Traffic incident warning method, device, equipment and storage medium

Abstract

The present application discloses a traffic event early warning method, device, equipment and storage medium, the method includes: determining the characteristic parameters of the target traffic event according to a preset multi-dimensional portrait library; obtaining the monitoring data of each characteristic parameter of the target traffic event monitored in real time; obtaining the early warning model of the target traffic event, and calculating the early warning threshold boundaries of different early warning levels according to the early warning model; when the monitoring data of any characteristic parameter monitored in real time exceeds the early warning threshold boundary of the lowest early warning level, issuing an early warning prompt of the corresponding early warning level. According to the traffic event early warning method, by using a preset multi-dimensional portrait library, matching the multi-dimensional portrait library with the target traffic event, determining the characteristic parameters corresponding to the target traffic event, quickly obtaining and determining whether the monitoring data exceeds the early warning threshold boundary calculated by the early warning model, and issuing corresponding early warning prompts, a scientific basis is provided for road classification management and control, and the level of road traffic safety management and control is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a traffic incident early warning method, device, equipment and storage medium. Background Art

[0002] The high incidence of traffic incidents seriously threatens the safety of road traffic and restricts the rapid development of highways. Preventing and reducing traffic incidents and improving road traffic safety are major issues facing traffic management departments. Traffic incident warning is to give early warnings of dangers and disasters based on historical laws or observed precursor information before dangerous disasters occur. Traffic management departments prepare in advance based on the warnings and formulate response measures to reduce the occurrence of traffic incidents and their consequences.

[0003] In the existing technology, whether a road is congested is generally predicted through neural network models, machine learning models and other methods, or dangerous sections are predicted only based on map data. There is no traffic incident warning method for different scenarios and monitoring objects, and it is impossible to provide a scientific basis for the overall management and control of road traffic safety. Summary of the invention

[0004] The embodiments of the present application provide a method, device, equipment and storage medium for warning of traffic incidents. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to determine the key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description below.

[0005] In a first aspect, an embodiment of the present application provides a method for early warning of a traffic incident, comprising:

[0006] Determine characteristic parameters of target traffic events based on a preset multi-dimensional portrait library;

[0007] Acquire monitoring data of various characteristic parameters of target traffic events monitored in real time;

[0008] Obtain the warning model of the target traffic event, and calculate the warning threshold boundaries of different warning levels according to the warning model;

[0009] When the monitoring data of any characteristic parameter monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt of the corresponding warning level will be issued.

[0010] In one embodiment, determining characteristic parameters of a target traffic event according to a preset multi-dimensional portrait library includes:

[0011] Searching at least one type of image library matching the target traffic event from the multi-dimensional image library;

[0012] The characteristic parameters corresponding to each type of image library are used to determine the characteristic parameters of the target traffic event.

[0013] In one embodiment, the training process of the warning model of the target traffic event is as follows:

[0014] Acquire multiple historical traffic event samples corresponding to the target traffic event type, wherein each historical traffic event sample includes an event level label and characteristic parameters affecting the historical traffic event;

[0015] According to the event level, the characteristic parameters of traffic events of different levels are classified and calculated to obtain the warning thresholds of the characteristic parameters of traffic events of different levels. The warning threshold is the threshold boundary composed of multiple characteristic parameters.

[0016] In one embodiment, the construction of a preset multi-dimensional portrait library includes:

[0017] Acquire multi-source heterogeneous data, including event data, location data, road condition data, toll data, weather data, freight data, map data, and video data;

[0018] Extract characteristic parameters from multi-source heterogeneous data, including spatial characteristics, traffic characteristics, vehicle type characteristics, speed characteristics, meteorological characteristics, event aggregation characteristics, event statistical characteristics, OD characteristics, static attribute characteristics, travel characteristics, route characteristics, safety characteristics, congestion characteristics and efficiency characteristics;

[0019] Obtain event portraits based on spatial features, traffic features, vehicle type features, speed features, meteorological features, event aggregation features, and event statistical features;

[0020] Obtain a road section profile based on spatial features, traffic features, vehicle type features, speed features, OD features, meteorological features, and event statistical features;

[0021] Obtain vehicle portraits based on static attribute features, travel features, route features, OD features, and safety features;

[0022] Get a toll station portrait based on spatial characteristics, flow characteristics, congestion characteristics, vehicle type characteristics, and efficiency characteristics;

[0023] Obtain personnel portraits based on static attribute characteristics and efficiency characteristics.

[0024] In one embodiment, it further includes:

[0025] Normalize the characteristic parameters corresponding to the obtained image of each object to obtain a normalized parameter value;

[0026] The normalized parameter values ​​are displayed in the radar chart corresponding to the image of the object, and the fusion features of the object are represented by the color of the center point of the radar chart.

