An airport ground traffic passenger flow prediction method

Abstract

This invention discloses a method for predicting passenger flow in airport ground transportation. By establishing a model relationship between the number of people in each area, the method obtains the relationship between the number of people in each area and the overall airport. This allows for the calculation of the total number of people in the airport by only counting the number of people in one area, thus providing a basis for subsequent passenger flow statistics. In addition, by combining the number of people in each area with the flow over time, this invention can obtain the passenger flow situation within the airport. Furthermore, by combining the attributes of each area, it can predict the overall passenger flow of the airport. When predicting the ratio of people in each area, this invention can obtain the proportion of people in each area by comparing the cleanliness of the roads in each area, and create a dynamic image to represent this relationship, providing some intuitive insights into airport passenger flow statistics.

Description

Technical Field

[0001] This invention relates to the field of airport ground management, and in particular to a method for predicting passenger flow in airport ground transportation. Background Technology

[0002] With the continuous improvement of air transport services, airport passenger traffic has increased significantly. To improve passenger service quality, it is necessary to predict passenger flow, which requires statistical analysis of passenger volume at multiple time points. Then, based on the statistical data, a model is built to predict the subsequent total number of passengers, thus completing the passenger flow forecast. This forecasting process requires extensive statistical work; each statistical analysis involves counting the total number of people at the entire airport. Therefore, significant upfront investment is needed to ensure the smooth operation of the forecasting process, necessitating substantial effort in model building. Summary of the Invention

[0003] The purpose of this invention is to overcome the problems existing in the prior art and provide a method for predicting passenger flow in airport ground transportation. This method only requires counting the number of people in a portion of the area to obtain the number of people in the entire area, thereby predicting the passenger flow in the airport.

[0004] Therefore, the present invention provides a method for predicting passenger flow in airport ground transportation, comprising:

[0005] Step 1: Divide the entire airport into various zones, with adjacent zones having different functions;

[0006] Step 2: Obtain the number of people in each region at each time point within the past set time points;

[0007] Step 3: Create a regional map for each time point. The regional map is formed by stitching together the various regions according to their corresponding positions and shapes.

[0008] Step 4: At each time point, the number of people in each area is evenly displayed as pixels in the area map corresponding to their respective areas. Each person corresponds to a set pixel value to obtain a time distribution map of people, and the current time point is stamped as a timestamp.

[0009] Step 5: Divide the personnel distribution map corresponding to each time node into personnel distribution maps of each region according to the boundaries of each region, calculate the gray value of the personnel distribution map of each region, and generate a gray value ratio sequence of each region to the overall airport.

[0010] Step 6: Cluster the time nodes according to the period to obtain periodic time points. At least one periodic time point constitutes a period. Analyze at least one gray value ratio sequence corresponding to each periodic time point after clustering to obtain its corresponding gray value ratio sequence.

[0011] Step 7: Count the number of people in any of the areas, obtain the corresponding gray value ratio sequence based on the periodic time points of the statistical time cluster, and calculate the total number of people in the airport.

[0012] Furthermore, in step 6, the proportion sequence of time nodes corresponding to each of the said periodic time points is T. m ={x 1m ,x 2m ,…,x nm ,X m}, x nm Let X be the number of people in the nth region at the mth time point. m Let m be the total number of passengers at the airport at the m-th time point, where m, n, and x are the total number of passengers. nm and X m All are positive integers;

[0013] After analyzing the grayscale value ratio sequence of each time node corresponding to the periodic time point, the desired result is only the grayscale value ratio sequence corresponding to the periodic time point being T. z ={x 1z ,x 2z ,…,x nz ,X z};

[0014] in,

[0015]

[0016] Where m∈z, z is a positive integer, x nz Let X be the number of people in the nth region at the z-th time point. z Let x be the total number of passengers at the airport at the z-th time point. nz and X z All are positive integers.

[0017] Furthermore, in step 5, after generating the grayscale value ratio sequence of each region and the overall airport, the obtained grayscale value ratio sequence is verified.

