CN120088989B - Traffic road network multi-mode sensing abnormal event early warning method and system - Google Patents
Traffic road network multi-mode sensing abnormal event early warning method and system Download PDFInfo
- Publication number
- CN120088989B CN120088989B CN202510562109.2A CN202510562109A CN120088989B CN 120088989 B CN120088989 B CN 120088989B CN 202510562109 A CN202510562109 A CN 202510562109A CN 120088989 B CN120088989 B CN 120088989B
- Authority
- CN
- China
- Prior art keywords
- abnormal event
- early warning
- feature
- prototype
- cluster
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Probability & Statistics with Applications (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a traffic road network multi-mode sensing abnormal event early warning method and system, and relates to the technical field of data processing, wherein the method comprises the steps of obtaining a path accident frequent data set of a target traffic road network, constructing an abnormal event prototype set, obtaining vehicle state information of a target vehicle based on V2X communication, carrying out probability matching in combination with multi-mode sensing data, and determining a target abnormal event prototype; and calling the historical abnormal event triggering characteristic set to perform characteristic enhancement on the historical abnormal event triggering characteristic set, taking the enhanced abnormal event triggering characteristic as an early warning condition, and performing real-time abnormal event early warning on the running process of the target vehicle. The invention solves the technical problems of incomplete detection and early warning delay of traffic abnormal events caused by the fact that the prior art relies on a single data source and lacks of real-time dynamic response capability, and achieves the technical effects of comprehensive and real-time detection and dynamic early warning of traffic abnormal events through multi-mode sensing and V2X communication technologies.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a traffic road network multi-mode sensing abnormal event early warning method and system.
Background
In an intelligent traffic system, abnormal event detection and early warning are key links for guaranteeing road safety. However, the prior art mainly relies on a single data source (such as a camera or a sensor), is difficult to comprehensively capture complex traffic scenes, and lacks real-time dynamic response capability, so that detection accuracy is low and early warning delay is large. In addition, the conventional method is insufficient in utilization of historical data, and potential risks cannot be effectively identified. With the development of the internet of vehicles (V2X, vehicle to everything) communication technology and the multi-mode sensing technology, how to integrate multi-source data and realize real-time dynamic early warning is a problem to be solved.
Disclosure of Invention
The application provides a traffic road network multi-mode sensing abnormal event early warning method and system, which are used for solving the technical problems of incomplete traffic abnormal event detection and early warning delay caused by the fact that the prior art relies on a single data source and lacks of real-time dynamic response capability.
The application provides a traffic road network multi-mode sensing abnormal event early warning method, which comprises the steps of obtaining a path accident frequent data set of a target traffic road network, traversing the path accident frequent data set to conduct abnormal event prototype clustering construction on the target traffic road network to obtain an abnormal event prototype set, obtaining vehicle state information of a target vehicle based on V2X communication, executing linkage multi-mode sensing data collection to obtain a multi-mode sensing data set, conducting probability matching on the abnormal event prototype set based on the vehicle state information and the multi-mode sensing data set to obtain a target abnormal event prototype, calling a historical abnormal event triggering feature set of the target abnormal event prototype, conducting feature enhancement on the historical abnormal event triggering feature set to determine enhanced abnormal event triggering features, and conducting abnormal event early warning on a running process of the target vehicle by taking the enhanced abnormal event triggering features as early warning triggering conditions.
The system comprises an abnormal event prototype construction module, a vehicle state information acquisition module, a multi-mode sensing data acquisition module, a target abnormal event matching module, an event triggering characteristic enhancement module and an early warning feature enhancement module, wherein the abnormal event prototype construction module is used for acquiring a path accident frequent data set of a target traffic network, traversing the path accident frequent data set to perform abnormal event prototype clustering construction on the target traffic network to obtain an abnormal event prototype set, the vehicle state information acquisition module is used for acquiring vehicle state information of a target vehicle based on V2X communication, the multi-mode sensing data acquisition module is used for executing linkage multi-mode sensing data acquisition to obtain a multi-mode sensing data set, the target abnormal event matching module is used for performing probability matching on the abnormal event prototype set to obtain a target abnormal event prototype, the event triggering characteristic enhancement module is used for calling a historical abnormal event triggering characteristic set of the target abnormal event, and the abnormal event triggering characteristic enhancement module is used for determining that the abnormal event triggering characteristic of the vehicle is an abnormal driving condition, and the abnormal event early warning feature is used for the early warning of the abnormal event.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The application provides a traffic road network multi-mode sensing abnormal event early warning method and system, which relate to the technical field of data processing, and are characterized in that an abnormal event prototype set is constructed by acquiring traffic road network accident data, V2X communication and multi-mode sensing data are combined, a target abnormal event prototype is matched, a history triggering characteristic set is called and enhanced, the enhanced characteristic is taken as an early warning condition, the running process of a target vehicle is monitored in real time, the accurate abnormal event early warning is realized, the technical problems that the prior art depends on a single data source and lacks real-time dynamic response capability, so that traffic abnormal event detection is incomplete and early warning is delayed are solved, and the technical effects of comprehensive and real-time traffic abnormal event detection and dynamic early warning are realized through the multi-mode sensing and V2X communication technology, and the traffic safety and management efficiency are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will briefly explain the drawings needed in the description of the embodiments, which are merely examples of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a traffic road network multi-mode sensing abnormal event early warning method according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of an abnormal event early warning system with multi-mode perception for a traffic road network according to an embodiment of the present application.
The reference numerals illustrate an abnormal event prototype construction module 11, a vehicle state information acquisition module 12, a multi-mode sensing data acquisition module 13, a target abnormal event matching module 14, an event triggering characteristic enhancement module 15 and an abnormal event early warning module 16.