[0027] In one embodiment, when the monitoring data of any characteristic parameter monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt corresponding to the warning level is issued, including:

[0028] When the monitoring data of the characteristic parameters in real-time monitoring exceeds any warning threshold in the threshold boundary of the lowest warning level;

[0029] An early warning prompt of the corresponding level is issued according to the level corresponding to the early warning threshold area where the monitoring data is located.

[0030] In one embodiment, the target traffic event type includes at least one of monitoring accident-prone sections, monitoring traffic flow-prone sections, monitoring safety levels of construction sections, monitoring toll station traffic efficiency, and monitoring important vehicles.

[0031] In a second aspect, an embodiment of the present application provides a traffic incident warning device, including:

[0032] A determination module, used to determine characteristic parameters of a target traffic event based on a preset multi-dimensional portrait library;

[0033] An acquisition module, used to acquire monitoring data of various characteristic parameters of a target traffic event monitored in real time;

[0034] A calculation module is used to obtain a warning model of a target traffic event and calculate warning threshold boundaries of different warning levels according to the warning model;

[0035] The early warning module is used to issue an early warning prompt of the corresponding warning level when the monitoring data of any characteristic parameter monitored in real time exceeds the warning threshold boundary of the lowest warning level.

[0036] In a third aspect, an embodiment of the present application provides a traffic event warning device, including a processor and a memory storing program instructions, wherein the processor is configured to execute the traffic event warning method provided in the above embodiment when executing the program instructions.

[0037] In a fourth aspect, an embodiment of the present application provides a computer-readable medium having computer-readable instructions stored thereon, and the computer-readable instructions are executed by a processor to implement a traffic event warning method provided in the above embodiment.

[0038] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0039] According to a traffic event warning method provided by an embodiment of the present application, the characteristic parameters of the target traffic event are determined according to a preset multidimensional portrait library, the monitoring data of each characteristic parameter of the target traffic event monitored in real time is obtained, the warning model of the target traffic event is obtained, and the warning threshold boundaries of different warning levels are calculated according to the warning model. When the monitoring data of any characteristic parameter monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt corresponding to the warning level is issued. By using a preset multidimensional portrait library, the multidimensional portrait library is matched with the target traffic event, and the characteristic parameters of the target traffic event are quickly obtained from the multidimensional portrait library. According to the real-time monitoring data of the characteristic parameters, it is determined whether the monitoring data exceeds the warning threshold boundary calculated by the warning model, and a corresponding warning prompt is issued. Provide a scientific basis for road classification management and control, and improve road traffic safety.

[0040] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0042] Figure 1 is a flow chart of a method for early warning of a traffic incident according to an exemplary embodiment;

[0043] Figure 2 is a schematic diagram of a traffic incident warning method according to an exemplary embodiment;

[0044] Figure 3 is a schematic diagram of a traffic incident warning method according to an exemplary embodiment;

[0045] Figure 4 is a schematic diagram of a method for determining a warning threshold according to an exemplary embodiment;

[0046] Figure 5 is a schematic diagram showing a method for determining a warning threshold boundary according to an exemplary embodiment;

[0047] Figure 6 is a schematic diagram of a portrait radar chart according to an exemplary embodiment;

[0048] Figure 7 is a schematic structural diagram of a warning device for traffic events according to an exemplary embodiment;

[0049] Figure 8 is a schematic structural diagram of a warning device for traffic incidents according to an exemplary embodiment;

[0050] Fig. 9 It is a schematic diagram of a computer storage medium according to an exemplary embodiment. DETAILED DESCRIPTION

[0051] The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them.

[0052] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are only examples of systems and methods consistent with some aspects of the present invention as detailed in the attached claims.

[0054] An embodiment of the present application provides a method for warning of traffic events. First, characteristic parameters of a target traffic event are determined based on a preset multi-dimensional portrait library, and monitoring data of each characteristic parameter of the target traffic event is acquired in real time, and a warning model of the target traffic event is acquired. Warning threshold boundaries of different warning levels are calculated according to the warning model. When the monitoring data of any characteristic parameter monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt of the corresponding warning level is issued.

[0055] The following will describe in detail the traffic incident warning method provided by the embodiment of the present application in conjunction with the accompanying drawings. Figure 1 , the method specifically includes the following steps.

[0056] S101 determines characteristic parameters of a target traffic event according to a preset multi-dimensional portrait library.

[0057] In one embodiment, determining characteristic parameters of a target traffic event based on a preset multidimensional portrait library includes: searching the multidimensional portrait library for at least one type of portrait library that matches the target traffic event; and determining characteristic parameters of the target traffic event using characteristic parameters corresponding to each type of portrait library.

[0058] It can be understood that in one embodiment of the present application, the multidimensional portrait library is matched with the target traffic event according to the first preset rule. According to the first preset rule, a portrait library matching the target traffic event is searched from the multidimensional portrait library. The matched portrait library can be one category, or two categories or more categories. The characteristic parameters of the matched portrait library are determined as the characteristic parameters of the target traffic event.

