CN115359658A - Method, device, equipment, storage medium and program product for detecting traffic incident - Google Patents
Method, device, equipment, storage medium and program product for detecting traffic incident Download PDFInfo
- Publication number
- CN115359658A CN115359658A CN202210989463.XA CN202210989463A CN115359658A CN 115359658 A CN115359658 A CN 115359658A CN 202210989463 A CN202210989463 A CN 202210989463A CN 115359658 A CN115359658 A CN 115359658A
- Authority
- CN
- China
- Prior art keywords
- traffic
- detection
- data
- road area
- event
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000001514 detection method Methods 0.000 claims abstract description 284
- 230000008447 perception Effects 0.000 claims abstract description 68
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000012549 training Methods 0.000 claims description 45
- 230000002159 abnormal effect Effects 0.000 claims description 39
- 230000006399 behavior Effects 0.000 claims description 27
- 238000004458 analytical method Methods 0.000 claims description 24
- 206010000117 Abnormal behaviour Diseases 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 5
- 206010039203 Road traffic accident Diseases 0.000 description 9
- 238000010276 construction Methods 0.000 description 8
- 238000011144 upstream manufacturing Methods 0.000 description 8
- 238000013145 classification model Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- 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
-
- 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
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- 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
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
- G08G1/054—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed photographing overspeeding vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Traffic Control Systems (AREA)
Abstract
The application provides a method, a device, equipment, a storage medium and a program product for detecting traffic incidents, wherein the method comprises the following steps: acquiring perception data in a target road area, wherein the perception data comprises at least one of the following items: the driving track of each vehicle in the target road area and the traffic flow data of the target road area; processing the perception data through at least one target detection model in a plurality of detection models to obtain detection results corresponding to the detection models; and determining the type of the target traffic event in the target road area according to each detection result. The method and the device can accurately identify the type of the traffic incident, and further can timely and effectively reduce the loss caused by the incident.
Description
Technical Field
The present application relates to the field of intelligent traffic technologies, and in particular, to a method, an apparatus, a device, a storage medium, and a program product for detecting a traffic incident.
Background
An Automatic Identification (AID) System for traffic events is an important component of an Intelligent Transportation System (ITS), and people can find traffic events occurring on a highway as early as possible by means of the AID System, so that loss caused by the events is reduced.
At present, a traffic incident automatic detection method generally finds a traffic incident according to changes of traffic flow parameters, but the traffic flow similarity of various traffic incidents is high, and specific traffic incident types cannot be identified.
Therefore, the prior art cannot accurately identify the type of the traffic incident, and further cannot timely and effectively reduce the loss caused by the incident.
Disclosure of Invention
A primary objective of embodiments of the present application is to provide a method, an apparatus, a device, a storage medium, and a program product for detecting a traffic incident, which can identify a type of the traffic incident more accurately, so as to reduce loss caused by the incident timely and effectively.
In a first aspect, an embodiment of the present application provides a method for detecting a traffic event, including:
acquiring perception data in a target road area, wherein the perception data comprises at least one of the following items: the driving track of each vehicle in the target road area and the traffic flow data of the target road area;
processing the perception data through at least one target detection model in a plurality of detection models to obtain a detection result corresponding to each detection model;
and determining the type of the target traffic event in the target road area according to each detection result.
Optionally, the processing the sensing data through at least one target detection model in a plurality of detection models to obtain a detection result corresponding to each detection model includes:
determining at least one target detection model for processing the perception data from the plurality of detection models according to the perception data; the detection models comprise a track abnormity detection model and a traffic flow abnormity detection model;
and obtaining detection results corresponding to the perception data through corresponding target detection models respectively according to the perception data.
Optionally, obtaining detection results corresponding to the sensing data through corresponding target detection models respectively according to the sensing data, including:
if the perception data comprise the driving tracks, inputting the driving tracks of all vehicles into a track abnormity detection model to obtain detection results corresponding to the driving tracks; the detection result corresponding to the driving track comprises the traffic event type of abnormal overspeed and/or abnormal lane change caused by the traffic event;
if the perception data comprises traffic flow data, inputting the traffic flow data into a traffic flow abnormity detection model to obtain a detection result corresponding to the traffic flow data; the detection result corresponding to the traffic flow data comprises the traffic event type of traffic flow abnormity caused by the traffic event.
Optionally, the method further includes:
if the perception data further comprises video stream data and the detection models further comprise video stream detection models, inputting the video stream data into the video stream detection models to obtain detection results corresponding to the video stream data; the detection result corresponding to the video stream data is a traffic event type;
the type of the target traffic event is determined by screening detection results corresponding to all perception data.
Optionally, the method further includes:
acquiring a traffic event type corresponding to each road area in a plurality of road areas with historical traffic events;
obtaining historical vehicle running track points under different cameras in the same road area, and connecting the same historical vehicle running track points under different cameras in series to generate historical running tracks of the historical vehicles;
and constructing a training sample according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area, and training a network model to obtain the track abnormity detection model.
Optionally, the training sample is constructed according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area, and the training network model includes:
aiming at each training sample in the same road area, determining whether overspeed behaviors and/or lane changing behaviors exist in each historical vehicle in the same road area through network model analysis;
determining whether abnormal overspeed and/or abnormal lane change behaviors exist in the same road area according to the vehicle proportion of overspeed behaviors and/or lane change behaviors in the same road area;
if the abnormal behavior exists, predicting to obtain a traffic event type corresponding to the abnormal behavior through a network model according to the historical driving tracks of all historical vehicles in the same road area;
and adjusting network model parameters according to the traffic incident types corresponding to the same road area and the predicted traffic incident types.
