CN111275957A - Traffic accident information acquisition method, system and camera - Google Patents
Traffic accident information acquisition method, system and camera Download PDFInfo
- Publication number
- CN111275957A CN111275957A CN201811478921.3A CN201811478921A CN111275957A CN 111275957 A CN111275957 A CN 111275957A CN 201811478921 A CN201811478921 A CN 201811478921A CN 111275957 A CN111275957 A CN 111275957A
- Authority
- CN
- China
- Prior art keywords
- accident
- traffic
- accident point
- video
- camera
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—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
- G08G1/20—Monitoring the location of vehicles belonging to a group, e.g. fleet of vehicles, countable or determined number of vehicles
- G08G1/205—Indicating the location of the monitored vehicles as destination, e.g. accidents, stolen, rental
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention provides a traffic accident information acquisition method, a system and a camera, wherein when a traffic accident is detected by the camera, a first relative position of an accident point of the traffic accident in an acquired video is acquired, the physical position of the accident point under a world coordinate system is converted according to the first relative position, first traffic information of each lane in a preset area range of the accident point is acquired according to the physical position of the accident point and a preset time interval, the video is acquired, the first traffic information acquired at the accident point and the acquired video are sent to a platform server, the platform server acquires second traffic information of each lane at an upstream and downstream intersection of the accident point according to the physical position of the accident point and the preset time interval, and the physical position of the accident point and the first traffic information acquired at each preset time interval are converted into a video, And the second stream information and the video are recorded in an associated manner. Through the scheme, the traffic accident information can be acquired in real time.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to a traffic accident information acquisition method, a system and a camera.
Background
Along with the explosive growth of motor vehicles, the urban traffic pressure is higher and higher, traffic accidents such as vehicle anchor dropping and vehicle scratch can be inevitably caused in urban road traffic, and the normal traffic of the vehicles is seriously influenced by the occurrence of the traffic accidents. The traffic accident needs to be recorded and analyzed, a traffic management basis is provided for urban traffic management departments, and the influence of the traffic accident on road traffic is reduced as much as possible.
At present, the influence of traffic accidents on road traffic is to analyze the temporal-spatial distribution rule of the accidents based on the macroscopic data such as the types, the grades, the influence degrees, the time, the space and the like of historical traffic accidents, and to present the influence of the traffic accidents by using a statistical table or to study and judge the situations of the accidents.
However, road traffic has the characteristics of real-time performance, mobility, nonlinearity and the like, particularly, traffic accidents such as vehicle breakdown, scraping and the like have strong randomness and unpredictability in early and late rush hours, and the occurrence time and place are difficult to regularly follow. Therefore, how to realize the real-time collection of the traffic accident information becomes a problem to be solved urgently in intelligent traffic control in order to enable the analysis result and the research and judgment result of the influence rule of the traffic accident to have higher accuracy.
Disclosure of Invention
The embodiment of the invention aims to provide a traffic accident information acquisition method, a system and a camera so as to realize real-time acquisition of traffic accident information. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a traffic accident information acquisition method, which is applied to a camera, and the method includes:
when a traffic accident is detected, acquiring a first relative position of an accident point of the traffic accident in a collected video;
converting the physical position of the accident point in a world coordinate system according to the first relative position;
acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video;
and sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server, so that the platform server acquires the second traffic information of each lane at the upstream and downstream intersections of the accident point according to the physical position of the accident point and the preset time interval, and performs associated recording on the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information and the video.
Optionally, the converting the physical position of the accident point in the world coordinate system according to the first relative position includes:
determining the relative positions of at least three reference targets which are closest to the accident point in the video according to the first relative position;
searching the physical positions of the at least three reference targets in a world coordinate system according to the relative positions of the at least three reference targets and the pre-stored corresponding relationship between the relative positions and the physical positions of the reference targets;
establishing a transformation matrix of a coordinate system of the video and a world coordinate system according to the relative positions and the physical positions of the at least three reference targets;
and converting the physical position of the accident point in the world coordinate system according to the first relative position and the transformation matrix.
Optionally, the converting the physical position of the accident point in the world coordinate system according to the first relative position includes:
determining a target calibration area to which the accident point belongs according to the first relative position and calibration areas divided in advance based on the relative positions of the reference targets in the video;
and converting to obtain the physical position of the accident point in a world coordinate system according to the first relative position and the homography matrix corresponding to the target calibration area acquired in advance.
Optionally, the method further includes:
acquiring equipment parameters of the camera;
determining a high-precision map matched with the camera according to the equipment parameters;
identifying a target located at a specified relative position in the video, and acquiring a physical position of the target in the high-precision map;
and determining the relative position of each reference target in the video according to the position relation between the target and the peripheral reference target.
Optionally, the converting the physical position of the accident point in the world coordinate system according to the first relative position includes:
acquiring equipment parameters of the camera, wherein the equipment parameters comprise a field angle, an erection height value and longitude and latitude;
determining a PT coordinate when the camera is over against the accident point according to the first relative position and the field angle, wherein the PT coordinate is used as a first P coordinate and a first T coordinate;
acquiring a P coordinate of the camera when the camera points to a specified direction, and taking the P coordinate as a second P coordinate;
calculating the difference between the first P coordinate and the second P coordinate to serve as a horizontal included angle between the accident point and the designated direction;
calculating the product of the tangent value of the first T coordinate and the erection height value as the horizontal distance between the accident point and the camera;
calculating the longitude and latitude distance between the accident point and the camera through a trigonometric function according to the horizontal included angle and the horizontal distance;
and calculating the physical position of the accident point in a world coordinate system according to the longitude and latitude of the camera and the longitude and latitude distance.
Optionally, the first traffic information includes a queuing length of each lane in a preset area range of the accident point;
the acquiring first traffic information of each lane in a preset area range of the accident point includes:
according to the physical position of the accident point, second relative positions of a plurality of vehicles on each lane in a preset area range of the accident point in the video are obtained;
converting the physical positions of a plurality of vehicles on each lane in the preset area range under a world coordinate system according to the second relative position;
and extracting and differentiating the physical position of the tail vehicle and the physical position of the head vehicle according to the physical positions of the head vehicle and the tail vehicle in the vehicle queue on each lane in the preset area range to obtain the queuing length of each lane in the preset area range.
In a second aspect, an embodiment of the present invention provides a traffic accident information acquisition method, which is applied to a platform server, and the method includes:
receiving the physical position of an accident point sent by a camera, and first traffic information and videos acquired at each preset time interval;
acquiring second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera of the upstream and downstream road junctions of the accident point, according to the physical position of the accident point and the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
Optionally, after the second traffic information of each lane of the upstream and downstream road junctions, which is acquired by the second camera at the upstream and downstream road junctions of the accident point, is acquired according to the physical location of the accident point and the preset time interval, the method further includes:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
if so, sending a trigger signal to a pan-tilt camera so that the pan-tilt camera tracks and collects accident videos of the traffic accident according to the preset time interval, stopping tracking and collecting until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value, and sending the collected accident videos to the platform server;
the performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval includes:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
Optionally, after the performing associated recording on the physical location of the accident point, the first traffic information, the second traffic information, and the video acquired at each preset time interval, the method further includes:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, and the first traffic information, the second traffic information and the video which are acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
In a third aspect, an embodiment of the present invention provides a method for acquiring traffic accident information, where the method includes:
acquiring a video sent by a camera when a traffic accident is detected;
identifying a first relative location of an accident point in the video at which the traffic accident occurred;
acquiring the physical position of the accident point in a world coordinate system according to the first relative position;
according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera;
and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
Optionally, after the first traffic information of each lane in a preset area range of the accident point, the second traffic information of each lane at an upstream/downstream intersection of the accident point, and the video acquired by the camera are acquired according to the physical position of the accident point and a preset time interval, the method further includes:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
if so, sending a trigger signal to a pan-tilt camera so as to enable the pan-tilt camera to track and acquire the accident video of the traffic accident according to the preset time interval, and stopping tracking and acquisition until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value;
the performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the videos acquired at each preset time interval includes:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
Optionally, after the performing associated recording on the physical location of the accident point, the first traffic information, the second traffic information, and the video acquired at each preset time interval, the method further includes:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, and the first traffic information, the second traffic information and the video which are acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
In a fourth aspect, an embodiment of the present invention provides a camera, including a processor and a memory;
the memory is used for storing a computer program;
the processor is configured to execute the computer program stored in the memory, so as to implement the method for acquiring traffic accident information according to the first aspect of the embodiment of the present invention.
In a fifth aspect, an embodiment of the present invention provides a platform server, including a processor and a memory;
the memory is used for storing a computer program;
the processor is configured to execute the computer program stored in the memory, so as to implement the traffic accident information collection method provided by the second aspect of the embodiment of the present invention, or to implement the traffic accident information collection method provided by the third aspect of the embodiment of the present invention.
