granularities, realizing the optimization adjustment of the trend smoothing coefficient, and obtaining the accurate time sequence prediction result.
Because the acquired market passenger flow is a time sequence data sequence, the randomness of data fluctuation exists, and the accurate change rule of the passenger flow is not beneficial to be effectively analyzed. In the process of time sequence prediction by utilizing the Holter index smoothing algorithm, the fluctuation degree of the passenger flow data under different time granularity often shows different remarkable characteristics, and a single trend smoothing coefficient often corresponds to a larger time granularity, so that accurate analysis of the passenger flow change rule and passenger flow prediction at the target moment are difficult to realize.
Therefore, the embodiment achieves time division of different historical passenger flow data by analyzing the fluctuation significance degree of the historical passenger flow data on different time granularities, quantifies the influence of the historical passenger flow data on trend smoothing coefficients in the time period according to the fluctuation significance degree difference of the time period, achieves adjustment of the trend smoothing coefficients of the Hall reference number smoothing algorithm, improves the prediction accuracy of the real-time passenger flow data, and completes subsequent management strategy optimization, resource allocation and other processing.
Therefore, the process of processing the acquired market passenger flow to obtain the accurate time sequence prediction result in the embodiment is as follows.
The historical passenger flow data in one sampling period is considered to have strong correlation, the influence of the environment, the climate change and the like where the historical passenger flow data is located can be considered to be consistent, and the fluctuation characteristic of the historical passenger flow data is mainly influenced by the random flow of customer staff and market activities.
Thus, the passenger flow data at a certain sampling moment is represented as a sample point in a sample space, the significance of which is different at different time granularities, such as 1h,1d,7d, etc., and the significance of the passenger flow data of the sample point in different time periods is mainly determined by the reason that the passenger flow data changes, and can be divided into regular passenger flow caused by the daily behavior habit of local customers and variable passenger flow of random flow of customer personnel and market activity. The conventional passenger flow volume determines the main part of the passenger flow volume prediction result, and the variable passenger flow volume determines the accuracy degree of the passenger flow volume prediction. The conventional passenger flow volume always presents a relatively stable change characteristic on a larger time granularity, and the variable passenger flow volume is concentrated and obviously presented on a certain time granularity. Thus, the types of sample points can be primarily divided for obtaining the fluctuation significance degree on different time granularity.
According to the acquired historical passenger flow data, the passenger flow data at each sampling moment is expressed as one sample point in a sample space, the historical passenger flow data can be expressed as a time sequence sample point sequence, and different time granularity segmentation is performed on the number of time sequence sample points, namely, the set different time granularity 1h,1D,7D and 15D are used as segmentation conditions, and can be defined by an implementer according to specific scenes, and the same processing mode is adopted for each time granularity:
the acquired time sequence sample point sequence is evenly divided by utilizing the time granularity, a plurality of sample point subsequences are acquired at different time granularities, each sample point corresponds to one sample point subsequence at different time granularities, the fluctuation degree of each sample point subsequence in the subsequence is different in performance, and the difference of the fluctuation degree at the plurality of time granularities is finally expressed as the fluctuation significance of the sample point, so that the fluctuation significance of the sample point is obtainedThe calculation formula of (2) is as follows:
；
wherein,indicate->Degree of fluctuation significance of individual sample points, +.>Representing the number of temporal granularity, +.>Indicate->The sample points are at->Sample point number of sub-sequence of sample points at each time granularity, +.>Representing the +.>Passenger flow at individual sample pointsVolume data value->Representing the +.>The sample points are at->Sample Point sub-sequence at time granularity except +.>Sample out of the first->Passenger flow data value of individual sample points, +.>Representing a maximum minimum normalization operation, wherein +.>Is indicated at +.>Fluctuation coefficient of variation of sub-sequence of sample points at each time granularity, +.>Indicate->Granularity of time and->Absolute value of difference between fluctuation change coefficients at each time granularity.
It should be noted that, the measurement of the degree of fluctuation significance of a sample point is represented by accumulation of the fluctuation degree in a sub-sequence of the sample point at different time granularities where the sample point is located and the fluctuation degree difference at a plurality of time granularities as a whole, and the larger the accumulation value is, the more significant the fluctuation of the sample point is.
So far, the fluctuation significance degree of each sampling point is obtained through the data distribution of the collected historical passenger flow data on different time granularities.