Optionally, the method further includes:
if the reported event data is received, generating a related event related to the reported event data according to the reported event data, wherein the related event is used for representing a related event caused by a traffic event type matched with the reported event data;
determining the type of the target traffic event in the target road area according to each detection result, wherein the determining comprises the following steps:
and obtaining the type of the target traffic event through an event analysis model according to the associated event and each detection result.
In a second aspect, an embodiment of the present application further provides a device for detecting a traffic event, where the device includes:
a data acquisition module, configured to acquire perception data in a target road region, where the perception data includes at least one of: the driving track of each vehicle in the target road area and the traffic flow data of the target road area;
the first detection module is used for processing the perception data through at least one target detection model in a plurality of detection models to obtain detection results corresponding to the detection models;
and the second detection model is used for determining the type of the target traffic event in the target road area according to each detection result.
In a third aspect, an embodiment of the present application provides an electronic device, including:
at least one processor; and
a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to cause the electronic device to perform the method of any of the above aspects.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when a processor executes the computer-executable instructions, the method according to any one of the above aspects is implemented.
In a fifth aspect, the present application provides a computer program product, including a computer program, which when executed by a processor, implements the method of any one of the above aspects.
The method, the device, the equipment, the storage medium and the program product for detecting the traffic incident provided by the embodiment of the application can be used for acquiring perception data in a target road area, wherein the perception data comprises at least one of the following items: driving tracks of vehicles in the target road area and traffic flow data of the target road area; further, the sensing data is processed through at least one target detection model in the multiple detection models, and detection results corresponding to the detection models are obtained; further, determining the type of the target traffic event in the target road area according to each detection result. According to the method and the device, at least one of the driving track and the traffic flow data can be adopted, the initial detection analysis is carried out based on the corresponding detection model, the final detection result (namely the target traffic event type) is determined based on the initial detection analysis result (namely the detection result), therefore, compared with the detection means of the driving track and the traffic flow data to the traffic event type, the occurrence of the event is identified only by means of the change of the traffic flow parameters, the traffic event type can be identified more accurately, the occurrence of the event is not only identified, and further the loss caused by the event can be timely and effectively reduced.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic view of a scene of a method for detecting a traffic event according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a method for detecting a traffic event according to an embodiment of the present disclosure;
fig. 3 is a schematic flow chart illustrating a method for detecting a traffic event according to another embodiment of the present application;
fig. 4 is a schematic flow chart illustrating a method for detecting a traffic event according to yet another embodiment of the present application;
fig. 5 is a schematic structural diagram of a device for detecting a traffic event according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Specific embodiments of the present application have been shown by way of example in the drawings and will be described in more detail below. These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims of the present application and in the above-described drawings (if any) are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of including other sequential examples in addition to those illustrated or described. 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 apparatus 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 elements not expressly listed or inherent to such process, method, article, or apparatus. At present, the automatic traffic incident detection method generally finds traffic incidents by the change of traffic flow parameters, but the traffic flow similarity of various traffic incidents is high, and specific traffic incident types cannot be identified. Therefore, the prior art cannot accurately identify the type of the traffic incident, and further cannot timely and effectively reduce the loss caused by the incident.
In order to solve the above problems, the inventive concept of the present application is: the method comprises the steps of processing at least one item of sensing data in the driving track and traffic flow data by adopting a plurality of detection models to obtain a primary detection analysis result, predicting a detection result corresponding to the corresponding sensing data to serve as the primary detection analysis result, and then determining the type of a target traffic event based on each primary detection analysis result, so that not only can an incident be identified, but also the type of the traffic event of the incident can be identified more accurately, and further, a specific traffic event can be issued timely and effectively to reduce the loss caused by the incident.
Fig. 1 is a scene schematic diagram of a method for detecting a traffic event according to an embodiment of the present application. The execution main body of the detection method of the traffic incident can be a terminal device or a server, taking the server as an example, the method can monitor the traffic road conditions of a certain or some road areas in real time, and if abnormal traffic road conditions occur in the certain or some road areas, whether the reason causing the abnormal traffic road conditions is the occurrence of the traffic incident and the type of the traffic incident occurring can be timely and accurately detected.
For example, a sensing device is installed on a road side of a certain road area or certain road areas, for example, a camera is installed on a certain road section in the certain road area or certain road areas, sensing data of the certain road area or certain road areas can be obtained through shooting by the camera, for example, vehicle sensing data, traffic flow data and video sensing data are obtained based on a video detection technology, a vehicle track is restored through the vehicle sensing data, and a driving track is generated; data can also be reported through related equipment, such as an accident vehicle or a vehicle passing through a traffic incident scene; based on at least one of the driving track, traffic flow data, video sensing data and reported data, the traffic incident types are respectively detected and analyzed through at least one target detection model in the multiple detection models, and then the target traffic incident types which accord with the current road condition scene are screened out through the incident analysis model. And the correlation events are combined with the traffic event types respectively detected by the plurality of detection models to carry out comprehensive analysis, so that the target traffic event type is obtained.
Specifically, the sensing data comprises at least one item of data, one detection model is used for processing one item of sensing data, the detection model for processing the corresponding data is selected based on the sensing data, and then the sensing data is respectively input into the corresponding detection models to obtain corresponding detection results. Thus, the detection result may include one or more. And when the number of the traffic events is more than one, the detection result is screened to obtain the type of the target traffic event.
Therefore, different detection modes are provided according to different acquired sensing data, diversification is achieved, and traffic incidents can be comprehensively and accurately detected.
The technical solution of the present application will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 2 is a schematic flowchart of a method for detecting a traffic event according to an embodiment of the present disclosure. The embodiment can be applied to any device capable of realizing detection, such as a terminal device or a server. As shown in fig. 2, the method may include:
s201, obtaining perception data in the target road area. Wherein the perception data comprises at least one of: the running track of each vehicle in the target road area and the traffic flow data of the target road area.