[0059] Specifically, a multidimensional portrait library is first constructed based on multi-source heterogeneous data.

[0060] Acquire multi-source heterogeneous data, for example, event data, location data, road condition data, toll data, video data, heavy-load freight data, weather data, map data, etc.

[0061] Furthermore, feature parameters of different objects are extracted based on the acquired multi-source heterogeneous data. This includes extracting spatial features, traffic features, vehicle type features, speed features, meteorological features, event aggregation features, event statistical features, OD features, static attribute features, travel features, route features, safety features, congestion features, and efficiency features from multi-source heterogeneous data. Among them, static attribute features include user name, user ID, gender, department ID, work number, user name, etc., and efficiency features include completion rate, punctuality rate, number of complaints, average task completion time, etc.

[0062] Furthermore, a multidimensional portrait library is formed according to the extracted characteristic parameters. In a possible implementation method, an event portrait is obtained according to spatial characteristics, flow characteristics, vehicle type characteristics, speed characteristics, meteorological characteristics, event aggregation characteristics, and event statistical characteristics; a road section portrait is obtained according to spatial characteristics, flow characteristics, vehicle type characteristics, speed characteristics, OD characteristics, meteorological characteristics, and event statistical characteristics; a vehicle portrait is obtained according to static attribute characteristics, travel characteristics, route characteristics, OD characteristics, and safety characteristics; a toll station portrait is obtained according to spatial characteristics, flow characteristics, congestion characteristics, vehicle type characteristics, and efficiency characteristics; and a personnel portrait is obtained according to static attribute characteristics and efficiency characteristics.

[0063] According to this step, a systematic and comprehensive method for establishing a multi-dimensional portrait library can be provided. According to the target traffic event, a multi-dimensional portrait matching the target traffic event is selected from the multi-dimensional portrait library to obtain characteristic parameters of each type of traffic event. Common traffic event warning scenarios include monitoring and warning of accident-prone sections, monitoring and warning of sections with high traffic volume, monitoring and warning of safety levels of construction sections, monitoring and warning of toll station traffic efficiency, and monitoring and warning of important vehicles.

[0064] For example, based on the historical accident data of a road section, we can correlate and analyze multi-dimensional data such as traffic flow, speed, weather, and geographical and spatial characteristics of the road section during the accident period, establish a safety risk warning model for accident-prone sections, and implement safety level assessment and warning for accident-prone sections.

[0065] For example, based on the historical traffic flow event data of the road section, the multi-dimensional data such as traffic flow, speed, weather, and geographic spatial characteristics during the traffic flow period are correlated and analyzed to establish a safety risk warning model for the road section with the most traffic flow. The safety level of the road section with the most traffic flow is evaluated, and the safety level of the road section with the most traffic flow is monitored and warned. Visual warning prompts are provided to the dispatching personnel at the interface of the monitoring system, providing a scientific basis for the selection and implementation of subsequent disposal decisions.

[0066] For example, taking the time interval between vehicles passing through the toll station's lanes as the core feature, we correlate and analyze data such as toll station traffic, congestion, and spatial characteristics, establish a multi-dimensional evaluation system for toll station lane-level traffic efficiency, and realize monitoring and early warning of toll station traffic efficiency.

[0067] For example, based on the gantry charging records and with license plates as search criteria, key security vehicles and fleets can be tracked and monitored, supporting real-time query of vehicle locations and traffic conditions.

[0068] According to the first preset rule, from the portraits corresponding to each target traffic event in the multi-dimensional portrait library, the feature parameters corresponding to the portraits are determined as the feature parameters corresponding to each target traffic event, so that the feature parameters corresponding to each traffic event can be obtained.

[0069] In a specific implementation, at least one type of portrait corresponding to the traffic event type is searched and determined from a multi-dimensional portrait library, and characteristic parameters corresponding to all types of portraits found and determined are determined as characteristic parameters corresponding to the target traffic event.

[0070] In a specific implementation, each type of portrait is stored in a portrait sub-library. When determining the characteristic parameters corresponding to the target traffic event, at least one type of portrait sub-library is determined, and the characteristic parameters corresponding to all types of portraits are determined as the characteristic parameters corresponding to the target traffic event. It can be understood that during training, multiple samples in the portrait sub-library corresponding to the target type event are found from the portrait library as training samples.

[0071] Specifically, the characteristic parameters in the road section portrait and event portrait are used as the characteristic parameters of monitoring events on accident-prone sections; the characteristic parameters in the road section portrait and event portrait are used as the characteristic parameters of monitoring events on sections with high traffic volume; the characteristic parameters in the road section portrait and event portrait are used as the characteristic parameters of monitoring events on the safety level of construction sections; the characteristic parameters in the toll station portrait and personnel portrait are used as the characteristic parameters of monitoring events on toll station traffic efficiency; the characteristic parameters in the vehicle portrait are used as the characteristic parameters of monitoring events on important vehicles.