Detailed Description
The application provides a traffic road network multi-mode sensing abnormal event early warning method and system, which are used for solving the technical problems of incomplete traffic abnormal event detection and early warning delay caused by the fact that the prior art relies on a single data source and lacks of real-time dynamic response capability.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be noted that, the terms "first," "second," and the like in the description of the present application and the above drawings are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus.
In a first embodiment, as shown in fig. 1, the present application provides a traffic road network multi-mode sensing abnormal event early warning method, which includes:
And P10, acquiring a path accident frequent data set of a target traffic path network, traversing the path accident frequent data set, and constructing an abnormal event prototype cluster of the target traffic path network to obtain an abnormal event prototype set.
Further, step P10 of the embodiment of the present application further includes:
The method comprises the steps of P11, performing multiple cluster center identification authentication from the path accident frequent data set to determine a cluster center set, P12, performing cluster analysis on the path accident frequent data set based on the cluster center set from two dimensions of time similarity and feature similarity to obtain clustered path accident frequent data clusters, wherein each cluster center corresponds to one clustered path accident frequent data set in the clustered path accident frequent data clusters, and P13, traversing the clustered path accident frequent data clusters to perform embedded feature averaging processing to construct an abnormal event prototype set.
It should be appreciated that obtaining a set of path accident frequency data for the target traffic network, the set comprising historical accident data, such as time, place, weather conditions, vehicle status, road conditions, etc. of the occurrence of the accident, is the basis for constructing the prototype of the abnormal event.
First, in order to ensure the accuracy and representativeness of the clustering result, multiple clustering center identification authentications are performed from the path accident frequent data set. And randomly selecting a plurality of cluster centers each time, and calculating the similarity between any two cluster centers. If the similarity between all the cluster centers is smaller than or equal to a preset similarity threshold, the cluster centers are considered to pass authentication and are uniformly divided into cluster center sets, so that the cluster centers are uniformly distributed in the feature space, and the clustering results are prevented from being too concentrated or overlapped.
Then, after the clustering center set is determined, clustering analysis is carried out on the path accident frequency data set from two dimensions of the time similarity and the feature similarity. The temporal similarity reflects the temporal proximity of the accident, while the feature similarity relates to the similarity of the features of the accident type, the vehicle state, the road environment, etc. Through comprehensive analysis of the two dimensions, the system divides accident data into a plurality of cluster path accident frequency data clusters, and each cluster center corresponds to one cluster path accident frequency data set. This process is similar to the clustering process of the K-Means algorithm, with each cluster corresponding to a cluster center, and the data within the clusters having higher similarity in time and feature dimensions. The purpose of this step is to group the historical incident data by time and feature, providing a basis for subsequent prototype construction.
And finally, traversing each cluster path accident frequent data cluster, and carrying out embedded feature averaging treatment on the data in the cluster, namely carrying out mean value calculation on the features of each data in the cluster, such as speed, acceleration, weather index and the like, so as to generate a feature vector. The mean value of the features of each data in the cluster is calculated to generate a feature vector representing the typical features of the cluster. For example, if a cluster contains a plurality of accidents, the average speed of which is 60km/h and the average visibility of which is 500 meters, the generated feature vector is [60,500]. Finally, the feature vectors of each cluster constitute an abnormal event prototype (i.e., a typical abnormal event pattern constituted by feature vectors), all prototypes forming an abnormal event prototype set. The purpose of this step is to abstract the typical features of each cluster into a prototype, providing the basis for subsequent abnormal event detection and early warning. The embedded feature averaging process not only improves the representativeness of the features, but also reduces the influence of noise data on the clustering result.
Through the steps, a representative abnormal event prototype is extracted from complex accident frequency data, and an accurate reference is provided for subsequent real-time monitoring and early warning.
Further, step P12 of the embodiment of the present application further includes:
p12-1, acquiring a cluster quality recognition function, wherein the cluster quality recognition function is as follows: Wherein, the method comprises the steps of, As a quality factor of the clusters,As a total number of cluster centers,Is a positive integer which is used for the preparation of the high-voltage power supply,Is the first cluster center of M clustersThe first cluster center corresponds toThe number of path incident frequency data in the individual clustered path incident frequency data sets,Is the firstThe path accident frequency data of the cluster centers,Is the firstThe first cluster center corresponds toThe first cluster path accident frequent data setThe frequency of the data of the path accidents,As a parameter of the weight-bearing element,For controlling the extent to which temporal similarity affects the loss calculation,Is the firstThe point in time at which the path accident frequency data of the cluster centers occurs,Is the firstThe first cluster center corresponds toThe first cluster path accident frequent data setThe point in time at which the path accident frequency data occurs,For the preset adjustment of the scale parameters of the time distance,The method comprises the steps of determining the attenuation speed of time similarity, performing cluster quality recognition on cluster path accident frequent data clusters by using a cluster quality recognition function to obtain a cluster quality factor, and performing cluster center screening again if the cluster quality factor is larger than or equal to a preset cluster quality factor by P12-2.
Optionally, the quality evaluation is further performed on the cluster path accident frequent data clusters through a cluster quality recognition function so as to ensure the accuracy and reliability of cluster analysis.
Firstly, a cluster quality recognition function is obtained, and the function can comprehensively quantify the quality of clusters by comprehensively considering the feature similarity and the time similarity. And then, calculating a clustering quality factor of each cluster by using the function, calculating the characteristic similarity and the time similarity between the data points and the cluster centers by traversing each cluster center and the corresponding data set, and finally obtaining the clustering quality factor by weighting and summing according to a formula. I.e., a LOSS value is calculated for each cluster, the smaller this value, the more similar the data points within the cluster in characteristics and time, and the higher the cluster quality. In this way, tight and consistent clusters can be identified, which represent a typical pattern of accident occurrences in traffic networks.