In the embodiment of the application, the sensing device installed on the target road area can acquire the sensing data in the road area, for example, the sensing device can acquire the vehicle sensing data, the traffic flow data and the like of the road area by shooting through a camera. The vehicle track can be restored through the vehicle perception data, and the driving track is generated.
S202, processing the sensing data through at least one target detection model in the plurality of detection models to obtain detection results corresponding to the detection models.
In the embodiment of the present application, the detection result may be used to indicate the type of the predicted traffic event, such as a traffic accident, a spill, an abnormal parking, a construction, and the like.
Specifically, different sensing data are processed through a plurality of trained detection models, a detection result corresponding to the sensing data is predicted, for example, the sensing data comprises a driving track, the driving track is input into the detection models for processing the driving track in the plurality of detection models, and a traffic event type corresponding to the driving track is predicted; and if the perception data comprises traffic flow data, inputting the traffic flow data into a plurality of detection models for processing the traffic flow data, and predicting the type of the traffic event corresponding to the traffic flow data. And if the perception data comprises the driving track and the traffic flow data, inputting the two data into corresponding detection models respectively, and predicting the traffic event type corresponding to the driving track and the traffic event type corresponding to the traffic flow data respectively. Therefore, the type of the traffic event can be predicted by a single detection model or a plurality of detection models, and if a plurality of types of the traffic events are predicted, the final type of the traffic event, namely the target type of the traffic event, can be determined by screening.
S203, determining the type of the target traffic event in the target road area according to each detection result.
In the embodiment of the application, if the detection result is a traffic event type, the traffic event type can be used as a target traffic event type; and if the detection result is at least two traffic event types, determining the target traffic event type through weight calculation or model reasoning. Therefore, the traffic incident type can be accurately identified by fusing the detection results of one detection model or a plurality of detection models.
Therefore, the embodiment of the application can obtain the perception data in the target road area, where the perception data includes at least one of the following: the driving track of each vehicle in the target road area and the traffic flow data of the target road area; further, processing the perception data through at least one target detection model in a plurality of detection models to obtain detection results corresponding to the detection models; further, determining the type of the target traffic event in the target road area according to each detection result. According to the method and the device, at least one of the driving track and the traffic flow data can be adopted, the initial detection analysis is carried out based on the corresponding detection model, the final detection result (namely the target traffic event type) is determined based on the initial detection analysis result (namely the detection result), therefore, compared with the detection means of the driving track and the traffic flow data to the traffic event type, the traffic event type can be accurately identified only by means of the change of traffic flow parameters, the occurrence of the traffic event is not only identified, and then the specific traffic event can be timely and effectively issued, so that the loss caused by the event is reduced.
Optionally, the processing the sensing data through at least one target detection model in a plurality of detection models to obtain a detection result corresponding to each detection model includes:
a1, determining at least one target detection model for processing the perception data from the plurality of detection models according to the perception data; the detection models comprise a track abnormity detection model and a traffic flow abnormity detection model;
and a2, obtaining detection results corresponding to all the perception data through corresponding target detection models according to all the perception data.
In the embodiment of the application, based on the sensing data, the detection model for processing the corresponding data is selected, and then the sensing data is respectively input into the corresponding detection models to obtain corresponding detection results. Therefore, the detection result may include one or two. And when the number of the detection results is two, the detection results are screened to obtain the target traffic incident type.
Specifically, one detection model is used for processing one perception data, and a target detection model for current detection can be determined from a plurality of detection models according to the perception data including driving track and/or traffic flow data. The plurality of detection models comprise a track abnormity detection model and a traffic flow abnormity detection model, the track abnormity detection model is used for processing the driving track, and the traffic flow abnormity detection model is used for processing the traffic flow data. Inputting the running track into a track abnormity detection model, and predicting a detection result corresponding to the running track; and inputting the traffic flow data into a traffic flow abnormity detection model, and predicting a detection result corresponding to the traffic flow data.
Therefore, different detection modes exist according to different acquired sensing data, diversification is achieved, and traffic incidents can be comprehensively and accurately detected.
Optionally, obtaining detection results corresponding to the sensing data through corresponding target detection models respectively according to the sensing data, including:
b1, if the sensing data comprise the driving tracks, inputting the driving tracks of all vehicles into a track abnormity detection model to obtain detection results corresponding to the driving tracks; the detection result corresponding to the driving track comprises the type of the traffic event which causes abnormal overspeed and/or abnormal lane change due to the traffic event;
b2, if the perception data comprise traffic flow data, inputting the traffic flow data into a traffic flow abnormity detection model to obtain a detection result corresponding to the traffic flow data; the detection result corresponding to the traffic flow data comprises a traffic event type of traffic flow abnormity caused by a traffic event;
wherein the traffic event type includes at least one of: traffic accidents, object throwing, abnormal parking and construction. The traffic event type may also include a normal status type where no actual traffic event has occurred.
In the embodiment of the application, different sensing data are input into different detection models, and the different detection models are used for detecting the traffic event type according to the different sensing data. By adopting the track anomaly detection model and/or the traffic flow anomaly detection model, the detection results obtained by prediction are respectively output, one model can accurately detect the type of the traffic incident, and when the two models are detected, the two models are comprehensively analyzed based on the two detection results, so that the type of the traffic incident is accurately detected.