[0072] According to this step, characteristic parameters corresponding to each traffic event can be obtained.

[0073] In one embodiment, the method further includes: normalizing the characteristic parameters corresponding to the obtained portrait of each object to obtain the normalized parameter values; displaying the normalized parameter values ​​in the radar chart corresponding to the portrait of the object, and representing the fusion characteristics of the object by the color of the center point of the radar chart. The object refers to an event, a road section, a vehicle, a toll station, a person, etc.

[0074] In order to intuitively display the values ​​and influence of different characteristic parameters of the analysis object, a radar chart is used to display the characteristic values ​​of the analysis object. Since the value type and value range of each dimension vary greatly, the unprocessed characteristic values ​​cannot be directly represented in a radar chart. For example, the spatial feature can be "number of bridges" or "downhill section", the flow value can be "2000veh / h", the meteorological feature can be "heavy rain", and the event statistical feature can be "accident-prone section". In order to simultaneously display the contribution of each dimension in the radar chart, the characteristic parameters are normalized to values ​​between 0 and 1. In addition, the color of the center point of the radar chart represents the fusion feature of the object.

[0075] The following uses event portrait as an example to explain how to determine the characteristic value R of each dimension of the radar chart. Figure 6 As shown, the dimensions of the event portrait radar chart include spatial features, traffic features, speed features, vehicle type features, meteorological features, event statistical features, and event aggregation features (center point).

[0076] (1) Spatial characteristics

[0077] The spatial characteristics of the road section where the event is located are represented by the number of bridges / tunnels / interchanges / toll stations / service areas and the slope (uphill / flat / downhill). The spatial characteristics of the radar chart are R = (number of bridges R1 + number of tunnels R2 + number of interchanges R3 + number of toll stations R4 + number of service areas R5) * 0.3 + slope score R6. The slope score R6 is determined by the following rules: if the slope = "downhill", then R6 = 0.3; otherwise, R6 = 0.

[0078] When (number of bridges R1 + number of tunnels R2 + number of interchanges R3 + number of toll stations R4 + number of service areas R5)*0.3+slope score R6>1, R=1.

[0079] (2) Traffic characteristics

[0080] Due to the differences in road capacity of different road grades, the road service levels corresponding to the flow values ​​are quite different. In order to eliminate the impact of road grade on flow characteristics, saturation is used to represent the flow characteristic value (saturation is equal to flow divided by capacity). When saturation ≤ 1, the flow characteristic value of the radar chart is R = saturation; when saturation > 1, R = 1.

[0081] The traffic flow characteristics are described according to the saturation. For example, when the saturation is ≤0.35, it is described as "the traffic flow is in a free flow state"; when 0.35<saturation≤0.55, it is described as "the traffic flow is basically in a free flow state"; when 0.55<saturation≤0.75, it is described as "the traffic flow is in the upper half of a stable flow"; when 0.75<saturation≤0.9, it is described as "the traffic flow is in the lower half of a stable flow"; when 0.9<saturation≤1, it is described as "the traffic flow is in a congested flow state"; when saturation>1, it is described as "the traffic flow is in a congested state".

[0082] (3) Vehicle features

[0083] Since the truck flow and truck proportion have a great impact on the sections with frequent incidents, the truck proportion is used to represent the vehicle type characteristics. Taking Liaoning Province as an example, the analysis found that the proportion of trucks in Liaoning Province was at a relatively high level when it was 80%. In order to make the vehicle type characteristic value of the radar chart take the value of 1 when the proportion of trucks exceeds 80%, the proportion of trucks is multiplied by 1.25. Therefore, the vehicle type characteristic value of the radar chart R = truck proportion * 1.25; when the proportion of trucks * 1.25 is greater than 1, R = 1.

[0084] The vehicle type characteristics of the road section are described according to the proportion of trucks: when the proportion of trucks is ≤0.4, it is described as "there are more passenger cars on this section"; when 0.4<the proportion of trucks is ≤0.6, it is described as "the passenger and freight flow on this section is balanced"; when the proportion of trucks is >0.6, it is described as "there are more trucks on this section".

[0085] (4) Speed ​​characteristics

[0086] Since speed fluctuation has a significant impact on traffic incidents, the speed fluctuation index of a road section is used to represent the fluctuation characteristics of speed. The speed fluctuation index (SFI) is defined as the normalized result (between 0 and 1) of the sum of the absolute values ​​of acceleration of all vehicles on the road section. i It represents the speed fluctuation index of road section i, and the calculation formula is as follows:

[0087]

[0088] In the formula, represents the sum of the absolute values ​​of acceleration of all vehicles on section i in the kth time period, and They represent the maximum and minimum values ​​of the sum of the absolute values ​​of acceleration of all vehicles on multiple road sections obtained based on historical data statistics:

[0089]

[0090]

[0091]

[0092] where a i,k,j represents the acceleration of the j-th vehicle on section i in the k-th time period, m represents the number of vehicles on section i, n represents the number of sections, and K represents the number of time periods.