Finally, judging whether the calculated cluster quality factor (namely LOSS value) is larger than or equal to a preset cluster quality factor threshold value. If the threshold value is not reached, the current cluster quality is not satisfied, and the system needs to carry out cluster center screening again. For example, the cluster center identification authentication step is re-executed, a new cluster center is selected, cluster analysis and cluster quality identification are re-performed until the cluster quality factor reaches a preset threshold value, so that the accuracy and reliability of cluster analysis are improved.
Through the process, not only is the selection of the clustering center optimized, but also the accuracy and the reliability of clustering analysis are remarkably improved through quantitative evaluation and dynamic adjustment, and a solid foundation is laid for subsequent abnormal event early warning.
And P20, acquiring vehicle state information of the target vehicle based on V2X communication.
Specifically, based on a V2X (Vehicle-to-evaluation) communication technology, vehicle state information of a target Vehicle is acquired in real time, and key data support is provided for subsequent traffic monitoring, abnormal event detection and early warning. The V2X communication technology is an advanced technology for information interaction between a vehicle and the surrounding environment (including other vehicles, infrastructures, pedestrians and the like), and can realize low-delay and high-reliability data transmission and ensure real-time performance and accuracy of vehicle state information.
First, a connection is established with a target Vehicle through V2X communication, and a plurality of communication modes such as V2V (Vehicle-to-Vehicle), V2I (Vehicle-to-Infrastructure), and V2P (Vehicle-to-PEDESTRIAN) are utilized to cover the interactive scene between the Vehicle and the surrounding environment. By the technology, the system can acquire dynamic data of the target vehicle in real time, including position information (such as longitude and latitude coordinates), speed information, acceleration information, direction information, vehicle states (such as engine states, brake states and light states) and environment sensing data (such as surrounding vehicles, pedestrians, traffic lights and the like).
Then, the acquired vehicle state information is transmitted to a system background in real time through a V2X communication network. The system analyzes and preprocesses the received data, and ensures the integrity and accuracy of the data. Meanwhile, the vehicle state information, the historical data and the environmental data are subjected to association analysis, and support is provided for subsequent abnormal event detection and early warning.
The vehicle state information obtained through the V2X communication technology can be widely applied to various scenes. For example, in real-time traffic monitoring, the system can monitor the running state of a target vehicle and timely find abnormal behaviors, in collision early warning, the system predicts potential collision risks according to vehicle state information and gives early warning to a driver, in path planning, the vehicle state and traffic environment are combined to provide optimal path suggestions for the driver, in accident analysis, the vehicle state information is utilized to conduct accident cause analysis, and basis is provided for accident handling.
Through the steps, the vehicle state information of the target vehicle can be efficiently and accurately acquired, the real-time performance and the reliability of the system are improved, and important data support is provided for traffic monitoring, abnormal event detection and early warning.
And P30, performing linked multi-mode sensing data acquisition to obtain a multi-mode sensing data set.
Further, step P30 of the embodiment of the present application further includes:
the method comprises the steps of P31, utilizing a traffic camera to collect video flow and extract characteristics of a target traffic network to obtain a road network visual characteristic set, P32, utilizing a geomagnetic sensor to extract lane occupation characteristics of the target traffic network to obtain lane occupation characteristics, P33, utilizing an environment perception sensor array to conduct environment perception on the target traffic network to obtain an environment perception characteristic set, and P34, summarizing the road network visual characteristic set, the lane occupation characteristics and the environment perception characteristic set to obtain the multi-mode perception data set.
It should be understood that by integrating multiple sensing technologies, the target traffic network is subjected to omnibearing and multi-angle data acquisition, and finally a multi-mode sensing data set is generated. The real-time state of the traffic network can be comprehensively reflected by integrating the data of various sensors, and abundant data support is provided for subsequent traffic monitoring, abnormal event detection and decision analysis.
Firstly, a traffic camera deployed at a key position is utilized to collect video streams of a target traffic network. These cameras are able to capture the dynamic behavior of the vehicle, traffic flow and possible anomalies. Key visual features such as vehicle type, color, license plate number, travel track, etc. can be extracted from the video stream by advanced image processing and computer vision techniques. These visual feature sets provide intuitive and rich information for traffic monitoring.
And then, deploying a geomagnetic sensor below the lane of the target traffic network, and extracting lane occupation characteristics of the target traffic network by using the geomagnetic sensor. The geomagnetic sensor can detect a magnetic field change caused when a vehicle passes by, so that the lane occupation situation can be accurately identified. The sensors can provide real-time lane occupation data, including the number, the type, the residence time and the like of vehicles on each lane, provide important basis for traffic flow analysis and lane management, and help the system to grasp the service condition of the lanes in real time.
In addition, an environment-aware sensor array is deployed in the target traffic network, and the environment-aware sensor array is utilized to perform environment awareness on the target traffic network. The sensors can monitor environmental factors such as weather conditions (such as rain, fog and snow), road conditions (such as wet skid and ice), visibility and the like, generate an environment perception feature set, provide important references for traffic management and safety early warning, and help the system to cope with complex environmental changes.
And finally, summarizing a road network visual feature set, lane occupation features and an environment perception feature set obtained from the video stream, the geomagnetic sensor and the environment perception sensor array to form a complete multi-mode perception data set. The method comprises the steps of firstly aligning data of different modes according to time stamps and space positions to ensure consistency and relevance of the data, secondly organically combining visual features, lane occupation features and environment features through multi-mode data fusion to form comprehensive traffic road network state description, and finally storing the fused multi-mode perception data set into a system database to provide a data base for subsequent traffic analysis, abnormal event detection and decision support. The data set integrates information in multiple aspects such as vision, physics, environment and the like, and provides comprehensive data support for detection and early warning of traffic abnormal events.
And P40, carrying out probability matching on the abnormal event prototype set based on the vehicle state information and the multi-mode perception data set to obtain a target abnormal event prototype.
