CN115359443A - Traffic accident detection method and device, electronic device and storage medium - Google Patents

Traffic accident detection method and device, electronic device and storage medium Download PDF

Info

Publication number
CN115359443A
CN115359443A CN202211160293.0A CN202211160293A CN115359443A CN 115359443 A CN115359443 A CN 115359443A CN 202211160293 A CN202211160293 A CN 202211160293A CN 115359443 A CN115359443 A CN 115359443A
Authority
CN
China
Prior art keywords
traffic accident
vehicle
current road
detection
determining
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.)
Pending
Application number
CN202211160293.0A
Other languages
Chinese (zh)
Inventor
张上鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhidao Network Technology Beijing Co Ltd
Original Assignee
Zhidao Network Technology Beijing Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhidao Network Technology Beijing Co Ltd filed Critical Zhidao Network Technology Beijing Co Ltd
Priority to CN202211160293.0A priority Critical patent/CN115359443A/en
Publication of CN115359443A publication Critical patent/CN115359443A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
    • 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
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Social Psychology (AREA)
  • Psychiatry (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

The application discloses a traffic accident detection method and device, electronic equipment and a storage medium, wherein the method is executed by road side equipment and comprises the following steps: acquiring a road image acquired by road side equipment at a current road section and detecting the road image; determining the running speed of the current road section according to the road image detection result; determining a target vehicle set according to the running speed, and determining whether a traffic accident occurs in the current road section according to the target vehicle set, a road image detection result and a preset traffic accident detection condition; and if so, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system for confirmation. The method and the device determine the target vehicle which is possibly subjected to the traffic accident or is possibly influenced by the traffic accident based on the running speed of the current road section, further judge whether the traffic accident occurs on the current road section or not by combining the road image detection result and the traffic accident detection condition, greatly reduce the false detection rate, and can be suitable for detecting the traffic accidents without obvious image characteristics, such as scratching and rear-end collision.