In a sixth aspect, an embodiment of the present invention provides a traffic accident information acquisition system, including multiple cameras and a platform server;
the camera is used for acquiring a first relative position of an accident point of the traffic accident in the acquired video when the traffic accident is detected; converting the physical position of the accident point in a world coordinate system according to the first relative position; acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video; sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server;
the platform server is used for receiving the physical position of the accident point sent by the camera, and first traffic flow information and videos acquired at each preset time interval; acquiring second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera of the upstream and downstream road junctions of the accident point, according to the physical position of the accident point and the preset time interval; and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
In a seventh aspect, an embodiment of the present invention provides a traffic accident information acquisition system, including multiple cameras and a platform server;
the camera is used for acquiring a video;
the platform server is used for acquiring a video sent by the camera when a traffic accident is detected; identifying a first relative location of an accident point in the video at which the traffic accident occurred; acquiring the physical position of the accident point in a world coordinate system according to the first relative position; according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera; and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
According to the traffic accident information acquisition method, the system and the camera provided by the embodiment of the invention, when a traffic accident is detected by the camera, a first relative position of an accident point of the traffic accident in an acquired video is acquired, the physical position of the accident point under a world coordinate system is converted according to the first relative position, first traffic information of each lane in a preset area range of the accident point is acquired according to the physical position of the accident point and a preset time interval, the video is acquired, the first traffic information acquired by the physical position of the accident point, the first traffic information acquired by each preset time interval and the acquired video are sent to a platform server, the platform server acquires second traffic information of each lane at an upstream and downstream intersection of the accident point according to the physical position of the accident point and the preset time interval, and the physical position of the accident point and the first traffic information acquired by each preset time interval are transmitted to the platform server, And the second stream information and the video are recorded in an associated manner. The camera detects traffic accidents occurring in the collected video in real time, once the traffic accidents are detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a traffic accident information acquisition method applied to a camera according to an embodiment of the present invention;
fig. 2 is a schematic flow chart illustrating a traffic accident information collecting method applied to a platform server according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a traffic accident information acquisition method in which a camera interacts with a platform server according to an embodiment of the present invention;
FIG. 4 is a schematic view of pan-tilt-zoom camera tracking and acquisition according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a simulation process according to an embodiment of the present invention;
fig. 6 is a schematic flow chart illustrating a traffic accident information collecting method applied to a platform server according to another embodiment of the present invention;
fig. 7 is a schematic structural diagram of a traffic accident information collecting device applied to a camera according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a traffic accident information collecting device applied to a platform server according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a traffic accident information collecting device applied to a platform server according to another embodiment of the present invention;
fig. 10 is a schematic structural diagram of a camera according to an embodiment of the present invention;
FIG. 11 is a block diagram of a platform server according to an embodiment of the present invention;
FIG. 12 is a block diagram of a platform server according to another embodiment of the present invention;
fig. 13 is a schematic structural diagram of a traffic accident information collection system according to an embodiment of the present invention;
fig. 14 is a schematic structural diagram of a traffic accident information collecting system according to another embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, 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 invention.
In order to realize the real-time acquisition of traffic accident information, the embodiment of the invention provides a traffic accident information acquisition method, a traffic accident information acquisition device, a camera, a platform server, a machine-readable storage medium and a machine-readable storage system.
The terms in the examples of the present invention are explained as follows:
high-precision position: the method is divided into different grades, the grades are commonly used in the prior art and are centimeter, decimeter and meter, and the corresponding application fields are different in different precision data acquisition modes.
Traffic accidents: the vehicle is in the event of personal injury or property loss caused by mistake or accident on the road. Traffic accidents are not only caused by unspecified persons violating road traffic safety regulations; or due to irresistible natural disasters such as earthquake, typhoon, mountain torrents, lightning stroke and the like.
The method for acquiring the traffic accident information provided by the embodiment of the invention can be applied to a camera, and as shown in fig. 1, the method can comprise the following steps.
S101, when a traffic accident is detected, a first relative position of an accident point of the traffic accident in the collected video is obtained.
And S102, converting the physical position of the accident point in a world coordinate system according to the first relative position.
S103, acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video.
And S104, sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server, so that the platform server acquires the second traffic information of each lane at the upstream and downstream intersections of the accident point according to the physical position of the accident point and the preset time interval, and performing associated recording on the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information and the video.
The method for acquiring traffic accident information provided by the embodiment of the invention can also be applied to a platform server, and as shown in fig. 2, the method can comprise the following steps.
S201, receiving the physical position of the accident point sent by the camera, and the first traffic information and the video acquired at each preset time interval.
S202, according to the physical position of the accident point and the preset time interval, second traffic information of each lane of the upstream and downstream intersections acquired by the second cameras of the upstream and downstream intersections of the accident point is acquired.
S203, the physical position of the accident point, the first traffic information, the second traffic information and the video which are acquired at each preset time interval are recorded in a correlated mode.
When a traffic accident is detected by a camera, a first relative position of an accident point of the traffic accident in an acquired video is acquired, a physical position of the accident point under a world coordinate system is converted according to the first relative position, first traffic information of each lane in a preset area range of the accident point is acquired according to the physical position of the accident point and a preset time interval, the video is acquired, the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video are sent to a platform server, the platform server acquires second traffic information of each lane at an upstream and downstream intersection of the accident point according to the physical position of the accident point and the preset time interval, and the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information and the video are recorded in a correlation mode. The camera detects traffic accidents occurring in the collected video in real time, once the traffic accidents are detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
For easier understanding, the traffic accident information collection method provided by the embodiment of the present invention is described in detail below from the perspective of interaction between the camera and the platform server. The interaction flow chart of the camera and the platform server is shown in fig. 3, and comprises the following steps.
S301, when the camera detects a traffic accident, the camera acquires a first relative position of an accident point of the traffic accident in the acquired video.
In the urban traffic monitoring system, a plurality of cameras are erected at each intersection and above each road and used for monitoring the vehicle passing condition of the road, the cameras have a target detection function, and whether a traffic accident occurs in a monitored area or not is detected through algorithms such as a deep learning network and feature matching. Each camera is provided with a certain visible angle range, the cameras detect traffic accidents within the visible angle range of the cameras, and if the traffic accidents occur outside the visible angle range, the cameras cannot detect the traffic accidents.
Once the camera detects that a traffic accident occurs, the first relative position of the accident point where the traffic accident occurs can be acquired from the acquired video. The first relative position of the accident point in the video is position information of the accident point in the video compared with reference points such as edges and centers, and a corresponding coordinate system can also be established for the video, and the first relative position is a coordinate position of the accident point in the coordinate system.
And S302, converting the physical position of the accident point in a world coordinate system by the camera according to the first relative position.
The first relative position is the relative position of the accident point in the video collected by the camera, and a certain mapping relation exists between the coordinate system of the video collected by the camera and the world coordinate system, so that the physical position of the accident point in the world coordinate system can be converted through the conversion of the coordinate system. The physical location represents detailed information such as a specific geographical location of an accident point, occurrence time of a traffic accident and the like, and the physical location can be a GPS location.
Optionally, S302 may specifically be:
determining the relative positions of at least three reference targets which are closest to the accident point in the video according to the first relative position;
searching the physical positions of the at least three reference targets in a world coordinate system according to the relative positions of the at least three reference targets and the pre-stored corresponding relationship between the relative positions and the physical positions of the reference targets;
establishing a transformation matrix of a coordinate system of the video and a world coordinate system according to the relative positions and the physical positions of at least three reference targets;
and converting the physical position of the accident point in a world coordinate system according to the first relative position and the transformation matrix.
The relative positions of at least three reference targets such as the mark line and the street lamp around the first relative position can be extracted through the first relative position, and the physical positions of the reference targets can be found out because the corresponding relation between the relative position and the physical position of each reference target is stored in advance, so that a transformation matrix of a coordinate system of a video and a world coordinate system can be established based on the physical position and the relative position of the reference targets. Because the transformation matrix is established based on the reference target near the accident point, the first relative position is brought into the transformation matrix through coordinate transformation, and the physical position of the accident point can be obtained.
Optionally, S302 may specifically be:
determining a target calibration area to which the accident point belongs according to the first relative position and calibration areas divided in advance based on the relative positions of the reference targets in the video;
and converting to obtain the physical position of the accident point in a world coordinate system according to the first relative position and the homography matrix corresponding to the target calibration area acquired in advance.
Because a plurality of reference targets exist in the visual range of the camera, the positions of the reference targets can be divided in advance, a plurality of calibration areas can be divided in the video correspondingly, at least three vertexes of each calibration area are relative position points of at least three reference targets with short distances in the video, a target calibration area to which an accident point belongs can be determined according to a first relative position, a corresponding homography matrix is preset for each calibration area, the mapping relation between the physical position of each reference target in the calibration area and the relative position in the video is recorded in the homography matrix, and the position mapping relation between the accident point and each reference target in the target calibration area is the closest to the position mapping relation between each reference target in the target calibration area because the accident point belongs to the target calibration area, the first relative position can be directly brought into the homography matrix corresponding to the target calibration area, and obtaining the physical position of the accident point through conversion.