Further, step P40 of the embodiment of the present application further includes:
The method comprises the steps of using vehicle state information and the multi-mode perception data set as matching vectors, using probability functions to carry out probability analysis on the matching vectors and the abnormal event prototype set respectively to obtain a matching probability set, and using an abnormal event prototype corresponding to the maximum value in the matching probability set as a target abnormal event prototype, wherein the P41 is used for carrying out probability analysis on the matching vectors and the abnormal event prototype set by using the probability functions.
Wherein the probability function is: Wherein, the method comprises the steps of, For the probability that the matching vector belongs to the abnormal event prototype,In order for the vector to be a match,To match events for which the vector belongs to the abnormal event prototype,For the purpose of the exception event prototype,As a distance measurement function, and。
Optionally, the probability matching is performed on the set of abnormal event prototypes by comprehensively analyzing the vehicle state information and the multimodal perception data set to identify the most likely abnormal event.
Firstly, fusing real-time acquired vehicle state information and multi-mode sensing data set into a comprehensive matching vector. This vector contains the current speed, acceleration, direction of travel, position information of the vehicle, as well as visual features extracted from the traffic cameras, lane occupancy features obtained from the geomagnetic sensors, and environmental awareness features collected from the environmental awareness sensor array. The information together form a comprehensive traffic condition description, and rich data support is provided for the identification of abnormal events.
Next, a probability function is used to calculate the probability of matching between the matching vector and each of the anomaly event prototypes. Here the number of the elements is the number,Is a distance metric function used to calculate the distance of the matching vector from the anomaly event prototype in feature space. The Euclidean distance is adopted by the distance measurement function, so that the similarity degree of the distance measurement function and the Euclidean distance in the multidimensional feature space can be intuitively reflected. The probability function converts the distance into probability through an exponential function, and the smaller the distance, the higher the matching probability, which indicates that the matching vector is more likely to belong to the abnormal event prototype.
During the calculation, the system will traverse each prototype in the set of abnormal event prototypes, calculating their distance from the matching vector and the corresponding matching probability, respectively. The probability values reflect the matching degree of different abnormal event prototypes and the current traffic condition, and can provide basis for subsequent abnormal event identification.
Finally, an abnormal event prototype with the maximum value is selected from the calculated matching probability set to be used as a target abnormal event prototype. The matching probability of this prototype to the current traffic situation is highest and thus most likely represents an impending anomaly. In this way, the most relevant one can be identified from a number of potential abnormal events, providing an explicit direction for real-time early warning and preventive measure adoption. By comprehensively analyzing the vehicle state and the multi-mode sensing data, the accuracy and timeliness of abnormal event detection can be remarkably improved.
And P50, calling a historical abnormal event triggering characteristic set of the target abnormal event prototype, carrying out characteristic enhancement on the historical abnormal event triggering characteristic set, and determining enhanced abnormal event triggering characteristics.
Further, step P50 of the embodiment of the present application further includes:
The method comprises the steps of P51, randomly extracting a first historical abnormal event triggering characteristic and a second historical abnormal event triggering characteristic from the historical abnormal event triggering characteristic set, P52, carrying out inner product mapping on the first historical abnormal event triggering characteristic and the second historical abnormal event triggering characteristic to obtain a first characteristic similarity set, P53, carrying out characteristic enhancement on the second historical abnormal event triggering characteristic based on the first characteristic similarity set to obtain a first enhanced historical abnormal event triggering characteristic, P54, randomly extracting a third historical abnormal event triggering characteristic from the historical abnormal event triggering characteristic set again, enhancing the third historical abnormal event triggering characteristic based on the first enhanced historical abnormal event triggering characteristic to obtain a second enhanced historical abnormal event triggering characteristic, and so on to obtain the enhanced abnormal event triggering characteristic.
It should be appreciated that by the feature enhancement method, features are extracted and enhanced from a set of historical abnormal event triggering features of a target abnormal event prototype to determine a set of more robust and representative abnormal event triggering features, thereby improving the accuracy of abnormal event detection and the reliability of the early warning system.
First, two different historical abnormal event trigger features are randomly extracted from a historical abnormal event trigger feature set, and the two different historical abnormal event trigger features are respectively called a first historical abnormal event trigger feature and a second historical abnormal event trigger feature. These feature samples typically include a variety of factors associated with the abnormal event, such as environmental conditions (e.g., rain and snow weather), vehicle status (e.g., vehicle speed, acceleration), road conditions (e.g., degree of congestion), and the like. By randomly extracting feature samples, the system is able to provide diversified input data for subsequent feature enhancements.
Then, the two extracted historical abnormal event trigger features are subjected to inner product mapping. And calculating the similarity between the two. The inner product mapping is a method for measuring the relevance between the features, and can help a system to judge the similarity degree between different feature samples. For example, the system may analyze the correlation between environmental factors (e.g., rain and snow weather) and vehicle conditions (e.g., vehicle speed) via the inner product map to determine the triggering cause of the abnormal event. Finally, the system generates a first feature similarity set for recording similarity values between the feature samples.
Then, based on the first feature similarity set, feature enhancement is performed on the second historical abnormal event triggering feature. For example, the second historical abnormal event triggering characteristic is weighted and adjusted by taking the similarity value as a weight, thereby generating the first enhanced historical abnormal event triggering characteristic. The process can highlight important information in the characteristic sample, inhibit noise data and improve the expression capability and distinguishing degree of the characteristics.
And finally, randomly extracting a third historical abnormal event triggering characteristic from the historical abnormal event triggering characteristic set again, and enhancing the newly extracted characteristic based on the first enhanced historical abnormal event triggering characteristic obtained before, so as to obtain a second enhanced historical abnormal event triggering characteristic. This process may be iterated, each iteration enhancing the new feature based on the latest enhanced feature until a complete set of enhanced anomaly event triggered features is obtained. Through multiple iterations, the system can gradually optimize the feature samples and mine potential rules in the historical data, so that the accuracy and reliability of the triggering feature of the abnormal event are improved. The method provides a more accurate and reliable characteristic basis for subsequent abnormal event detection and early warning, thereby remarkably improving the safety and efficiency of the intelligent traffic system.