Description

Traffic accident detection method and device, electronic device and storage medium
Technical Field
The present application relates to the field of traffic accident detection technologies, and in particular, to a traffic accident detection method and apparatus, an electronic device, and a storage medium.
Background
The intelligent traffic roadside monitoring system is developed more and more mature, most functions of the existing roadside monitoring system are limited to detection and display, and the detection data are not fully utilized to detect traffic accidents and report or send the traffic accidents to subsequent vehicles which are about to pass through accident areas. The main reason for this problem is that traffic accidents are various, the requirement for detection accuracy is high, and if false alarm occurs, on one hand, user experience is poor, and on the other hand, manpower and material resources are wasted to check and process detection results.
Some schemes for detecting traffic accidents by using a road side camera mostly detect whether a vehicle is on fire or directly use a trained recognition model to perform classification judgment on whether the traffic accidents occur or not, however, most of the traffic accidents in an actual scene are only accidents such as scratch and rear-end collision and have no obvious image characteristics, so that if the accidents are judged by directly using an image recognition model obtained by deep learning, the misdetection rate is high easily.
Disclosure of Invention
The embodiment of the application provides a traffic accident detection method and device, electronic equipment and a storage medium, so as to improve the accuracy of traffic accident detection.
The embodiment of the application adopts the following technical scheme:
in a first aspect, an embodiment of the present application provides a traffic accident detection method, which is performed by a roadside device, where the method includes:
acquiring a road image acquired by the road side equipment at the current road section, and detecting the road image to obtain a road image detection result of the current road section;
determining the running speed of the current road section according to the road image detection result of the current road section;
determining a target vehicle set of the current road section according to the running speed of the current road section, and determining whether a traffic accident occurs on the current road section according to the target vehicle set, the road image detection result and a preset traffic accident detection condition;
and under the condition that the traffic accident occurs in the current road section, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system so that the traffic accident confirmation system confirms the traffic accident detection result.
Optionally, the road image detection result includes a vehicle position, the travel speed includes a lane travel speed and a vehicle travel speed, and the determining the travel speed of the current road segment according to the road image detection result of the current road segment includes:
determining the driving distance and the driving time of each vehicle on the current road section according to the vehicle positions corresponding to the multi-frame road images;
determining the vehicle running speed of each vehicle according to the running distance and the running time of each vehicle on the current road section;
and determining the lane where each vehicle is located according to the vehicle position of each vehicle, and determining the lane driving speed of each lane according to the lane where each vehicle is located and the vehicle driving speed of each vehicle.
Optionally, the driving speed includes a lane driving speed and a vehicle driving speed, and the determining the target vehicle set of the current road segment according to the driving speed of the current road segment includes:
determining whether a target lane exists in a current road section or not according to the lane driving speed of each lane, wherein the lane driving speed of the target lane is lower than a first preset speed threshold;
under the condition that a target lane exists in the current road section, determining whether a target vehicle exists on the target lane, wherein the vehicle running speed of the target vehicle is lower than a second preset speed threshold;
and under the condition that a target vehicle exists on the target lane, marking the target vehicle and adding the target vehicle into the target vehicle set.
Optionally, the preset traffic accident detection condition includes a basic detection condition and an additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and the preset traffic accident detection condition includes:
determining whether the current road section triggers the basic detection condition according to the target vehicle set and the road image detection result, wherein the basic detection condition comprises a plurality of basic detection conditions;
if the current road section triggers each basic detection condition, determining whether the current road section triggers the additional detection conditions or not according to the target vehicle set and the road image detection result, wherein the number of the additional detection conditions is multiple;
and if the current road section triggers at least one additional detection condition, determining that a traffic accident occurs on the current road section.
Optionally, the road image detection result includes a vehicle position, the basic detection condition includes a first basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining traffic light information corresponding to the current road section, wherein the traffic light information comprises a traffic light position and a traffic light state;
if the traffic light state corresponding to the current road section is a green light state and the vehicle position of the target vehicle in the target vehicle set is located before the traffic light position corresponding to the current road section, determining that the current road section triggers the first basic detection condition;
otherwise, determining that the current road section does not trigger the first basic detection condition.
Optionally, the road image detection result includes a vehicle position, the basic detection condition includes a second basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining the running tracks of a plurality of target vehicles in the target vehicle set according to the vehicle positions;
if the running track triggering the shape of the preset overtaking track exists in the running tracks of the target vehicles, determining that the current road section triggers the second basic detection condition;
otherwise, determining that the current road section does not trigger the second basic detection condition.
Optionally, the road image detection result includes a vehicle position, the basic detection condition includes a third basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining a plurality of target vehicles in the target vehicle set, wherein the target vehicles are located before a preset position;
determining the relative distance between two adjacent target vehicles in the plurality of target vehicles positioned in front of the preset position according to the vehicle positions;
if the relative distance between two adjacent target vehicles in the plurality of target vehicles located in front of the preset position is smaller than a preset distance threshold, determining that the current road section triggers the third basic detection condition;
otherwise, determining that the current road section does not trigger the third basic detection condition.
Optionally, the road image detection result includes a vehicle detection frame, the basic detection condition includes a fourth basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining whether a pedestrian detection frame is detected according to a road image detection result of the multi-frame road image;
in the case where a pedestrian detection frame is detected, determining a duration of the pedestrian detection frame and a relative positional relationship of the pedestrian detection frame and a vehicle detection frame of a target vehicle in the set of target vehicles;
if the duration of the pedestrian detection frame reaches a preset time threshold, the pedestrian detection frame and the vehicle detection frame of the target vehicle have an overlapping area, and the earliest detected pedestrian detection frame is determined to be located on the left side of the vehicle detection frame of the target vehicle based on the vehicle driving direction, determining that the current road section triggers the fourth basic detection condition;
otherwise, determining that the current road section does not trigger the fourth basic detection condition.
Optionally, the road image detection result includes a vehicle position, the additional detection condition includes a first additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining a lane where a target vehicle is located according to the vehicle position of the target vehicle in the target vehicle set;
if the lane where the target vehicle is located is not the rightmost lane of the current road section, determining that the current road section triggers the first additional detection condition;
otherwise, determining that the current road segment does not trigger the first additional detection condition.
Optionally, the road image detection result includes a vehicle position, the additional detection condition includes a second additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining whether a traffic accident sign is detected at a vehicle rear position of a target vehicle in the set of target vehicles according to the vehicle position of the target vehicle;
if the traffic accident sign is detected at the vehicle rear position of the target vehicle, determining whether the movement of a pedestrian detection frame is detected between the vehicle rear position of the target vehicle and the position of the traffic accident sign in the process that the traffic accident sign is placed on the ground;
if the movement of the pedestrian detection frame is detected, determining that the current road section triggers the second additional detection condition;
otherwise, determining that the current road segment does not trigger the second additional detection condition.
Optionally, the determining whether a traffic accident occurs in the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes:
determining relative positions of the pedestrian detection frame and vehicle detection frames of target vehicles in the set of target vehicles;
if the relative position of the pedestrian detection frame and the vehicle detection frame of the target vehicle meets the requirement of a preset relative position, performing gesture action recognition on the pedestrian detection frame by using a preset gesture action recognition model;
if the gesture corresponding to the pedestrian detection frame is taken as a preset gesture, determining that the current road section triggers the third additional detection condition;
otherwise, determining that the current road segment does not trigger the third additional detection condition.
In a second aspect, an embodiment of the present application further provides a traffic accident detection device, which is applied to road side equipment, wherein the device includes:
the acquisition unit is used for acquiring a road image acquired by the road side equipment on the current road section, and detecting the road image to obtain a road image detection result of the current road section;
a first determination unit for determining a driving speed of the current road section according to a road image detection result of the current road section;
the second determining unit is used for determining a target vehicle set of the current road section according to the running speed of the current road section and determining whether a traffic accident happens to the current road section according to the target vehicle set, the road image detection result and preset traffic accident detection conditions;