Optionally, the method provided in the embodiment of the present invention may further implement the following steps:
acquiring the equipment parameters of the camera;
determining a high-precision map matched with the camera according to the equipment parameters;
identifying a target located at a specified relative position in a video, and acquiring a physical position of the target in a high-precision map;
and determining the relative position of each reference target in the video according to the position relation between the target and the peripheral reference target.
In the above methods for converting physical positions, the mapping relationship between the relative position of the reference target in the video and the physical position in the world coordinate system needs to be recorded in advance, such recording requires matching and fusing a high-precision map and a video in advance, and the orientation angle, longitude and latitude value and other device parameters of the camera can be known in advance, for example, can be obtained through calibration or through equipment such as a GPS chip, a gyroscope/electronic compass and the like carried by the camera, determining a high-precision map which can be matched according to the orientation, physical position and other equipment parameters of the camera, determining a target point by the camera through image recognition, e.g., the intersection center point, the physical location of this target point in the high-precision map is also unambiguous, and then determining the relative position of each reference target in the video one by one according to the position relationship between the target point and the peripheral reference targets. The selected reference target can be a road sign line, a street lamp, a well cover, an isolation guardrail and the like, and the precision of the physical position obtained by conversion can meet the high-precision requirement, for example, the precision within the range of 200 meters can be within 1 meter, so that the lane-level application can be met.
Optionally, S302 may specifically be:
acquiring equipment parameters of the camera, wherein the equipment parameters comprise a field angle, an erection height value and longitude and latitude;
determining a PT coordinate when the camera is over against the accident point according to the first relative position and the field angle, and taking the PT coordinate as a first P coordinate and a first T coordinate;
acquiring a P coordinate of the camera when the camera points to the designated direction as a second P coordinate;
calculating the difference between the first P coordinate and the second P coordinate to be used as a horizontal included angle between the accident point and the designated direction;
calculating the product of the tangent value of the first T coordinate and the erection height value as the horizontal distance between the accident point and the camera;
calculating the longitude and latitude distance between the accident point and the camera through a trigonometric function according to the horizontal included angle and the horizontal distance;
and calculating the physical position of the accident point in a world coordinate system according to the longitude and latitude and the longitude and latitude distance of the camera.
The PT coordinates of the accident point shot by the camera can be read firstly, and then the read PT coordinates are converted into the PT coordinates of the accident point when the accident point is just aligned to the accident point according to the relative position of the accident point in the video shot by the camera and the field angle of the accident point shot by the camera, and the PT coordinates are used as a first P coordinate and a first T coordinate. Assuming that the relative position of the accident point in the video captured by the camera is (X, Y), the first P coordinate and the first T coordinate can be obtained by using the following equations:
Pan_tar=Pan_cur+arctan((2*X/L1-1)*tan(θ1/2));
Tilt_tar=Tilt_cur+arctan((2*Y/L2-1)*tan(θ2/2));
wherein Pan _ tar represents a first P coordinate, Tilt _ tar represents a first T coordinate, Pan _ cur represents a horizontal direction angle of the current camera in the PT coordinate system, Tilt _ cur represents a vertical direction angle of the current camera in the PT coordinate system, (Pan _ cur, Tilt _ cur) corresponds to a center position of the current video, L _ cur represents a center position of the current video, and1representing the total number of pixels in the horizontal direction of the video, L2Representing the total number of pixels, theta, in the vertical direction of the video1Expressed as the horizontal field angle, theta, corresponding to the current video2Representing the vertical field angle corresponding to the current video; the XY coordinate system uses the top left corner of the video as the origin and pixel as the unit.
The coordinates of the camera P when the camera points in the north, south, east, west, and the like directions can be acquired through an electronic compass of the camera, and are referred to as second P coordinates for descriptive purposes. The difference between the first P coordinate and the second P coordinate is the horizontal included angle between the accident point and the designated direction.
The horizontal distance between the accident point and the camera can be calculated according to the value of tanT h, wherein h represents the erection height value of the camera, and L represents the horizontal distance between the accident point and the camera. The horizontal distance is the distance from the camera to the accident point, assuming that the camera is at the same height as the accident point.
Assuming that the designated direction is the north direction, the direction may be defined by L sin θ Llon,L*cosθ=LlatCalculating longitude and latitude distances between the accident point and the camera, wherein L represents the horizontal distance between the accident point and the camera, theta represents the horizontal included angle between the accident point and the due north direction, and LlonIndicating the longitudinal distance, L, of the point of accident from the cameralatFurther, assuming that the designated direction is the east-ward direction, it may be determined that the accident point is the latitudinal distance from the camera, for example, by L sin α -Llon,L*cosα=LlatCalculating the longitude and latitude distance between the accident point and the camera, wherein L represents the horizontal distance between the accident point and the camera, α represents the horizontal included angle between the accident point and the east direction, and L represents the horizontal included angle between the accident point and the east directionlonIndicating the longitudinal distance, L, of the point of accident from the cameralatIndicating the latitude distance of the accident point from the camera. For specified directions of true south and true westThe specific calculation process is similar and will not be described herein again.
The camera is usually provided with a GPS positioning device, and the longitude and latitude of the camera can be obtained based on the GPS positioning device, so that the longitude and latitude of the camera and the longitude and latitude distance between the camera and an accident point are obtained, the longitude and latitude of the accident point can be obtained through calculation, and the physical position of the accident point in a world coordinate system is also obtained.
And S303, acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval by using the camera, and acquiring a video.
After the physical position of the accident point is determined, the accident point can cause the traffic conditions of all lanes in the surrounding area and all lanes at the upstream and downstream intersections to change, and the change of the traffic conditions can be reflected through the traffic information, and the traffic information changes at each moment, so that the first traffic information of the lanes in the surrounding area of the accident point can be obtained according to the physical position of the accident point and the preset time interval, and in order to observe the traffic flow and the vehicle queuing condition more intuitively, videos can be collected according to the preset time interval. Assuming that the frame rate of the video captured by the camera is N, the preset time interval may be 1s/N at minimum, and the preset time interval may be set according to actual situations, for example, may be set to be an integer multiple of 1s/N, and may also be set to be 1s, and the like.
The camera that detects the traffic accident may be a fixed camera, and may also be a pan-tilt camera. The fixed camera is erected above a road, the installation position is fixed, zooming is not available, the range of a visible angle is small, and shooting parameters are not adjustable; the pan-tilt camera is a camera with a pan-tilt, is provided with a device for bearing the camera to rotate in the horizontal direction and the vertical direction, can make the camera shoot from a plurality of angles by mounting the camera on the pan-tilt, and has larger zoom and visible angle range.
Optionally, the first traffic information may include a queuing length of each lane in a preset area range of the accident point;
correspondingly, the step of obtaining the first traffic information of each lane in the preset area range of the accident point may specifically be:
according to the physical position of the accident point, second relative positions of a plurality of vehicles on each lane in the preset area range of the accident point in the video are obtained;
converting the physical positions of a plurality of vehicles on each lane in the preset area range under a world coordinate system according to the second relative position;
and extracting and differentiating the physical position of the tail vehicle and the physical position of the head vehicle according to the physical positions of the head vehicle and the tail vehicle in the vehicle queue on each lane in the preset area range to obtain the queuing length of each lane in the preset area range.
The video collected by the camera comprises vehicles of all lanes (including left and right lanes in the same direction, lanes where accidents occur, left-turning lanes in the same direction, right-turning lanes and the like) in an accident point preset area range, the second relative positions of the vehicles can be obtained, the physical positions of a plurality of vehicles on all the lanes in the accident point preset area range can be obtained based on a mode of calculating the physical positions of the accident points, the physical positions of the same vehicle in different videos can be obtained due to continuous shooting, traffic flow information, average speed of the vehicles and the like can be obtained through calculation, the head vehicles and the tail vehicles of a vehicle queue can be identified through identifying the vehicle queue, coordinates of the two vehicles are differed, and traffic flow information of the queuing length can be obtained.
S304, the camera sends the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to the platform server.
The physical position of the accident point is obtained through conversion of the camera, after the first traffic information and the collected video are obtained at each preset time interval, the information can be sent to the platform server, and the platform server can record the traffic accident information conveniently. It should be noted that, when the camera acquires a first traffic information and acquires a video at the same time, the camera sends the acquired first traffic information and the acquired video to the platform server.
S305, the platform server acquires second traffic information of each lane of the upstream and downstream intersections acquired by the second cameras of the upstream and downstream intersections of the accident point according to the physical position of the accident point and the preset time interval.
Because traffic accidents may cause congestion at the upstream and downstream intersections, in order to grasp road conditions of the upstream and downstream intersections, second traffic information of each lane of the upstream and downstream intersections acquired by a second camera at the upstream and downstream intersection at an accident point needs to be acquired, and the second camera may be a fixed camera or a pan-tilt camera. The platform server sends an acquisition instruction to the second camera every time the platform server receives a piece of first traffic information, so that the second camera sends the second traffic information to the platform server, and therefore the platform server can acquire the first traffic information and the second traffic information synchronously. Of course, the platform server may also notify the second camera to acquire the second traffic information at a preset time interval after acquiring the physical location of the accident point.