[0093] According to the above formula, the calculation formula for the speed fluctuation index is as follows:

[0094]

[0095] The value of the speed characteristic of the radar chart R = the speed fluctuation index of the section. When R ≤ 0.2, it is described as "stable speed"; when 0.2 < R ≤ 0.4, it is described as "small speed fluctuation"; when 0.4 < R ≤ 0.6, it is described as "large speed fluctuation"; when 0.6 < R ≤ 0.8, it is described as "very large speed fluctuation"; when R > 0.8, it is described as "severe speed fluctuation".

[0096] (5) Meteorological characteristics

[0097] The meteorological characteristics are represented by the meteorological grade. When the meteorological grade is level 1, the value of the meteorological characteristic of the radar chart R = 0 (normal meteorological conditions); when the meteorological grade is level 2, R = 0.2 (general meteorological conditions); when the meteorological grade is level 3, R = 0.4 (poor meteorological conditions); when the meteorological grade is level 4, R = 0.7 (severe meteorological conditions); when the meteorological grade is 5, R = 1 (extremely severe meteorological conditions).

[0098] According to different meteorological grades, the meteorological characteristics are described as "the meteorological conditions of this section are normal / general / poor / severe / extremely severe".

[0099] (6) Event statistical characteristics

[0100] The event statistical characteristics are the types of frequently-occurring events (such as sections with frequent accidents, sections with frequent congestion, sections with high traffic flow), frequently-occurring time periods, and the frequency of event occurrences statistically obtained based on historical data. The value of the event statistical characteristic of the radar chart is based on the frequency of event occurrences. For example, when the event type is an accident: when the frequency of event occurrences ≤ θ, R = the frequency of event occurrences / θ; when the frequency of event occurrences > θ, R = 1. When the event type is congestion or high traffic flow: R = the frequency of event occurrences / the total number of days. θ represents the threshold of the frequency of accident occurrences. For example, for the expressways in Liaoning Province, according to the statistical analysis results of the frequency of traffic events in 2020, the value of θ can be taken as 60.

[0101] Based on the R value, the event statistical features can be described as follows: When R = 0, it describes that "this road section is a normal road section"; when 0 < R ≤ 0.25, it describes that "this road section is an accident / congestion / high traffic volume low-incidence road section"; when 0.25 < R ≤ 0.5, it describes that "this road section is an accident / congestion / high traffic volume easy-incidence road section"; when 0.5 < R ≤ 0.75, it describes that "this road section is an accident / congestion / high traffic volume frequent-incidence road section"; when 0.75 < R ≤ 1, it describes that "this road section is an accident / congestion / high traffic volume high-incidence road section".

[0102] (7) Event aggregation feature (center point)

[0103] The event aggregation feature describes the detailed information of the event, including event ID, event type, event description, event level, event occurrence time period, number of closed lanes, duration, influence distance, number of accident vehicles, number of injured people, road damage amount, loss description, accident license plate ID, accident vehicle type, etc.

[0104] The event level is used to represent the comprehensive severity of the event, that is, the color of the center point of the radar chart is set according to the event level. For example Figure 6 As shown, when the event level is a minor accident, R = 0.1 (the center line point is green); when the event level is a general accident, R = 0.4 (the center point is yellow); when the event level is a major accident, R = 0.8 (the center point is orange); when the event level is an extremely serious accident, R = 1 (the center point is red).

[0105] S102 Obtain the monitoring data of each characteristic parameter of the target traffic event under real-time monitoring. According to the type of characteristic parameters of the target traffic event determined by S101, determine the data interface of the characteristic parameters, and obtain the monitoring data of each characteristic parameter of the target traffic event under real-time monitoring. For example, if the target traffic event is an accident-prone road section, then obtain the monitoring data of the characteristic parameters corresponding to the traffic event of the accident-prone road section.

[0106] S103 Obtain the warning model of the target traffic event, and calculate the warning threshold boundaries of different warning levels according to the warning model.

[0107] In terms of setting the warning threshold of traffic events, an inappropriate warning time will reduce the trust of drivers and highway management personnel in the warning system and affect the warning effect. Therefore, it is necessary to set a reasonable warning threshold.

[0108] Specifically, obtain multiple historical traffic event samples corresponding to the target traffic event type, where each historical traffic event sample includes an event level label and the characteristic parameters affecting the historical traffic event; use a classification algorithm to perform classification calculations on the historical traffic event samples to obtain the boundary thresholds of the characteristic parameters of different warning levels of the target type traffic event.

[0109] In one possible implementation, historical data of each type of traffic event can be obtained; based on the historical data, traffic events are divided into traffic events of different types and levels; characteristic parameters of traffic events of different types and levels are clustered to obtain warning thresholds of characteristic parameters of traffic events of different types and levels, and the warning threshold is a threshold boundary composed of multiple characteristic parameters.