Specifically, referring to fig. 3, fig. 3 is a schematic flow chart of a method for detecting a traffic event according to another embodiment of the present application. And if the perception data comprises the driving tracks, inputting the driving tracks of the vehicles into a trained track abnormity detection model, and predicting whether abnormal behaviors exist in the target road area based on the driving tracks of the vehicles and the type of the traffic incident pointed by the abnormal behaviors, namely, the abnormal behaviors of the vehicles in the target road area caused by the traffic incident of which type. If the perception data comprises traffic flow data, inputting the traffic flow data of the target road area into a trained traffic flow abnormity detection model, predicting whether the traffic flow data in the target road area is abnormal or not, and predicting the type of a traffic event pointed by the traffic flow abnormity, namely the traffic flow abnormity caused by the traffic event of which type. And then, comprehensively analyzing based on the two detection results, and accurately detecting the type of the target traffic event.
Optionally, the method further includes:
step c1, acquiring traffic incident types corresponding to all road areas in a plurality of road areas with historical traffic incidents;
step c2, obtaining historical vehicle running track points under different cameras in the same road area, and connecting the same historical vehicle running track points under different cameras in series to generate historical running tracks of the historical vehicles;
and c3, constructing a training sample according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area, and training a network model to obtain the track abnormity detection model.
In this embodiment of the application, the modeling process of the track anomaly detection model may be: firstly, a training sample is obtained, wherein the sample is obtained based on a restored track, namely videos of different road sections can be shot through a plurality of cameras arranged in a target road area, and based on vehicle perception data detected by videos, the same vehicle under one camera uses the same id, and the same vehicles between different cameras are different in possible id, so that vehicle running track points under different cameras are connected in series according to the same vehicle characteristics (the vehicle characteristics comprise id, license plate, color, vehicle body appearance and the like) under different cameras to generate a corresponding running track; and then, training a network model according to the traffic event type of the driving track and the corresponding label to obtain a track abnormity detection model.
Because the track can intuitively reflect the change caused by the traffic incident, the traffic incident occurring in the target road area and the traffic incident type to which the traffic incident belongs can be effectively detected under the condition that the acquired perception data is the driving track by taking the track and the marked traffic incident type as samples.
Optionally, the training sample is constructed according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area, and the training network model includes:
d1, aiming at each training sample in the same road area, determining whether each historical vehicle in the same road area has overspeed behavior and/or lane changing behavior through network model analysis;
d2, determining whether abnormal behaviors of abnormal overspeed and/or abnormal lane change exist in the same road area according to the vehicle proportion of overspeed behavior and/or lane change behavior in the same road area;
d3, if the abnormal behaviors exist, predicting the traffic event type corresponding to the abnormal behaviors through a network model according to the historical driving tracks of all historical vehicles in the same road area;
and d4, adjusting network model parameters according to the traffic incident types corresponding to the same road area and the predicted traffic incident types.
In the embodiment of the application, the abnormal lane change and/or abnormal overspeed behaviors are all corresponding classification models constructed by taking historical traffic events as samples. Each sample may include a driving track of a vehicle and a corresponding traffic event type, and whether the vehicle generating the track has overspeed or lane change behavior is analyzed for a single track, and then whether frequent overspeed or lane change behavior exists is statistically analyzed for multiple tracks (for example, all tracks of traffic events of the same type occur at close time points in the same road area) in the same road area (that is, the same interval), for example, if a vehicle occupancy ratio of overspeed and/or lane change behavior in the same road area is greater than a predefined occupancy threshold (for example, 0.95), it is determined that abnormal overspeed and/or abnormal lane change behavior exists in the road area, and if it is determined that abnormal behavior exists, the set of samples is adopted, and the traffic event type corresponding to the abnormal behavior is obtained through network model prediction.
In order to enable the constructed network model to have high precision and stability, the abnormal behaviors of abnormal overspeed and/or abnormal lane change in the road area are analyzed through the network model, instead of the overspeed behaviors and/or the lane change behaviors of individual vehicles, after the abnormal behaviors are determined, the abnormal behaviors are predicted to be corresponding traffic event types through the network model, parameters of the network model are adjusted based on the labels and the prediction results, and iteration is carried out continuously until the training is finished.
Optionally, the method further includes:
acquiring traffic event types and traffic flow data corresponding to each road area in a plurality of road areas with historical traffic events; the traffic flow data comprises the speed, the flow, the upstream and downstream speed difference and the upstream and downstream flow difference of the traffic flow;
and constructing a training sample according to the traffic event type and the traffic flow data corresponding to each road area, and training a network model to obtain the traffic flow abnormity detection model.
In the embodiment of the application, the traffic flow changes are caused by and/or related to the occurrence of the event, and if the detected traffic flow parameters change more than a certain degree relative to time or space, the occurrence of the event can be indicated. Specifically, the macroscopic speed and flow data of the traffic section can be sensed through the sensing device, wherein the speed can comprise the speed of the traffic flow, the difference between the upstream speed and the downstream speed; the flow rate may include a flow rate of a traffic flow, an upstream and downstream flow rate difference.
The traffic flow data can be processed by adopting a comparison method, a statistical method, a traffic and theoretical model method, a low-flow event detection algorithm and a classification detection algorithm, and whether a traffic event occurs or not and the type of the occurring traffic event are judged. Taking a classification model corresponding to a classification detection algorithm as an example, a certain type of historical events and non-events are respectively positive and negative samples, and the speed, flow, upstream and downstream speed difference, upstream and downstream flow difference and the like of traffic flow are taken as characteristics, machine learning technologies such as logistic regression and eXtreme Gradient Boosting (XGB) are introduced to be used as the classification model, and traffic flow abnormity and the type of traffic events causing the abnormal events are identified. The traffic event type can be accurately predicted through the traffic flow abnormity detection model obtained through training.