According to the fluctuation significance degree of the obtained sample points, the sensitivity degree of the sample points to different time granularities is reflected, the sample point types are divided to a certain extent, namely, the influence degree of the sample points on the conventional passenger flow is different from the influence degree of the variable passenger flow, the influence degree of the different sample points on the trend smoothing parameter acquisition is different in the trend smoothing coefficient acquisition process of the Hall index smoothing algorithm, the influence degree can be represented by the distribution characteristics of the fluctuation significance degree of the sample points in a certain time neighborhood, the division of time periods can be realized through the clustering result of the clustering algorithm, and the trend influence credibility degree of different time periods is constructed according to the fluctuation significance degree overall characteristics in the time periods and is used for adjusting the follow-up trend smoothing coefficients.
According to the fluctuation significance degree of the obtained sample points, which corresponds to the time sequence sample point sequence, a one-dimensional fluctuation significance degree sequence is formed, the sample points are clustered by using a Gaussian mixture model, the similarity of the sample points is measured by using the fluctuation significance degree difference of the sample points, and therefore a plurality of clusters are obtained, wherein each cluster at least comprises one sample point, and if the number of sample points is a plurality of sample points, the sample points are arranged in time sequence.
The time span of the continuous sample points contained in each cluster is taken as each divided time period, wherein the isolated sample points are discarded in each cluster. When the hall reference number smoothing algorithm predicts the time series data, the deadline data is generally selected for determining the trend smoothing coefficient, and the deadline data is divided time periods in the embodiment.
Thereby obtaining the degree of confidence that the trend of each time period affectsThe calculation formula of (2) is as follows:
；
wherein,indicate->Trend of individual time periods affects the degree of confidence, +.>Representing an exponential function based on a natural constant e, < ->Indicate->Mean value of fluctuation significance of all sample points in each time period, +.>Representing the number of time periods contained in the deadline data selected by the Hall reference number smoothing algorithm, wherein +_>For the first mean>Is the second mean.
It should be noted that the number of the substrates,indicate->The greater the value of the degree of difference between the time period and the time period of all the selected time periods, the more the change trend of the sample points in the time period probably plays a main role by the influence of the variability error of the variable passenger flow, and when the degree of reliability of the time period is smaller when the time period is used for determining the trend smoothing coefficient, the trend of the time period influences the degree of reliability>The smaller.
By the method, different time periods are divided through the obtained fluctuation significance degree on different time granularity, and the credibility degree is affected by the trend of the different time periods.
According to the obtained trend influence credibility of different time periods, the reliability of the passenger flow data of the sample points in the time period is reflected when the trend smoothing coefficient is obtained, so that the trend smoothing coefficient can be used as the weight coefficient of the passenger flow data of the sample points in the time period, the adjustment of the trend smoothing coefficient obtaining process is realized, the construction of a follow-up Hall reference number smoothing algorithm on a prediction model is completed, and a precise prediction result is obtained.
According to the obtained trend influence credibility of different time periods, when the trend smoothing coefficient is calculated by utilizing the data of a plurality of time periods, the trend influence credibility of different time periods is used as a weighted average weight coefficient of the obtained trend smoothing coefficient, namely the obtained trend smoothing coefficient is a weighted average of a plurality of calculation results, and the weighted average weight coefficient can be expressed as:
；
wherein,representing the trend smoothing coefficient after overall adjustment, +.>The number of time periods contained in the deadline data representing the choice of the Hall reference number smoothing algorithm,/for each time period>Indicate->Trend of individual time periods affects the degree of confidence, +.>Indicate->Trend smoothing coefficients for each time period.
And the construction of the Hall index smoothing algorithm on the passenger flow prediction model is completed according to the trend smoothing coefficient after the integral adjustment.
And taking the passenger flow data of the previous time period of the passenger flow needing to be predicted as the input of a passenger flow prediction model, and outputting the predicted data of the next time period, wherein the predicted data is the accurate time sequence predicted result needing to be obtained.
So far, the reliability is affected by the acquired trends in different time periods, the adjustment of the trend smoothing coefficient is realized, and the accurate prediction result is obtained. The flow chart of the prediction of the passenger flow volume prediction data is shown in fig. 2.
Step S003, the subsequent operation strategy adjustment and resource allocation are completed through the obtained accurate time sequence prediction result.