and the sending unit is used for generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system under the condition that the traffic accident of the current road section is detected, so that the traffic accident confirmation system confirms the traffic accident detection result.
In a third aspect, an embodiment of the present application further provides an electronic device, including:
a processor; and
a memory arranged to store computer executable instructions that, when executed, cause the processor to perform any of the methods described above.
In a fourth aspect, embodiments of the present application further provide a computer-readable storage medium storing one or more programs that, when executed by an electronic device including a plurality of application programs, cause the electronic device to perform any of the methods described above.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects: the traffic accident detection method is executed by road side equipment, firstly, road images collected by the road side equipment on a current road section are obtained, and the road images are detected to obtain a road image detection result of the current road section; then determining the running speed of the current road section according to the road image detection result of the current road section; then determining a target vehicle set of the current road section according to the running speed of the current road section, and determining whether a traffic accident occurs in the current road section according to the target vehicle set and the road image detection result; and finally, under the condition that the traffic accident happens on the current road section is detected, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system so that the traffic accident confirmation system confirms the traffic accident detection result. The traffic accident detection method determines the target vehicle which is possibly subjected to the traffic accident or is possibly influenced by the traffic accident based on the running speed of the current road section, and further judges whether the traffic accident occurs on the current road section or not by combining the continuous road image detection result and a plurality of traffic accident detection conditions, so that the false detection rate of the traffic accident is greatly reduced, and the traffic accident detection method can be suitable for traffic accidents without obvious image characteristics, such as scratch and rear-end collision and the like.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic flow chart of a traffic accident detection method in an embodiment of the present application;
FIG. 2 is a schematic structural diagram of a traffic accident detection apparatus according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of an electronic device in an embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be clearly and completely described below with reference to specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some 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 technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
The embodiment of the present application provides a traffic accident detection method, which is executed by roadside devices, and as shown in fig. 1, provides a flow diagram of the traffic accident detection method in the embodiment of the present application, where the method at least includes the following steps S110 to S140:
step S110, acquiring a road image acquired by the road side equipment on the current road section, and detecting the road image to obtain a road image detection result of the current road section.
The traffic accident detection method provided by the embodiment of the application can be executed by road side equipment, and when the traffic accident detection is carried out, a road image of a current road section acquired by the road side equipment needs to be acquired first, and targets such as vehicles and pedestrians in the road image are detected by using a preset target detection algorithm, so that a road image detection result of the current road section is obtained.
In order to ensure the accuracy of the detection of the traffic accident, the road condition of the current road section can be continuously monitored, namely, the embodiment of the application can continuously acquire the road image at the current road section and continuously detect and identify the image.
And step S120, determining the running speed of the current road section according to the road image detection result of the current road section.
The running speed of the current road section can reflect whether the current road section is possible to have traffic accidents or not to a certain extent, for example, when a certain road section has traffic accidents, traffic jam to a certain extent is often caused, so that the running speed of a vehicle affected by the traffic accidents or the whole running speed of the current road section is reduced to 0 even to a certain extent. Based on this, the embodiment of the application needs to determine the driving speed of the current road section according to the road image detection result of the current road section.
As described above, the road image detection result of the current link may include the vehicle detection result, and the vehicle detection result may include, for example, the positions of the vehicle at the respective image frame times, so that the traveling speed of the vehicle itself and the overall traveling speed of the respective lanes may be calculated from the positions of the vehicle at the respective image frame times.
Step S130, determining a target vehicle set of the current road section according to the running speed of the current road section, and determining whether a traffic accident occurs on the current road section according to the target vehicle set, the road image detection result and preset traffic accident detection conditions.
After the running speed of the current road section is calculated, vehicles which are possibly subjected to traffic accidents or are influenced by the traffic accidents in the current road section can be determined according to the running speed of the current road section, the vehicles are used as target vehicles and added into the target vehicle set, and whether the traffic accidents occur in the current road section is judged according to the target vehicle set, real-time road image detection results and preset traffic accident detection conditions.
The preset traffic accident detection condition may include a plurality of detection conditions defined from different dimensions, for example, the preset traffic accident detection condition may include a traffic light state of a current street lamp, a relative position relationship between a target vehicle and the traffic light, a pedestrian detection, a relative position relationship between a pedestrian and the target vehicle, a driving track of the target vehicle, and the like, so as to ensure accuracy of the traffic accident detection. In addition, due to the arrangement of a plurality of detection conditions, the detection of the traffic accident does not require that obvious characteristics such as ignition and the like exist in the image, and further the method and the device can be suitable for detecting the traffic accident without the obvious image characteristics such as scratch and rear-end collision and the like.
And step S140, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system under the condition that the traffic accident of the current road section is detected, so that the traffic accident confirmation system confirms the traffic accident detection result.
If the traffic accident occurs in the current road section, the data such as the road image collected in the period, the corresponding detection result and the like can be further sent to a traffic accident confirmation system, so that the traffic accident confirmation system can further confirm the traffic accident, and the accuracy of traffic accident detection is ensured.
The traffic accident detection method determines the target vehicle which is possibly subjected to the traffic accident or is possibly influenced by the traffic accident based on the running speed of the current road section, and further judges whether the traffic accident occurs on the current road section or not by combining the continuous road image detection result and a plurality of traffic accident detection conditions, so that the false detection rate of the traffic accident is greatly reduced, and the traffic accident detection method can be suitable for traffic accidents without obvious image characteristics, such as scratch and rear-end collision and the like.
In some embodiments of the present application, the road image detection result includes a vehicle position, the travel speed includes a lane travel speed and a vehicle travel speed, and the determining the travel speed of the current link from the road image detection result of the current link includes: determining the driving distance and the driving time of each vehicle on the current road section according to the vehicle positions corresponding to the multi-frame road images; determining the vehicle running speed of each vehicle according to the running distance and the running time of each vehicle on the current road section; and determining the lane where each vehicle is located according to the vehicle position of each vehicle, and determining the lane driving speed of each lane according to the lane where each vehicle is located and the vehicle driving speed of each vehicle.
The driving speed of the current road section mainly comprises two dimensions, wherein one dimension is a vehicle dimension, namely the average driving speed of each vehicle, and the other dimension is a lane dimension, namely the average driving speed of each lane.
For example, if a vehicle is detected in ten consecutive images, the vehicle travel distance may be calculated from the vehicle position detected in the first image and the vehicle position detected in the tenth image, and then divided by the time interval between the ten images to obtain the vehicle travel speed of the vehicle.
For the driving speed of the lane dimension, which vehicles correspond to each lane may be determined first, and similarly, the driving speed may be determined based on the vehicle positions detected from the road image, and then the driving speeds of all vehicles corresponding to each lane may be averaged, respectively, to obtain the driving speed of each lane. For example, if there are 5 vehicles in a certain lane, the vehicle travel speeds of the 5 vehicles may be averaged.
In some embodiments of the present application, the driving speed includes a lane driving speed and a vehicle driving speed, and the determining the target vehicle set of the current road segment according to the driving speed of the current road segment includes: determining whether a target lane exists in a current road section or not according to the lane driving speed of each lane, wherein the lane driving speed of the target lane is lower than a first preset speed threshold; under the condition that a target lane exists in the current road section, determining whether a target vehicle exists on the target lane, wherein the vehicle running speed of the target vehicle is lower than a second preset speed threshold; and under the condition that a target vehicle exists on the target lane, marking the target vehicle and adding the target vehicle into the target vehicle set.
When determining the target vehicle set of the current road section according to the driving speed of the current road section, first determining whether an abnormal lane exists according to the driving speed of the lane of each lane, for example, comparing the driving speed of the lane of each lane with a first preset speed threshold, if the driving speed of the lane is lower than the first preset speed threshold, it is indicated that the overall driving speed of the lane does not meet the normal driving speed requirement, then further determining the driving speed of each vehicle corresponding to the lane, if the driving speed of each vehicle corresponding to the lane is lower than a second preset speed threshold, it is indicated that a vehicle with a slow driving speed exists on the lane, and marking the vehicles as S respectively 1 ,...S n ,S n+1 \8230; \8230andrecorded in the target vehicle collection.
