Railway passenger identification method, device, electronic device and computer-readable storage medium

Abstract

The present invention provides a railway passenger identification method, device, electronic device, and computer-readable storage medium, relating to the field of big data technology. This method utilizes the signaling set within a target area within a specified timeframe and the signaling interaction range of a railway station to initially identify potential users who have stopped at the railway station and generated signaling interactions. Based on the potential users' stay information and the travel behavior characteristics of railway passengers, non-railway passengers who have only stopped at a single railway station, such as station attendants or railway staff, are then excluded. Finally, various misjudgment scenarios are comprehensively considered to exclude the itineraries of non-railway travelers, thereby accurately identifying each railway passenger's journey within the specified timeframe.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a railway passenger identification method, device, electronic device and computer-readable storage medium. Background Art

[0002] With the rapid development of my country's railways, railways have become an important part of the national economy. In addition, railway passenger flow statistics are very important in railway planning, layout and service optimization.

[0003] Railway passenger flow statistics rely on accurate railway passenger identification. Therefore, how to accurately identify railway passengers is an urgent problem that needs to be solved. Summary of the Invention

[0004] The object of the present invention is to provide a railway passenger identification method, device, electronic device and computer-readable storage medium to improve the problems existing in the prior art.

[0005] The embodiments of the present invention can be implemented as follows:

[0006] In a first aspect, an embodiment of the present invention provides a railway passenger identification method, comprising:

[0007] Get the signaling set in the target area within the specified time;

[0008] Obtaining the signaling interaction range corresponding to each railway station in the target area;

[0009] Based on the signaling set and the signaling interaction range corresponding to each of the train stations, determining the residency information of each pending user; the pending user is a user who has performed signaling interaction in any of the signaling interaction ranges;

[0010] Based on the residency information of each of the pending users, identifying each candidate user who has performed signaling interactions within at least two signaling interaction ranges from all the pending users, and determining the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey;

[0011] Based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey, misjudged non-railway journeys are filtered out from all candidate journeys of all candidate users to obtain each railway journey of each railway passenger.

[0012] In a second aspect, an embodiment of the present invention further provides a railway passenger identification device, comprising:

[0013] A signaling acquisition module is used to acquire a signaling set within a specified time in a target area; the signaling set includes multiple signaling data of several users;

[0014] A range determination module, configured to obtain a signaling interaction range corresponding to each railway station in the target area;

[0015] a residency determination module, configured to determine residency information of each pending user based on the signaling set and the signaling interaction range corresponding to each of the train stations; the pending user is a user who has performed signaling interaction in any of the signaling interaction ranges;

[0016] a first identification module configured to identify, from all the pending users, each candidate user who has performed signaling interactions within at least two signaling interaction ranges based on the residency information of each of the pending users, and determine an inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey;

[0017] The second identification module is configured to filter out misjudged non-railway journeys from all candidate journeys of all candidate users based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey, to obtain each railway journey of each railway passenger.

[0018] In a third aspect, an embodiment of the present invention further provides an electronic device comprising: a memory and a processor, wherein the memory stores a software program, and when the electronic device is running, the processor executes the software program to implement the railway passenger identification method as described in the first aspect above.

[0019] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the railway passenger identification method described in the first aspect is implemented.

[0020] Compared to the prior art, embodiments of the present invention provide a railway passenger identification method, apparatus, electronic device, and computer-readable storage medium. The method first obtains a signaling set within a specified time period within a target area, comprising multiple pieces of signaling data for several users. The method then obtains the signaling interaction range corresponding to each train station within the target area. Based on the signaling set and the signaling interaction range corresponding to each train station, the method determines the residence information of each candidate user. A candidate user is a user that has interacted with signals within any signaling interaction range. This method filters out all candidate users who have stayed within the signaling interaction range of a train station. Based on the residence information of each candidate user, each candidate user who has interacted with signals within at least two signaling interaction ranges is identified from all candidate users. This method eliminates non-railway passengers who have only stayed at a single train station, such as station attendants or railway staff. Finally, based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate trip, each railway journey of each candidate user is identified from all candidate trips of all candidate users, ultimately accurately identifying each railway journey of each railway passenger within the specified time period. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 This is one of the flow charts of a railway passenger identification method provided by an embodiment of the present invention.

[0023] Figure 2 Schematic diagram of the process of constructing Thiessen polygons.

[0024] Figure 3 This is an example diagram of integrating multiple dwell point data provided in an embodiment of the present invention.

[0025] Figure 4 The second flowchart of a railway passenger identification method provided by an embodiment of the present invention.

[0026] Figure 5 A schematic diagram of segmenting a trajectory sequence provided by an embodiment of the present invention.

[0027] Figure 6 A schematic structural diagram of a railway passenger identification device provided in an embodiment of the present invention.

[0028] Figure 7A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0030] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0031] Railway passenger flow is an important basis for optimizing the layout of the transportation network, rationally planning and layout of stations, and intermodal transport analysis. It usually includes indicators such as the number of passengers sent from the station, passenger flow direction and proportion, passenger flow during peak hours and peak hour coefficient, the maximum number of people gathered at the station, the source of departing passenger flow, and the destination of arriving passenger flow.

[0032] Railway passenger flow depends on accurate railway passenger data. Although existing railway real-name ticketing data can accurately grasp the transportation volume between station pairs and a small number of passenger attribute records, this data is difficult to obtain and almost impossible to obtain. In addition, it is difficult to obtain more comprehensive data such as when departing passengers go to the station, where they go to the station from, how they go to the station, when arriving passengers leave the station, where they go after leaving the station, how they leave the station, as well as the passengers' complete travel trajectories, permanent residence, place of origin, gender, age, etc.

[0033] Therefore, existing technologies can only achieve railway passenger identification through other means, such as:

[0034] The first method involves collecting travel information through traditional questionnaires. However, questionnaire surveys are limited in scope and are only suitable for small-scale studies. Due to labor and time constraints, they are difficult to conduct on a large scale. Furthermore, the quality of questionnaire samples and the reliability of results depend heavily on the respondents' personal preferences, making it difficult for questionnaire distributors to effectively control these factors. Different respondents may interpret the questions differently, leading to inconsistent responses. Some may even answer based on social expectations rather than expressing their own thoughts, further compromising the authenticity and accuracy of the questionnaire data.

[0035] The second method is to build a railway station basic database and a railway line base station database based on mobile phone signaling data. According to the passenger single trip signaling data, the entry, exit, transfer station identification method and travel route matching method are used to extract the spatiotemporal data of the passenger single trip trajectory. This method makes up for the shortcomings of the traditional questionnaire survey method in terms of low data dimensional scalability, high acquisition cost, long survey period, and small survey samples to a certain extent. However, it still has certain shortcomings, mainly including: (1) The identification results of this method are mixed with samples of non-real railway travel, such as station staff (including staff directly related to station travel and personnel selling goods in the station), station pick-up and drop-off personnel, and passing through the station, which will cause great interference to the results; (2) Passenger identification is not closely combined with the behavioral characteristics of railway passenger travel, and the analysis results may have some uncertainty.