And presenting the accurate time sequence prediction result to a user by using a visualization scheme, helping the user analyze the passenger flow change condition of the market in a prediction time period, and completing subsequent operation strategy adjustment and resource allocation according to an operation strategy formulated by the user.
Based on the same inventive concept as the above method, the embodiment of the invention further provides a passenger flow volume prediction system, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any one of the above passenger flow volume prediction methods.
According to the embodiment of the invention, the fluctuation significance degree of each sample point is obtained by analyzing the difference between the historical passenger flow data on different time granularities, the sensitivity degree of the sample point on the values of different time granularities is reflected, and the classification of the sample point types is realized to a certain extent;
meanwhile, as the influence degree of the conventional passenger flow and the variable passenger flow on each sample point is different, the influence degree on the acquisition of the trend smoothing parameters is different, the sample points are clustered according to the fluctuation significance, each time period obtained by dividing the sample points of the same type is realized, the influence of historical passenger flow data on the trend smoothing coefficient in the time period is quantized according to the fluctuation significance difference in the time period, the adjustment of the trend smoothing coefficient of the Hall specific number smoothing algorithm is realized, the prediction accuracy of the Hall specific number smoothing algorithm on the real-time passenger flow data is improved, and the subsequent management strategy optimization, resource allocation and other treatments are completed.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.
Claims (10)
1. A method for predicting passenger flow, the method comprising the steps of:
acquiring historical passenger flow data of each entrance and exit position; recording historical passenger flow data as a time sequence sample point sequence;
dividing the time sequence sample point sequence according to each preset time granularity to obtain a sample point sub-sequence in which each sample point is positioned under each time granularity, wherein the sample points are passenger flow data at sampling time; for each sample point of the time sequence sample point sequence, obtaining the fluctuation significance degree of the sample point according to the relation between the data fluctuation conditions of the sub-sequence of the sample point where the sample point is located under the adjacent time granularity; the fluctuation significance of each sample point is formed into a fluctuation significance sequence according to the time sequence; clustering sample points in the fluctuation significance level sequence to obtain clustering clusters;
taking the time span of the continuous sample points with the number larger than 1 in each cluster as each time period of the fluctuation saliency sequence; obtaining the credibility of the trend influence of the time period according to the difference condition of the fluctuation significance of the sample points in the time period; the trend smoothing coefficients of all time periods are adjusted according to the trend influence credibility degree of all time periods to obtain overall adjusted trend smoothing coefficients;
and (5) according to the trend smoothing coefficient after the integral adjustment, adopting a Hall index smoothing algorithm to finish passenger flow prediction.
2. The method for predicting passenger flow volume as set forth in claim 1, wherein said obtaining historical passenger flow volume data for each doorway location comprises:
inputting video frame images of the monitoring video of each entrance and exit position into a semantic segmentation neural network, outputting a personnel segmentation graph in the video frame images by the semantic segmentation neural network, and counting personnel in the segmentation graph to obtain historical passenger flow data;
the historical passenger flow data are passenger flow data of all sampling moments collected under a preset sampling period, and statistics is carried out only once in the time of a preset sampling interval for the same personnel area.
3. The passenger flow volume prediction method as set forth in claim 1, wherein the dividing the time sequence sample point sequence according to each preset time granularity to obtain a sample point sub-sequence in which each sample point is located at each time granularity comprises:
uniformly dividing the time sequence sample point sequence according to different time granularities to obtain each sample point subsequence under each time granularity;
for each sample point of the time sequence of sample points, a sub-sequence of sample points in which the sub-sequence of sample points is located at each time granularity can be obtained.
4. A method for predicting passenger traffic as recited in claim 3, wherein said deriving the degree of fluctuation significance of the sample points from the relationship between the data fluctuation conditions of the sub-sequence of sample points where the sample points are located at adjacent time granularity comprises:
for each time granularity, obtaining a fluctuation change coefficient according to the data fluctuation condition of the sample points in the sample point subsequence where the sample points are located under the time granularity;
and calculating the absolute value of the difference value of the fluctuation change coefficient under any two adjacent time granularities, and taking the normalized value of the sum value of the absolute value of the difference values of the sample points under all time granularities as the fluctuation significance of the sample points.