The size of the first preset speed threshold value can be flexibly set according to actual requirements, and the driving speed of each lane can be measured to a certain extent. The second preset speed threshold may be set to 0 or a value close to 0, and when the speed of the vehicle is reduced to 0, it indicates that the vehicle may not be able to travel due to a traffic accident, and thus, the vehicle may be further determined as the target vehicle.
Of course, the target vehicle may be determined by continuous detection, for example, when a trend of decreasing the driving speed of the lane in each lane is detected, the driving speed of some vehicles in the lane is decreased to 0, and the driving speed of the following vehicles is also rapidly decreased to 0, then the vehicles with the speed decreased to 0 may be used as the target vehicle.
In some embodiments of the present application, the preset traffic accident detection condition includes a basic detection condition and an additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and the preset traffic accident detection condition includes: determining whether the current road section triggers the basic detection conditions or not according to the target vehicle set and the road image detection result, wherein the basic detection conditions comprise a plurality of basic detection conditions; if the current road section triggers each basic detection condition, determining whether the current road section triggers the additional detection conditions or not according to the target vehicle set and the road image detection result, wherein the number of the additional detection conditions is multiple; and if the current road section triggers at least one additional detection condition, determining that a traffic accident occurs on the current road section.
In order to ensure the accuracy of detecting traffic accidents of types such as scratch and rear-end collisions, the preset traffic accident detection conditions defined in the embodiment of the present application may include a plurality of basic detection conditions and additional detection conditions, the basic detection conditions may be regarded as necessary conditions for detecting whether a traffic accident occurs on a current road segment, the traffic accident may be considered to occur on the current road segment only when all the basic detection conditions are satisfied, the additional detection conditions may be regarded as optional conditions for detecting whether a traffic accident occurs on the current road segment, and the traffic accident may be considered to occur on the current road segment when all the basic detection conditions are satisfied and any one of the additional detection conditions is satisfied.
Of course, the basic detection condition and the additional detection condition are specifically defined from which dimensions, and those skilled in the art can flexibly set the conditions according to the actual scene and the actual requirement, which are not specifically limited herein.
In some embodiments of the present application, the road image detection result includes a vehicle location, the basic detection condition includes a first basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition includes: determining traffic light information corresponding to the current road section, wherein the traffic light information comprises a traffic light position and a traffic light state; if the traffic light state corresponding to the current road section is a green light state and the vehicle position of the target vehicle in the target vehicle set is located before the traffic light position corresponding to the current road section, determining that the current road section triggers the first basic detection condition; otherwise, determining that the current road section does not trigger the first basic detection condition.
The embodiment of the application defines a basic detection condition of a traffic accident according to actual scene requirements, namely a first basic detection condition, wherein the first basic detection condition is mainly used for judging whether the traffic accident is possible to happen on the current road section by combining traffic light information of the current road section, position information of a target vehicle and the like.
Specifically, the road side device can acquire traffic light information of the road section in real time, including a traffic light state, a traffic light position and the like, and certainly, the road side device can also determine the traffic light information of the road section in an image detection manner. If the traffic light state corresponding to the lane where the target vehicle is located is the green light state and the traffic light position is located at the position of the target vehicle, it is indicated that the vehicle on the lane can normally run and runs towards the traffic light, but the average running speed on the lane is reduced or the running speeds of some vehicles are already reduced to 0, it is indicated that a traffic accident is likely to occur on the lane, so that the vehicles cannot run at the normal speed and pass through the intersection, and therefore it can be considered that the current road section meets the first basic detection condition.
In some embodiments of the present application, the road image detection result includes a vehicle location, the basic detection condition includes a second basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes: determining the running tracks of a plurality of target vehicles in the target vehicle set according to the vehicle positions; if the running track triggering the shape of the preset overtaking track exists in the running tracks of the target vehicles, determining that the current road section triggers the second basic detection condition; otherwise, determining that the current road segment does not trigger the second basic detection condition.
According to the embodiment of the application, another basic detection condition of the traffic accident, namely a second basic detection condition, is defined according to the actual scene requirement, and the second basic detection condition is mainly used for judging whether the traffic accident possibly occurs in the current road section by detecting the running track of the target vehicle.
Under the scene of traffic accidents such as scratch and rear-end collision, target vehicles influenced by the accident are likely to bypass the vehicle with the accident in an overtaking lane changing mode and then continuously run, so the running tracks of the target vehicles with the speed reduced to 0 can be continuously detected, for example, the positions of the vehicles in continuous multi-frame road images can be continuously detected, then the positions of the vehicles in the road images are sequentially spliced to obtain the running tracks of the target vehicles, and if the running tracks of the target vehicles conform to the shape of the overtaking lane changing track, the current road section can be considered to meet the second basic detection condition.
In some embodiments of the present application, the road image detection result includes a vehicle location, the basic detection condition includes a third basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition includes: determining a plurality of target vehicles in the target vehicle set before a preset position; determining the relative distance between two adjacent target vehicles in the plurality of target vehicles positioned in front of the preset position according to the vehicle positions; if the relative distance between two adjacent target vehicles in the plurality of target vehicles located in front of the preset position is smaller than a preset distance threshold, determining that the current road section triggers the third basic detection condition; otherwise, determining that the current road section does not trigger the third basic detection condition.
According to the embodiment of the application, another basic detection condition of the traffic accident, namely a third basic detection condition, is defined according to the actual scene requirement, and the third basic detection condition is mainly used for judging whether the traffic accident possibly occurs in the current road section by detecting the relative distance between the target vehicles.
In a traffic accident situation such as a scratch and a rear-end collision, continuous collision between more than two vehicles is usually caused, so that for the vehicles with the scratch and the rear-end collision, the relative distance between the vehicles is much smaller than the normal running distance, and therefore the embodiment of the application can judge the relative distance between every two vehicles in the front vehicle set, namely calculate the relative distance between two adjacent vehicles according to the detected vehicle position of each target vehicle, and if the relative distance between two adjacent vehicles is smaller than a preset distance threshold value, for example, 20cm, then the current road section can be considered to meet the third basic detection condition.
In some embodiments of the present application, the road image detection result includes a vehicle detection frame, the basic detection condition includes a fourth basic detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result and a preset traffic accident detection condition includes: determining whether a pedestrian detection frame is detected according to a road image detection result of the multiple frames of road images; in the case that a pedestrian detection frame is detected, determining the duration of the pedestrian detection frame and the relative positional relationship of the pedestrian detection frame and a vehicle detection frame of a target vehicle in the set of target vehicles; if the duration of the pedestrian detection frame reaches a preset time threshold value, the pedestrian detection frame and the vehicle detection frame of the target vehicle have an overlapping area, and it is determined that the earliest detected pedestrian detection frame is located on the left side of the vehicle detection frame of the target vehicle based on the vehicle driving direction, it is determined that the current road section triggers the fourth basic detection condition; otherwise, determining that the current road section does not trigger the fourth basic detection condition.
According to the embodiment of the application, another basic detection condition of the traffic accident, namely a fourth basic detection condition, is defined according to the actual scene requirement, and the fourth basic detection condition is mainly used for judging whether the traffic accident is possible to happen on the current road section by detecting the relative position relation between the target vehicle and the pedestrian.
In a scene of traffic accidents such as scratch and rear-end collision, related drivers usually get off and deal with the problems of accident cause, compensation and the like, so the continuous detection of the road image of the current road section in the embodiment of the application can further include detection of pedestrians, if the detection of a pedestrian detection frame suddenly appearing on the current road section through the continuous detection can further determine the duration of the pedestrian detection frame and the relative position of the pedestrian detection frame and a vehicle detection frame of a target vehicle, and if the duration of the pedestrian detection frame reaches a preset time threshold value and an overlapping region exists between the pedestrian detection frame and the vehicle detection frame of the target vehicle, it is described that the pedestrian stably exists and the distance from the target vehicle is very close, that is, the drivers who get off from the target vehicle are possible. In order to further improve the judgment accuracy, whether the pedestrian detection frame which is detected firstly appears at the left side position of the vehicle detection frame or not can be determined based on the vehicle driving direction, if so, the pedestrian is the driver who gets off from the target vehicle, and further, the situation that the driver cannot get off from the target vehicle for accident treatment due to traffic accidents of the current vehicle is verified, and the traffic jam and the like are eliminated.