Optionally, the method further includes:
if the perception data further comprises video stream data and the detection models further comprise video stream detection models, inputting the video stream data into the video stream detection models to obtain detection results corresponding to the video stream data;
the detection result corresponding to the video stream data is a traffic event type; the type of the target traffic event is determined by screening detection results corresponding to all perception data.
In the embodiment of the application, a video detection technology is added, and if the sensing data comprises video stream data, the video stream is output and input into a video stream detection model in a plurality of detection models, and the traffic event type corresponding to the video stream is predicted.
Specifically, the detection based on the video stream detection model is combined with traffic flow detection and/or track detection, compared with the detection method which only uses a video image processing mode in the prior art, namely, the detection method which uses an image processing technology to find the traffic event of an event vehicle, the traffic event is actually seen, but the identification accuracy is low and the direct use is difficult to achieve under a complex scene (such as sheltered, sight (weather) and other objective reasons, for example, construction, the phenomenon that pedestrians enter a motor vehicle lane, construction vehicles (vehicle traffic accidents, faults) and the like) depending on the precision of the image technology, experts are required to perform manual secondary confirmation in most cases, and the combination of the video stream detection and the traffic flow detection and/or the track detection is more accurate and does not need manual verification.
Optionally, the method further includes:
acquiring traffic event types and video stream data corresponding to each road area in a plurality of road areas with historical traffic events; the video stream data comprises video images of traffic events occurring in various road areas;
and constructing a training sample according to the traffic event type and the video image corresponding to each road area, training a network model, and obtaining the video stream detection model.
In the embodiment of the application, the classification model is established by calibrating different types of event video images and then performing deep learning, such as a CNN model. In order to avoid frequent alarm of the same traffic incident, alarm fusion can be performed through time and space based on the same traffic incident occurring in the same space in the same time period, corresponding types of incidents are output for alarm, the type of the alarm traffic incident can be used as a label of a training sample, a video image corresponding to the traffic incident occurs as a feature of the training sample, a network model is trained, a classification model is obtained, and the classification model is used as a video stream detection model.
Optionally, the method further includes:
step e1, if the reported event data is received, generating a related event related to the reported event data according to the reported event data, wherein the related event is used for representing a related event caused by a traffic event type matched with the reported event data;
step e2, determining the type of the target traffic event in the target road area according to each detection result, wherein the step comprises the following steps:
and e3, obtaining the type of the target traffic incident through an incident analysis model according to the associated incident and each detection result.
In the embodiment of the application, the accessed reported data can be fused according to time and space by accessing multi-source data including but not limited to accident data, construction data and the like, and the events of different types generate the associated events. For example, a traffic accident occurs on a certain road segment at a certain time point, and the lane change behavior is caused by the accident; and combining the associated event with at least one item of the driving track, the traffic flow data and the video stream data to serve as characteristic input of an event analysis model, detecting the type of the traffic event in a mode of combining various characteristics, and outputting an emergency event if a traffic accident, a sprinkled object, an abnormal parking and the like are detected. The emergency event includes traffic accident, throwing thing, abnormal parking, etc.
Therefore, the event types are comprehensively analyzed by combining the reported data; the reported data is direct and has a large occupation weight, and the detection results obtained by each detection model are combined, so that the errors of the detection models can be reduced, and the finally detected target traffic incident types are more accurate.
Optionally, the method further includes:
acquiring historical reported event data corresponding to each road area in a plurality of road areas with historical traffic events, generating historical associated events associated with the historical reported event data according to the historical reported event data,
acquiring each historical detection result and traffic incident type corresponding to each road area;
and constructing a training sample according to the historical associated events, the historical detection results corresponding to the road areas respectively and the traffic event types corresponding to the road areas respectively, and training a network model to obtain an event analysis model.
Specifically, referring to fig. 4, fig. 4 is a schematic flow chart of a method for detecting a traffic event according to yet another embodiment of the present application. Reporting and fusing accident reporting data and construction reporting data to determine a first traffic incident type; performing track restoration through track perception, judging whether abnormal overspeed and/or abnormal lane change exist or not, and predicting a second traffic event type causing the abnormal overspeed and/or the abnormal lane change; acquiring speed and flow to form traffic flow data through sensing equipment, judging whether the traffic flow is abnormal or not through a statistical model and/or a classification model, and predicting a third traffic event type causing the traffic flow to be abnormal; the method comprises the steps of determining the type of a traffic incident through video perception, for example, an accident (namely, a traffic accident), construction, abnormal parking, a throwing object and the like, giving an alarm, performing space-time fusion on the alarm incident, and generating alarm information, wherein the alarm information comprises a fourth traffic incident type.
The first traffic event type, the second traffic event type, the third traffic event type and the fourth traffic event type are input into an event analysis model, a target traffic event type is combined and identified, when the target traffic event type comprises a traffic accident, an abnormal parking, a spill object and the like, the target traffic event type is determined to be an emergency and is reported to relevant departments, a specific traffic event is issued timely and effectively, and loss caused by the event is reduced.
Therefore, the method and the device have the advantages that the characteristics of video perception, track abnormity, traffic flow abnormity, reporting alarm and the like are integrated, the accuracy and the robustness of the model are enhanced, and the subdivision events can be accurately identified. The traffic incident is automatically detected based on the combined model (which can be referred to as an incident analysis model), multi-source data is fused, the combined model is constructed, the problems of low precision and poor robustness of the existing traffic incident detection method are effectively solved, the labor cost is greatly reduced, and the loss caused by the incident is reduced.