Further, step P53 of the embodiment of the present application further includes:
And P53-1, performing normalization processing on the first feature similarity set to obtain a first feature similarity normalization value set, and P53-2, adding the first feature similarity normalization value set into an initial empty matrix, and further performing convolution operation on the first feature similarity normalization value set and the second historical abnormal event triggering feature by using a graph convolution network to obtain the first enhanced historical abnormal event triggering feature.
In one possible embodiment of the application, the feature enhancement process may be further refined to ensure the accuracy and effectiveness of feature enhancement.
First, a normalization process is performed on the first feature similarity set. Normalization is a data preprocessing technique that scales feature values to a specific range (typically 0 to 1) to eliminate the effects of dimension between different features and ensure that they have the same weight in subsequent processing. This step generates a first feature similarity normalization value set that provides a normalized similarity measure for feature enhancement.
Next, the normalized feature similarity set is added to an initially empty matrix. This matrix will be used as an input to the graph convolution network to convolve with the second historical anomaly event trigger feature. The graph rolling network (Graph Convolutional Network, GCN) is a deep learning model that enables efficient feature learning on graph structure data. By integrating the feature similarity information into the graph rolling network, complex relationships between features can be captured and the expressive power of the features can be enhanced.
In the convolution operation process, the graph convolution network adjusts the weight of the second historical abnormal event triggering characteristic by utilizing the characteristic similarity normalization value set, so that the first enhanced historical abnormal event triggering characteristic is obtained. The enhanced features not only contain the information of the original features, but also integrate the normalized information of the feature similarity, so that the features are more accurate and comprehensive when expressing abnormal events. The feature enhancement method not only improves the accuracy of the features, but also enhances the recognition capability of the model on the abnormal events, and provides a more reliable feature basis for subsequent abnormal event early warning.
And P60, carrying out abnormal event early warning on the running process of the target vehicle by taking the enhanced abnormal event triggering characteristic as an early warning triggering condition.
Specifically, the enhanced abnormal event triggering characteristic is used as an early warning triggering condition, the abnormal event early warning is carried out on the running process of the target vehicle, and the early warning information can be immediately received once the vehicle approaches the potential abnormal event condition in the running process.
Illustratively, the system continuously tracks the driving state of the target vehicle and collects relevant multimodal data in real time. These data are used to update the matching vector of the vehicle reflecting the current traffic conditions. This real-time updated match vector is then compared to the enhanced anomaly event trigger feature. If there is a match, i.e., the current condition meets the pre-alarm trigger condition defined by any of the enhanced features, the system will evaluate the degree of match. When the matching degree exceeds a preset threshold, the system judges that an early warning needs to be sent out. Thereupon, early warning information including the type of abnormal event that may occur, suggested countermeasures, etc. is generated and rapidly transmitted to the driver through a central control screen of the vehicle, an acoustic alarm, a mobile device, etc.
The early warning mechanism aims at giving the driver or an automatic driving system enough time and information to take the risk avoidance measures, so as to avoid or alleviate potential traffic accidents. Through the process, a closed-loop abnormal event detection and response system is realized, real-time monitoring and early warning of the running process of the target vehicle can be realized, and the intelligent level and safety of traffic management are improved.
Further, the embodiment of the application further comprises a step P70 of acquiring early warning trigger time and continuously feeding back and monitoring the target vehicle in an early warning feedback monitoring window.
Optionally, the specific time of early warning triggering can be further obtained, and continuous feedback monitoring is performed on the target vehicle in a set early warning feedback monitoring window, so that the system can provide continuous monitoring and necessary follow-up guidance before and after an abnormal event possibly occurs, and the reliability and the effectiveness of the early warning system are enhanced.
For example, a specific time point of the early warning trigger is recorded first as the starting time of the subsequent monitoring. And continuously collecting real-time driving data of the target vehicle in a preset early warning feedback monitoring window, wherein the real-time driving data comprise information such as vehicle speed, acceleration, position, direction and the like, and dynamically comparing the real-time driving data with the triggering characteristics of the enhanced abnormal event to evaluate the state change of the vehicle after early warning. If an abnormal event is detected to be persistent or further worsening, the system may upgrade the early warning level and take more aggressive intervention, such as sending an emergency alert to the driver or linking directly with the traffic management center. Meanwhile, the system can record all data during monitoring for subsequent analysis and optimization of the early warning model. Through the step, the system can realize real-time evaluation and dynamic adjustment of the early warning effect, ensure timely and effective treatment of abnormal events, and further improve the safety and reliability of vehicle running.
In summary, the embodiment of the application has at least the following technical effects:
the method comprises the steps of constructing an abnormal event prototype set by obtaining a path accident frequent data set of a target traffic network, and obtaining vehicle state information of a target vehicle based on V2X communication. And (3) carrying out probability matching on the abnormal event prototype set by combining the vehicle state information and the multi-mode sensing data set through linkage multi-mode sensing data acquisition, and determining a target abnormal event prototype. And calling the historical abnormal event triggering characteristic set and carrying out characteristic enhancement, taking the enhanced abnormal event triggering characteristic as an early warning condition, and carrying out real-time abnormal event early warning on the running process of the target vehicle.
The technical effects of comprehensive and real-time traffic abnormal event detection and dynamic early warning and traffic safety and management efficiency improvement are achieved through the multi-mode sensing and V2X communication technology.