The above-mentioned several basic detection conditions can already judge to a certain extent whether the current road section is likely to have traffic accidents such as scratch and rear-end collision or the like based on the driving speed of the vehicle, the traffic light information and the position information of the target vehicle, the driving track of the target vehicle, the relative distance between the target vehicles, the relative position relationship between the target vehicle and the pedestrian, and the like.
In some embodiments of the present application, the road image detection result includes a vehicle location, the additional detection condition includes a first additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition includes: determining a lane where a target vehicle is located according to the vehicle position of the target vehicle in the target vehicle set; if the lane where the target vehicle is located is not the rightmost lane of the current road section, determining that the current road section triggers the first additional detection condition; otherwise, determining that the current road segment does not trigger the first additional detection condition.
According to the embodiment of the application, an additional detection condition of the traffic accident, namely a first additional detection condition, is defined according to actual scene requirements, and the first additional detection condition is mainly used for confirming whether the traffic accident is possible to occur on the current road section by detecting the parking position of a target vehicle.
Based on the current traffic regulations, the vehicle parking needs to follow the rule of parking on the right side, if the vehicle is a multi-lane road, the vehicle needs to park on the rightmost lane, and under the scene of traffic accidents such as scratch rear-end collision, the vehicle may be forced parking caused by accidents such as scratch rear-end collision and the like in the driving process, and the vehicle is likely not to park on the rightmost lane, so the embodiment of the application can detect and judge the lane where the target vehicle is located, and if the lane where the target vehicle is located is not the rightmost lane, the traffic accident can be considered to occur on the current road section.
In some embodiments of the present application, the road image detection result includes a vehicle location, the additional detection condition includes a second additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and a preset traffic accident detection condition includes: determining whether a traffic accident sign is detected at a vehicle rear position of a target vehicle in the set of target vehicles according to the vehicle position of the target vehicle; if the traffic accident sign is detected at the position behind the target vehicle, determining whether movement of a pedestrian detection frame is detected between the position behind the target vehicle and the position of the traffic accident sign in the process that the traffic accident sign is placed on the ground; if the movement of the pedestrian detection frame is detected, determining that the current road section triggers the second additional detection condition; otherwise, determining that the current road segment does not trigger the second additional detection condition.
According to the embodiment of the application, another additional detection condition of the traffic accident, namely a second additional detection condition, is defined according to the actual scene requirement, and the second additional detection condition is mainly used for confirming whether the traffic accident possibly occurs in the current road section or not by detecting a traffic accident sign.
Based on the current traffic regulations, in a scene of a traffic accident, vehicle personnel need to place a warning sign of the traffic accident at a certain distance behind a vehicle, and the warning sign of the traffic accident is usually a triangular sign board in a uniform mode, so that the traffic accident sign in a road image can be detected by using a pre-trained target detection algorithm, and if the traffic accident sign is detected, the traffic accident is very likely to happen to a target vehicle. Of course, in order to further improve the determination accuracy, the movement of the pedestrian may be continuously detected while the traffic accident sign is placed on the ground, and if the pedestrian detection frame is detected to move between the position of the target vehicle and the position where the traffic accident sign is placed during the detection, the traffic accident sign may be considered to be placed on the ground by the person of the target vehicle, and thus the traffic accident may be considered to have occurred at the current road section.
In some embodiments of the application, the road image detection result includes a vehicle detection frame and a pedestrian detection frame, the additional detection condition includes a third additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result and a preset traffic accident detection condition includes: determining relative positions of the pedestrian detection frame and vehicle detection frames of target vehicles in the set of target vehicles; if the relative position of the pedestrian detection frame and the vehicle detection frame of the target vehicle meets the requirement of a preset relative position, performing gesture action recognition on the pedestrian detection frame by using a preset gesture action recognition model; if the gesture corresponding to the pedestrian detection frame is taken as a preset gesture, determining that the current road section triggers the third additional detection condition; otherwise, determining that the current road segment does not trigger the third additional detection condition.
According to the embodiment of the application, another additional detection condition of the traffic accident, namely a third additional detection condition is defined according to the actual scene requirement, and the third additional detection condition is mainly used for confirming whether the traffic accident possibly occurs in the current road section by detecting the position and the action of a pedestrian.
In a situation of traffic accidents such as tailgating, relevant drivers usually get off to take pictures of a traffic accident scene to obtain evidence, so that the embodiment of the application can continuously detect pedestrian detection frames of a current road section and compare the positions of the detected vehicle detection frames of target vehicles, if the movement of the pedestrian detection frames is detected among a plurality of target vehicles, the relevant drivers can possibly check the accident scene, at the moment, the gesture actions of the pedestrians can be further detected by using a pre-trained gesture action detection model, and if the gesture actions of the pedestrians are detected as preset gesture actions such as picture taking actions, the pedestrians are shown to take pictures of the accident scene to obtain evidence, and further, the traffic accidents of the current road section are shown.
It should be noted that, the model for detecting the vehicle, the pedestrian, the traffic accident sign, the gesture action, and the like in the embodiments of the present application may be obtained based on the existing convolutional neural network, such as yolo network training, and of course, a person skilled in the art may determine how to train the model in combination with the prior art, which is not described herein again.
The multiple basic detection conditions and the multiple additional detection conditions defined in the embodiment of the application have no strict sequence requirement, detection can be performed simultaneously, the detection precision and the detection efficiency are higher in traffic accident scenes such as scratch and rear-end collision, and when the method is suitable for other traffic accident scenes, a person skilled in the art can flexibly adjust the detection conditions according to actual requirements.
It should be noted that, the detection conditions related to the above embodiments mainly aim at traffic accident scenes where image features are not obvious, such as scratch and rear-end collision, and for traffic accident scenes with obvious image features, such as vehicle ignition, an attribute detection of whether the vehicle is on fire can be directly added through the existing image detection model, and since the vehicle has obvious features and a false detection rate is low, if the vehicle ignition features are detected in the image, the vehicle can be directly determined as a traffic accident, and data such as videos collected before and after and detection results are uploaded to a traffic accident confirmation system for confirmation.
The embodiment of the present application further provides a traffic accident detection apparatus 200, which is applied to roadside devices, and as shown in fig. 2, provides a schematic structural diagram of the traffic accident detection apparatus in the embodiment of the present application, where the apparatus 200 includes: an obtaining unit 210, a first determining unit 220, a second determining unit 230, and a sending unit 240, wherein:
the acquiring unit 210 is configured to acquire a road image acquired by the roadside device on the current road segment, and detect the road image to obtain a road image detection result of the current road segment;
a first determining unit 220 for determining a driving speed of the current road segment according to a road image detection result of the current road segment;
a second determining unit 230, configured to determine a target vehicle set of the current road segment according to the driving speed of the current road segment, and determine whether a traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result, and preset traffic accident detection conditions;
a sending unit 240, configured to generate a traffic accident detection result and send the traffic accident detection result to a traffic accident confirmation system when detecting that a traffic accident occurs on the current road segment, so that the traffic accident confirmation system confirms the traffic accident detection result.
In some embodiments of the present application, the road image detection result includes a vehicle position, the driving speed includes a lane driving speed and a vehicle driving speed, and the first determining unit 220 is specifically configured to: determining the driving distance and the driving time of each vehicle on the current road section according to the vehicle positions corresponding to the multi-frame road images; determining the vehicle running speed of each vehicle according to the running distance and the running time of each vehicle on the current road section; and determining the lane where each vehicle is located according to the vehicle position of each vehicle, and determining the lane driving speed of each lane according to the lane where each vehicle is located and the vehicle driving speed of each vehicle.