[0036] Based on the discovery of the above technical problems, the inventors have proposed the following technical solutions after creative work to solve or improve the above problems. It should be noted that the defects existing in the solutions in the above prior art are the results obtained by the inventors after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed in the embodiments of this application below for the above problems should all be the contributions made by the inventors to this application in the process of invention and creation, and should not be understood as technical contents known to those skilled in the art.

[0037] In light of this, embodiments of the present invention provide a railway passenger identification method that utilizes the signaling set of a target area and the signaling interaction range of each railway station within the target area, fully integrating the travel behavior characteristics of railway passengers, and accurately identifying each railway journey of each railway passenger in the target area within a specified time period. The following examples and accompanying drawings provide a detailed explanation of this method.

[0038] Please refer to Figure 1 , Figure 1 This is a flow chart of a railway passenger identification method provided by an embodiment of the present invention. The method may be performed by, but is not limited to, a smart phone, a personal notebook, a personal computer, a server, or other electronic device with computing capabilities. The method includes the following steps S101 to S105:

[0039] S101: Acquire a signaling set within a specified time in a target area.

[0040] In this embodiment, the target area may be a geographical area of ​​a province, a geographical area of ​​multiple provinces, or a national area, and the designated time may be an exact day, several days, a week, a month, or the like.

[0041] Among them, the signaling set includes multiple signaling data of several users. One signaling data includes but is not limited to a user code (representing the user terminal used by the user), a timestamp, a base station code (representing a public mobile communication base station, referred to as a base station), and a base station location, etc. Therefore, the signaling data can reflect the object (i.e., the base station) and the interaction time of the user's command interaction using the terminal device.

[0042] It should be understood that during the collection, transmission and processing of signaling data, due to some force majeure factors, redundant data, noise data, drift data, ping-pong data and other dirty data will be generated. Therefore, after initially receiving all the signaling data, it is also necessary to eliminate dirty data such as time and space conflicts and invalid data to form the required signaling set.

[0043] S102: Obtain the signaling interaction range corresponding to each railway station in the target area.

[0044] In this embodiment, the term "train station" is a general term, which can be: an ordinary train station serving conventional trains (such as K, T, and Z trains), or a high-speed railway station specifically serving high-speed railways (G and D trains).

[0045] It should be noted that the signaling interaction range is not directly equivalent to the geographical boundary of the railway station, but is an area determined by combining the base station distribution of the railway station and the geographical boundary of the railway station.

[0046] S103: Determine the residence information of each pending user based on the signaling set and the signaling interaction range corresponding to each railway station.

[0047] In this embodiment, the pending users are users who have conducted signaling interactions in any signaling interaction range, that is, the pending users are only suspected railway passengers. All pending users also include railway staff, pick-up and drop-off personnel, passers-by, people traveling by other means, etc., so further identification is required.

[0048] S104: Based on the residency information of each pending user, identify each candidate user who has performed signaling interactions within at least two signaling interaction ranges from all pending users, and determine the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey.

[0049] In this embodiment, the residence information of a pending user reflects the user's movement trajectory within a specified time.

[0050] It should be understood that within a given timeframe, a user must have signaled within the signaling interaction range of at least two train stations to qualify as a railway passenger. Therefore, it is necessary to identify each candidate user who has signaled within at least two signaling interaction ranges from all pending users, excluding passengers such as those picking up passengers at the station or passing by the train station.

[0051] S105 : Based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey, filter out misjudged non-railway journeys from all candidate journeys of all candidate users to obtain each railway journey of each railway passenger.

[0052] Among them, candidate users are not necessarily all railway passengers, and a candidate trip represents a travel journey between two railway stations. Therefore, each candidate trip of a candidate user is not necessarily a railway journey. It is also possible that signaling interaction occurs within the signaling interaction range of two railway stations within a specified time, but the actual travel mode is non-railway travel, such as road travel or air travel.

[0053] Therefore, it is also necessary to identify each railway journey of real railway passengers based on the inter-station trajectory sequence and travel information of each candidate user on each candidate journey, so as to facilitate the subsequent statistics of railway travel passenger flow indicators such as the number of railway travelers and the number of railway trips within the specified time based on all railway journeys of all railway passengers within the specified time.

[0054] The railway passenger identification method provided by an embodiment of the present invention first obtains a signaling set within a target area within a specified time period, which includes multiple pieces of signaling data for several users. It then obtains the signaling interaction range corresponding to each train station in the target area. Based on the signaling set and the signaling interaction range corresponding to each train station, the method determines the residence information of each pending user. A pending user is a user who has interacted with signals within any signaling interaction range. This method can filter out all pending users who have stayed within the signaling interaction range of a train station. Based on the residence information of each pending user, each candidate user who has interacted with signals within at least two signaling interaction ranges is identified from all pending users. This method can exclude non-railway passengers who have only stayed at a single train station, such as station attendants or railway staff. Finally, based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey, each railway journey of each railway passenger is identified from all candidate journeys of all candidate users, ultimately accurately identifying each railway journey of each railway passenger within the specified time period.

[0055] The following takes the target area as the national area as an example to introduce in detail the implementation principle of the railway passenger identification method provided by the embodiment of the present invention.

[0056] There are over 3,600 railway stations across China, each with its own distinct scale, location, and socioeconomic development, leading to significant variations in communications infrastructure. Large, recently built railway stations typically have micro base stations installed within them, while smaller, older stations typically rely on peripheral macro base stations for communication services.

[0057] Therefore, regarding S102 above, the signaling interaction range of a railway station is determined by appropriately expanding the geographical boundaries of the railway station to accommodate the base station installation conditions of the two types of railway stations mentioned above and to avoid missing out on railway passengers from smaller railway stations built earlier. Specifically, the sub-steps of step S102 may include S1021 to S1024.

[0058] S1021. Obtain geographic boundary information of each railway station in the target area.

[0059] In this embodiment, the geographic boundary information of the train station may include the positions (such as longitude and latitude coordinates) of several boundary points on the boundary line of the train station.

[0060] Step S1021 can be obtained in real time from an external system or read from the station information stored within the system. Since train stations may be added, cancelled, renovated, or expanded, the geographical boundary information of each train station in the station information needs to be updated regularly, with a focus on newly added / removed train stations and train stations with changed geographical boundaries.

[0061] S1022. Obtain the longitude and latitude coordinates of each base station in the target area.

[0062] S1023. Determine the signal coverage range of each base station based on the latitude and longitude coordinates of each base station.

[0063] Optionally, the Voronoi Diagram tool provided by ArcGIS can be called to generate the Thiessen polygons of each base station based on the latitude and longitude coordinates of each base station, so that the Thiessen polygon area of ​​each base station can be used as the signal coverage range of the base station.

[0064] For example, taking 12 base stations as an example, please combine Figure 2 , you need to first connect every two adjacent base stations to form a triangulated network, and then construct Thiessen polygons based on the triangulated network, and finally you can get the signal coverage range of each base station. The signal coverage range of base station A is the area surrounded by the red line. It should be noted that Figure 2 The figure is only an example, and the shape of the Thiessen polygon determined by each base station in the target area is not limited here.