[0110] Figure 4 is a schematic diagram showing a method for determining a warning threshold according to an exemplary embodiment. Figure 4 As shown, first, historical data of each type of traffic event is obtained, for example, data of each type of traffic event in the past three months is obtained. Then, characteristic parameters of traffic events of different types and levels are obtained according to the multi-dimensional portraits corresponding to each traffic event.

[0111] Finally, the classification algorithm is used to extract the characteristic parameter threshold and determine the threshold boundary of traffic incident warning. Figure 5 As shown, traffic events are classified according to type and level, characteristic parameters of each type and level of traffic events are clustered, and threshold boundaries of characteristic parameters corresponding to each type and level of traffic events are extracted. The threshold boundaries are composed of warning thresholds of multiple characteristic parameters.

[0112] Among them, each type and level of event needs to be trained, each type and level of event has a warning threshold, traffic events are classified according to type and level, the characteristic parameters of each type and level of traffic events are clustered, and the threshold boundaries of the characteristic parameters corresponding to each type and level of traffic events are extracted.

[0113] S104: When the monitoring data of any one of the characteristic parameters monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt corresponding to the warning level is issued.

[0114] In one embodiment, when the monitoring data of the characteristic parameter in real-time monitoring exceeds any warning threshold in the threshold boundary of the lowest warning level, a warning prompt of the corresponding level is issued according to the level corresponding to the warning threshold area where the monitoring data is located.

[0115] Taking the monitoring and early warning of sections with heavy traffic as an example, the early warning levels can be divided into level 1, level 2, and level 3 from low to high. Level 1 corresponds to heavy traffic, level 2 corresponds to heavy traffic, and level 3 corresponds to extremely heavy traffic. The early warning levels are divided into accident event levels. When any type of monitoring data of the characteristic parameters of the target traffic event section with heavy traffic in real-time monitoring exceeds the level 1 early warning range, an early warning is issued, and an early warning prompt of the corresponding level is issued according to the early warning area where the monitoring data is located.

[0116] By providing monitoring and early warning for accident-prone sections, monitoring and early warning for sections with high traffic volume, monitoring and early warning for safety levels of construction sections, monitoring and early warning for traffic efficiency at toll stations, and monitoring and early warning for important vehicles, visual early warning prompts are provided to dispatch personnel at the monitoring system interface, providing a scientific basis for the selection and implementation of subsequent disposal decisions.

[0117] An embodiment of the present application provides an early warning method for traffic events. The method determines characteristic parameters of a target traffic event according to a preset multidimensional portrait library, obtains monitoring data of each characteristic parameter of the target traffic event monitored in real time, obtains an early warning model for the target traffic event, calculates early warning threshold boundaries of different early warning levels according to the early warning model, and issues an early warning prompt of the corresponding early warning level when the monitoring data of any characteristic parameter monitored in real time exceeds the early warning threshold boundary of the lowest early warning level. The preset multidimensional portrait library is adopted to match the multidimensional portrait library with the target traffic event, and the characteristic parameters of the target traffic event are quickly obtained from the multidimensional portrait library. The real-time monitoring data of the characteristic parameters is used to determine whether the monitoring data exceeds the early warning threshold boundary calculated by the early warning model, and a corresponding early warning prompt is issued. The preset multidimensional portrait library is adopted to match the multidimensional portrait library with the target traffic event, and the characteristic parameters of the target traffic event are quickly obtained from the multidimensional portrait library. The real-time monitoring data of the characteristic parameters is used to determine whether the monitoring data exceeds the early warning threshold boundary calculated by the early warning model, and a corresponding early warning prompt is issued.

[0118] In order to facilitate understanding of the early warning method for traffic incidents provided in the embodiment of the present application, the following is a Figure 2 For explanation. Figure 2 As shown, the method includes the following steps.

[0119] First, obtain multi-source heterogeneous data, which includes event data, Internet location data, road condition data, toll data, video data, heavy-load freight data, meteorological data, map data, etc.

[0120] Furthermore, feature parameters are extracted from multi-source heterogeneous data, and multidimensional portraits are constructed based on the extracted feature parameters, including road section portraits, vehicle portraits, event portraits, toll plaza portraits, lane portraits, organization portraits, personnel portraits, etc.

[0121] Furthermore, according to the characteristic parameters corresponding to traffic events of different types and levels, the warning threshold is calculated, the characteristic parameters of each type and level of traffic events are clustered, and the threshold boundary of the characteristic parameters corresponding to each type and level of traffic events is extracted. According to the determined warning thresholds of different types of events, monitoring and warning of accident-prone sections, monitoring and warning of sections with high traffic volume, monitoring and warning of safety levels of construction sections, monitoring and warning of toll station traffic efficiency, and monitoring and warning of important vehicles are carried out.