In some embodiments of the application, the driving speeds include a lane driving speed and a vehicle driving speed, and the second determining unit 230 is specifically configured to: determining whether a target lane exists in a current road section or not according to the lane driving speed of each lane, wherein the lane driving speed of the target lane is lower than a first preset speed threshold; under the condition that a target lane exists in the current road section, determining whether a target vehicle exists on the target lane, wherein the vehicle running speed of the target vehicle is lower than a second preset speed threshold; and under the condition that the target vehicles exist on the target lane, marking the target vehicles and adding the target vehicles into the target vehicle set.
In some embodiments of the application, the preset traffic accident detection condition includes a basic detection condition and an additional detection condition, and the second determining unit 230 is specifically configured to: determining whether the current road section triggers the basic detection condition according to the target vehicle set and the road image detection result, wherein the basic detection condition comprises a plurality of basic detection conditions; if the current road section triggers each basic detection condition, determining whether the current road section triggers the additional detection conditions or not according to the target vehicle set and the road image detection result, wherein the additional detection conditions comprise a plurality of conditions; and if the current road section triggers at least one additional detection condition, determining that a traffic accident occurs on the current road section.
In some embodiments of the application, the road image detection result includes a vehicle position, the basic detection condition includes a first basic detection condition, and the second determining unit 230 is specifically configured to: determining traffic light information corresponding to the current road section, wherein the traffic light information comprises a traffic light position and a traffic light state; if the traffic light state corresponding to the current road section is a green light state and the vehicle position of the target vehicle in the target vehicle set is located before the traffic light position corresponding to the current road section, determining that the current road section triggers the first basic detection condition; otherwise, determining that the current road section does not trigger the first basic detection condition.
In some embodiments of the application, the road image detection result includes a vehicle position, the basic detection condition includes a second basic detection condition, and the second determining unit 230 is specifically configured to: determining the running tracks of a plurality of target vehicles in the target vehicle set according to the vehicle positions; if the running track triggering the shape of the preset overtaking track exists in the running tracks of the target vehicles, determining that the current road section triggers the second basic detection condition; otherwise, determining that the current road section does not trigger the second basic detection condition.
In some embodiments of the application, the road image detection result includes a vehicle position, the basic detection condition includes a third basic detection condition, and the second determining unit 230 is specifically configured to: determining a plurality of target vehicles in the target vehicle set before a preset position; determining the relative distance between two adjacent target vehicles in the plurality of target vehicles positioned in front of the preset position according to the vehicle positions; if the relative distance between two adjacent target vehicles in the plurality of target vehicles located in front of the preset position is smaller than a preset distance threshold, determining that the current road section triggers the third basic detection condition; otherwise, determining that the current road segment does not trigger the third basic detection condition.
In some embodiments of the present application, the road image detection result includes a vehicle detection frame, the basic detection condition includes a fourth basic detection condition, and the second determining unit 230 is specifically configured to: determining whether a pedestrian detection frame is detected according to a road image detection result of the multi-frame road image; in the case that a pedestrian detection frame is detected, determining the duration of the pedestrian detection frame and the relative positional relationship of the pedestrian detection frame and a vehicle detection frame of a target vehicle in the set of target vehicles; if the duration of the pedestrian detection frame reaches a preset time threshold value, the pedestrian detection frame and the vehicle detection frame of the target vehicle have an overlapping area, and it is determined that the earliest detected pedestrian detection frame is located on the left side of the vehicle detection frame of the target vehicle based on the vehicle driving direction, it is determined that the current road section triggers the fourth basic detection condition; otherwise, it is determined that the fourth basic detection condition is not triggered by the current road segment.
In some embodiments of the application, the road image detection result includes a vehicle position, the additional detection condition includes a first additional detection condition, and the second determining unit 230 is specifically configured to: determining a lane where a target vehicle is located according to the vehicle position of the target vehicle in the target vehicle set; if the lane where the target vehicle is located is not the rightmost lane of the current road section, determining that the current road section triggers the first additional detection condition; otherwise, determining that the current road segment does not trigger the first additional detection condition.
In some embodiments of the present application, the road image detection result includes a vehicle position, the additional detection condition includes a second additional detection condition, and the second determining unit 230 is specifically configured to: determining whether a traffic accident sign is detected at a vehicle rear position of a target vehicle in the set of target vehicles according to the vehicle position of the target vehicle; if the traffic accident sign is detected at the position behind the target vehicle, determining whether movement of a pedestrian detection frame is detected between the position behind the target vehicle and the position of the traffic accident sign in the process that the traffic accident sign is placed on the ground; if the movement of the pedestrian detection frame is detected, determining that the current road section triggers the second additional detection condition; otherwise, determining that the current road segment does not trigger the second additional detection condition.
In some embodiments of the present application, the road image detection result includes a vehicle detection frame and a pedestrian detection frame, the additional detection condition includes a third additional detection condition, and the second determining unit 230 is specifically configured to: determining relative positions of the pedestrian detection frame and vehicle detection frames of target vehicles in the set of target vehicles; if the relative position of the pedestrian detection frame and the vehicle detection frame of the target vehicle meets the requirement of a preset relative position, performing gesture action recognition on the pedestrian detection frame by using a preset gesture action recognition model; if the gesture corresponding to the pedestrian detection frame is taken as a preset gesture, determining that the current road section triggers the third additional detection condition; otherwise, determining that the current road segment does not trigger the third additional detection condition.
It can be understood that the above-mentioned traffic accident detection apparatus can implement the steps of the traffic accident detection method provided in the foregoing embodiment, and the related explanations about the traffic accident detection method are applicable to the traffic accident detection apparatus, and are not described herein again.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Referring to fig. 3, at a hardware level, the electronic device includes a processor, and optionally further includes an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory, such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.
The processor, the network interface, and the memory may be connected to each other by an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one double-headed arrow is shown in FIG. 3, but this does not indicate only one bus or one type of bus.
And the memory is used for storing programs. In particular, the program may include program code comprising computer operating instructions. The memory may include both memory and non-volatile storage and provides instructions and data to the processor.
The processor reads a corresponding computer program from the nonvolatile memory into the memory and then runs the computer program to form the traffic accident detection device on a logic level. The processor is used for executing the program stored in the memory and is specifically used for executing the following operations:
acquiring a road image acquired by the road side equipment at the current road section, and detecting the road image to obtain a road image detection result of the current road section;
determining the running speed of the current road section according to the road image detection result of the current road section;
determining a target vehicle set of the current road section according to the running speed of the current road section, and determining whether a traffic accident occurs in the current road section according to the target vehicle set, the road image detection result and a preset traffic accident detection condition;
and under the condition that the traffic accident happens to the current road section, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system so that the traffic accident confirmation system confirms the traffic accident detection result.
The method performed by the traffic accident detection apparatus according to the embodiment shown in fig. 1 of the present application may be applied to or implemented by a processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, etc. as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and combines hardware thereof to complete the steps of the method.
The electronic device may further execute the method executed by the traffic accident detection apparatus in fig. 1, and implement the functions of the traffic accident detection apparatus in the embodiment shown in fig. 1, which are not described herein again in this embodiment of the present application.
An embodiment of the present application further provides a computer-readable storage medium storing one or more programs, where the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, enable the electronic device to perform the method performed by the traffic accident detection apparatus in the embodiment shown in fig. 1, and are specifically configured to perform:
acquiring a road image acquired by the road side equipment at the current road section, and detecting the road image to obtain a road image detection result of the current road section;
determining the running speed of the current road section according to the road image detection result of the current road section;
determining a target vehicle set of the current road section according to the running speed of the current road section, and determining whether a traffic accident occurs on the current road section according to the target vehicle set, the road image detection result and a preset traffic accident detection condition;
and under the condition that the traffic accident occurs in the current road section, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system so that the traffic accident confirmation system confirms the traffic accident detection result.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that 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 phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (14)