The step of acquiring second traffic information of each lane at the upstream and downstream intersections of the accident point by the second camera may specifically be:
acquiring third relative positions of a plurality of vehicles on each lane in the acquired video in the video;
converting the physical positions of a plurality of vehicles on each lane of the upstream and downstream road junctions under a world coordinate system according to the third relative position;
and extracting and differentiating the physical position of the tail vehicle from the physical position of the head vehicle according to the physical positions of the head vehicle and the tail vehicle in the vehicle queue on each lane of the upstream and downstream intersections to obtain the queue length of each lane of the upstream and downstream intersections.
The second cameras are respectively erected at the upstream and downstream intersections of the accident point, the second cameras can acquire the third relative positions of the vehicles through target identification, the physical positions of a plurality of vehicles on each lane of the upstream and downstream intersections of the accident point can be acquired based on the mode of calculating the physical positions of the accident point, the physical positions of the same vehicle in different videos can be acquired due to continuous shooting, traffic information such as traffic flow, average speed of the vehicles and the like can be acquired through calculation, the head vehicles and the tail vehicles of the vehicle queues are identified through identifying the vehicle queues, and the traffic information of the queuing length can be acquired by differentiating the coordinates of the two vehicles.
Optionally, after executing S305, the platform server may further execute the following steps:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
and if so, sending a trigger signal to the pan-tilt camera so that the pan-tilt camera tracks and collects accident videos of the traffic accident according to a preset time interval, stopping tracking and collecting until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value, and sending the collected accident videos to the platform server.
For a camera, the shooting range is limited, when the camera recognizes a traffic accident, the camera needs to judge whether the traffic accident causes vehicle queuing, if the length of the queued vehicles exceeds a first preset length threshold, congestion is caused, at this time, a trigger signal is sent to a pan-tilt camera, after the pan-tilt camera receives the trigger signal, because the trigger signal can include information such as accident point physical position, queuing length and the like, the pan-tilt camera can track and shoot accident videos after analyzing the data until road traffic is recovered to be normal (the queuing length of the vehicles is less than a second preset length threshold). Of course, the instruction for stopping the tracking and collecting of the pan-tilt camera may also be sent by the fixed camera, the fixed camera may recognize the change of the queuing length in real time, if the queuing length is smaller than the second preset length threshold, the instruction for stopping the shooting is sent to the pan-tilt camera, and at this time, the pan-tilt camera may stop the tracking and collecting of the traffic accident.
Wherein, the step of cloud platform camera tracking collection the accident video of traffic accident specifically can be: positioning the tail vehicle of the accident point or the vehicle queue; and acquiring an accident video of the traffic accident by taking the physical position of the accident point or the tail vehicle as a focusing central point.
The way of shooting the accident video by the pan-tilt camera in a tracking manner can be to position the accident point or the tail vehicle of the vehicle queue, take the physical position of the accident point or the tail vehicle as a focusing central point, as shown in fig. 4, shoot along with the accident point or the tail vehicle all the time, and record the complete process from the occurrence to the extinction of the queuing length in real time according to the change of the queuing length.
S306, the platform server performs associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
The platform server can perform associated recording on the data and the physical position of the accident point every time when the first traffic information, the second traffic information and the video are acquired, records the information which can visually reflect the influence of the traffic accident on the road traffic in real time, and feeds the information back to an urban traffic management department or directly performs traffic accident simulation, so that the traffic management department can acquire the road traffic condition in real time, timely perform traffic control such as accident evacuation, signal lamp control and the like, and reduce the influence of the traffic accident on the normal traffic of the road traffic as much as possible.
Optionally, S306 may specifically be:
the method comprises the steps that a platform server obtains accident videos collected by a pan-tilt camera according to a preset time interval;
and the platform server performs associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
If the pan-tilt camera is triggered to track and shoot the traffic accident, accident videos collected by the pan-tilt camera according to preset time intervals can be obtained, and the physical position of the accident point, the first traffic information, the second traffic information, the videos collected by the camera and the accident videos collected by the pan-tilt camera are recorded in a correlated mode, so that the traffic accident rule is recorded more completely.
Optionally, after executing S306, the platform server may further execute the following steps:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
The associated stored information can determine the selection and optimization of the timing strategy, so that signal lamps of the accident point intersection and the upstream and downstream intersections can be controlled, and congestion can be dredged in time. As shown in fig. 5, for a traffic accident collected in real time, first traffic information of each lane in a preset area range of an accident point and second traffic information of each lane at an upstream intersection and a downstream intersection are stored in a data resource pool based on a physical position of the accident point of the traffic accident, a corresponding timing strategy is determined, and the corresponding timing strategy is stored in an expert database for researching the timing strategy, wherein the timing strategy of each time period is stored in the expert database. Meanwhile, the timing strategy is evaluated through a signal lamp control evaluation algorithm, whether the quality requirement is met (whether the traffic efficiency can be improved) is judged, and the timing strategy is operated to control the signal lamp if the quality requirement is met. If the quality requirement is not met, the timing strategy is diagnosed and model optimization is carried out by combining the associated and stored traffic accident information, the optimized timing strategy is determined, the previous process is repeated, and the quality of the timing strategy is improved through continuous learning. The diagnosis mode comprises the judgment of whether the conditions of the air release and the overflow occur, the model optimization can be based on the data extraction model optimization parameters (such as queuing length), and the specific model optimization mode can be a traditional model algorithm or a machine learning algorithm.
By applying the embodiment, when a traffic accident is detected, the camera acquires a first relative position of an accident point of the traffic accident in the acquired video, converts the physical position of the accident point under a world coordinate system according to the first relative position, acquires first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and preset time intervals, acquires the video, transmits the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to the platform server, the platform server acquires second traffic information of each lane at an upstream and downstream intersection of the accident point according to the physical position of the accident point and the preset time intervals, and performs associated recording on the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information and the video. The camera detects traffic accidents occurring in the collected video in real time, once the traffic accidents are detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
The embodiment of the invention also provides a traffic accident information acquisition method, as shown in fig. 6, the method is applied to a platform server, and comprises the following steps.
S601, acquiring a video sent by the camera when the traffic accident is detected.
And once the camera detects that a traffic accident occurs, sending the acquired video to the platform server.
S602, identifying a first relative position of an accident point of the traffic accident in the video.
The platform server has a target identification function and can identify a first relative position of an accident point of a traffic accident in the video.
And S603, acquiring the physical position of the accident point in a world coordinate system according to the first relative position.
After the platform server identifies the first relative position, the physical position of the accident point can be converted based on the stored conversion relation, the first relative position can be sent to the camera, the physical position of the accident point is converted by the camera, and the physical position of the accident point is fed back to the platform server. The specific conversion process is shown in the embodiment shown in fig. 3, and is not described herein again.
S604, according to the physical position of the accident point and the preset time interval, acquiring first traffic information of each lane in the preset area range of the accident point, second traffic information of each lane at the upstream and downstream intersections of the accident point and videos acquired by the camera.
The platform server can respectively send an acquisition instruction to the camera which detects the traffic accident and the second camera at the upstream and downstream intersections of the accident point according to the physical position of the accident point and the preset time interval, so that the camera sends first traffic information and collected videos of all lanes in the preset area range of the accident point to the platform server, and the second camera sends second traffic information of all lanes at the upstream and downstream intersections of the accident point to the platform server.
And S605, performing related recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
The platform server can perform associated recording on the data and the physical position of the accident point every time when the first traffic information, the second traffic information and the video are acquired, records the information which can visually reflect the influence of the traffic accident on the road traffic in real time, and feeds the information back to an urban traffic management department or directly performs traffic accident simulation, so that the traffic management department can acquire the road traffic condition in real time, timely perform traffic control such as accident evacuation, signal lamp control and the like, and reduce the influence of the traffic accident on the normal traffic of the road traffic as much as possible.
Optionally, after executing S604, the embodiment of the present invention may further execute the following steps:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
and if so, sending a trigger signal to the pan-tilt camera so that the pan-tilt camera tracks and collects accident videos of the traffic accident according to a preset time interval, and stopping tracking and collecting until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value.
Then, S605 may specifically be:
acquiring an accident video acquired by a pan-tilt camera according to a preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
For a camera, the shooting range is limited, when the camera recognizes a traffic accident, the camera needs to judge whether the traffic accident causes vehicle queuing, if the length of the queued vehicles exceeds a first preset length threshold, congestion is caused, at this time, a trigger signal is sent to a pan-tilt camera, after the pan-tilt camera receives the trigger signal, because the trigger signal can include information such as accident point physical position, queuing length and the like, the pan-tilt camera can track and shoot accident videos after analyzing the data until road traffic is recovered to be normal (the queuing length of the vehicles is less than a second preset length threshold). Of course, the instruction for stopping the tracking and collecting of the pan-tilt camera may also be sent by the fixed camera, the fixed camera may recognize the change of the queuing length in real time, if the queuing length is smaller than the second preset length threshold, the instruction for stopping the shooting is sent to the pan-tilt camera, and at this time, the pan-tilt camera may stop the tracking and collecting of the traffic accident.