[0018] Furthermore, the verification of the grayscale value ratio sequence includes the following steps:

[0019] Step 1: Obtain the functional attributes of each area;

[0020] Step 2: Obtain the personnel data evaluation value corresponding to each region based on the functional attributes of each region. The personnel data evaluation value and the functional attribute are in one-to-one correspondence.

[0021] Step 3: Obtain the verification ratio sequence based on the personnel data evaluation values ​​corresponding to each region. The arrangement order of each region corresponding to the verification ratio sequence is consistent with the arrangement order of each region corresponding to the gray value ratio sequence.

[0022] Step 4: Compare the grayscale value ratio sequence with the verification ratio sequence;

[0023] Step 5: Output the median of the grayscale value ratio sequence and the verification ratio sequence as the grayscale value ratio sequence.

[0024] Furthermore, the functional attributes include area category and area degree, where the area degree refers to the population density.

[0025] Furthermore, the degree of area assessment includes the cleanliness of the ground and the population density of the area; the calculation of the degree of area assessment includes the following steps:

[0026] Step A: Obtain the number of times the cleaning staff cleaned the area, and determine the cleanliness of the floor in the area based on the number of cleanings;

[0027] Step B: Obtain surveillance video of the area and determine the population density of the area based on the surveillance video;

[0028] Step C: The cleanliness of the ground and the population density of the area are used to obtain the area's degree of cleanliness by using a set weighting parameter.

[0029] Furthermore, in step B, determining the population density of the area based on the surveillance video includes the following steps:

[0030] Step B1: Decompose the surveillance video into individual video frames;

[0031] Step B2: Randomly select at least one video frame and process the selected video frame to obtain the grayscale value of the video frame;

[0032] Step B3: Average all the obtained grayscale values ​​to obtain the video grayscale;

[0033] Step B4: Obtain the population density of the region based on the video grayscale, where the population density of the region corresponds one-to-one with the video grayscale.

[0034] Furthermore, in step 6, when clustering the various time nodes according to the period to obtain the periodic time points, the following steps are included:

[0035] Step 6.1: Arrange all the aforementioned time points in chronological order;

[0036] Step 6.2: Determine the number of time points to be captured based on the cycle setting to obtain the cycle time points;

[0037] Step 6.3: Sequentially extract time points from the arranged time nodes according to the number of periodic time points, and the extracted time nodes form the periodic time points.

[0038] Furthermore, it also includes: arranging the personnel time distribution map in chronological order according to the timestamps of the personnel time distribution map, synthesizing the personnel flow video, and displaying the personnel flow video on the display screen.

[0039] The present invention provides a method for predicting passenger flow in airport ground transportation, which has the following beneficial effects:

[0040] This invention establishes a model relationship between the number of people in each region, thereby obtaining the relationship between the number of people in each region and the overall airport. This allows for the calculation of the number of people in the overall airport by only counting the number of people in one region each time, thus laying the foundation for subsequent passenger flow statistics.

[0041] In addition, by combining the number of people in each area with the flow of time, this invention can obtain the flow of people in the airport, and by combining the attributes of each area, it can predict the overall flow of people in the airport.

[0042] When predicting the ratio of people in different areas, this invention can obtain the ratio of people in each area by comparing the cleanliness of the roads in each area, and create a dynamic picture to represent this relationship. This provides some inspiration for the statistics of airport passenger flow from an intuitive perspective. Attached Figure Description

[0043] Figure 1 This is a schematic flowchart illustrating the overall process of the present invention;

[0044] Figure 2 This is a schematic flowchart illustrating the process of verifying the grayscale value ratio sequence according to the present invention.

[0045] Figure 3 This is a flowchart illustrating the calculation of the degree of the region according to the present invention;

[0046] Figure 4 This is a schematic flowchart illustrating the overall process of determining the population density of a given area according to the present invention. Detailed Implementation

[0047] The following detailed description of a specific embodiment of the present invention is provided in conjunction with the accompanying drawings. However, it should be understood that the scope of protection of the present invention is not limited to the specific embodiment.

[0048] Specifically, such as Figure 1-4 As shown, this embodiment of the invention provides a method for predicting passenger flow in airport ground transportation, including:

[0049] Step 1: Divide the entire airport into various zones, with adjacent zones having different functions;

[0050] Step 2: Obtain the number of people in each region at each time point within the past set time points;

[0051] Step 3: Create a regional map for each time point. The regional map is formed by stitching together the various regions according to their corresponding positions and shapes.