[0122] In order to facilitate understanding of the early warning method for traffic incidents provided in the embodiment of the present application, the following is a Figure 3 For explanation. Figure 3 As shown, the method includes the following steps.

[0123] First, obtain multi-source heterogeneous data, which includes event data, Internet location data, road condition data, toll data, video data, heavy-load freight data, meteorological data, map data, etc.

[0124] Furthermore, feature parameters are extracted from multi-source heterogeneous data, and a multidimensional portrait library is constructed based on the extracted feature parameters, including road portraits, vehicle portraits, event portraits, etc. Among them, road portraits include spatial features, flow features, vehicle model features, speed features, OD features, meteorological features, and event statistical features; vehicle portraits include static attribute features, travel features, route features, OD features, safety features, etc.; event portraits include spatial features, flow features, vehicle model features, speed features, meteorological features, event aggregation features, and event statistical features.

[0125] Furthermore, according to the target traffic event type, at least one type of portrait corresponding to it is searched and determined from the multi-dimensional portrait library, and the characteristic parameters corresponding to all types of portraits found and determined are determined as the characteristic parameters corresponding to the target traffic event.

[0126] Furthermore, monitoring data of each characteristic parameter of the target traffic event monitored in real time is obtained.

[0127] Furthermore, the warning model of the target traffic event is obtained, and the warning threshold boundaries of different warning levels are calculated according to the warning model. When training the warning model, multiple samples in the image sub-library corresponding to the target type event are found from the image library as training samples. Training is performed based on the training samples, and the characteristic parameters of traffic events of different types and levels are clustered to obtain the warning thresholds of the characteristic parameters of traffic events of different types and levels. The warning threshold is the threshold boundary composed of multiple characteristic parameters.

[0128] Finally, when the monitoring data of any characteristic parameter monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt of the corresponding warning level is issued.

[0129] According to the traffic incident warning method provided in the embodiment of the present application, a traffic incident warning system can be established for the warning needs of different scenes and monitoring objects such as accident-prone sections, sections with high traffic volume, construction sections, toll booths, and important vehicles; it can extract characteristic parameters that describe different objects such as roads, vehicles, and events, and establish a multidimensional portrait library; it can extract characteristic parameters that reflect the specified traffic incident type, and determine the traffic incident warning threshold. It provides a scientific basis for road classification management and control, and improves road traffic safety.

[0130] The present application also provides a traffic incident warning device, which is used to execute the traffic incident warning method of the above embodiment, such as Figure 7 As shown, the device comprises:

[0131] A determination module 701 is used to determine characteristic parameters of a target traffic event according to a preset multi-dimensional portrait library;

[0132] An acquisition module 702 is used to acquire monitoring data of various characteristic parameters of a target traffic event monitored in real time;

[0133] The calculation module 703 is used to obtain the warning model of the target traffic event and calculate the warning threshold boundaries of different warning levels according to the warning model;

[0134] The early warning module 704 is used to issue an early warning prompt of the corresponding early warning level when the monitoring data of any characteristic parameter monitored in real time exceeds the early warning threshold boundary of the lowest early warning level.

[0135] It should be noted that the warning device for traffic incidents provided in the above embodiment only uses the division of the above functional modules as an example when executing the warning method for traffic incidents. In actual applications, the above functional distribution can be completed by different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the warning device for traffic incidents provided in the above embodiment and the warning method embodiment for traffic incidents belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0136] The embodiment of the present application also provides an electronic device corresponding to the warning method for traffic events provided in the above embodiment, so as to execute the above warning method for traffic events.

[0137] Please refer to Figure 8 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 8As shown, the electronic device includes: a processor 800, a memory 801, a bus 802 and a communication interface 803, and the processor 800, the communication interface 803 and the memory 801 are connected via the bus 802; the memory 801 stores a computer program that can be run on the processor 800, and when the processor 800 runs the computer program, it executes the warning method for traffic events provided in any of the aforementioned embodiments of the present application.

[0138] The memory 801 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 803 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0139] The bus 802 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 801 is used to store programs, and the processor 800 executes the programs after receiving the execution instructions. The traffic incident warning method disclosed in any of the embodiments of the present application may be applied to the processor 800, or implemented by the processor 800.

[0140] The processor 800 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 800. The above processor 800 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a readily available programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 801, and the processor 800 reads the information in the memory 801 and completes the steps of the above method in combination with its hardware.

[0141] The electronic device provided in the embodiment of the present application and the traffic incident warning method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0142] The present application also provides a computer-readable storage medium corresponding to the traffic incident warning method provided in the above embodiment. Fig. 9 The computer-readable storage medium shown is a CD 900 on which a computer program (ie, a program product) is stored. When the computer program is executed by a processor, the early warning method for traffic events provided in any of the aforementioned embodiments will be executed.

[0143] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0144] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the traffic incident warning method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0145] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the attached claims.