If the pan-tilt camera is triggered to track and shoot the traffic accident, accident videos collected by the pan-tilt camera according to preset time intervals can be obtained, and the physical position of the accident point, the first traffic information, the second traffic information, the videos collected by the camera and the accident videos collected by the pan-tilt camera are recorded in a correlated mode, so that the traffic accident rule is recorded more completely.
Optionally, after executing S605, the embodiment of the present invention may further execute the following steps:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the signal lamps of the upstream and downstream intersections of the accident point based on the optimized timing strategy.
The associated stored information can determine the selection and optimization of the timing strategy, so that signal lamps of the accident point intersection and the upstream and downstream intersections can be controlled, and congestion can be dredged in time.
By applying the embodiment, when a traffic accident is detected by the camera, the acquired video is sent to the platform server, the platform server can identify the first relative position of the accident point in the video, and further acquire the physical position of the accident point, and then based on the physical position of the accident point, the first traffic information, the second traffic information and the acquired video can be acquired according to the preset time interval, and the first traffic information, the second traffic information and the acquired video can visually and timely reflect the influence rule of the traffic accident on road traffic.
Corresponding to the above method embodiment, an embodiment of the present invention provides a traffic accident information collecting device, as shown in fig. 7, applied to a camera, where the device may include:
the acquisition module 710 is configured to acquire, when a traffic accident is detected, a first relative position of an accident point where the traffic accident occurs in a collected video;
a conversion module 720, configured to convert, according to the first relative position, a physical position of the accident point in a world coordinate system;
the obtaining module 710 is further configured to obtain first traffic information of each lane within a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and collect a video;
the sending module 730 is configured to send the physical position of the accident point, the first traffic information acquired at each preset time interval, and the collected videos to a platform server, so that the platform server acquires, according to the physical position of the accident point and according to the preset time interval, the second traffic information of each lane at the upstream and downstream intersection of the accident point, and performs associated recording on the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information, and each video.
Optionally, the scaling module 720 may be specifically configured to:
determining the relative positions of at least three reference targets which are closest to the accident point in the video according to the first relative position;
searching the physical positions of the at least three reference targets in a world coordinate system according to the relative positions of the at least three reference targets and the pre-stored corresponding relationship between the relative positions and the physical positions of the reference targets;
establishing a transformation matrix of a coordinate system of the video and a world coordinate system according to the relative positions and the physical positions of the at least three reference targets;
and converting the physical position of the accident point in the world coordinate system according to the first relative position and the transformation matrix.
Optionally, the scaling module 720 may be specifically configured to:
determining a target calibration area to which the accident point belongs according to the first relative position and calibration areas divided in advance based on the relative positions of the reference targets in the video;
and converting to obtain the physical position of the accident point in a world coordinate system according to the first relative position and the homography matrix corresponding to the target calibration area acquired in advance.
Optionally, the scaling module 720 may be further configured to:
acquiring equipment parameters of the camera;
determining a high-precision map matched with the camera according to the equipment parameters;
identifying a target located at a specified relative position in the video, and acquiring a physical position of the target in the high-precision map;
and determining the relative position of each reference target in the video according to the position relation between the target and the peripheral reference target.
Optionally, the scaling module 720 may be specifically configured to:
acquiring equipment parameters of the camera, wherein the equipment parameters comprise a field angle, an erection height value and longitude and latitude;
determining a PT coordinate when the camera is over against the accident point according to the first relative position and the field angle, wherein the PT coordinate is used as a first P coordinate and a first T coordinate;
acquiring a P coordinate of the camera when the camera points to a specified direction, and taking the P coordinate as a second P coordinate;
calculating the difference between the first P coordinate and the second P coordinate to serve as a horizontal included angle between the accident point and the designated direction;
calculating the product of the tangent value of the first T coordinate and the erection height value as the horizontal distance between the accident point and the camera;
calculating the longitude and latitude distance between the accident point and the camera through a trigonometric function according to the horizontal included angle and the horizontal distance;
and calculating the physical position of the accident point in a world coordinate system according to the longitude and latitude of the camera and the longitude and latitude distance.
Optionally, the first traffic information includes a queuing length of each lane in a preset area range of the accident point;
the obtaining module 710 may be specifically configured to:
according to the physical position of the accident point, second relative positions of a plurality of vehicles on each lane in a preset area range of the accident point in the video are obtained;
converting the physical positions of a plurality of vehicles on each lane in the preset area range under a world coordinate system according to the second relative position;
and extracting and differentiating the physical position of the tail vehicle and the physical position of the head vehicle according to the physical positions of the head vehicle and the tail vehicle in the vehicle queue on each lane in the preset area range to obtain the queuing length of each lane in the preset area range.
An embodiment of the present invention further provides a traffic accident information collecting device, as shown in fig. 8, which is applied to a platform server, and the device may include:
the receiving module 810 is configured to receive the physical location of the accident point sent by the camera, the first traffic information acquired at each preset time interval, and each video;
an obtaining module 820, configured to obtain second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera at the upstream and downstream road junctions of the accident point, according to the physical position of the accident point;
the recording module 830 is configured to perform associated recording on the physical location of the accident point, the first traffic information, the second traffic information, and each video acquired at each preset time interval.
Optionally, the apparatus may further include:
the judging module is used for judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated or not according to the first traffic information and the second traffic information acquired at the preset time interval aiming at each preset time interval;
the sending module is used for sending a trigger signal to the pan-tilt camera if the judgment result of the judging module is generated, so that the pan-tilt camera tracks and collects the accident video of the traffic accident according to the preset time interval until the fact that the queuing length of the vehicle queue is smaller than a second preset length threshold value is determined, the tracking and collecting are stopped, and the collected accident video is sent to the platform server;
the recording module may be specifically configured to:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
Optionally, the apparatus may further include:
the simulation module is used for carrying out real-time road traffic simulation on the basis of a traffic simulation algorithm according to the physical position of the accident point, the first traffic information, the second traffic information and the video which are acquired at the preset time interval aiming at each preset time interval;
the optimization module is used for diagnosing and optimizing a model for a preset timing strategy according to a simulation result and determining an optimized timing strategy;
and the control module is used for controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
By applying the embodiment, the camera detects the traffic accident in the acquired video in real time, once the traffic accident is detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
An embodiment of the present invention further provides a traffic accident information collecting device, as shown in fig. 9, the device may include:
an obtaining module 910, configured to obtain a video sent by a camera when a traffic accident is detected;
an identifying module 920, configured to identify a first relative position of an accident point of the video where the traffic accident occurs;
the obtaining module 910 is further configured to obtain a physical location of the accident point in a world coordinate system according to the first relative location; according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera;
the recording module 930 is configured to perform associated recording on the physical location of the accident point, the first traffic information, the second traffic information, and each video acquired at each preset time interval.
Optionally, the apparatus may further include:
the judging module is used for judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated or not according to the first traffic information and the second traffic information acquired at the preset time interval aiming at each preset time interval;
the sending module is used for sending a trigger signal to the pan-tilt camera if the judgment result of the judging module is generated, so that the pan-tilt camera tracks and collects the accident video of the traffic accident according to the preset time interval until the tracking and collection are stopped when the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold;
the recording module may be specifically configured to:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
Optionally, the apparatus may further include:
the simulation module is used for carrying out real-time road traffic simulation on the basis of a traffic simulation algorithm according to the physical position of the accident point, the first traffic information, the second traffic information and the video which are acquired at the preset time interval aiming at each preset time interval;
the optimization module is used for diagnosing and optimizing a model for a preset timing strategy according to a simulation result and determining an optimized timing strategy;
and the control module is used for controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
By applying the embodiment, when a traffic accident is detected by the camera, the acquired video is sent to the platform server, the platform server can identify the first relative position of the accident point in the video, and further acquire the physical position of the accident point, and then based on the physical position of the accident point, the first traffic information, the second traffic information and the acquired video can be acquired according to the preset time interval, and the first traffic information, the second traffic information and the acquired video can visually and timely reflect the influence rule of the traffic accident on road traffic.
Corresponding to the above method embodiment, the embodiment of the present invention provides a camera, as shown in fig. 10, including a processor 1001 and a memory 1002;
the memory 1002 is used for storing computer programs;
the processor 1001 is configured to execute the computer program stored in the memory 1002, and implement the following steps:
when a traffic accident is detected, acquiring a first relative position of an accident point of the traffic accident in a collected video;
converting the physical position of the accident point in a world coordinate system according to the first relative position;
acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video;
and sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server, so that the platform server acquires the second traffic information of each lane at the upstream and downstream intersections of the accident point according to the physical position of the accident point and the preset time interval, and performs associated recording on the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information and the video.