In the second embodiment, based on the same inventive concept as the traffic road network multi-mode sensing abnormal event early warning method in the previous embodiment, as shown in fig. 2, the present application provides a traffic road network multi-mode sensing abnormal event early warning system, and the system and method embodiments in the embodiments of the present application are based on the same inventive concept. Wherein the system comprises:
The abnormal event prototype construction module 11 is configured to obtain a path accident frequent data set of a target traffic path network, traverse the path accident frequent data set, and perform abnormal event prototype clustering construction on the target traffic path network to obtain an abnormal event prototype set.
A vehicle state information acquisition module 12, wherein the vehicle state information acquisition module 12 is configured to acquire vehicle state information of a target vehicle based on V2X communication.
The multi-mode sensing data acquisition module 13 is used for executing linked multi-mode sensing data acquisition to obtain a multi-mode sensing data set.
The target abnormal event matching module 14 is configured to perform probability matching on the abnormal event prototype set based on the vehicle state information and the multimodal perception data set, so as to obtain a target abnormal event prototype.
The event triggering characteristic enhancement module 15 is configured to invoke a historical abnormal event triggering characteristic set of the target abnormal event prototype, perform characteristic enhancement on the historical abnormal event triggering characteristic set, and determine an enhanced abnormal event triggering characteristic.
The abnormal event early warning module 16 is configured to early warn the abnormal event during the driving process of the target vehicle by using the enhanced abnormal event triggering feature as an early warning triggering condition by the abnormal event early warning module 16.
Further, the abnormal event prototype construction module 11 is further configured to perform the following steps:
The method comprises the steps of carrying out multiple times of cluster center identification authentication from a path accident frequent data set to determine a cluster center set, carrying out cluster analysis on the path accident frequent data set based on the cluster center set from two dimensions of time similarity and feature similarity to obtain clustered path accident frequent data clusters, wherein each cluster center corresponds to one clustered path accident frequent data set in the clustered path accident frequent data clusters, traversing the clustered path accident frequent data clusters to carry out embedded feature averaging treatment, and constructing an abnormal event prototype set.
Further, the abnormal event prototype construction module 11 is further configured to perform the following steps:
acquiring a cluster quality recognition function, wherein the cluster quality recognition function is as follows: Wherein, the method comprises the steps of, As a quality factor of the clusters,As a total number of cluster centers,Is a positive integer which is used for the preparation of the high-voltage power supply,Is the first cluster center of M clustersThe first cluster center corresponds toThe number of path incident frequency data in the individual clustered path incident frequency data sets,Is the firstThe path accident frequency data of the cluster centers,Is the firstThe first cluster center corresponds toThe first cluster path accident frequent data setThe frequency of the data of the path accidents,As a parameter of the weight-bearing element,For controlling the extent to which temporal similarity affects the loss calculation,Is the firstThe point in time at which the path accident frequency data of the cluster centers occurs,Is the firstThe first cluster center corresponds toThe first cluster path accident frequent data setThe point in time at which the path accident frequency data occurs,For the preset adjustment of the scale parameters of the time distance,The method comprises the steps of determining the attenuation speed of time similarity, carrying out cluster quality recognition on cluster path accident frequent data clusters by using a cluster quality recognition function to obtain a cluster quality factor, judging whether the cluster quality factor is greater than or equal to a preset cluster quality factor, and if not, carrying out cluster center screening again.
Further, the multi-modal sensing data collection module 13 is further configured to perform the following steps:
the method comprises the steps of utilizing a traffic camera to collect video flow and extract characteristics of a target traffic network to obtain a road network visual characteristic set, utilizing a geomagnetic sensor to extract lane occupation characteristics of the target traffic network to obtain lane occupation characteristics, utilizing an environment perception sensor array to conduct environment perception on the target traffic network to obtain an environment perception characteristic set, and summarizing the road network visual characteristic set, the lane occupation characteristics and the environment perception characteristic set to obtain the multi-mode perception data set.
Further, the objective abnormal event matching module 14 is further configured to perform the following steps:
The vehicle state information and the multi-mode perception data set are used as matching vectors, probability analysis is carried out on the matching vectors and the abnormal event prototype set respectively by using probability functions to obtain a matching probability set, and an abnormal event prototype corresponding to the maximum value in the matching probability set is used as a target abnormal event prototype.
Further, the objective abnormal event matching module 14 is further configured to perform the following steps:
The probability function is: Wherein, the method comprises the steps of, For the probability that the matching vector belongs to the abnormal event prototype,In order for the vector to be a match,To match events for which the vector belongs to the abnormal event prototype,For the purpose of the exception event prototype,As a function of the distance measure,。
Further, the event trigger feature enhancement module 15 is further configured to perform the following steps:
The method comprises the steps of randomly extracting a first historical abnormal event trigger feature and a second historical abnormal event trigger feature from a historical abnormal event trigger feature set, carrying out inner product mapping on the first historical abnormal event trigger feature and the second historical abnormal event trigger feature to obtain a first feature similarity set, carrying out feature enhancement on the second historical abnormal event trigger feature based on the first feature similarity set to obtain a first enhanced historical abnormal event trigger feature, randomly extracting a third historical abnormal event trigger feature from the historical abnormal event trigger feature set again, carrying out enhancement on the third historical abnormal event trigger feature based on the first enhanced historical abnormal event trigger feature to obtain a second enhanced historical abnormal event trigger feature, and so on to obtain the enhanced abnormal event trigger feature.
Further, the event trigger feature enhancement module 15 is further configured to perform the following steps:
And adding the first feature similarity normalization value set into an initial empty matrix, and further carrying out convolution operation on the first feature similarity set and the second historical abnormal event triggering feature by using a graph convolution network to obtain the first enhanced historical abnormal event triggering feature.
Further, the system further comprises:
the early warning feedback monitoring module is used for acquiring early warning trigger time and continuously feeding back and monitoring the target vehicle in the early warning feedback monitoring window.
It should be noted that the sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing description of the preferred embodiments of the application is not intended to limit the application to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the application are intended to be included within the scope of the application.