[0052] Step 4: At each time point, the number of people in each area is evenly displayed as pixels in the area map corresponding to their respective areas. Each person corresponds to a set pixel value to obtain a time distribution map of people, and the current time point is stamped as a timestamp.

[0053] Step 5: Divide the personnel distribution map corresponding to each time node into personnel distribution maps of each region according to the boundaries of each region, calculate the gray value of the personnel distribution map of each region, and generate a gray value ratio sequence of each region to the overall airport.

[0054] Step 6: Cluster the time nodes according to the period to obtain periodic time points. At least one periodic time point constitutes a period. Analyze at least one gray value ratio sequence corresponding to each periodic time point after clustering to obtain its corresponding gray value ratio sequence.

[0055] Step 7: Count the number of people in any of the areas, obtain the corresponding gray value ratio sequence based on the periodic time points of the statistical time cluster, and calculate the total number of people in the airport.

[0056] In this embodiment of the invention, the entire airport is divided into various areas. The limitation of each area is that its function differs. The entire airport is divided according to the function of each area, such as passenger waiting areas, restroom areas, and airport security checkpoints. By establishing a model of the population distribution in each area and obtaining the population ratio between each area and the entire airport, the population of the entire airport can be obtained by counting the number of people in only one area, thus laying the foundation for subsequent passenger flow statistics.

[0057] Meanwhile, this invention divides time nodes according to a cycle, for example, a day, with time nodes being the hourly intervals within that day. Furthermore, this invention combines image and population statistics, using the grayscale value of the image as the basis for population proportions. While airports are typically monitored by cameras with no blind spots, making image data acquisition easy, population statistics are not as readily available. Therefore, this invention cleverly utilizes airport image data to determine the population relationship between different areas and the overall airport. This allows for the calculation of population in only one area to obtain the overall airport population, thus laying the foundation for subsequent passenger flow statistics and effectively reducing the workload of population counting.

[0058] In this embodiment, in step 6, the proportion sequence of time nodes corresponding to each of the periodic time points is T. m ={x 1m ,x 2m ,…,x nm ,X m}, x nm Let X be the number of people in the nth region at the mth time point. m Let m be the total number of passengers at the airport at the m-th time point, where m, n, and x are the total number of passengers. nm and X m All are positive integers;

[0059] After analyzing the grayscale value ratio sequence of each time node corresponding to the periodic time point, the desired result is only the grayscale value ratio sequence corresponding to the periodic time point being T. z ={x 1z ,x 2z ,…,x nz ,X z};

[0060] in,

[0061]

[0062] Where m∈z, z is a positive integer, x nz Let X be the number of people in the nth region at the z-th time point. z Let x be the total number of passengers at the airport at the z-th time point. nz and X z All are positive integers.

[0063] The above technical solution makes the proportional sequence of time nodes in each period of the present invention universal. In the present invention, that is, T m ={x 1m ,x 2m ,…,x nm ,Xm The array represented by} is x 1m x is the number of people at 2 PM on the first day. 2m x is the number of people at 2 PM on the second day. nm Let m represent the number of people at 2 PM on day n, and the maximum value of m is the number of days, which is the number of cycles.

[0064] The periodic time points of this invention are the same time nodes that exist in every cycle; for example, each day has 2 PM. Therefore, T is obtained through the above formula. z ={x 1z ,x 2z ,…,x nz ,X z This ensures that the grayscale ratio sequence at each periodic time point is obtained based on past data and has universality.

[0065] In this embodiment, after generating the grayscale value ratio sequence of each region and the overall airport in step 5, the obtained grayscale value ratio sequence is verified. The present invention proposes a method for verifying the grayscale value ratio sequence as follows.

[0066] As a preferred technical solution, the verification of the grayscale value ratio sequence includes the following steps:

[0067] Step 1: Obtain the functional attributes of each area;

[0068] Step 2: Obtain the personnel data evaluation value corresponding to each region based on the functional attributes of each region. The personnel data evaluation value and the functional attribute are in one-to-one correspondence.