Optionally, when the step of converting the physical position of the accident point in the world coordinate system according to the first relative position is implemented, the processor 1001 may be specifically configured to implement the following steps:
determining the relative positions of at least three reference targets which are closest to the accident point in the video according to the first relative position;
searching the physical positions of the at least three reference targets in a world coordinate system according to the relative positions of the at least three reference targets and the pre-stored corresponding relationship between the relative positions and the physical positions of the reference targets;
establishing a transformation matrix of a coordinate system of the video and a world coordinate system according to the relative positions and the physical positions of the at least three reference targets;
and converting the physical position of the accident point in the world coordinate system according to the first relative position and the transformation matrix.
Optionally, when the step of converting the physical position of the accident point in the world coordinate system according to the first relative position is implemented, the processor 1001 may be specifically configured to implement the following steps:
determining a target calibration area to which the accident point belongs according to the first relative position and calibration areas divided in advance based on the relative positions of the reference targets in the video;
and converting to obtain the physical position of the accident point in a world coordinate system according to the first relative position and the homography matrix corresponding to the target calibration area acquired in advance.
Optionally, the processor 1001 may be further configured to implement the following steps:
acquiring equipment parameters of the camera;
determining a high-precision map matched with the camera according to the equipment parameters;
identifying a target located at a specified relative position in the video, and acquiring a physical position of the target in the high-precision map;
and determining the relative position of each reference target in the video according to the position relation between the target and the peripheral reference target.
Optionally, when the step of converting the physical position of the accident point in the world coordinate system according to the first relative position is implemented, the processor 1001 may be specifically configured to implement the following steps:
acquiring equipment parameters of the camera, wherein the equipment parameters comprise a field angle, an erection height value and longitude and latitude;
determining a PT coordinate when the camera is over against the accident point according to the first relative position and the field angle, wherein the PT coordinate is used as a first P coordinate and a first T coordinate;
acquiring a P coordinate of the camera when the camera points to a specified direction, and taking the P coordinate as a second P coordinate;
calculating the difference between the first P coordinate and the second P coordinate to serve as a horizontal included angle between the accident point and the designated direction;
calculating the product of the tangent value of the first T coordinate and the erection height value as the horizontal distance between the accident point and the camera;
calculating the longitude and latitude distance between the accident point and the camera through a trigonometric function according to the horizontal included angle and the horizontal distance;
and calculating the physical position of the accident point in a world coordinate system according to the longitude and latitude of the camera and the longitude and latitude distance.
Optionally, the first traffic information may include a queuing length of each lane in a preset area range of the accident point;
when the step of obtaining the first traffic information of each lane in the preset area range of the accident point is implemented, the processor 1001 may be specifically configured to implement the following steps:
according to the physical position of the accident point, second relative positions of a plurality of vehicles on each lane in a preset area range of the accident point in the video are obtained;
converting the physical positions of a plurality of vehicles on each lane in the preset area range under a world coordinate system according to the second relative position;
and extracting and differentiating the physical position of the tail vehicle and the physical position of the head vehicle according to the physical positions of the head vehicle and the tail vehicle in the vehicle queue on each lane in the preset area range to obtain the queuing length of each lane in the preset area range.
An embodiment of the present invention further provides a platform server, as shown in fig. 11, including a processor 1101 and a memory 1102;
the memory 1102 is used for storing computer programs;
the processor 1101 is configured to execute the computer program stored in the memory 1102, and implement the following steps:
receiving the physical position of an accident point sent by a camera, and first traffic information and videos acquired at each preset time interval;
acquiring second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera of the upstream and downstream road junctions of the accident point, according to the physical position of the accident point and the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
Optionally, the processor 1101 may further be configured to implement the following steps:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
if so, sending a trigger signal to a pan-tilt camera so that the pan-tilt camera tracks and collects accident videos of the traffic accident according to the preset time interval, stopping tracking and collecting until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value, and sending the collected accident videos to the platform server;
when the processor 1101 performs the step of performing associated recording on the physical location of the accident point, the first traffic information, the second traffic information, and the video acquired at each preset time interval, the processor 1101 may be specifically configured to perform the following steps:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
Optionally, the processor 1101 may further be configured to implement the following steps:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, and the first traffic information, the second traffic information and the video which are acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
Data transmission may be performed between the memory 1002 and the processor 1001, and between the memory 1102 and the processor 1101 by wired connection or wireless connection, and the camera may communicate with a device such as a platform server through a wired communication interface or a wireless communication interface.
The Memory may include a RAM (Random Access Memory) or an NVM (Non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.
In this embodiment, the processor can realize that: the camera detects traffic accidents occurring in the collected video in real time, once the traffic accidents are detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
In addition, an embodiment of the present invention provides a machine-readable storage medium for storing a computer program that causes a processor to execute all the steps of the traffic accident information collecting method applied to a camera provided by an embodiment of the present invention.
Embodiments of the present invention provide a machine-readable storage medium for storing a computer program, which causes a processor to execute all the steps of the traffic accident information collection method applied to a platform server provided by embodiments of the present invention.
In this embodiment, the machine-readable storage medium stores a computer program that executes the traffic accident information collection method applied to the camera and the platform server provided in the embodiment of the present invention when running, and thus can implement: the camera detects traffic accidents occurring in the collected video in real time, once the traffic accidents are detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
An embodiment of the present invention provides a platform server, as shown in fig. 12, including a processor 1201 and a memory 1202;
the memory 1202 for storing computer programs;
the processor 1201 is configured to execute the computer program stored in the memory 1202, and implement the following steps:
acquiring a video sent by a camera when a traffic accident is detected;
identifying a first relative location of an accident point in the video at which the traffic accident occurred;
acquiring the physical position of the accident point in a world coordinate system according to the first relative position;
according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera;
and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
Optionally, the processor 1201 may be further configured to implement the following steps:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
if so, sending a trigger signal to a pan-tilt camera so as to enable the pan-tilt camera to track and acquire the accident video of the traffic accident according to the preset time interval, and stopping tracking and acquisition until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value;
when the step of performing associated recording on the physical location of the accident point, the first traffic information, the second traffic information, and each video acquired at each preset time interval is implemented, the processor 1201 may be specifically configured to implement the following steps:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
Optionally, the processor 1201 may be further configured to implement the following steps:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, and the first traffic information, the second traffic information and the video which are acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
The memory 1202 and the processor 1201 may be connected by wire or wireless, and the platform server may communicate with a device such as a camera through a wire communication interface or a wireless communication interface.
The memory may include RAM, or may include NVM, such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The processor can be a general processor, including a CPU, an NP, etc.; but also DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
In this embodiment, the processor can realize that: when the camera detects a traffic accident, the acquired video is sent to the platform server, the platform server can identify a first relative position of an accident point in the video, and further acquire a physical position of the accident point, and then based on the physical position of the accident point, the first traffic information, the second traffic information and the acquired video can be acquired according to a preset time interval, and the first traffic information, the second traffic information and the acquired video can visually and timely reflect the influence rule of the traffic accident on road traffic, so that the first traffic information, the second traffic information and the video acquired through the physical position of the accident point and each preset time interval are subjected to associated recording, and real-time acquisition of traffic accident information is realized.
In addition, an embodiment of the present invention provides a machine-readable storage medium for storing a computer program, which causes a processor to execute all the steps of the traffic accident information collection method provided by the embodiment of the present invention.
In this embodiment, the machine-readable storage medium stores a computer program that executes the traffic accident information collection method provided by the embodiment of the present invention when running, so that it is possible to implement: when the camera detects a traffic accident, the acquired video is sent to the platform server, the platform server can identify a first relative position of an accident point in the video, and further acquire a physical position of the accident point, and then based on the physical position of the accident point, the first traffic information, the second traffic information and the acquired video can be acquired according to a preset time interval, and the first traffic information, the second traffic information and the acquired video can visually and timely reflect the influence rule of the traffic accident on road traffic, so that the first traffic information, the second traffic information and the video acquired through the physical position of the accident point and each preset time interval are subjected to associated recording, and real-time acquisition of traffic accident information is realized.