The specification and figures are merely exemplary illustrations of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (8)
1. The utility model provides a traffic road network multimode perception abnormal event early warning method which is characterized in that the method comprises the following steps:
Acquiring a path accident frequent data set of a target traffic path network, traversing the path accident frequent data set, and constructing an abnormal event prototype cluster for the target traffic path network to acquire an abnormal event prototype set;
The method comprises the following steps:
performing multiple clustering center identification authentication from the path accident frequent data set to determine a clustering center set;
Clustering analysis is carried out on the path accident frequent data sets based on the clustering center set from two dimensions of the time similarity and the feature similarity to obtain clustered path accident frequent data clusters, wherein each clustering center corresponds to one clustered path accident frequent data set in the clustered path accident frequent data clusters;
traversing the clustered path accident frequent data clusters to perform embedded feature averaging treatment, and constructing an abnormal event prototype set;
Acquiring vehicle state information of a target vehicle based on V2X communication;
performing linked multi-modal sensing data acquisition to obtain a multi-modal sensing data set;
Based on the vehicle state information and the multi-mode perception data set, probability matching is carried out on the abnormal event prototype set, and a target abnormal event prototype is obtained;
Invoking a historical abnormal event triggering feature set of the target abnormal event prototype, performing feature enhancement on the historical abnormal event triggering feature set, and determining enhanced abnormal event triggering features;
The method comprises the following steps:
Randomly extracting a first historical abnormal event trigger feature and a second historical abnormal event trigger feature from the historical abnormal event trigger feature set;
Performing inner product mapping on the first historical abnormal event triggering characteristic and the second historical abnormal event triggering characteristic to obtain a first characteristic similarity set;
respectively carrying out feature enhancement on the second historical abnormal event triggering features based on the first feature similarity set to obtain first enhanced historical abnormal event triggering features;
Randomly extracting a third historical abnormal event triggering characteristic from the historical abnormal event triggering characteristic set again, enhancing the third historical abnormal event triggering characteristic based on the first enhanced historical abnormal event triggering characteristic to obtain a second enhanced historical abnormal event triggering characteristic, and so on to obtain the enhanced abnormal event triggering characteristic;
and carrying out abnormal event early warning on the running process of the target vehicle by taking the enhanced abnormal event triggering characteristic as an early warning triggering condition.
2. The traffic road network multi-mode perception abnormal event early warning method according to claim 1, wherein clustering analysis is performed on the path accident frequent data set based on the clustering center set from two dimensions of time similarity and feature similarity to obtain a clustered path accident frequent data cluster, and the method comprises the following steps:
acquiring a cluster quality recognition function, wherein the cluster quality recognition function is as follows:
;
Wherein, the As a quality factor of the clusters,As a total number of cluster centers,Is a positive integer which is used for the preparation of the high-voltage power supply,Is the first cluster center of M clustersThe first cluster center corresponds toThe number of path incident frequency data in the individual clustered path incident frequency data sets,Is the firstThe path accident frequency data of the cluster centers,Is the firstThe first cluster center corresponds toThe first cluster path accident frequent data setThe frequency of the data of the path accidents,As a parameter of the weight-bearing element,For controlling the extent to which temporal similarity affects the loss calculation,Is the firstThe point in time at which the path accident frequency data of the cluster centers occurs,Is the firstThe first cluster center corresponds toThe first cluster path accident frequent data setThe point in time at which the path accident frequency data occurs,For the preset adjustment of the scale parameters of the time distance,A decay rate for determining temporal similarity;
performing cluster quality recognition on the cluster path accident frequent data clusters by using a cluster quality recognition function to obtain a cluster quality factor;
judging whether the clustering quality factor is larger than or equal to a preset clustering quality factor, and if not, re-screening the clustering center.
3. The traffic road network multi-modal aware abnormal event early warning method of claim 1, wherein probability matching is performed on the abnormal event prototype set based on the vehicle state information and the multi-modal aware dataset to obtain a target abnormal event prototype, comprising:
taking the vehicle state information and the multi-mode sensing data set as matching vectors;
respectively carrying out probability analysis on the matching vector and the abnormal event prototype set by using a probability function to obtain a matching probability set;
And taking the abnormal event prototype corresponding to the maximum value in the matching probability set as a target abnormal event prototype.
4. The traffic road network multi-mode perceived abnormal event early warning method according to claim 3, wherein the probability function is:
;
Wherein, the For the probability that the matching vector belongs to the abnormal event prototype,In order for the vector to be a match,To match events for which the vector belongs to the abnormal event prototype,For the purpose of the exception event prototype,As a function of the distance measure,。
5. The traffic road network multi-mode perceived abnormal event early warning method according to claim 1, wherein the feature enhancement is performed on the second historical abnormal event triggering features based on the first feature similarity set, respectively, to obtain first enhanced historical abnormal event triggering features, and the method comprises the following steps:
normalizing the first feature similarity set to obtain a first feature similarity normalization value set;
And adding the first feature similarity normalization value set into an initial empty matrix, and further carrying out convolution operation on the first feature similarity normalization value set and the second historical abnormal event triggering feature by using a graph convolution network to obtain the first enhanced historical abnormal event triggering feature.
6. The traffic road network multi-modal sensing abnormal event early warning method of claim 1, wherein the step of performing linked multi-modal sensing data collection to obtain a multi-modal sensing data set comprises:
The traffic cameras are utilized to collect video streams and extract features of the target traffic road network, and a road network visual feature set is obtained;
extracting lane occupation characteristics of the target traffic network by utilizing a geomagnetic sensor to obtain lane occupation characteristics;
Performing environment sensing on the target traffic network by using an environment sensing sensor array to obtain an environment sensing characteristic set;
summarizing the road network visual feature set, the lane occupation feature and the environment perception feature set to obtain the multi-mode perception data set.