[0065] S1024. Perform spatial correlation analysis on the signal coverage of all base stations and the geographical boundary information of each railway station to determine the signaling interaction range of each railway station.

[0066] In this embodiment, for any railway station, the geographical boundary information of the railway station is superimposed with the signal coverage range of all base stations to filter out the signal coverage range of each base station that intersects with the geographical boundary of the railway station, which is used as the signaling interaction range of the railway station.

[0067] In an optional implementation, in step S103 above, the residence information of a pending user includes multiple residence point data, which can reflect the trajectory of the pending user within a specified time. Specifically, the sub-steps of step S103 can include S1031 to S1034.

[0068] S1031 : For any signaling data in the signaling set, if the base station location of the signaling data is within any signaling interaction range, determine that the user corresponding to the signaling data is a pending user to be identified.

[0069] In this embodiment, if all signaling data in the signaling set is classified according to user code, a signaling subset corresponding to each user can be obtained. For any user, if the base station location in any piece of signaling data in the signaling subset corresponding to that user falls within the signaling interaction range of any train station in the target area, the user can be determined as a potential railway passenger. For each potential railway passenger, the residency information can be determined through the following steps S1032-S1034.

[0070] S1032: For each pending user, determine at least one dwelling point data and at least one waypoint data of the pending user based on all signaling data of the pending user in the signaling set.

[0071] In this embodiment, the stay point data includes the location and stay period of the stay point, and the stay period can reflect the stay duration; the waypoint data includes the location and pass time of the waypoint.

[0072] Among them, based on the signaling subset of a pending user, all the residence point data and transit point data of the pending user within a specified time can be analyzed and determined. The specific analysis process is existing technology and will not be described in detail here.

[0073] S1033: Perform stretching processing on the passing time in each waypoint data to convert each waypoint data into residence point data.

[0074] In this embodiment, converting the waypoint data into the stay point data is actually just converting the passage time into the stay period, so that the waypoint is regarded as the stay point.

[0075] For example, assuming the transit time is t and the minimum dwell time ∆t is preset (∆t can be 0.5s, 1s, 2s, etc.), the dwell time [t1, t2] converted from the transit time t can be calculated in the following three ways:

[0076]

[0077] S1034. Integrate all the residence point data corresponding to the pending user, the residence points that are located within the signaling interaction range corresponding to the same train station and have consecutive residence periods, into one piece of residence point data with the train station as the residence point, to obtain multiple residence point data of the pending user.

[0078] In this embodiment, all the residence point data of a pending user can be sorted in ascending order according to the start time of the residence period, and then the residence points that are located within the signaling interaction range corresponding to the same railway station and have continuous residence periods are integrated into a residence point data with the railway station as the residence point. The residence period of the integrated residence point data is the minimum start time to the maximum end time of the original periods.

[0079] For example, assume that a pending user includes 11 dwell point data (i.e., there are 11 dwell points and their dwell time periods), which are sorted by dwell time period into dwell points 1 to 11. Figure 3 Assuming that dwell points 3 to 6 are within the signaling interaction range of station A, and dwell points 7 to 10 are within the signaling interaction range of station B, the integration is as follows:

[0080] (1) Integrate the stay points 3 to 6 into one stay point A (the location of stay point A is the location of station A), and its stay period is from the start time of stay point 3 to the end time of stay point 6;

[0081] (2) Integrate stop points 7 to 10 into one stop point B (the location of stop point B is the location of station B), and its stop period is from the start time of stop point 7 to the end time of stop point 10.

[0082] from Figure 3 It can be seen that after the integration, the number of residence points has been reduced from 11 to 5. It should be noted that this example is only an example and is not limiting.

[0083] In the residence information of each pending user obtained through the above steps S1031 to S1034, at least one residence point in the residence point data is a train station, which can reflect the travel trajectory.

[0084] For the above step S104 , the candidate users may be screened out in combination with the travel characteristics of railway passengers, and the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey may be determined.

[0085] The travel characteristics of railway passengers are mainly as follows:

[0086] Feature 1: Point-to-point travel, that is, travel from station to station, will generate signaling data at the departure station, transit stations, and arrival stations, and the waiting time at the departure station is generally longer than the exit time at the arrival station;

[0087] Feature 2: During a continuous railway trip, users will not stay in areas other than stations for a long time;

[0088] Feature 3: Users may have non-consecutive train trips in a day;

[0089] Feature 4: Transfers may be required during rail travel.

[0090] Based on the above travel characteristics, each pending user can be judged. Figure 1 Based on Figure 4 The sub-steps of the above step S104 may include S1041~S1045.

[0091] S1041 , for each pending user, sort all the dwell point data of the pending user in ascending order according to the start time of the dwell period to obtain a trajectory sequence of each pending user within a specified time period.

[0092] Among them, the trajectory sequence of each pending user within the specified time can clearly reflect the user's spatial movement path and time distribution within the entire specified time.

[0093] S1042: If any trajectory sequence contains an off-station dwell point, the trajectory sequence is divided into at least one trajectory sequence based on all the off-station dwell points.

[0094] In this embodiment, the off-station dwell point is a dwell point where the dwell time corresponding to the dwell period exceeds the preset dwell time, and the dwell point is not a train station. The preset dwell time may be 1 hour or 1.5 hours.

[0095] It can be understood that if there is no trajectory point sequence in a trajectory sequence, there is no need to split it. If there is one off-station residence point in a trajectory sequence, it can be split into two new trajectory sequences. If there are two off-station residence points in a trajectory sequence, it can be split into three new trajectory sequences, and so on.

[0096] For example, assume that a trajectory sequence includes 15 dwell points (dwell points 1 to 15), see Figure 5 , where the locations of the four dwelling points 2, 6, 10, and 14 are the locations of the four railway stations A, B, C, and D, respectively, and the dwelling time of dwelling point 8 exceeds the preset dwelling time, which means it belongs to an out-of-station dwelling point. Then the two trajectory sequences after segmentation are Figure 5 The track sequence 1 and track sequence 2 are shown. It should be noted that this example is only an example and is not limiting.

[0097] S1043. From all trajectory sequences, remove non-railway trajectory sequences that include only one train station or no train station, and remove trajectory sequences of non-railway passengers whose two stops are at the same train station.

[0098] In this embodiment, among all the trajectory sequences obtained through steps S1041 and S1042 above, the following three types of trajectory sequences need to be eliminated:

[0099] (1) Non-railway track sequences that only include one train station: If this type is not segmented, it may be the journey of station personnel, train station staff, or passers-by passing through the train station; if this type is segmented, it may be the journey of a railway passenger before or after a railway trip, and the journey may just pass through the signaling interaction range of a train station;

[0100] (2) Non-railway track sequences that do not include train stations: This type is usually segmented and may be the journey of a railway passenger before or after a railway trip, and the journey does not pass through the signaling interaction range of any train station;

[0101] (3) Trajectory sequences of non-railway passengers with two stops at the same railway station: This type of trajectory sequence is usually the itinerary of the staff of the railway train crew (train conductor, crew members, catering staff, cleaning staff, etc.).

[0102] For example, in combination Figure 5 ,if Figure 5 If stop 10 in the example is not at station C, then trajectory sequence 2 needs to be removed. If stations A and B in trajectory sequence 1 are actually the same train station, and stations C and D in trajectory sequence 1 are also the same train station, then the trajectory sequence belongs to the itinerary of the train crew and needs to be deleted. It should be noted that this example is for illustrative purposes only and is not intended to be limiting.

[0103] S1044: The undetermined user corresponding to each remaining trajectory sequence is regarded as a candidate user, and each remaining trajectory sequence is regarded as a candidate journey to be identified.

[0104] Therefore, step S1043 can exclude train station staff, pick-up and drop-off personnel, and passers-by from all pending users. Furthermore, the non-railway track sequences corresponding to a railway passenger's non-railway itinerary can be excluded from all pending users. Thus, all the pending users corresponding to the remaining track sequences are candidate users who have interacted with signaling within the signaling interaction range of at least two railway stations and are likely to be railway passengers.

[0105] S1045. For each candidate trip of each candidate user, based on the first and last dwell point data of the corresponding trajectory sequence where the dwell point is a train station, determine the inter-station trajectory sequence and travel information of the candidate user on the candidate trip.

[0106] In this embodiment, the inter-station trajectory sequence of a candidate user on a candidate journey only reflects the travel trajectory between two train stations.

[0107] Optionally, the travel information may include the departure station, arrival station, travel time, travel distance, and travel speed. Therefore, for each candidate trip of each candidate user, based on the corresponding trajectory sequence, the process of determining the inter-station trajectory sequence and travel information of the candidate user on the candidate trip may include the following steps S1 to S6:

[0108] S1. From the trajectory sequence, extract the data from the first to the last dwelling point at the train station to obtain the inter-station trajectory sequence of the candidate user on the candidate journey.

[0109] It can be understood that the first and last dwelling points in the inter-station trajectory sequence of a candidate user on a candidate journey are both train stations.

[0110] S2. The first train station and the last train station in the inter-station trajectory sequence are used as the departure station and arrival station of the candidate journey, respectively.

[0111] In this embodiment, the inter-station trajectory sequence on each candidate journey includes a plurality of dwell point data, wherein the first dwell point and the last dwell point are the departure station and the arrival station of the candidate journey, respectively.

[0112] Among them, if there is a stop at a train station between the departure station and the arrival station in the inter-station trajectory sequence and the stay time at the stop is longer than the preset transfer time (for example, 20 minutes or 30 minutes), the intermediate train station can be recorded as the transfer station of the candidate journey.

[0113] S3. Determine the travel time of the candidate journey based on the dwell periods corresponding to the departure station and the arrival station of the candidate journey.

[0114] For example, please combine Figure 5 ,by Figure 5 Taking trajectory sequence 1 in the example, the inter-station trajectory sequence of this candidate trip is from stop 2 to stop 6. The departure station of this candidate trip is station A and the arrival station is station B. The travel time of this candidate trip is the time difference between the end time of the stop period at stop 6 and the start time of the stop period at stop 2. It should be noted that this example is only illustrative and not limiting.

[0115] S4. If the inter-station trajectory sequence includes other train stations besides the departure station and the arrival station, the sum of the distances between each two adjacent train stations in the inter-station trajectory sequence is taken as the travel distance.

[0116] S5. If the inter-station trajectory sequence only includes two train stations, the departure station and the arrival station, the product of the straight-line distance between the departure station and the arrival station and the preset non-linear coefficient is used as the travel distance.

[0117] Among them, by investigating the ratio of some actual train travel distances to the straight-line distances between train stations, the average value of the non-linear coefficient was calculated to be 1.4.

[0118] S6. Calculate the ratio of travel distance to travel time to obtain travel speed.

[0119] In this embodiment, the travel speed only represents the speed at which the user moves between the departure station and the arrival station of a candidate journey.

[0120] The above method of screening out each candidate journey of each candidate user is mainly based on the judgment of spatial position relationship.

[0121] In reality, among all candidate journeys of all candidate users, there are three types of misjudgments that lead to the introduction of non-railway passengers' railway journeys:

[0122] Misidentification scenario 1: Two train stations are close or even overlap, and users passing through are mistakenly counted. For example, Huadu Station and Guangzhou North Station actually overlap geographically. In this case, if a user happens to pass through the signaling interaction range of two train stations and generates signaling, the user's passing trip will be mistakenly counted as a candidate trip. Such users' non-railway trips (actually passing trips) need to be excluded.

[0123] Misjudgment Case 2: With the advancement of the coordinated and integrated development of comprehensive transportation, the development of transportation corridors is showing a trend from single to integrated, and from two-dimensional to three-dimensional. The line coordination and cross-sectional spatial integration of linear infrastructure such as railways and highways are constantly deepening. Therefore, in the actual distribution of transportation routes, railways and highways in the same transportation corridor are often located in close proximity, such as the Shanghai-Kunming Corridor (the Shanghai-Kunming Expressway and the Shanghai-Kunming High-Speed ​​Railway are close to each other) and the Beijing-Harbin Corridor (the Beijing-Harbin Expressway and the Beijing-Harbin High-Speed ​​Railway are close to each other). If a user traveling on the Shanghai-Kunming Expressway happens to have signaling interactions with two train stations on the Shanghai-Kunming High-Speed ​​Railway before or after traveling on the Shanghai-Kunming Expressway, they will be mistakenly counted as candidate users. It is necessary to identify the non-railway trips (actually road trips) of such road users from all candidate trips.

[0124] Misjudgment scenario three: In reality, there are situations where there are train stations at both ends of the airport. For example, a user's flight journey is from Nanning Wuxu International Airport (nearby Wuxu Airport Station) to Lanzhou Zhongchuan International Airport (nearby Zhongchuan Airport East Station), and the user happens to enter the signaling interaction range of the train stations at both ends and conduct signaling interaction. In this case, this user's flight journey will also be mistakenly counted as a candidate user's candidate journey. It is necessary to identify the non-railway journeys (actually flight journeys) of this type of flight user from all the candidate users' candidate journeys.

[0125] Therefore, the sub-steps of the above step S105 may include S1051 to S1054:

[0126] S1051. Filter out target candidate trips from all candidate trips, whose travel speed does not exceed a preset speed range and whose travel time is greater than a preset travel duration.

[0127] In this embodiment, the non-railway trips indicated by the above-mentioned misjudgment case 1 generally do not conform to the speed and time characteristics of real railway travel. Therefore, the non-railway trips corresponding to the above-mentioned misjudgment case 1 can be excluded by determining whether the travel speed and travel time of each candidate user's candidate trip meet the speed and time characteristics of real railway travel.

[0128] Among them, the preset travel time can be set based on the shortest travel time between adjacent stations in the target area, for example, it can be set to 5 minutes, 10 minutes, etc., which is not limited here.

[0129] Optionally, the preset speed range can be set directly or determined by:

[0130] A PDF (Probability Density Function) is constructed based on a large number of historical travel speeds of various railway trains in the target area. The PDF function reflects the probability distribution of travel speeds. Then, based on the PDF function, it is determined that the probability of travel speeds being in the range of 30km / h to 350km / h is 85%. Therefore, the preset speed range can be set to 30km / h to 350km / h.

[0131] Among them, the probability density function is: , is the mean driving speed, is the standard deviation of driving speed; calculate the speed range The corresponding probability density formula is: ; Standardized conversion: , , .

[0132] S1052: Based on the inter-station trajectory sequence and travel information corresponding to each target candidate trip, filter out the first type of non-railway trips that are misjudged due to the co-linearity of railways and highways from all target candidate trips.

[0133] In this embodiment, to exclude non-railway trips corresponding to the second misjudgment scenario, it is necessary to identify whether the departure and arrival stations of each candidate trip are collinear station pairs. If so, the candidate trip is further determined to be a non-railway trip of the first type, where a railway and a highway are collinear, leading to misjudgment. A collinear station pair refers to a situation where the railway line and highway are close together.

[0134] Therefore, an optional implementation of step S1052 includes:

[0135] Step a: The departure station and arrival station in the inter-station trajectory sequence corresponding to each target candidate journey are regarded as a station pair;

[0136] Step b, grouping the inter-station trajectory sequences corresponding to all target candidate journeys with the same station pair into a journey set;

[0137] Step c: Based on each set of journeys, determine the route category of each station pair, where the route category of a station pair is either a collinear station pair or a non-collinear station pair;

[0138] Step d: For any candidate trip corresponding to each co-linear station pair, if the dwell time at the departure station in the inter-station trajectory sequence corresponding to the candidate trip is less than the preset waiting time threshold, then the candidate trip is determined to be a first-class non-railway trip misclassified due to the co-linearity of railway and highway.

[0139] In this embodiment, a station pair represents two train stations with a fixed travel direction. For example, a station pair A→B indicates that the departure station is station A and the arrival station is station B; while a station pair B→A indicates that the departure station is station B and the arrival station is station A.

[0140] Among them, due to the need for security checks, ticket checks and waiting times after entering the station, real railway journeys usually require a certain amount of time to stay at the departure station. Therefore, after determining the line category of each station pair, for each candidate journey of the collinear station pair: if the dwell time at the departure station in the inter-station trajectory sequence corresponding to this candidate journey is less than the preset waiting time threshold (for example, 10 minutes), then this candidate journey is determined to belong to the first type of non-railway journey caused by the collinearity of railway and highway; if the dwell time at the departure station is not less than, then this candidate journey is actually a real railway journey.

[0141] Therefore, in practice, a typical characteristic of non-collinear station pairs is that the dwell time at the departure station in each inter-station trajectory sequence in the corresponding journey set is generally no less than a preset waiting time threshold (e.g., 10 minutes). A typical characteristic of collinear station pairs is that the dwell time at the departure station in each inter-station trajectory sequence in the corresponding journey set is generally less than a preset waiting time threshold (e.g., 10 minutes). This type of trajectory sequence is not a railway trajectory sequence, but rather a road trip trajectory.

[0142] In an optional implementation, a clustering approach may be used to determine the route category of each station pair. That is, in step c, the implementation of "determining the route category of each station pair based on each journey set" includes:

[0143] Step c-1: Determine the feature vector of each station pair based on the dwell time at the departure station of each inter-station trajectory sequence in each journey set;

[0144] Step c-2: cluster the feature vectors of all station pairs to obtain two cluster sets;

[0145] Step c-3: If the first vector value in the feature vector corresponding to each station pair in one of the cluster sets is greater than the set threshold, then each station pair in the cluster set is determined to be a collinear station pair, and each station pair in the other cluster set is determined to be a non-collinear station pair.

[0146] The feature vector reflects the proportion of trips whose dwell time falls within multiple intervals, ranging from small to large. The first value of the feature vector corresponds to a duration interval from 0 to a preset waiting time threshold. The threshold can be 80%, 85%, 90%, etc.

[0147] For example, assume there are six duration intervals: less than 10 minutes, 10 minutes to 30 minutes, 30 minutes to 1 hour, 1 hour to 1 hour and 30 minutes, 1 hour and 30 minutes to 2 hours, and more than 2 hours. Assuming a station pair A→B (the departure station is station A and the arrival station is station B), the corresponding journey set includes 100 inter-station trajectory sequences. If the number of dwell times at the departure station in these 100 inter-station trajectory sequences in each duration interval and the proportion of the journeys are as follows:

[0148]

[0149] Therefore, the eigenvector corresponding to the station pair A→B is 0.87, 0.09, 0.02, 0.02, 0, 0, which is a 6-dimensional vector. If the threshold is set to 85%, and the first vector value of the eigenvector corresponding to each station pair in the cluster set where the station pair A→B is located is greater than 85%, it can be confirmed that the line category of each station pair in the cluster set where the station pair A→B is located is a collinear station pair.

[0150] It should be noted that this example is only an example, and the present invention does not limit the number of inter-station trajectory sequences in the journey set, the granularity and number of divisions of the duration intervals, etc.

[0151] In another optional implementation, the route category of each station pair can be determined by using data statistics. That is, in step c, the implementation of "determining the route category of each station pair based on each journey set" includes:

[0152] Step cA: For each station pair, count the proportion of trips in the corresponding trip set whose dwell time at the departure station is less than a preset waiting time threshold;

[0153] Step cB: If the proportion of the number of trips is greater than a set threshold, the station pair is determined to be a collinear station pair;

[0154] Step cC: If the proportion of the number of trips is not greater than the set threshold, the station pair is determined to be a non-collinear station pair.

[0155] In practical applications, whether to use clustering or data statistics to determine the route category of each station pair can be determined based on the size of each journey set, and a method with low computing power consumption or fast calculation speed can be selected.

[0156] S1053. Filter out the second type of non-railway trips that are misjudged due to proximity between the airport and the train station from all target candidate trips.

[0157] For the non-railway journeys corresponding to the three aforementioned misjudgment situations, to ensure flight safety, the user's mobile phone is in flight mode or off mode during the flight. No signaling data is actually generated during the flight, and the user's stay point data will not be counted during the flight period. In addition, under normal circumstances, the train stations at both ends overlap with the airport, that is, the train stations are located below the ground level of the airport. For example, Hongqiao Railway Station and Hongqiao Airport T2 Terminal are horizontally connected within the same building, and Zhengding Airport Station is located on the basement floor of the terminal building of Shijiazhuang Zhengding Airport.

[0158] Therefore, in the trajectory sequence between stations corresponding to a target candidate trip, if there is no dwell point between the departure station and the arrival station and the departure station and the arrival station are located below or close to the airport, then the end time of the dwell period corresponding to the departure station and the start time of the dwell period corresponding to the arrival station should actually be a flight period. This can be considered that this target candidate trip belongs to the second type of non-railway trip that is misjudged due to the proximity of the airport and the railway station, and needs to be eliminated.

[0159] S1054. Delete each type of non-railway journey from all target candidate journeys to obtain each railway journey of each railway passenger.

[0160] Through the above steps S1051 to S1054, the non-railway journeys in the above three misjudgment situations are eliminated, and the remaining ones are each railway journey of each railway passenger.

[0161] Ultimately, the system outputs each passenger's user code, along with their departure and arrival stations, transfer stations, time of entry to the departure station (i.e., the start time of their stay at the departure station), time of departure from the departure station (i.e., the end time of their stay at the departure station), time of entry to the arrival station (i.e., the start time of their stay at the arrival station), time of departure from the arrival station (i.e., the end time of their stay at the arrival station), travel time, distance, and speed. This information is then output as a detailed passenger list. This facilitates subsequent analysis of passenger flow metrics within a target area within a specified timeframe, as well as analysis of origin, destination, and transfers.

[0162] It should be noted that the execution order of the steps in the above method embodiment is not limited to that shown in the drawings, and the execution order of the steps shall be based on actual application conditions.

[0163] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0164] By adopting the signaling set of the target area and the signaling interaction range of each railway station in the target area, fully combining the travel behavior characteristics of railway passengers, and comprehensively considering various misjudgment situations, we can ultimately accurately identify each railway journey of each railway passenger in the target area within the specified time.

[0165] In order to execute the corresponding steps in the above method embodiment and various possible implementations, an implementation of a railway passenger identification device is provided below.

[0166] See Figure 6 , Figure 6 The structure diagram of the railway passenger identification device provided by an embodiment of the present invention is shown. The railway passenger identification device 200 includes: a signaling acquisition module 210, a range determination module 220, a residence determination module 230, a first identification module 240 and a second identification module 250, wherein:

[0167] The signaling acquisition module 210 is used to acquire a signaling set within a specified time in a target area; the signaling set includes multiple signaling data of several users;

[0168] Range determination module 220, for obtaining the signaling interaction range corresponding to each railway station in the target area;

[0169] The residency determination module 230 is used to determine the residency information of each pending user based on the signaling set and the signaling interaction range corresponding to each railway station; the pending user is a user who has performed signaling interaction in any signaling interaction range;

[0170] A first identification module 240 is configured to identify, from all the pending users, each candidate user who has performed signaling interactions within at least two signaling interaction ranges based on the residency information of each pending user, and determine the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey;

[0171] The second identification module 250 is configured to filter out misidentified non-railway journeys from all candidate journeys of all candidate users based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey, and obtain each railway journey of each railway passenger.

[0172] Optionally, the range determination module 220 can be specifically used to: obtain the geographical boundary information of each railway station in the target area; obtain the latitude and longitude coordinates of each base station in the target area; determine the signal coverage range of each base station based on the latitude and longitude coordinates of each base station; perform spatial correlation analysis on the signal coverage range of all base stations and the geographical boundary information of each railway station to determine the signaling interaction range of each railway station.

[0173] Optionally, the signaling set includes multiple signaling data of several users, the signaling data including the base station location; the residence information includes multiple residence point data, and at least one residence point in the residence point data is a train station. The residence determination module 230 can be specifically used to:

[0174] For any signaling data in the signaling set, if the base station location of the signaling data is within any signaling interaction range, the user corresponding to the signaling data is determined to be a pending user to be identified; for each pending user, based on all the signaling data of the pending user in the signaling set, at least one residence point data and at least one waypoint data of the pending user are determined; the residence point data includes the location and residence period of the residence point; the waypoint data includes the location and passage time of the waypoint; the passage time in each waypoint data is stretched to convert each waypoint data into residence point data; among all the residence point data corresponding to the pending user, the residence points that are within the signaling interaction range corresponding to the same train station and have continuous residence periods are integrated into a set of residence point data with the residence point being the train station, thereby obtaining multiple residence point data of the pending user.

[0175] Optionally, the residence information includes multiple residence point data, the residence point data includes the location and residence period of the residence point, and at least one residence point in each residence information is a train station. The first identification module 240 can be specifically used to:

[0176] For each pending user, all dwell point data of the pending user are sorted in ascending order according to the start time of the dwell period to obtain the trajectory sequence of each pending user within the specified time period. If any trajectory sequence contains an off-station dwell point, the trajectory sequence is divided into at least one trajectory sequence based on all the off-station dwell points. An off-station dwell point is a dwell point whose dwell time corresponding to the dwell period exceeds a preset dwell time. From all trajectory sequences, non-railway trajectory sequences that include only one train station or no train station are eliminated, and trajectory sequences of non-railway passengers with two dwell points at the same train station are eliminated. The pending user corresponding to each remaining trajectory sequence is regarded as a candidate user, and each remaining trajectory sequence is regarded as a candidate journey to be identified. For each candidate journey of each candidate user, the inter-station trajectory sequence and travel information of the candidate user on that candidate journey are determined based on the data of the first and last dwell points in the corresponding trajectory sequence that have a dwell point at a train station.

[0177] Optionally, the travel information includes the departure station, arrival station, travel time, travel distance, and travel speed. The first identification module 240, in the process of "determining the inter-station trajectory sequence and travel information of the candidate user on the candidate journey based on the first and last dwell point data of the corresponding trajectory sequence where the dwell point is a train station," can specifically be used to:

[0178] From the trajectory sequence, the portion from the first dwell point data to the last dwell point data at a train station is intercepted to obtain the inter-station trajectory sequence of the candidate user on the candidate journey. The first and last train stations in the inter-station trajectory sequence are used as the departure station and arrival station of the candidate journey, respectively. The travel time of the candidate journey is determined based on the dwell periods corresponding to the departure station and arrival station of the candidate journey. If the inter-station trajectory sequence includes train stations other than the departure station and arrival station, the sum of the distances between each two adjacent train stations in the inter-station trajectory sequence is used as the travel distance. If the inter-station trajectory sequence only includes the departure station and arrival station, the travel distance is the product of the straight-line distance between the departure station and the arrival station and the preset non-linear coefficient. The travel speed is calculated by calculating the ratio of travel distance to travel time.

[0179] Optionally, the inter-station trajectory sequence includes multiple dwell point data, the dwell point data including the location and dwell period of the dwell point, the dwell period reflecting the dwell duration, the first dwell point and the last dwell point in the inter-station trajectory sequence being the departure station and arrival station of a candidate journey, respectively; the travel information includes the departure station, arrival station, travel time, travel distance, and travel speed. The second identification module 250 can be specifically used to:

[0180] From all candidate trips, target candidate trips are selected whose travel speed does not exceed a preset speed range and whose travel time is greater than a preset travel duration. Based on the inter-station trajectory sequence and travel information corresponding to each target candidate trip, the first type of non-railway trips that are misidentified due to the co-linearity of railways and highways are selected from all target candidate trips. The second type of non-railway trips that are misidentified due to the proximity of airports and railway stations are selected from all target candidate trips. In the inter-station trajectory sequence of the second type of non-railway trips, there is no stop between the departure station and the arrival station, and the departure station and the arrival station are located below or immediately adjacent to the airport. Each type of non-railway trip is deleted from all target candidate trips to obtain each railway trip of each railway passenger.

[0181] Optionally, the second identification module 250, in the process of "screening out the first type of non-railway trips that are misidentified due to the co-linearity of railways and highways from all target candidate trips based on the inter-station trajectory sequence and travel information corresponding to each target candidate trip," can specifically be used to:

[0182] The departure and arrival stations in the inter-station trajectory sequence corresponding to each target candidate trip are considered a station pair. The inter-station trajectory sequences corresponding to all target candidate trips with the same station pair are grouped into a trip set. Based on each trip set, the route category of each station pair is determined, which can be either a collinear station pair or a non-collinear station pair. For any candidate trip corresponding to a collinear station pair, if the dwell time at the departure station in the inter-station trajectory sequence corresponding to that candidate trip is lower than a preset waiting time threshold, the candidate trip is determined to be a Class 1 non-railway trip, which is misclassified due to the collinearity of railways and highways.

[0183] Optionally, the second identification module 250 , in the process of “determining the route category of each station pair based on each journey set”, may be specifically configured to:

[0184] Based on the dwell time at the departure station of each inter-station trajectory sequence in each journey set, a feature vector is determined for each station pair; the feature vector reflects the proportion of journeys whose dwell time falls within multiple duration intervals from small to large; the feature vectors of all station pairs are clustered to obtain two cluster sets; if the first vector value in the feature vector corresponding to each station pair in one cluster set is greater than a set threshold, then each station pair in the cluster set is determined to be a collinear station pair, and each station pair in the other cluster set is determined to be a non-collinear station pair; or, specifically, it can also be used for:

[0185] For each station pair, the proportion of trips in the corresponding trip set whose dwell time at the departure station is lower than the preset time threshold is counted; if the proportion of trips is greater than the set threshold, the station pair is determined to be a collinear station pair; if the proportion of trips is not greater than the set threshold, the station pair is determined to be a non-collinear station pair.

[0186] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the railway passenger identification device 200 described above can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0187] See Figure 7 , Figure 7 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 300 includes a processor 310 , a memory 320 , and a bus 330 , wherein the processor 310 is connected to the memory 320 via the bus 330 .

[0188] The memory 320 may be used to store software programs or firmware, for example, the software programs or firmware corresponding to the railway passenger identification device 200. The processor 310 executes the software programs stored in the memory 320 to perform various functional applications and data processing to implement the railway passenger identification method provided in the embodiment of the present invention.

[0189] Among them, the memory 320 can be but is not limited to: RAM (Random Access Memory), ROM (Read Only Memory), FLASH (Flash Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electric Erasable Programmable Read-Only Memory), etc.

[0190] The processor 310 can be an integrated circuit chip with signal processing capabilities and can be used to execute software programs, such as the software program corresponding to the railway passenger identification device 200. The processor 310 can be a general-purpose processor, including a CPU (Central Processing Unit), an NP (Network Processor), or an SoC (System on Chip). It can also be a DSP (Digital Signal Processing), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0191] I understand. Figure 7 The structure shown is for illustration only. The electronic device 300 may also include Figure 7 More or fewer components than shown, or with Figure 7 Different configurations shown. Figure 7 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0192] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the railway passenger identification method disclosed in the above embodiment. The computer-readable storage medium may be, but is not limited to, a USB flash drive, a mobile hard drive, ROM, RAM, PROM, EPROM, EEPROM, a FLASH disk, or an optical disk, among other media capable of storing program code.

[0193] In summary, embodiments of the present invention provide a railway passenger identification method, apparatus, electronic device, and computer-readable storage medium. The method first obtains a signaling set within a specified time period within a target area, comprising multiple pieces of signaling data for several users. The method then obtains the signaling interaction range corresponding to each train station within the target area. Based on the signaling set and the signaling interaction range corresponding to each train station, the method then determines the residency information of each pending user. A pending user is a user that has interacted with signals within any signaling interaction range. This method allows screening out all pending users who have stayed within the signaling interaction range of a train station. Based on the residency information of each pending user, each candidate user who has interacted with signals within at least two signaling interaction ranges is identified from all pending users. This method eliminates non-railway passengers who have only stayed at a single train station, such as station attendants or railway staff. Finally, based on the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey, each railway journey of each railway passenger is identified from all candidate journeys of all candidate users, ultimately accurately identifying each railway journey of each railway passenger within the specified time period.

[0194] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A railway passenger identification method, characterized in that: include: Get the signaling set in the target area within the specified time; Obtaining the signaling interaction range corresponding to each railway station in the target area; Determining the residency information of each pending user based on the signaling set and the signaling interaction range corresponding to each of the train stations; The pending user is a user who has performed signaling interaction in any of the signaling interaction ranges; Based on the dwell information of each of the pending users, each candidate user who has conducted signaling interactions within at least two signaling interaction ranges is identified from all the pending users, and an inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey are determined; wherein the inter-station trajectory sequence includes a plurality of dwell point data, the dwell point data includes the location and dwell period of the dwell point, the dwell period reflects the dwell duration, the first dwell point and the last dwell point in the inter-station trajectory sequence are the departure station and arrival station of a candidate journey, respectively; the travel information includes the departure station, the arrival station, the travel time, and the travel speed; Filtering, from all candidate trips, target candidate trips whose travel speed does not exceed a preset speed range and whose travel time is greater than a preset travel time; Based on the inter-station trajectory sequence and travel information corresponding to each target candidate trip, the first type of non-railway trips that are misjudged due to the co-linearity of railways and highways are screened out from all target candidate trips; Filter out the second type of non-railway trips that are misjudged due to the proximity of an airport to a train station from all candidate target trips; in the inter-station trajectory sequence of the second type of non-railway trips, there is no stop between the departure station and the arrival station, and the departure station and the arrival station are located below or immediately adjacent to the airport; Each type of non-railway journey is deleted from all target candidate journeys to obtain each railway journey of each railway passenger.

2. The railway passenger identification method according to claim 1, characterized in that: The step of obtaining the signaling interaction range corresponding to each railway station in the target area includes: Obtaining geographic boundary information of each railway station in the target area; Obtaining the longitude and latitude coordinates of each base station in the target area; Determining the signal coverage range of each base station based on the latitude and longitude coordinates of each base station; A spatial correlation analysis is performed on the signal coverage ranges of all the base stations and the geographical boundary information of each of the railway stations to determine the signaling interaction range of each of the railway stations.

3. The railway passenger identification method according to claim 1, characterized in that: The signaling set includes multiple signaling data of several users, the signaling data including base station locations; the residency information includes multiple residency point data, and the residency point in at least one residency point data is the train station; the step of determining the residency information of each pending user based on the signaling set and the signaling interaction range corresponding to each train station includes: For any signaling data in the signaling set, if the base station location of the signaling data is within any of the signaling interaction ranges, determining that the user corresponding to the signaling data is a pending user to be identified; For each of the pending users, determining at least one dwelling point data and at least one waypoint data of the pending user based on all signaling data of the pending user in the signaling set; the dwelling point data includes the location and dwelling period of the dwelling point; the waypoint data includes the location and passing time of the waypoint; Performing stretching processing on the passing time in each of the waypoint data to convert each of the waypoint data into dwell point data; Among all the residence point data corresponding to the pending user, the residence points whose residence points are located within the signaling interaction range corresponding to the same train station and whose residence periods are continuous are integrated into one piece of residence point data whose residence point is the train station, thereby obtaining multiple residence point data of the pending user.

4. The railway passenger identification method according to claim 1, characterized in that: The residence information includes a plurality of residence point data, wherein the residence point data includes the location and residence period of the residence point, and at least one residence point in each of the residence information is the railway station; The step of identifying each candidate user who has performed signaling interactions within at least two signaling interaction ranges from all the candidate users based on the residency information of each of the candidate users, and determining the inter-station trajectory sequence and travel information of each candidate user on at least one candidate journey includes: For each of the pending users, sorting all the residence point data of the pending user in ascending order of the start time of the residence period to obtain a trajectory sequence of each of the pending users within the specified time; If any of the trajectory sequences contains an off-station dwell point, the trajectory sequence is divided into at least one trajectory sequence based on all the off-station dwell points; the off-station dwell point is a dwell point where the dwell time corresponding to the dwell period exceeds a preset dwell time; From all trajectory sequences, remove non-railway trajectory sequences that include only one train station or no train station, and remove trajectory sequences of non-railway passengers whose two stops are at the same train station; The undetermined user corresponding to each remaining trajectory sequence is regarded as the candidate user, and each remaining trajectory sequence is regarded as a candidate journey to be identified; For each candidate journey of each candidate user, based on the first and last dwell point data of the corresponding trajectory sequence in which the dwell point is the train station, the inter-station trajectory sequence and travel information of the candidate user on the candidate journey are determined.

5. The railway passenger identification method according to claim 4, characterized in that: The travel information also includes travel distance; The step of determining the inter-station trajectory sequence and travel information of the candidate user on the candidate journey based on the first dwelling point data and the last dwelling point data of the corresponding trajectory sequence in which the dwelling point is the train station includes: From the trajectory sequence, intercept the portion from the first dwell point data to the last dwell point data of the train station, to obtain the inter-station trajectory sequence of the candidate user on the candidate journey; The first train station and the last train station in the inter-station trajectory sequence are respectively used as the departure station and the arrival station of the candidate journey; Determine the travel time of the candidate trip based on the dwell periods corresponding to the departure and arrival stations of the candidate trip; If the inter-station trajectory sequence includes other train stations in addition to the departure station and the arrival station, the sum of the distances between each two adjacent train stations in the inter-station trajectory sequence is used as the travel distance; If the inter-station trajectory sequence only includes two train stations, the departure station and the arrival station, the product of the straight-line distance between the departure station and the arrival station and a preset non-linear coefficient is used as the travel distance; The ratio of the travel distance to the travel time is calculated to obtain the travel speed.

6. The railway passenger identification method according to claim 1, characterized in that: The step of screening out the first type of non-railway trips that are misjudged due to the co-linearity of railways and highways from all target candidate trips based on the inter-station trajectory sequence and travel information corresponding to each target candidate trip includes: The departure station and the arrival station in the inter-station trajectory sequence corresponding to each target candidate journey are regarded as a station pair; Grouping the inter-station trajectory sequences corresponding to all target candidate trips with the same station pair into a trip set; Determining a route category of each station pair based on each of the journey sets, wherein the route category is a collinear station pair or a non-collinear station pair; For any candidate journey corresponding to each co-linear station pair, if the dwell time at the departure station in the inter-station trajectory sequence corresponding to this candidate journey is lower than the preset waiting time threshold, then this candidate journey is considered to be a Class I non-railway journey due to misidentification caused by the co-linearity of railways and highways.

7. The railway passenger identification method according to claim 6, characterized in that: The step of determining the route category of each station pair based on each of the journey sets comprises: Determining a feature vector for each station pair based on the dwell time at the departure station of each inter-station trajectory sequence in each of the trip sets; the feature vector reflects the proportion of trips whose dwell time falls within a plurality of duration intervals from small to large; Clustering the feature vectors of all the station pairs to obtain two cluster sets; If the first vector value in the feature vector corresponding to each station pair in one of the cluster sets is greater than the set threshold, then each station pair in the cluster set is determined to be a collinear station pair, and each station pair in the other cluster set is determined to be a non-collinear station pair.

8. The railway passenger identification method according to claim 6, characterized in that: The step of determining the route category of each station pair based on each of the journey sets comprises: For each station pair, counting the proportion of trips in the corresponding trip set whose dwell time at the departure station is less than a preset time threshold; If the proportion of the number of trips is greater than a set threshold, determining that the station pair is the collinear station pair; If the proportion of the number of journeys is not greater than the set threshold, the station pair is determined to be a non-collinear station pair.

9. A railway passenger identification device, characterized in that: include: A signaling acquisition module is used to acquire a signaling set within a specified time in a target area; the signaling set includes multiple signaling data of several users; A range determination module, configured to obtain a signaling interaction range corresponding to each railway station in the target area; A residency determination module, configured to determine residency information of each pending user based on the signaling set and the signaling interaction range corresponding to each of the train stations; The pending user is a user who has performed signaling interaction in any of the signaling interaction ranges; a first identification module configured to identify, from all the pending users, each candidate user who has conducted signaling interactions within at least two signaling interaction ranges based on the dwell information of each of the pending users, and determine an inter-station trajectory sequence and travel information for each candidate user on at least one candidate journey; wherein the inter-station trajectory sequence includes a plurality of dwell point data, the dwell point data includes a location and a dwell period of the dwell point, the dwell period reflects the dwell duration, the first dwell point and the last dwell point in the inter-station trajectory sequence are respectively the departure station and arrival station of a candidate journey; and the travel information includes the departure station, the arrival station, the travel time, and the travel speed; The second identification module is used to: Filtering, from all candidate trips, target candidate trips whose travel speed does not exceed a preset speed range and whose travel time is greater than a preset travel time; Based on the inter-station trajectory sequence and travel information corresponding to each target candidate trip, the first type of non-railway trips that are misjudged due to the co-linearity of railways and highways are screened out from all target candidate trips; Filter out the second type of non-railway trips that are misjudged due to the proximity of an airport to a train station from all candidate target trips; in the inter-station trajectory sequence of the second type of non-railway trips, there is no stop between the departure station and the arrival station, and the departure station and the arrival station are located below or immediately adjacent to the airport; Each type of non-railway journey is deleted from all target candidate journeys to obtain each railway journey of each railway passenger.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a software program, and when the electronic device is running, the processor executes the software program to implement the railway passenger identification method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the railway passenger identification method according to any one of claims 1 to 8 is implemented.

Images