1. A traffic accident detection method performed by a roadside apparatus, wherein the method comprises:
acquiring a road image acquired by the road side equipment at the current road section, and detecting the road image to obtain a road image detection result of the current road section;
determining the running speed of the current road section according to the road image detection result of the current road section;
determining a target vehicle set of the current road section according to the running speed of the current road section, and determining whether a traffic accident occurs on the current road section according to the target vehicle set, the road image detection result and a preset traffic accident detection condition;
and under the condition that the traffic accident happens to the current road section, generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system so that the traffic accident confirmation system confirms the traffic accident detection result.
2. The method of claim 1, wherein the road image detection result includes a vehicle position, the travel speed includes a lane travel speed and a vehicle travel speed, and the determining the travel speed of the current link from the road image detection result of the current link includes:
determining the driving distance and the driving time of each vehicle on the current road section according to the vehicle positions corresponding to the multi-frame road images;
determining the vehicle running speed of each vehicle according to the running distance and the running time of each vehicle on the current road section;
and determining the lane where each vehicle is located according to the vehicle position of each vehicle, and determining the lane driving speed of each lane according to the lane where each vehicle is located and the vehicle driving speed of each vehicle.
3. The method of claim 1, wherein the travel speed comprises a lane travel speed and a vehicle travel speed, and wherein determining the set of target vehicles for the current road segment based on the travel speed for the current road segment comprises:
determining whether a target lane exists in a current road section or not according to the lane driving speed of each lane, wherein the lane driving speed of the target lane is lower than a first preset speed threshold;
under the condition that a target lane exists in the current road section, determining whether a target vehicle exists on the target lane, wherein the vehicle running speed of the target vehicle is lower than a second preset speed threshold;
and under the condition that the target vehicles exist on the target lane, marking the target vehicles and adding the target vehicles into the target vehicle set.
4. The method of claim 1, wherein the preset traffic accident detection condition comprises a basic detection condition and an additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and the preset traffic accident detection condition comprises:
determining whether the current road section triggers the basic detection conditions or not according to the target vehicle set and the road image detection result, wherein the basic detection conditions comprise a plurality of basic detection conditions;
if the current road section triggers each basic detection condition, determining whether the current road section triggers the additional detection conditions or not according to the target vehicle set and the road image detection result, wherein the number of the additional detection conditions is multiple;
and if the current road section triggers at least one additional detection condition, determining that a traffic accident occurs on the current road section.
5. The method of claim 4, wherein the road image detection result comprises a vehicle position, the base detection condition comprises a first base detection condition, and the determining whether the traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition comprises:
determining traffic light information corresponding to the current road section, wherein the traffic light information comprises a traffic light position and a traffic light state;
if the traffic light state corresponding to the current road section is a green light state and the vehicle position of the target vehicle in the target vehicle set is located before the traffic light position corresponding to the current road section, determining that the current road section triggers the first basic detection condition;
otherwise, determining that the current road section does not trigger the first basic detection condition.
6. The method of claim 4, wherein the road image detection result comprises a vehicle location, the base detection condition comprises a second base detection condition, and the determining whether the traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition comprises:
determining the running tracks of a plurality of target vehicles in the target vehicle set according to the vehicle positions;
if the running track of the plurality of target vehicles has the running track which triggers the shape of the preset overtaking track, determining that the current road section triggers the second basic detection condition;
otherwise, determining that the current road segment does not trigger the second basic detection condition.
7. The method of claim 4, wherein the road image detection result comprises a vehicle location, the base detection condition comprises a third base detection condition, and the determining whether the traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition comprises:
determining a plurality of target vehicles in the target vehicle set, wherein the target vehicles are located before a preset position;
determining the relative distance between two adjacent target vehicles in the plurality of target vehicles positioned in front of the preset position according to the vehicle positions;
if the relative distance between two adjacent target vehicles in the plurality of target vehicles located in front of the preset position is smaller than a preset distance threshold, determining that the current road section triggers the third basic detection condition;
otherwise, determining that the current road section does not trigger the third basic detection condition.
8. The method of claim 4, wherein the road image detection result comprises a vehicle detection box, the base detection condition comprises a fourth base detection condition, and the determining whether the traffic accident occurs on the current road segment according to the target vehicle set, the road image detection result and a preset traffic accident detection condition comprises:
determining whether a pedestrian detection frame is detected according to a road image detection result of the multiple frames of road images;
in the case that a pedestrian detection frame is detected, determining the duration of the pedestrian detection frame and the relative positional relationship of the pedestrian detection frame and a vehicle detection frame of a target vehicle in the set of target vehicles;
if the duration of the pedestrian detection frame reaches a preset time threshold value, the pedestrian detection frame and the vehicle detection frame of the target vehicle have an overlapping area, and it is determined that the earliest detected pedestrian detection frame is located on the left side of the vehicle detection frame of the target vehicle based on the vehicle driving direction, it is determined that the current road section triggers the fourth basic detection condition;
otherwise, it is determined that the fourth basic detection condition is not triggered by the current road segment.
9. The method of claim 4, wherein the road image detection result comprises a vehicle position, the additional detection condition comprises a first additional detection condition, and the determining whether the traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition comprises:
determining a lane where a target vehicle is located according to the vehicle position of the target vehicle in the target vehicle set;
if the lane where the target vehicle is located is not the rightmost lane of the current road section, determining that the current road section triggers the first additional detection condition;
otherwise, determining that the current road segment does not trigger the first additional detection condition.
10. The method of claim 4, wherein the road image detection result comprises a vehicle position, the additional detection condition comprises a second additional detection condition, and the determining whether a traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition comprises:
determining whether a traffic accident sign is detected at a vehicle rear position of a target vehicle in the set of target vehicles according to the vehicle position of the target vehicle;
if the traffic accident sign is detected at the vehicle rear position of the target vehicle, determining whether the movement of a pedestrian detection frame is detected between the vehicle rear position of the target vehicle and the position of the traffic accident sign in the process that the traffic accident sign is placed on the ground;
if the movement of the pedestrian detection frame is detected, determining that the current road section triggers the second additional detection condition;
otherwise, determining that the current road segment does not trigger the second additional detection condition.
11. The method of claim 4, wherein the road image detection result comprises a vehicle detection box and a pedestrian detection box, the additional detection condition comprises a third additional detection condition, and the determining whether the traffic accident occurs on the current road segment according to the target vehicle set and the road image detection result and a preset traffic accident detection condition comprises:
determining relative positions of the pedestrian detection frame and vehicle detection frames of target vehicles in the set of target vehicles;
if the relative position of the pedestrian detection frame and the vehicle detection frame of the target vehicle meets the requirement of a preset relative position, performing gesture action recognition on the pedestrian detection frame by using a preset gesture action recognition model;
if the gesture corresponding to the pedestrian detection frame is taken as a preset gesture, determining that the current road section triggers the third additional detection condition;
otherwise, determining that the current road segment does not trigger the third additional detection condition.
12. A traffic accident detection apparatus applied to roadside equipment, wherein the apparatus comprises:
the acquisition unit is used for acquiring a road image acquired by the road side equipment on the current road section and detecting the road image to obtain a road image detection result of the current road section;
the first determining unit is used for determining the running speed of the current road section according to the road image detection result of the current road section;
the second determining unit is used for determining a target vehicle set of the current road section according to the running speed of the current road section and determining whether a traffic accident happens to the current road section according to the target vehicle set, the road image detection result and preset traffic accident detection conditions;
and the sending unit is used for generating a traffic accident detection result and sending the traffic accident detection result to a traffic accident confirmation system under the condition that the traffic accident of the current road section is detected, so that the traffic accident confirmation system confirms the traffic accident detection result.
13. An electronic device, comprising:
a processor; and
a memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method of any one of claims 1 to 11.
14. A computer readable storage medium storing one or more programs which, when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of any of claims 1-11.
CN202211160293.0A 2022-09-22 2022-09-22 Traffic accident detection method and device, electronic device and storage medium Pending CN115359443A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211160293.0A CN115359443A (en) 2022-09-22 2022-09-22 Traffic accident detection method and device, electronic device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211160293.0A CN115359443A (en) 2022-09-22 2022-09-22 Traffic accident detection method and device, electronic device and storage medium