5. A method for predicting passenger traffic as recited in claim 4, wherein the deriving the fluctuation variance factor from the fluctuation of data in the sub-sequence of sample points at the time granularity comprises:
marking each sample point in a sub-sequence of the sample points where the sample points are located at the time granularity as a first sample point, calculating the sum of absolute values of differences of all first sample points except the sample points in the sub-sequence of the sample points where the sample points are located at the time granularity and passenger flow data values of the sample points, and taking the sum as a fluctuation change coefficient.
6. The passenger flow volume prediction method according to claim 1, wherein the clustering of the sample points in the fluctuation saliency sequence to obtain clusters comprises:
the clustering distance is the fluctuation significance degree of each sample point, a Gaussian mixture model is adopted in the clustering process, and the sample points in each cluster obtained by the clustering result are ordered according to the time sequence.
7. A passenger flow volume prediction method as set forth in claim 1, wherein the obtaining the confidence level of the trend influence of the time period according to the difference of the significance level of the sample point fluctuation in the time period comprises:
acquiring the mean value of the fluctuation significance degree of all sample points in a time period as a first mean value;
acquiring a fluctuation significant degree mean value of all time periods in the fluctuation significant degree sequence as a second mean value;
calculating the absolute value of the difference between the first mean value and the second mean value, and calculating the ratio of the absolute value of the difference to the second mean value;
the inverse number of the ratio is taken as an index of an exponential function based on a natural constant, and the calculation result of the exponential function is taken as the trend of the time period to influence the credibility.
8. The passenger flow volume prediction method as set forth in claim 1, wherein the adjusting the trend smoothing coefficient of each time period according to the reliability degree of the trend influence of each time period to obtain the overall adjusted trend smoothing coefficient comprises:
calculating the mean value of the credibility of the trend influence in all time periods;
for each time period, calculating the ratio of the credibility of the trend influence of the time period to the average value of the credibility of the trend influence, and calculating the product of the ratio and the trend smoothing coefficient of the time period;
taking the average value of the products of all the time periods as the trend smoothing coefficient after overall adjustment.
9. The passenger flow volume prediction method as set forth in claim 8, wherein the passenger flow volume prediction is accomplished by adopting a holter index smoothing algorithm according to the overall-adjusted trend smoothing coefficient, comprising:
according to the trend smoothing coefficient after overall adjustment, the trend smoothing coefficient is used as a trend smoothing coefficient in a Hall index smoothing algorithm, and a passenger flow prediction model is constructed;
the passenger flow volume data of the previous time period of the passenger flow volume to be predicted is used as the input of the passenger flow volume prediction model, and is output as the passenger flow volume prediction data of the next time period.
10. A passenger flow volume prediction system comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-9 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410050982.9A CN117575684B (en) | 2024-01-15 | 2024-01-15 | Passenger flow volume prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410050982.9A CN117575684B (en) | 2024-01-15 | 2024-01-15 | Passenger flow volume prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117575684A true CN117575684A (en) | 2024-02-20 |
| CN117575684B CN117575684B (en) | 2024-04-05 |
Family
ID=89890383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410050982.9A Active CN117575684B (en) | 2024-01-15 | 2024-01-15 | Passenger flow volume prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117575684B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117792615A (en) * | 2024-02-28 | 2024-03-29 | 青岛克莱玛物联技术有限公司 | Data intelligent processing method based on intensive communication module |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110161261A1 (en) * | 2009-12-28 | 2011-06-30 | Nec(China) Co., Ltd. | Method and system for traffic prediction based on space-time relation |