[0069] Step 3: Obtain the verification ratio sequence based on the personnel data evaluation values ​​corresponding to each region. The arrangement order of each region corresponding to the verification ratio sequence is consistent with the arrangement order of each region corresponding to the gray value ratio sequence.

[0070] Step 4: Compare the grayscale value ratio sequence with the verification ratio sequence;

[0071] Step 5: Output the median of the grayscale value ratio sequence and the verification ratio sequence as the grayscale value ratio sequence.

[0072] When verifying the grayscale value ratio sequence, the number of people in each area is estimated based on the function of each area. For example, the number of people in the staff's work area and the number of people in the waiting hall are inconsistent according to the functional area assessment. The number of people in the waiting hall is definitely higher than the number of people in the staff's work area. It should be noted that before the comparison in step four, the grayscale value ratio sequence and the verification ratio sequence are normalized respectively before comparison. In step five, the result after neutralizing the two is the result after neutralizing the grayscale value ratio sequence and the verification ratio sequence.

[0073] In the above technical solution, the functional attributes include area category and area degree, where the area degree refers to the population density. The area category is the function and name of the area, serving as the basis for dividing the area.

[0074] In embodiments of the present invention, the area degree includes the cleanliness of the ground and the population density of the area; the calculation of the area degree includes the following steps:

[0075] Step A: Obtain the number of times the cleaning staff cleaned the area, and determine the cleanliness of the floor in the area based on the number of cleanings;

[0076] Step B: Obtain surveillance video of the area and determine the population density of the area based on the surveillance video;

[0077] Step C: The cleanliness of the ground and the population density of the area are used to obtain the area's degree of cleanliness by using a set weighting parameter.

[0078] The above technical solution combines the number of cleaning sessions in an area with the density of personnel. The order of steps A and B is not restricted, thus allowing for a more accurate evaluation of the area's condition. Obviously, the more cleaning sessions there are and the higher the personnel density, the greater the degree of personnel concentration. The number of cleaning sessions by cleaning personnel can be statistically analyzed through a worksheet or automatically recorded by capturing the cleaning personnel's cleaning sessions through surveillance video.

[0079] Meanwhile, in step B, determining the population density of the area based on the surveillance video includes the following steps:

[0080] Step B1: Decompose the surveillance video into individual video frames;

[0081] Step B2: Randomly select at least one video frame and process the selected video frame to obtain the grayscale value of the video frame;

[0082] Step B3: Average all the obtained grayscale values ​​to obtain the video grayscale;

[0083] Step B4: Obtain the population density of the region based on the video grayscale, where the population density of the region corresponds one-to-one with the video grayscale.

[0084] In order to ensure that the grayscale of the video reflects all the characteristics of the time period when the grayscale of the entire video is obtained, the present invention decomposes the video and processes the grayscale values ​​of each video frame to obtain a comprehensive description of the grayscale values ​​of the time node.

[0085] In this embodiment, step 6, when clustering the various time nodes according to the period to obtain the periodic time points, includes the following steps:

[0086] Step 6.1: Arrange all the aforementioned time points in chronological order;

[0087] Step 6.2: Determine the number of time points to be captured based on the cycle setting to obtain the cycle time points;

[0088] Step 6.3: Sequentially extract time points from the arranged time nodes according to the number of periodic time points, and the extracted time nodes form the periodic time points.

[0089] The above technical solution enables the present invention to classify different time points in a cycle, such as classifying the grayscale values ​​at 2 PM each day, and proposes the concept of a cycle time point, which facilitates subsequent processing. The cycle time point represents the average of the time points in each cycle, resulting in a larger number of cycle time points in a cycle. When used later, any cycle time point can be found to correspond to the current time, making the model established by the present invention have a finer time scale and store a larger amount of information.

[0090] In this embodiment, the method further includes: arranging the personnel time distribution map in chronological order according to the timestamps, synthesizing a personnel flow video, and displaying the personnel flow video on a display screen. This technical solution combines the advantages of the present invention in predicting the ratio of personnel in each area by comparing the cleanliness of the road surface in each area, and simultaneously creating a dynamic image to represent this relationship. This provides a more intuitive understanding of airport passenger flow statistics and better reflects the flow of people.