Publications (1)

Publication Number Publication Date
CN115359443A true CN115359443A (en) 2022-11-18

Family

ID=84005814

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211160293.0A Pending CN115359443A (en) 2022-09-22 2022-09-22 Traffic accident detection method and device, electronic device and storage medium

Country Status (1)

Country Link
CN (1) CN115359443A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117475642A (en) * 2023-12-28 2024-01-30 辽宁艾特斯智能交通技术有限公司 Road traffic state detection method and device, electronic equipment and storage medium

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117475642A (en) * 2023-12-28 2024-01-30 辽宁艾特斯智能交通技术有限公司 Road traffic state detection method and device, electronic equipment and storage medium
CN117475642B (en) * 2023-12-28 2024-03-01 辽宁艾特斯智能交通技术有限公司 Road traffic state detection method and device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN108022450B (en) Auxiliary driving method based on cellular network and traffic control unit
CN110400478A (en) A kind of road condition notification method and device
CN108932849B (en) Method and device for recording low-speed running illegal behaviors of multiple motor vehicles
CN108806272B (en) Method and device for reminding multiple motor vehicle owners of illegal parking behaviors
CN105809964A (en) Traffic warning method and device
CN113012436B (en) Road monitoring method and device and electronic equipment
CN110032947B (en) Method and device for monitoring occurrence of event
CN113345258B (en) Road traffic guarantee method, system, equipment and medium based on edge calculation
CN108932850B (en) Method and device for recording low-speed driving illegal behaviors of motor vehicle
CN104575043A (en) Automatic prompt system and method during passing of motor vehicle through pedestrian crosswalk
CN111492416A (en) Violation monitoring system and violation monitoring method
CN115049993B (en) Vehicle abnormal stop monitoring method based on deep learning
CN114333344A (en) Motor vehicle violation snapshot method and device and electronic equipment
CN115393803A (en) Vehicle violation detection method, device and system and storage medium
CN115359443A (en) Traffic accident detection method and device, electronic device and storage medium
CN112507874B (en) Method and device for detecting motor vehicle jamming behavior
CN113420714A (en) Collected image reporting method and device and electronic equipment
CN114170798B (en) Message reminding system and method
CN113112814B (en) Snapshot method and device without stopping right turn and computer storage medium
CN115762153A (en) Method and device for detecting backing up
CN114414259A (en) Anti-collision test method and device for vehicle, electronic equipment and storage medium
JP2022027305A (en) Risky driving detection device, risky driving detection system, and risky driving detection program
CN113470213A (en) Data processing method and device, vehicle-mounted terminal equipment and server
CN114120660B (en) Expressway ramp exit early warning method, system, equipment and storage medium
CN117037476A (en) Fatigue driving detection method and device for high-speed scene, electronic equipment and storage medium

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