Claims

1. A traffic incident warning method, characterized in that: include: Determine characteristic parameters of target traffic events based on a preset multi-dimensional portrait library; Acquire monitoring data of various characteristic parameters of target traffic events monitored in real time; Acquire the warning model of the target traffic event, and calculate the warning threshold boundaries of different warning levels according to the warning model; When the monitoring data of any of the characteristic parameters monitored in real time exceeds the warning threshold boundary of the lowest warning level, a warning prompt corresponding to the warning level is issued; The training process of the warning model of the target traffic event is as follows: obtaining multiple historical traffic event samples corresponding to the target traffic event type, wherein each historical traffic event sample includes an event level label and characteristic parameters affecting the historical traffic event; classifying and calculating the characteristic parameters of traffic events of different levels according to the event level, and obtaining warning thresholds of the characteristic parameters of traffic events of different levels, wherein the warning threshold is a threshold boundary composed of multiple characteristic parameters.

2. The method according to claim 1, characterized in that The characteristic parameters of the target traffic event determined based on the preset multi-dimensional image library include: Searching at least one type of image library matching the target traffic event from the multi-dimensional image library; The characteristic parameters corresponding to each type of image library are used to determine the characteristic parameters of the target traffic event.

3. The method according to claim 1, characterized in that The construction of the preset multi-dimensional portrait library includes: Acquiring multi-source heterogeneous data, wherein the multi-source heterogeneous data includes event data, location data, road condition data, toll data, weather data, freight data, map data, and video data; Extracting characteristic parameters from the multi-source heterogeneous data, the characteristic parameters including spatial characteristics, flow characteristics, vehicle type characteristics, speed characteristics, meteorological characteristics, event aggregation characteristics, event statistical characteristics, OD characteristics, static attribute characteristics, travel characteristics, route characteristics, safety characteristics, congestion characteristics and efficiency characteristics; Obtaining an event portrait according to the spatial characteristics, traffic characteristics, vehicle type characteristics, speed characteristics, meteorological characteristics, event aggregation characteristics, and event statistical characteristics; Obtaining a road section portrait according to the spatial characteristics, flow characteristics, vehicle type characteristics, speed characteristics, OD characteristics, meteorological characteristics, and event statistical characteristics; Obtaining a vehicle portrait according to the static attribute features, travel features, route features, OD features, and safety features; Obtain a toll station portrait based on the spatial characteristics, flow characteristics, congestion characteristics, vehicle type characteristics, and efficiency characteristics; A personnel portrait is obtained based on the static attribute characteristics and efficiency characteristics.

4. The method according to claim 3, characterized in that Also includes: Normalizing the characteristic parameters corresponding to the obtained image of each object to obtain normalized parameter values; The normalized parameter values ​​are displayed in a radar chart corresponding to the image of the object, and the fusion features of the object are represented by the color of the center point of the radar chart.

5. The method according to claim 1, characterized in that When the monitoring data of any one of the characteristic parameters monitored in real time exceeds the warning threshold boundary of the lowest warning level, issuing a warning prompt corresponding to the warning level includes: When the monitoring data of the characteristic parameter in real-time monitoring exceeds any one of the warning thresholds in the threshold boundaries of the lowest warning level; A warning prompt of a corresponding level is issued according to the level corresponding to the warning threshold area where the monitoring data is located.

6. The method according to claim 1, characterized in that The target traffic event types include at least one of monitoring accident-prone sections, monitoring traffic-prone sections with high traffic volume, monitoring safety levels of construction sections, monitoring traffic efficiency at toll stations, and monitoring important vehicles.

7. A traffic incident warning device, characterized in that: include: A determination module, used to determine characteristic parameters of a target traffic event based on a preset multi-dimensional portrait library; An acquisition module, used to acquire monitoring data of various characteristic parameters of a target traffic event monitored in real time; A calculation module is used to obtain the warning model of the target traffic event, and calculate the warning threshold boundaries of different warning levels according to the warning model; the training process of the warning model of the target traffic event is as follows: obtaining multiple historical traffic event samples corresponding to the target traffic event type, wherein each historical traffic event sample includes an event level label and characteristic parameters affecting the historical traffic event; classifying and calculating the characteristic parameters of traffic events of different levels according to the event level, and obtaining the warning thresholds of the characteristic parameters of traffic events of different levels, wherein the warning threshold is a threshold boundary composed of multiple characteristic parameters; The early warning module is used to issue an early warning prompt of the corresponding early warning level when the monitoring data of any one of the characteristic parameters monitored in real time exceeds the early warning threshold boundary of the lowest early warning level.

8. A traffic incident warning device, characterized in that: The system comprises a processor and a memory storing program instructions, wherein the processor is configured to execute the traffic event warning method according to any one of claims 1 to 6 when executing the program instructions.

9. A computer-readable medium, characterized in that Computer-readable instructions are stored thereon, and the computer-readable instructions are executed by a processor to implement a traffic event warning method as described in any one of claims 1 to 6.

Images