Corresponding to the foregoing traffic incident detection method, an embodiment of the present application provides a device for detecting a traffic incident, and fig. 5 is a schematic structural diagram of the device for detecting a traffic incident provided in the embodiment of the present application, where the device includes:
a data obtaining module 501, configured to obtain perception data in a target road region, where the perception data includes at least one of: the driving track of each vehicle in the target road area and the traffic flow data of the target road area;
a first detection module 502, configured to process the sensing data through at least one target detection model in multiple detection models to obtain a detection result corresponding to each detection model;
and the second detection model 503 is configured to determine the type of the target traffic event in the target road area according to each detection result.
The detection apparatus for a traffic incident provided in the embodiment of the present application may be used to implement the technical solutions in the embodiments shown in fig. 1 to fig. 4, which have similar implementation principles and technical effects, and this embodiment is not described herein again.
In the embodiment of the application, the data acquisition module 501 may acquire the sensing data in the target road area, where the sensing data includes at least one of the following: driving tracks of vehicles in the target road area and traffic flow data of the target road area; further, the first detection module 502 processes the sensing data through at least one target detection model of the plurality of detection models to obtain a detection result corresponding to each detection model; further, the type of the target traffic event in the target road area is determined according to each detection result through the second detection model 503. According to the method and the device, at least one of the driving track and the traffic flow data can be adopted, the initial detection analysis is carried out based on the corresponding detection model, the final detection result (namely the target traffic event type) is determined based on the initial detection analysis result (namely the detection result), therefore, compared with the detection means of the driving track and the traffic flow data to the traffic event type, the traffic event type can be accurately identified only by means of the change of traffic flow parameters, the occurrence of the traffic event is not only identified, and then the specific traffic event can be timely and accurately provided, so that the loss caused by the event is reduced.
Optionally, the first detection model is specifically configured to:
determining at least one target detection model for processing the perception data from the plurality of detection models according to the perception data; the detection models comprise a track abnormity detection model and a traffic flow abnormity detection model;
and obtaining detection results corresponding to the perception data through corresponding target detection models respectively according to the perception data.
Optionally, the first detection module is specifically configured to:
when the perception data comprise the driving tracks, inputting the driving tracks of all vehicles into a track abnormity detection model to obtain detection results corresponding to the driving tracks; the detection result corresponding to the driving track comprises the type of the traffic event which causes abnormal overspeed and/or abnormal lane change due to the traffic event;
when the perception data comprises traffic flow data, inputting the traffic flow data into a traffic flow abnormity detection model to obtain a detection result corresponding to the traffic flow data; the detection result corresponding to the traffic flow data comprises the traffic event type of traffic flow abnormity caused by the traffic event;
wherein the traffic event type includes at least one of: traffic accidents, sprinkled objects, abnormal parking and construction.
Optionally, the detection apparatus for a traffic event further includes a third detection module; a third detection module to:
when the perception data further comprises video stream data and the plurality of detection models further comprise video stream detection models, inputting the video stream data into the video stream detection models to obtain detection results corresponding to the video stream data;
the detection result corresponding to the video stream data is a traffic event type; the type of the target traffic event is determined by screening detection results corresponding to all perception data.
Optionally, the apparatus for detecting a traffic event further includes: a first training module; a first training module to:
acquiring a traffic event type corresponding to each road area in a plurality of road areas with historical traffic events;
obtaining historical vehicle running track points under different cameras in the same road area, and connecting the same historical vehicle running track points under different cameras in series to generate historical running tracks of the historical vehicles;
and constructing a training sample according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area, and training a network model to obtain the track anomaly detection model.
Optionally, the first training module is specifically configured to:
aiming at each training sample in the same road area, determining whether overspeed behaviors and/or lane changing behaviors exist in each historical vehicle in the same road area through network model analysis;
determining whether abnormal overspeed and/or abnormal lane change behaviors exist in the same road area according to the vehicle proportion of overspeed behaviors and/or lane change behaviors in the same road area;
if the abnormal behavior exists, predicting to obtain a traffic event type corresponding to the abnormal behavior through a network model according to the historical driving tracks of all historical vehicles in the same road area;
and adjusting network model parameters according to the traffic incident types corresponding to the same road area and the predicted traffic incident types.
Optionally, the device for detecting a traffic event further includes: a second training module; a second training module to:
acquiring traffic event types and traffic flow data corresponding to each road area in a plurality of road areas with historical traffic events; the traffic flow data comprises the speed, the flow, the upstream and downstream speed difference and the upstream and downstream flow difference of the traffic flow;
and constructing a training sample according to the traffic event type and the traffic flow data corresponding to each road area, and training a network model to obtain the traffic flow abnormity detection model.
Optionally, the device for detecting a traffic event further includes: a third training module; a third training module to:
acquiring traffic event types and video stream data corresponding to each road area in a plurality of road areas with historical traffic events; the video stream data comprises video images of traffic events in various road areas;
and constructing a training sample according to the traffic event type and the video image corresponding to each road area, training a network model, and obtaining the video stream detection model.
Optionally, the apparatus for detecting a traffic event further includes a fourth detecting module; a fourth detection module to: and when the reported event data is received, generating a related event related to the reported event data according to the reported event data, wherein the related event is used for representing a related event caused by the traffic event type matched with the reported event data.
Optionally, the second detection module is specifically configured to: and obtaining the type of the target traffic event through an event analysis model according to the associated event and each detection result.
Optionally, the device for detecting a traffic event further includes: a fourth training module; a fourth training module to:
acquiring historical reported event data corresponding to each road area in a plurality of road areas with historical traffic events, generating historical associated events associated with the historical reported event data according to the historical reported event data,
acquiring each historical detection result and traffic incident type respectively corresponding to each road area;
and constructing a training sample according to the historical associated events, the historical detection results corresponding to the road areas respectively and the traffic event types corresponding to the road areas respectively, and training a network model to obtain an event analysis model.
Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 6, the electronic device of the present embodiment may include:
at least one processor 601; and
a memory 602 communicatively coupled to the at least one processor;
wherein the memory 602 stores instructions executable by the at least one processor 601, the instructions being executable by the at least one processor 601 to cause the electronic device to perform the method according to any of the embodiments described above.
Alternatively, the memory 602 may be separate or integrated with the processor 601.
For the implementation principle and the technical effect of the electronic device provided by this embodiment, reference may be made to the foregoing embodiments, which are not described herein again.
The embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored, and when a processor executes the computer-executable instructions, the method described in any one of the foregoing embodiments is implemented.
The present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the method described in any of the foregoing embodiments.
In the technical scheme of the application, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the related user data and other information all accord with the regulations of related laws and regulations and do not violate the good customs of the public order.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice, for example, a plurality of modules may be combined or integrated into another system, or some features may be omitted, or not executed.
The integrated module implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a processor to execute some steps of the methods described in the embodiments of the present application.
It should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in the incorporated application may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor. The memory may comprise a high-speed RAM memory, and may further comprise a non-volatile storage NVM, such as at least one disk memory, and may also be a usb disk, a removable hard disk, a read-only memory, a magnetic or optical disk, etc.
The storage medium may be implemented by any type or combination of volatile and non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the storage medium may reside as discrete components in an electronic device or host device.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a … …" does not exclude the presence of another identical element in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present application are merely for description, and do not represent the advantages and disadvantages of the embodiments.
Through the description of the foregoing embodiments, it is clear to those skilled in the art that the method of the foregoing embodiments may be implemented by software plus a necessary general hardware platform, and certainly may also be implemented by hardware, but in many cases, the former is a better implementation. Based on such understanding, the technical solutions of the present application or portions thereof that contribute to the prior art may be embodied in the form of a software product, where the computer software product is stored in a storage medium (such as a ROM/RAM, a magnetic disk, and an optical disk), and includes several instructions for enabling a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are included in the scope of the present application.
Claims (11)
1. A method of detecting a traffic event, the method comprising:
acquiring perception data in a target road area, wherein the perception data comprises at least one of the following items: the driving track of each vehicle in the target road area and the traffic flow data of the target road area;
processing the perception data through at least one target detection model in a plurality of detection models to obtain a detection result corresponding to each detection model;
and determining the type of the target traffic event in the target road area according to each detection result.
2. The method according to claim 1, wherein processing the perception data through at least one target detection model of a plurality of detection models to obtain a detection result corresponding to each detection model comprises:
determining at least one target detection model for processing the perception data from the plurality of detection models according to the perception data; the detection models comprise a track abnormity detection model and a traffic flow abnormity detection model;
and obtaining detection results corresponding to the perception data through corresponding target detection models respectively according to the perception data.
3. The method of claim 2, wherein obtaining the detection result corresponding to each item of sensing data through the corresponding target detection model according to each item of sensing data comprises:
if the perception data comprise the driving tracks, inputting the driving tracks of all vehicles into a track abnormity detection model to obtain detection results corresponding to the driving tracks; the detection result corresponding to the driving track comprises the type of the traffic event which causes abnormal overspeed and/or abnormal lane change due to the traffic event;
if the perception data comprise traffic flow data, inputting the traffic flow data into a traffic flow abnormity detection model to obtain a detection result corresponding to the traffic flow data; the detection result corresponding to the traffic flow data comprises the traffic event type of traffic flow abnormity caused by the traffic event.
4. The method of claim 3, further comprising:
if the perception data further comprises video stream data and the detection models further comprise video stream detection models, inputting the video stream data into the video stream detection models to obtain detection results corresponding to the video stream data;
the detection result corresponding to the video stream data is a traffic event type; the type of the target traffic event is determined by screening detection results corresponding to all perception data.
5. The method according to any one of claims 2-4, further comprising:
acquiring a traffic event type corresponding to each road area in a plurality of road areas with historical traffic events;
obtaining historical vehicle running track points under different cameras in the same road area, and connecting the same historical vehicle running track points under different cameras in series to generate historical running tracks of the historical vehicles;
and constructing a training sample according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area, and training a network model to obtain the track abnormity detection model.
6. The method of claim 5, wherein training the network model according to the traffic event type corresponding to each road area and the historical driving track of each historical vehicle in each road area and constructing a training sample comprises:
aiming at each training sample in the same road area, determining whether each historical vehicle in the same road area has overspeed behavior and/or lane changing behavior through network model analysis;
determining whether abnormal overspeed and/or abnormal lane change behaviors exist in the same road area according to the vehicle proportion of overspeed behaviors and/or lane change behaviors in the same road area;
if the abnormal behavior exists, predicting to obtain a traffic event type corresponding to the abnormal behavior through a network model according to the historical driving tracks of all historical vehicles in the same road area;
and adjusting network model parameters according to the traffic incident types corresponding to the same road area and the predicted traffic incident types.
7. The method according to any one of claims 1-4, further comprising:
if the reported event data is received, generating a related event related to the reported event data according to the reported event data, wherein the related event is used for representing a related event caused by a traffic event type matched with the reported event data;
determining the type of the target traffic event in the target road area according to each detection result, wherein the determining comprises the following steps:
and obtaining the type of the target traffic event through an event analysis model according to the associated event and each detection result.
8. An apparatus for detecting a traffic event, the apparatus comprising:
a data acquisition module, configured to acquire perception data in a target road region, where the perception data includes at least one of: the driving track of each vehicle in the target road area and the traffic flow data of the target road area;
the first detection module is used for processing the perception data through at least one target detection model in a plurality of detection models to obtain detection results corresponding to the detection models;
and the second detection model is used for determining the type of the target traffic event in the target road area according to each detection result.
9. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor to cause the electronic device to perform the method of any of claims 1-7.
10. A computer-readable storage medium having computer-executable instructions stored thereon which, when executed by a processor, implement the method of any one of claims 1-7.
11. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210989463.XA CN115359658B (en) | 2022-08-17 | 2022-08-17 | Traffic event detection method, device, equipment, storage medium and program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210989463.XA CN115359658B (en) | 2022-08-17 | 2022-08-17 | Traffic event detection method, device, equipment, storage medium and program product |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115359658A true CN115359658A (en) | 2022-11-18 |
| CN115359658B CN115359658B (en) | 2024-06-11 |
Family
ID=84002207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210989463.XA Active CN115359658B (en) | 2022-08-17 | 2022-08-17 | Traffic event detection method, device, equipment, storage medium and program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115359658B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115797883A (en) * | 2023-02-06 | 2023-03-14 | 以萨技术股份有限公司 | Data processing system for determining abnormal event |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110473402A (en) * | 2019-08-20 | 2019-11-19 | 河北德冠隆电子科技有限公司 | A kind of accident detection early warning system based on target abnormal behaviour trajectory analysis |
| CN111402612A (en) * | 2019-01-03 | 2020-07-10 | 北京嘀嘀无限科技发展有限公司 | Traffic incident notification method and device |
| CN113688652A (en) * | 2020-05-18 | 2021-11-23 | 魔门塔(苏州)科技有限公司 | Method and device for processing abnormal driving behaviors |
| US20220051032A1 (en) * | 2020-12-11 | 2022-02-17 | Beijing Baidu Netcom Science Technology Co., Ltd. | Road event detection method, apparatus, device and storage medium |
| CN114446042A (en) * | 2020-11-04 | 2022-05-06 | 腾讯科技（深圳）有限公司 | Method, device, equipment and storage medium for early warning of traffic accidents |
| JP7092958B1 (en) * | 2022-03-09 | 2022-06-28 | あいおいニッセイ同和損害保険株式会社 | Information processing methods, information processing devices, and programs |
-
2022
- 2022-08-17 CN CN202210989463.XA patent/CN115359658B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111402612A (en) * | 2019-01-03 | 2020-07-10 | 北京嘀嘀无限科技发展有限公司 | Traffic incident notification method and device |
| CN110473402A (en) * | 2019-08-20 | 2019-11-19 | 河北德冠隆电子科技有限公司 | A kind of accident detection early warning system based on target abnormal behaviour trajectory analysis |
| CN113688652A (en) * | 2020-05-18 | 2021-11-23 | 魔门塔(苏州)科技有限公司 | Method and device for processing abnormal driving behaviors |
| CN114446042A (en) * | 2020-11-04 | 2022-05-06 | 腾讯科技（深圳）有限公司 | Method, device, equipment and storage medium for early warning of traffic accidents |
| US20220051032A1 (en) * | 2020-12-11 | 2022-02-17 | Beijing Baidu Netcom Science Technology Co., Ltd. | Road event detection method, apparatus, device and storage medium |
| JP7092958B1 (en) * | 2022-03-09 | 2022-06-28 | あいおいニッセイ同和損害保険株式会社 | Information processing methods, information processing devices, and programs |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115797883A (en) * | 2023-02-06 | 2023-03-14 | 以萨技术股份有限公司 | Data processing system for determining abnormal event |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115359658B (en) | 2024-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110428091B (en) | Risk identification method based on data analysis and related equipment | |
| US20200034958A1 (en) | Automatic Image Based Object Damage Assessment | |
| CN109190488B (en) | Front vehicle door opening detection method and device based on deep learning YOLOv3 algorithm | |
| Stadler et al. | A credibility assessment approach for scenario-based virtual testing of automated driving functions | |
| AU2021104243A4 (en) | Method of Pedestrian detection based on multi-layer feature fusion | |
| CN110930756A (en) | Method and device for judging roadside parking based on image and geomagnetic sensor | |
| CN115359658B (en) | Traffic event detection method, device, equipment, storage medium and program product | |
| CN111369792B (en) | Traffic incident analysis method and device and electronic equipment | |
| Kejriwal et al. | Vehicle detection and counting using deep learning basedYOLO and deep SORT algorithm for urban traffic management system | |
| CN106448161A (en) | Road monitoring method and road monitoring device | |
| CN114973211A (en) | Object identification method, device, equipment and storage medium | |
| CN113343880A (en) | Method, device, equipment and medium for identifying vehicle shielding number plate | |
| CN112800841A (en) | Pedestrian counting method, device and system and computer readable storage medium | |
| CN111667697B (en) | Abnormal vehicle identification method and device, and computer readable storage medium | |
| CN115731711B (en) | Automatic identification method and device for variable tidal lane applied to automatic driving | |
| CN114724357B (en) | Abnormal driving vehicle identification method and device and computer equipment | |
| CN114648718B (en) | Road occupation detection method, system, electronic equipment and readable storage medium | |
| CN116259173A (en) | D-S evidence theory-based multi-detection mode traffic event fusion method | |
| CN113902999A (en) | Tracking method, device, equipment and medium | |
| CN113592901A (en) | Image processing method, image processing device, computer equipment and computer readable storage medium | |
| CN112749239A (en) | Event map construction method and device and computing equipment | |
| Zhang et al. | Identification and screening of key traffic violations: based on the perspective of expressing driver’s accident risk | |
| CN111723604A (en) | Vehicle fake plate detection method and device | |
| Xu | Application of Bayes Theorem to Accident Analysis and Classification | |
| CN116758259B (en) | Highway asset information identification method and system |
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 |