| CN103984993A (en) * | 2014-05-13 | 2014-08-13 | 东南大学 | Rail transit passenger flow OD distribution real-time speculation method |
| CN104809787A (en) * | 2015-04-23 | 2015-07-29 | 中电科安（北京）系统集成有限公司 | Intelligent passenger flow volume statistics device based on camera |
| CN106951976A (en) * | 2016-10-12 | 2017-07-14 | 华南理工大学 | A kind of bus passenger flow Forecasting Methodology based on pattern classification |
| CN108492568A (en) * | 2018-04-25 | 2018-09-04 | 南京邮电大学 | A kind of Short-time Traffic Flow Forecasting Methods based on space-time characterisation analysis |
| CN108804731A (en) * | 2017-09-12 | 2018-11-13 | 中南大学 | Based on the dual evaluation points time series trend feature extracting method of vital point |
| CN110276474A (en) * | 2019-05-22 | 2019-09-24 | 南京理工大学 | A kind of track traffic station passenger flow forecasting in short-term |
| CN111126681A (en) * | 2019-12-12 | 2020-05-08 | 华侨大学 | Bus route adjusting method based on historical passenger flow |
| CN112465566A (en) * | 2020-12-14 | 2021-03-09 | 树蛙信息科技(南京)有限公司 | Shopping mall passenger flow volume prediction method based on historical data auxiliary scene analysis |
| CN113361810A (en) * | 2021-06-30 | 2021-09-07 | 佳都科技集团股份有限公司 | Passenger flow volume prediction method, device, equipment and storage medium |
| CN114626887A (en) * | 2022-03-18 | 2022-06-14 | 建信金融科技有限责任公司 | Passenger flow volume prediction method and device, computer equipment and storage medium |
| WO2022142413A1 (en) * | 2020-12-31 | 2022-07-07 | 深圳云天励飞技术股份有限公司 | Method and apparatus for predicting customer flow volume of mall, and electronic device and storage medium |
| CN115511179A (en) * | 2022-09-23 | 2022-12-23 | 五邑大学 | Passenger flow prediction method, device and medium |
| CN115860790A (en) * | 2022-11-11 | 2023-03-28 | 深圳海智创科技有限公司 | Method for predicting place passenger flow, method and device for training model |
| WO2023056696A1 (en) * | 2021-10-08 | 2023-04-13 | 南威软件股份有限公司 | Urban rail transit short-term passenger flow forecasting method based on recurrent neural network |
| CN116244609A (en) * | 2022-12-29 | 2023-06-09 | 深圳云天励飞技术股份有限公司 | Passenger flow volume statistics method and device, computer equipment and storage medium |
| CN117010537A (en) * | 2022-04-26 | 2023-11-07 | 腾讯科技（深圳）有限公司 | Target area prediction method, device, computer equipment and storage medium |
| CN117077846A (en) * | 2023-08-01 | 2023-11-17 | 中国移动通信集团内蒙古有限公司 | Passenger flow volume prediction method, equipment, storage medium and device |
| CN117132083A (en) * | 2023-10-20 | 2023-11-28 | 惠州市金雄城建筑科技有限公司 | Urban rail transit hub operation and maintenance management system based on BIM |
-
2024
- 2024-01-15 CN CN202410050982.9A patent/CN117575684B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110161261A1 (en) * | 2009-12-28 | 2011-06-30 | Nec(China) Co., Ltd. | Method and system for traffic prediction based on space-time relation |
| CN103984993A (en) * | 2014-05-13 | 2014-08-13 | 东南大学 | Rail transit passenger flow OD distribution real-time speculation method |
| CN104809787A (en) * | 2015-04-23 | 2015-07-29 | 中电科安（北京）系统集成有限公司 | Intelligent passenger flow volume statistics device based on camera |
| CN106951976A (en) * | 2016-10-12 | 2017-07-14 | 华南理工大学 | A kind of bus passenger flow Forecasting Methodology based on pattern classification |
| CN108804731A (en) * | 2017-09-12 | 2018-11-13 | 中南大学 | Based on the dual evaluation points time series trend feature extracting method of vital point |
| CN108492568A (en) * | 2018-04-25 | 2018-09-04 | 南京邮电大学 | A kind of Short-time Traffic Flow Forecasting Methods based on space-time characterisation analysis |
| CN110276474A (en) * | 2019-05-22 | 2019-09-24 | 南京理工大学 | A kind of track traffic station passenger flow forecasting in short-term |
| CN111126681A (en) * | 2019-12-12 | 2020-05-08 | 华侨大学 | Bus route adjusting method based on historical passenger flow |
| CN112465566A (en) * | 2020-12-14 | 2021-03-09 | 树蛙信息科技(南京)有限公司 | Shopping mall passenger flow volume prediction method based on historical data auxiliary scene analysis |
| WO2022142413A1 (en) * | 2020-12-31 | 2022-07-07 | 深圳云天励飞技术股份有限公司 | Method and apparatus for predicting customer flow volume of mall, and electronic device and storage medium |
| CN113361810A (en) * | 2021-06-30 | 2021-09-07 | 佳都科技集团股份有限公司 | Passenger flow volume prediction method, device, equipment and storage medium |
| WO2023056696A1 (en) * | 2021-10-08 | 2023-04-13 | 南威软件股份有限公司 | Urban rail transit short-term passenger flow forecasting method based on recurrent neural network |
| CN114626887A (en) * | 2022-03-18 | 2022-06-14 | 建信金融科技有限责任公司 | Passenger flow volume prediction method and device, computer equipment and storage medium |
| CN117010537A (en) * | 2022-04-26 | 2023-11-07 | 腾讯科技（深圳）有限公司 | Target area prediction method, device, computer equipment and storage medium |
| CN115511179A (en) * | 2022-09-23 | 2022-12-23 | 五邑大学 | Passenger flow prediction method, device and medium |
| CN115860790A (en) * | 2022-11-11 | 2023-03-28 | 深圳海智创科技有限公司 | Method for predicting place passenger flow, method and device for training model |
| CN116244609A (en) * | 2022-12-29 | 2023-06-09 | 深圳云天励飞技术股份有限公司 | Passenger flow volume statistics method and device, computer equipment and storage medium |
| CN117077846A (en) * | 2023-08-01 | 2023-11-17 | 中国移动通信集团内蒙古有限公司 | Passenger flow volume prediction method, equipment, storage medium and device |
| CN117132083A (en) * | 2023-10-20 | 2023-11-28 | 惠州市金雄城建筑科技有限公司 | Urban rail transit hub operation and maintenance management system based on BIM |
Non-Patent Citations (3)
| Title |
|---|
| 徐肖豪 等: "空中交通流量统计和预测系统的设计与实现", 中国民航学院学报, vol. 23, no. 04, 31 August 2005 (2005-08-31), pages 1 - 5 * |
| 王炜炜 等: "基于时间序列聚类方法的小长假铁路客流规律研究", 铁路计算机应用, vol. 24, no. 04, 31 December 2015 (2015-12-31), pages 23 - 27 * |
| 马超群 等: "基于不同时间粒度的城市轨道交通短时客流预测", 长安大学学报(自然科学版), vol. 40, no. 03, 31 May 2020 (2020-05-31), pages 75 - 83 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117792615A (en) * | 2024-02-28 | 2024-03-29 | 青岛克莱玛物联技术有限公司 | Data intelligent processing method based on intensive communication module |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117575684B (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117575684B (en) | Passenger flow volume prediction method and system | |
| Tsantekidis et al. | Forecasting stock prices from the limit order book using convolutional neural networks | |
| CN112381011B (en) | Non-contact heart rate measurement method, system and device based on face image | |
| CN109711890B (en) | User data processing method and system | |
| Jiang et al. | A new hybrid framework for probabilistic wind speed prediction using deep feature selection and multi-error modification | |
| CN110399835B (en) | Analysis method, device and system for personnel residence time | |
| CN113158909B (en) | Behavior recognition light-weight method, system and equipment based on multi-target tracking | |
| CN111339813B (en) | Face attribute recognition method and device, electronic equipment and storage medium | |
| CN111275479B (en) | People flow prediction method, device and system | |
| CN112446399A (en) | Label determination method, device and system | |
| CN117456428B (en) | Garbage throwing behavior detection method based on video image feature analysis | |
| CN112560827A (en) | Model training method, model training device, model prediction method, electronic device, and medium | |
| CN116418120A (en) | Intelligent early warning method applied to water-cooled power supply | |
| Liseune et al. | Leveraging latent representations for milk yield prediction and interpolation using deep learning | |
| CN114359787A (en) | Target attribute identification method and device, computer equipment and storage medium | |
| CN116843085B (en) | Freshwater fish growth monitoring method, device, equipment and storage medium | |
| Yu et al. | A diagnosis model of soybean leaf diseases based on improved residual neural network | |
| CN108802845B (en) | A kind of indoor occupant occupation rate estimation method based on infrared sensor array | |
| CN115115038B (en) | Model construction method based on single lead electrocardiosignal and gender identification method | |
| CN116861076A (en) | Sequence recommendation method and device based on user popularity preference | |
| CN113128452A (en) | Greening satisfaction acquisition method and system based on image recognition | |
| CN114492657A (en) | Plant disease classification method and device, electronic equipment and storage medium | |
| CN114491410A (en) | Motion mode identification method and system, intelligent wearable device and storage medium | |
| Emine et al. | Emergency Department Overcrowding Detection by a Multifractal Analysis | |
| CN117809124B (en) | Medical image association calling method and system based on multi-feature fusion |
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 |