[0091] The above-disclosed embodiments are merely a few specific examples of the present invention. However, the embodiments of the present invention are not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for predicting passenger flow in airport ground transportation, characterized in that, include: The airport is divided into various zones, with adjacent zones having different functions. Get the number of people in each region at each time point within the past set time points; A regional map is created for each time point, and the regional map is formed by stitching together the various regions according to their corresponding positions and shapes; At each time point, the number of people in each region is evenly displayed as pixels in the region map, with each person corresponding to a set pixel value, to obtain a time distribution map of the people, and the current time point is stamped as a timestamp. The personnel distribution map corresponding to each time point is divided into personnel distribution maps of each region according to the boundaries of each region. The gray value of the personnel distribution map of each region is calculated, and the gray value ratio sequence of each region and the overall airport is generated. Each time point is clustered according to the period to obtain periodic time points. At least one periodic time point constitutes a period. The gray value ratio sequence corresponding to each periodic time point after clustering is analyzed to obtain its corresponding gray value ratio sequence. The number of people in any one of the areas is counted, and the corresponding gray value ratio sequence is obtained based on the periodic time points of the statistical time cluster. The total number of people in the airport is then calculated. The personnel time distribution map is arranged in chronological order based on the timestamps of the map, and the personnel flow video is synthesized and displayed on the screen. The proportional sequence of time nodes corresponding to each of the aforementioned periodic time points is as follows: , For the first The first time point The number of people in each region For the first The total number of passengers at the airport at each time point, among which , , as well as All are positive integers; After analyzing the grayscale value ratio sequence at each time point corresponding to the periodic time point, the desired result is only the grayscale value ratio sequence corresponding to the periodic time point. ; in, ;in, , It is a positive integer. For the first The first time point The number of people in each region For the first The total number of passengers at the airport at each time point, among which and All are positive integers; After generating the grayscale value ratio sequence of each region and the overall airport, the obtained grayscale value ratio sequence is verified. When clustering the various time points according to the period to obtain the periodic time points, the following steps are included: Arrange all the aforementioned time points in chronological order; The number of time points to be captured is determined by setting the cycle; According to the number of periodic time points, the time points are sequentially extracted from the arranged time nodes, and the extracted time nodes form the periodic time points. The verification of the grayscale value ratio sequence includes the following steps: Obtain the functional attributes of each region; Based on the functional attributes of each region, the personnel data evaluation value corresponding to each region is obtained, and the personnel data evaluation value and the functional attribute correspond one-to-one. A verification ratio sequence is obtained based on the personnel data evaluation value corresponding to each region, and the arrangement order of each region corresponding to the verification ratio sequence is consistent with the arrangement order of each region corresponding to the gray value ratio sequence. Compare the grayscale value ratio sequence with the verification ratio sequence; The median of the grayscale value ratio sequence and the verification ratio sequence is output as the grayscale value ratio sequence.

2. The airport ground transportation passenger flow prediction method as described in claim 1, characterized in that, The functional attributes include area category and area degree, where the area degree refers to the population density.

3. The airport ground traffic passenger flow prediction method as described in claim 2, characterized in that, The area level includes the cleanliness of the ground and the population density of the area; the calculation of the area level includes the following steps: The number of times the cleaning staff cleaned the area is obtained, and the cleanliness of the ground in the area is judged based on the number of times the cleaning staff cleaned the area; Acquire surveillance video of the area and determine the population density of the area based on the surveillance video; The degree of cleanliness of the ground and the population density of the area are used to determine the degree of zoning of the area by using a set weighting parameter.

4. The airport ground transportation passenger flow prediction method as described in claim 3, characterized in that, Determining the population density of a given area based on surveillance video includes the following steps: The surveillance video is broken down into individual video frames; Randomly select at least one video frame and process the selected video frame to obtain the grayscale value of the video frame; The average value of all the obtained grayscale values ​​is then processed to obtain the video grayscale. The population density of the region is obtained based on the grayscale of the video, and the population density of the region corresponds one-to-one with the grayscale of the video.

Images