An embodiment of the present invention further provides a traffic accident information collecting system, as shown in fig. 13, including a plurality of cameras 1301 and a platform server 1302;
the camera 1301 is used for acquiring a first relative position of an accident point of the traffic accident in the acquired video when the traffic accident is detected; converting the physical position of the accident point in a world coordinate system according to the first relative position; acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video; sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server;
the platform server 1302 is configured to receive the physical location of the accident point sent by the camera, and the first traffic information and the video acquired at each preset time interval; acquiring second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera of the upstream and downstream road junctions of the accident point, according to the physical position of the accident point and the preset time interval; and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
By applying the embodiment, the camera detects the traffic accident in the acquired video in real time, once the traffic accident is detected, the first relative position of the accident point in the video can be acquired, further calculating the physical position of the accident point, acquiring the first traffic information and the collected video according to a preset time interval based on the physical position of the accident point, and the information is sent to a platform server, the platform server can also acquire second traffic information according to the physical position of the accident point, the first traffic information, the second traffic information and the collected video can intuitively and real-timely reflect the influence rule of the traffic accident on road traffic, therefore, the real-time acquisition of the traffic accident information is realized by performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
An embodiment of the present invention further provides a traffic accident information collecting system, as shown in fig. 14, including a plurality of cameras 1401 and a platform server 1402;
the camera 1401 is used for acquiring videos;
the platform server 1402 is configured to obtain a video sent by the camera when a traffic accident is detected; identifying a first relative location of an accident point in the video at which the traffic accident occurred; acquiring the physical position of the accident point in a world coordinate system according to the first relative position; according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera; and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
By applying the embodiment, when a traffic accident is detected by the camera, the acquired video is sent to the platform server, the platform server can identify the first relative position of the accident point in the video, and further acquire the physical position of the accident point, and then based on the physical position of the accident point, the first traffic information, the second traffic information and the acquired video can be acquired according to the preset time interval, and the first traffic information, the second traffic information and the acquired video can visually and timely reflect the influence rule of the traffic accident on road traffic.
For the embodiments of the device, the camera, the platform server, the machine-readable storage medium and the traffic accident information collecting system, the contents of the related methods are basically similar to those of the method embodiments, so that the description is simple, and relevant points can be referred to the partial description of the method embodiments.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the embodiments of the device, the camera, the platform server, the machine-readable storage medium and the traffic accident information collecting system, since they are basically similar to the embodiments of the method, the description is simple, and the relevant points can be referred to the partial description of the embodiments of the method.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (16)
1. A traffic accident information acquisition method is applied to a camera, and comprises the following steps:
when a traffic accident is detected, acquiring a first relative position of an accident point of the traffic accident in a collected video;
converting the physical position of the accident point in a world coordinate system according to the first relative position;
acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video;
and sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server, so that the platform server acquires the second traffic information of each lane at the upstream and downstream intersections of the accident point according to the physical position of the accident point and the preset time interval, and performs associated recording on the physical position of the accident point, the first traffic information acquired at each preset time interval, the second traffic information and the video.
2. The method of claim 1, wherein said converting the physical location of the incident point in a world coordinate system based on the first relative location comprises:
determining the relative positions of at least three reference targets which are closest to the accident point in the video according to the first relative position;
searching the physical positions of the at least three reference targets in a world coordinate system according to the relative positions of the at least three reference targets and the pre-stored corresponding relationship between the relative positions and the physical positions of the reference targets;
establishing a transformation matrix of a coordinate system of the video and a world coordinate system according to the relative positions and the physical positions of the at least three reference targets;
and converting the physical position of the accident point in the world coordinate system according to the first relative position and the transformation matrix.
3. The method of claim 1, wherein said converting the physical location of the incident point in a world coordinate system based on the first relative location comprises:
determining a target calibration area to which the accident point belongs according to the first relative position and calibration areas divided in advance based on the relative positions of the reference targets in the video;
and converting to obtain the physical position of the accident point in a world coordinate system according to the first relative position and the homography matrix corresponding to the target calibration area acquired in advance.
4. A method according to claim 2 or 3, characterized in that the method further comprises:
acquiring equipment parameters of the camera;
determining a high-precision map matched with the camera according to the equipment parameters;
identifying a target located at a specified relative position in the video, and acquiring a physical position of the target in the high-precision map;
and determining the relative position of each reference target in the video according to the position relation between the target and the peripheral reference target.
5. The method of claim 1, wherein said converting the physical location of the incident point in a world coordinate system based on the first relative location comprises:
acquiring equipment parameters of the camera, wherein the equipment parameters comprise a field angle, an erection height value and longitude and latitude;
determining a PT coordinate when the camera is over against the accident point according to the first relative position and the field angle, wherein the PT coordinate is used as a first P coordinate and a first T coordinate;
acquiring a P coordinate of the camera when the camera points to a specified direction, and taking the P coordinate as a second P coordinate;
calculating the difference between the first P coordinate and the second P coordinate to serve as a horizontal included angle between the accident point and the designated direction;
calculating the product of the tangent value of the first T coordinate and the erection height value as the horizontal distance between the accident point and the camera;
calculating the longitude and latitude distance between the accident point and the camera through a trigonometric function according to the horizontal included angle and the horizontal distance;
and calculating the physical position of the accident point in a world coordinate system according to the longitude and latitude of the camera and the longitude and latitude distance.
6. The method of claim 1, wherein the first traffic information includes a queue length of each lane within a preset area of the accident site;
the acquiring first traffic information of each lane in a preset area range of the accident point includes:
according to the physical position of the accident point, second relative positions of a plurality of vehicles on each lane in a preset area range of the accident point in the video are obtained;
converting the physical positions of a plurality of vehicles on each lane in the preset area range under a world coordinate system according to the second relative position;
and extracting and differentiating the physical position of the tail vehicle and the physical position of the head vehicle according to the physical positions of the head vehicle and the tail vehicle in the vehicle queue on each lane in the preset area range to obtain the queuing length of each lane in the preset area range.
7. A traffic accident information acquisition method is applied to a platform server, and comprises the following steps:
receiving the physical position of an accident point sent by a camera, and first traffic information and videos acquired at each preset time interval;
acquiring second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera of the upstream and downstream road junctions of the accident point, according to the physical position of the accident point and the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
8. The method according to claim 7, wherein after acquiring the second stream information of each lane of the upstream and downstream road junctions collected by the second camera at the upstream and downstream road junctions of the accident point according to the physical location of the accident point and the preset time interval, the method further comprises:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
if so, sending a trigger signal to a pan-tilt camera so that the pan-tilt camera tracks and collects accident videos of the traffic accident according to the preset time interval, stopping tracking and collecting until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value, and sending the collected accident videos to the platform server;
the performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval includes:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
9. The method according to claim 7, wherein after the recording of the physical location of the accident site, the first flow information, the second flow information and the video acquired at each preset time interval, the method further comprises:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, and the first traffic information, the second traffic information and the video which are acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
10. A method for collecting traffic accident information, the method comprising:
acquiring a video sent by a camera when a traffic accident is detected;
identifying a first relative location of an accident point in the video at which the traffic accident occurred;
acquiring the physical position of the accident point in a world coordinate system according to the first relative position;
according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera;
and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
11. The method according to claim 10, wherein after the acquiring first traffic information of each lane within a preset area range of the accident point, second traffic information of each lane at an intersection upstream and downstream of the accident point, and the video collected by the camera at preset time intervals according to the physical location of the accident point, the method further comprises:
aiming at each preset time interval, judging whether a vehicle queue with a queuing length larger than a first preset length threshold value is generated according to first traffic information and second traffic information acquired by the preset time interval;
if so, sending a trigger signal to a pan-tilt camera so as to enable the pan-tilt camera to track and acquire the accident video of the traffic accident according to the preset time interval, and stopping tracking and acquisition until the queuing length of the vehicle queue is determined to be smaller than a second preset length threshold value;
the performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the videos acquired at each preset time interval includes:
acquiring accident videos collected by the pan-tilt camera according to the preset time interval;
and performing associated recording on the physical position of the accident point, the first traffic flow information and the second traffic flow information acquired at each preset time interval, the video acquired by the camera and the accident video acquired by the pan-tilt camera.
12. The method of claim 10, wherein after the recording of the physical location of the accident site, the first flow information, the second flow information and the video obtained at each preset time interval, the method further comprises:
aiming at each preset time interval, performing real-time road traffic simulation based on a traffic simulation algorithm according to the physical position of the accident point, and the first traffic information, the second traffic information and the video which are acquired at the preset time interval;
according to the simulation result, diagnosing and optimizing a model for a preset timing strategy, and determining an optimized timing strategy;
and controlling the upstream and downstream intersection signal lamps of the accident point based on the optimized timing strategy.
13. A camera comprising a processor and a memory;
the memory is used for storing a computer program;
the processor is used for executing the computer program stored on the memory to realize the traffic accident information acquisition method according to the claims 1-6.
14. A platform server comprising a processor and a memory;
the memory is used for storing a computer program;
the processor is configured to execute the computer program stored in the memory, to implement the traffic accident information collection method according to any one of claims 7 to 9, or to implement the traffic accident information collection method according to any one of claims 10 to 12.