7. The traffic road network multi-mode sensing abnormal event early warning method according to claim 1, wherein early warning trigger time is obtained, and continuous feedback monitoring is carried out on a target vehicle in an early warning feedback monitoring window.
8. A traffic road network multi-modal aware abnormal event early warning system, characterized by the steps for implementing a traffic road network multi-modal aware abnormal event early warning method according to any one of claims 1 to 7, the system comprising:
The abnormal event prototype construction module is used for acquiring a path accident frequent data set of a target traffic path network, traversing the path accident frequent data set, and constructing an abnormal event prototype cluster of the target traffic path network to obtain an abnormal event prototype set;
the vehicle state information acquisition module is used for acquiring vehicle state information of a target vehicle based on V2X communication;
The multi-mode sensing data acquisition module is used for executing linkage multi-mode sensing data acquisition to obtain a multi-mode sensing data set;
The target abnormal event matching module is used for carrying out probability matching on the abnormal event prototype set based on the vehicle state information and the multi-mode perception data set to obtain a target abnormal event prototype;
the event triggering characteristic enhancement module is used for calling a historical abnormal event triggering characteristic set of the target abnormal event prototype, carrying out characteristic enhancement on the historical abnormal event triggering characteristic set and determining enhanced abnormal event triggering characteristics;
The abnormal event early warning module is used for carrying out abnormal event early warning on the running process of the target vehicle by taking the enhanced abnormal event triggering characteristic as an early warning triggering condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510562109.2A CN120088989B (en) | 2025-04-30 | 2025-04-30 | Traffic road network multi-mode sensing abnormal event early warning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510562109.2A CN120088989B (en) | 2025-04-30 | 2025-04-30 | Traffic road network multi-mode sensing abnormal event early warning method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120088989A CN120088989A (en) | 2025-06-03 |
| CN120088989B true CN120088989B (en) | 2025-07-11 |
Family
ID=95856401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510562109.2A Active CN120088989B (en) | 2025-04-30 | 2025-04-30 | Traffic road network multi-mode sensing abnormal event early warning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120088989B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120526599A (en) * | 2025-07-24 | 2025-08-22 | 长安大学 | Transportation operation monitoring, early warning and decision analysis method based on vehicle-road collaboration |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107103754A (en) * | 2017-05-10 | 2017-08-29 | 华南师范大学 | A kind of road traffic condition Forecasting Methodology and system |
| CN116738761A (en) * | 2023-07-25 | 2023-09-12 | 南京感动科技有限公司 | SUMO-based expressway network real-time microscopic traffic simulation method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1997020433A1 (en) * | 1995-12-01 | 1997-06-05 | Southwest Research Institute | Methods and apparatus for traffic incident detection |
| CN119495186A (en) * | 2024-10-09 | 2025-02-21 | 复旦大学 | A traffic prediction system based on historical trajectory collection |
| CN119049292B (en) * | 2024-11-04 | 2024-12-31 | 广州市声讯电子科技股份有限公司 | Road condition forecast and early warning method and system based on intelligent traffic |
| CN119625994B (en) * | 2025-02-14 | 2025-05-09 | 特微乐行(广州)技术有限公司 | Expressway toll station flow prediction method and system |
-
2025
- 2025-04-30 CN CN202510562109.2A patent/CN120088989B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107103754A (en) * | 2017-05-10 | 2017-08-29 | 华南师范大学 | A kind of road traffic condition Forecasting Methodology and system |
| CN116738761A (en) * | 2023-07-25 | 2023-09-12 | 南京感动科技有限公司 | SUMO-based expressway network real-time microscopic traffic simulation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120088989A (en) | 2025-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN120088989B (en) | Traffic road network multi-mode sensing abnormal event early warning method and system | |
| CN116504076A (en) | Expressway traffic flow prediction method based on ETC portal data | |
| CN118644990B (en) | Data control system for smart city based on big data | |
| CN118278841B (en) | Large-piece transportation supervision and early warning method, system and device based on driving trajectory | |
| CN116631186B (en) | Highway traffic accident risk assessment method and system based on dangerous driving event data | |
| CN120338516B (en) | New energy vehicle gathering prevention and control system based on cluster recognition and regional risk modeling | |
| CN118781815B (en) | Intelligent traffic information acquisition method and system based on big data | |
| CN117115752A (en) | Expressway video monitoring method and system | |
| CN118429912A (en) | Security scene optimization method and system based on big data analysis | |
| CN119312169A (en) | A risk prediction method and platform for operating vehicles based on multi-factor analysis | |
| CN120046903A (en) | Road transportation government affair safety supervision and management method | |
| CN118887808A (en) | A method for detecting traffic violations on high-speed roads | |
| CN119049152A (en) | Abnormal behavior monitoring system of vehicle based on pattern recognition | |
| CN119007131B (en) | Roadside digital video monitoring method and system based on deep learning | |
| CN118861696B (en) | Road traffic accident detection method and system based on holographic perception | |
| KR102494953B1 (en) | On-device real-time traffic signal control system based on deep learning | |
| CN119068659A (en) | A vehicle abnormal behavior detection method based on Beidou positioning | |
| CN114613134B (en) | Regional traffic flow condition prediction method based on visual identification | |
| CN113128847A (en) | Entrance ramp real-time risk early warning system and method based on laser radar | |
| CN120318793B (en) | Sensing method and system for road surface subsidence and jolt | |
| CN119339579B (en) | Traffic accident automatic monitoring and alarm system and method | |
| CN120472404B (en) | Self-adaptive context-aware vehicle target tracking and identifying system | |
| CN119889047B (en) | Traffic flow induction monitoring method and system based on convolutional denoising autoencoder | |
| CN120340256A (en) | Traffic conflict identification method and system for road weaving area | |
| CN120783549A (en) | Traffic gate intelligent management method and device based on vehicle state sensing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |