Search

CN-119942776-B - Intersection scene recognition method and computer readable storage medium

CN119942776BCN 119942776 BCN119942776 BCN 119942776BCN-119942776-B

Abstract

The embodiment of the application provides an intersection scene recognition method and a computer-readable storage medium, and relates to the technical field of traffic signal control. The method comprises the steps of obtaining target passing data of a target lane in each direction of a target intersection during a target green light period in a time period to be controlled, obtaining a first feature to be identified, a second feature to be identified and a third feature to be identified based on the target passing data, determining a traffic flow dispersion state of the target lane based on the first feature to be identified, determining a traffic demand state of the target lane based on the second feature to be identified and determining a traffic running state of the target lane based on the third feature to be identified, and determining scene types of the target intersection in the time period to be controlled according to the traffic flow dispersion state, the traffic demand state and the traffic running state of the target lane in each direction. In this way, the scene recognition can be performed on the intersection according to the passing data of the intersection.

Inventors

  • JIANG WEIHAO
  • YUAN SHUFEN
  • TAN XU
  • Yan lijing
  • HAO YONGGANG

Assignees

  • 杭州海康威视数字技术股份有限公司

Dates

Publication Date
20260505
Application Date
20231102

Claims (10)

  1. 1. A method for identifying an intersection scene, the method comprising: Aiming at a target lane of each direction of a target intersection, acquiring target passing data of the target lane in a target green light period in a time period to be controlled; Based on the target passing data, obtaining a first feature to be identified which represents the probability distribution of the headway in the target lane in the target green light period, a second feature to be identified which represents the number change condition of the target lane passing vehicles in each statistical time period in the target green light period, and a third feature to be identified which represents the probability distribution of the vehicle running speed in the target lane in the target green light period; Determining a traffic flow dispersion state of the target lane based on the first feature to be identified, determining a traffic demand state of the target lane based on the second feature to be identified, and determining a traffic running state of the target lane based on the third feature to be identified, wherein the traffic flow dispersion state is discrete, weakly followed or strongly followed; If at least one target lane in each direction of the target intersection meets a first screening condition, determining that the scene type of the target intersection in the time period to be controlled is a queuing balance type, wherein the first screening condition indicates that the traffic flow discrete state is strong following, the traffic demand state is high traffic demand, and the traffic running state is slow running; if no target lane meeting the first screening condition exists in the target lanes in each direction of the target intersection and at least one target lane meets a second screening condition, determining that the scene type of the target intersection in the time period to be controlled is traffic capacity type, wherein the second screening condition indicates that the traffic flow discrete state is strong-following or weak-following, the traffic demand state is high traffic demand, and the traffic running state is smooth; If the target lanes in all directions of the target intersection do not exist, which meet the first screening condition, and the target lanes do not exist, which meet the second screening condition, determining that the scene type of the target intersection in the time period to be controlled is an efficiency improvement type.
  2. 2. The method according to claim 1, characterized in that in the period to be controlled, the number of target lane-releasing vehicles in each lane of the road section in either direction in the target intersection is the largest.
  3. 3. The method of claim 1, wherein prior to determining the traffic flow dispersion state for the target lane based on the first feature to be identified, the method further comprises: Acquiring first sample passing data of a first sample lane in a first sample green light period when the first sample lane is in a preset traffic flow discrete state according to each preset traffic flow discrete state; Obtaining a first sample feature representing a probability distribution of headway in the first sample lane during the first sample green light based on the first sample passing data; The determining the traffic flow dispersion state of the target lane based on the first feature to be identified comprises the following steps: Calculating the similarity between the first sample characteristic of the preset traffic flow dispersion state and the first characteristic to be identified according to a first similarity algorithm aiming at each preset traffic flow dispersion state, wherein the first similarity algorithm comprises at least one of a bulldozing distance algorithm and a Markov distance algorithm; And determining the corresponding preset traffic flow dispersion state with the maximum similarity as the traffic flow dispersion state of the target lane.
  4. 4. A method according to claim 3, wherein said deriving a first sample feature representing a probability distribution of headway in the first sample lane during the first sample green light based on the first sample passing data comprises: Determining each headway for a specified position in the first sample lane during the first sample green light as a sample headway based on the first sample passing data; determining a preset headway interval to which each sample headway belongs, wherein the preset headway intervals are adjacent and have the same length; calculating the ratio of the sample headway belonging to each preset headway interval in all the sample headway according to each preset headway interval to obtain a first sample characteristic; The obtaining, based on the target passing data, a first feature to be identified representing a probability distribution of a headway in the target lane during the target green light includes: Based on the target passing data, determining each headway aiming at a designated position of the target lane in the target green light period as a headway to be processed; determining a preset headway interval to which each headway to be processed belongs; And calculating the ratio of the to-be-processed headway belonging to each preset headway interval in all the to-be-processed headway intervals according to each preset headway interval to obtain a first to-be-identified feature.
  5. 5. The method of claim 1, wherein prior to determining the traffic demand status of the target lane based on the second feature to be identified, the method further comprises: acquiring second sample passing data of a second sample lane in a second sample green light period when the second sample lane is in a preset traffic demand state according to each preset traffic demand state; obtaining a second sample characteristic representing the number change condition of the second sample lane release vehicles in each statistical time period during the green light of the second sample based on the second sample vehicle passing data; the determining the traffic demand state of the target lane based on the second feature to be identified includes: Calculating the similarity between the second sample characteristics of the preset traffic demand state and the second characteristics to be identified according to each preset traffic demand state; And determining the corresponding preset traffic demand state with the maximum similarity as the traffic demand state of the target lane.
  6. 6. The method of claim 5, wherein the second sample green light period comprises a plurality of sample green light periods; The obtaining, based on the second sample passing data, a second sample feature representing a change condition of the number of the second sample lane release vehicles in each statistical time period during the second sample green light includes: for each sample green light period, determining the number of the second sample lane release vehicles in each statistical time period of the sample green light period based on second sample vehicle passing data of the sample green light period, and taking the number as the release number of each statistical time period of the sample green light period; Calculating the average value of the number of vehicles passing in each sample green light period in each statistical period according to any statistical period to obtain a second sample characteristic representing the number change condition of the vehicles passing in the second sample lane in each statistical period in the second sample green light period; the obtaining, based on the target passing data, a second feature to be identified representing a number change condition of the target lane release vehicles in each statistical time period during the target green light includes: For each target green light period, determining the number of the target lane release vehicles in each statistical time period of the target green light period based on target passing data of the target green light period, and taking the number as the release number of each statistical time period of the target green light period; and calculating the average value of the release numbers of the statistical time periods of each target green light period according to any statistical time period to obtain a second feature to be identified, wherein the second feature to be identified represents the change condition of the number of the target lane release vehicles in each statistical time period of each target green light period.
  7. 7. The method of claim 1, wherein prior to determining the traffic operating status of the target lane based on the third feature to be identified, the method further comprises: Acquiring third sample passing data of a third sample lane in a third sample green light period when the third sample lane is in a preset traffic running state according to each preset traffic running state; obtaining a third sample feature representing a probability distribution of vehicle travel speed in the third sample lane during the third sample green light based on the third sample passing data; the determining the traffic running state of the target lane based on the third feature to be identified includes: Calculating the similarity between a third sample feature of the preset traffic running state and the third feature to be identified according to a second similarity algorithm aiming at each preset traffic running state, wherein the second similarity algorithm comprises at least one of a bulldozing distance algorithm and a Markov distance algorithm; And determining the corresponding preset traffic running state with the maximum similarity as the traffic running state of the target lane.
  8. 8. The method of claim 7, wherein the deriving a third sample feature representing a probability distribution of vehicle travel speed in the third sample lane during the third sample green light based on the third sample passing data comprises: Determining a driving speed of each vehicle in the third sample lane during the third sample green light based on the third sample passing data as a sample driving speed; determining a preset running speed interval to which each sample running speed belongs, wherein the preset running speed intervals are adjacent and have the same length; calculating the ratio of the sample running speed belonging to each preset running speed interval to all the sample running speeds aiming at each preset running speed interval to obtain a third sample characteristic; the obtaining, based on the target passing data, a third feature to be identified representing a probability distribution of a vehicle running speed in the target lane during the target green light includes: determining the running speed of each vehicle in the target lane during the target green light based on the target passing data as the running speed to be processed; Determining a preset running speed interval to which each running speed to be processed belongs; and calculating the ratio of the to-be-processed running speed belonging to each preset running speed interval in all to-be-processed running speeds aiming at each preset running speed interval to obtain a third to-be-identified characteristic.
  9. 9. The method according to claim 1, wherein the method further comprises: if the scene type of the target intersection in the time period to be controlled is a queuing balance type, adopting a control strategy of a first type of traffic signals at the target intersection, wherein the control strategy of the first type of traffic signals is used for shortening the queuing length of vehicles at the target intersection; if the scene type of the target intersection in the time period to be controlled is traffic capacity type, adopting a control strategy of a second type of traffic signals at the target intersection, wherein the control strategy of the second type of traffic signals is used for improving the traffic capacity of the target intersection; And if the scene type of the target intersection in the time period to be controlled is an efficiency improvement type, adopting a control strategy of a third type of traffic signals at the target intersection, wherein the control strategy of the third type of traffic signals is used for reducing the delay time of the vehicle at the target intersection.
  10. 10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-9.

Description

Intersection scene recognition method and computer readable storage medium Technical Field The present application relates to the field of traffic signal control technologies, and in particular, to a method for identifying an intersection scene and a computer readable storage medium. Background In the technical field of traffic signal control, the traffic state of the same intersection may be different in different time periods of the day under the influence of the human living rules. For example, in urban roads, the number of vehicles traveling at each intersection during an early peak period (e.g., 7-9 points) is typically greater than the number of vehicles traveling at each intersection during other periods (e.g., 9-11 points). However, if the same traffic signal is adopted for intersections in different traffic states, there may be a situation that the traffic signal is not matched with the traffic state, so that the traffic jam is caused by too short green time, or the problem that the empty space exists in too long green time. Therefore, there is a need for a method for identifying a scene of an intersection, which identifies the scene of the intersection according to the passing data of the intersection, and then, can adopt a proper control strategy of traffic signals at the intersection according to the scene of the intersection. Disclosure of Invention An object of an embodiment of the present application is to provide a method for identifying a scene of an intersection and a computer readable storage medium, so as to identify the scene of the intersection according to the driving data of the intersection. The specific technical scheme is as follows: in a first aspect of the embodiment of the present application, there is first provided a method for identifying an intersection scene, where the method includes: Aiming at a target lane of each direction of a target intersection, acquiring target passing data of the target lane in a target green light period in a time period to be controlled; Based on the target passing data, obtaining a first feature to be identified which represents the probability distribution of the headway in the target lane in the target green light period, a second feature to be identified which represents the number change condition of the target lane passing vehicles in each statistical time period in the target green light period, and a third feature to be identified which represents the probability distribution of the vehicle running speed in the target lane in the target green light period; Determining a traffic flow dispersion state of the target lane based on the first feature to be identified, determining a traffic demand state of the target lane based on the second feature to be identified, and determining a traffic running state of the target lane based on the third feature to be identified; and determining the scene type of the target intersection in the time period to be controlled according to the traffic flow dispersion state, the traffic demand state and the traffic running state of the target lane in each direction. In some embodiments, in the period to be controlled, the number of target lane-passing vehicles in each lane of the road section in any direction at the target intersection is the largest. In some embodiments, before determining the traffic flow dispersion state of the target lane based on the first feature to be identified, the method further comprises, for each preset traffic flow dispersion state, acquiring first sample passing data of a first sample lane during a first sample green light when the first sample lane is in the preset traffic flow dispersion state; The method comprises the steps of calculating the similarity between the first sample feature of each preset traffic flow dispersion state and the first feature to be identified according to a first similarity algorithm for each preset traffic flow dispersion state, wherein the first similarity algorithm comprises at least one of a bulldozing distance algorithm and a Markov distance algorithm, and determining the preset traffic flow dispersion state with the maximum corresponding similarity as the traffic flow dispersion state of the target lane. In some embodiments, the obtaining a first sample feature representing a probability distribution of headway in the first sample lane during the first sample green light based on the first sample passing data comprises determining each headway at a specified position in the first sample lane during the first sample green light based on the first sample passing data as a sample headway, determining a preset headway interval to which each sample headway belongs, wherein each preset headway interval is adjacent and the length of each preset headway interval is the same, calculating the ratio of the sample headway belonging to the preset headway interval in all sample headways for each preset headway interval, and obtaining the first sample feature;