15. A traffic accident information acquisition system is characterized by comprising a plurality of cameras and a platform server;
the camera is used for acquiring a first relative position of an accident point of the traffic accident in the acquired video when the traffic accident is detected; converting the physical position of the accident point in a world coordinate system according to the first relative position; acquiring first traffic information of each lane in a preset area range of the accident point according to the physical position of the accident point and a preset time interval, and acquiring a video; sending the physical position of the accident point, the first traffic information acquired at each preset time interval and the acquired video to a platform server;
the platform server is used for receiving the physical position of the accident point sent by the camera, and first traffic flow information and videos acquired at each preset time interval; acquiring second traffic information of each lane of the upstream and downstream road junctions, which is acquired by a second camera of the upstream and downstream road junctions of the accident point, according to the physical position of the accident point and the preset time interval; and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
16. A traffic accident information acquisition system is characterized by comprising a plurality of cameras and a platform server;
the camera is used for acquiring a video;
the platform server is used for acquiring a video sent by the camera when a traffic accident is detected; identifying a first relative location of an accident point in the video at which the traffic accident occurred; acquiring the physical position of the accident point in a world coordinate system according to the first relative position; according to the physical position of the accident point and a preset time interval, acquiring first traffic information of each lane in a preset area range of the accident point, second traffic information of each lane at an upstream intersection and a downstream intersection of the accident point and a video acquired by the camera; and performing associated recording on the physical position of the accident point, the first traffic information, the second traffic information and the video acquired at each preset time interval.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811478921.3A CN111275957A (en) | 2018-12-05 | 2018-12-05 | Traffic accident information acquisition method, system and camera |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811478921.3A CN111275957A (en) | 2018-12-05 | 2018-12-05 | Traffic accident information acquisition method, system and camera |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111275957A true CN111275957A (en) | 2020-06-12 |
Family
ID=71001487
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811478921.3A Pending CN111275957A (en) | 2018-12-05 | 2018-12-05 | Traffic accident information acquisition method, system and camera |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111275957A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111645705A (en) * | 2020-06-17 | 2020-09-11 | 广州小鹏车联网科技有限公司 | Method for issuing driving route adjustment and server |
CN111859291A (en) * | 2020-06-23 | 2020-10-30 | 北京百度网讯科技有限公司 | Traffic accident recognition method, device, equipment and computer storage medium |
CN112966941A (en) * | 2021-03-09 | 2021-06-15 | 宁波工程学院 | Accident black spot identification method and system based on traffic accident big data |
CN114969575A (en) * | 2022-05-25 | 2022-08-30 | 杭州海康威视系统技术有限公司 | Camera query method and device, electronic equipment and storage medium |
CN115171368A (en) * | 2022-05-26 | 2022-10-11 | 上海市政工程设计研究总院(集团)有限公司 | Lane dynamic control method and system for traffic incident management and control |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002190090A (en) * | 2000-10-13 | 2002-07-05 | Sumitomo Electric Ind Ltd | Traffic flow abnormal condition detector and its method (spectrum) |
JP2003296870A (en) * | 2002-03-29 | 2003-10-17 | Fujitsu Fip Corp | Method for preventing/dissolving traffic jam utilizing traffic flow prediction and example inference |
CN1645037A (en) * | 2004-12-31 | 2005-07-27 | 天津大学 | Partitioned pointing method for three-dimensional scanning measurement system by light band method |
CN101159098A (en) * | 2007-11-20 | 2008-04-09 | 东南大学 | Automatic detection confirmations method for urban traffic incident |
CN102270387A (en) * | 2011-04-02 | 2011-12-07 | 东南大学 | Highway traffic dispersion method |
CN104392610A (en) * | 2014-12-19 | 2015-03-04 | 山东大学 | Distributed video based highway traffic incident influence range dynamic monitoring method |
CN105489034A (en) * | 2015-09-21 | 2016-04-13 | 青岛智能产业技术研究院 | Main line full traffic control system and method |
CN105989593A (en) * | 2015-02-12 | 2016-10-05 | 杭州海康威视系统技术有限公司 | Method and device for measuring speed of specific vehicle in video record |
CN107742420A (en) * | 2017-09-22 | 2018-02-27 | 北京交通大学 | It is a kind of to be used for the method that road traffic flow is predicted under emergent traffic incident |
-
2018
- 2018-12-05 CN CN201811478921.3A patent/CN111275957A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002190090A (en) * | 2000-10-13 | 2002-07-05 | Sumitomo Electric Ind Ltd | Traffic flow abnormal condition detector and its method (spectrum) |
JP2003296870A (en) * | 2002-03-29 | 2003-10-17 | Fujitsu Fip Corp | Method for preventing/dissolving traffic jam utilizing traffic flow prediction and example inference |
CN1645037A (en) * | 2004-12-31 | 2005-07-27 | 天津大学 | Partitioned pointing method for three-dimensional scanning measurement system by light band method |
CN101159098A (en) * | 2007-11-20 | 2008-04-09 | 东南大学 | Automatic detection confirmations method for urban traffic incident |
CN102270387A (en) * | 2011-04-02 | 2011-12-07 | 东南大学 | Highway traffic dispersion method |
CN104392610A (en) * | 2014-12-19 | 2015-03-04 | 山东大学 | Distributed video based highway traffic incident influence range dynamic monitoring method |
CN105989593A (en) * | 2015-02-12 | 2016-10-05 | 杭州海康威视系统技术有限公司 | Method and device for measuring speed of specific vehicle in video record |
CN105489034A (en) * | 2015-09-21 | 2016-04-13 | 青岛智能产业技术研究院 | Main line full traffic control system and method |
CN107742420A (en) * | 2017-09-22 | 2018-02-27 | 北京交通大学 | It is a kind of to be used for the method that road traffic flow is predicted under emergent traffic incident |
Non-Patent Citations (2)
Title |
---|
孙宇臣 等: "采用线性分区方法对三维传感器的标定", 《光电子 激光》 * |
李晓洁: "提高激光三维人体扫描系统性能的关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111645705A (en) * | 2020-06-17 | 2020-09-11 | 广州小鹏车联网科技有限公司 | Method for issuing driving route adjustment and server |
CN111859291A (en) * | 2020-06-23 | 2020-10-30 | 北京百度网讯科技有限公司 | Traffic accident recognition method, device, equipment and computer storage medium |
US11328600B2 (en) | 2020-06-23 | 2022-05-10 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for identifying traffic accident, device and computer storage medium |
CN112966941A (en) * | 2021-03-09 | 2021-06-15 | 宁波工程学院 | Accident black spot identification method and system based on traffic accident big data |
CN112966941B (en) * | 2021-03-09 | 2024-04-19 | 宁波工程学院 | Accident black spot identification method and system based on traffic accident big data |
CN114969575A (en) * | 2022-05-25 | 2022-08-30 | 杭州海康威视系统技术有限公司 | Camera query method and device, electronic equipment and storage medium |
CN115171368A (en) * | 2022-05-26 | 2022-10-11 | 上海市政工程设计研究总院(集团)有限公司 | Lane dynamic control method and system for traffic incident management and control |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111275957A (en) | Traffic accident information acquisition method, system and camera | |
CN102945603B (en) | Method for detecting traffic event and electronic police device | |
Semertzidis et al. | Video sensor network for real-time traffic monitoring and surveillance | |
US9154741B2 (en) | Apparatus and method for processing data of heterogeneous sensors in integrated manner to classify objects on road and detect locations of objects | |
JP7070683B2 (en) | Deterioration diagnosis device, deterioration diagnosis system, deterioration diagnosis method, program | |
CN111833598B (en) | Automatic traffic incident monitoring method and system for unmanned aerial vehicle on highway | |
US20100128110A1 (en) | System and method for real-time 3-d object tracking and alerting via networked sensors | |
CN111754581A (en) | Camera calibration method, roadside sensing equipment and intelligent traffic system | |
CN112099040A (en) | Whole-course continuous track vehicle tracking system and method based on laser radar network | |
CN111275960A (en) | Traffic road condition analysis method, system and camera | |
CN103295396A (en) | Method and system for speedy ex-situ evidence collection of traffic accident | |
CN111914592B (en) | Multi-camera combined evidence obtaining method, device and system | |
CN101882374A (en) | System and method for analysis of comprehensive information on vehicle trip distribution | |
CN107360394A (en) | More preset point dynamic and intelligent monitoring methods applied to frontier defense video monitoring system | |
Wang et al. | Realtime wide-area vehicle trajectory tracking using millimeter-wave radar sensors and the open TJRD TS dataset | |
US11410371B1 (en) | Conversion of object-related traffic sensor information at roadways and intersections for virtual dynamic digital representation of objects | |
EA029104B1 (en) | Method for controlling a monitoring system and system for implementing same | |
CN115965655A (en) | Traffic target tracking method based on radar-vision integration | |
RU120270U1 (en) | PEDESTRIAN CROSSING CONTROL COMPLEX | |
CN114771548A (en) | Data logging for advanced driver assistance system testing and verification | |
EP3940666A1 (en) | Digital reconstruction method, apparatus, and system for traffic road | |
CN106960027B (en) | The UAV Video big data multidate association analysis method of spatial information auxiliary | |
CN114677640A (en) | Intelligent construction site safety monitoring system and method based on machine vision | |
CN111277791A (en) | Case event monitoring method and system | |
CN111383248A (en) | Method and device for judging red light running of pedestrian and electronic equipment |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200612 |