CN114333330B - Intersection event detection system based on road side edge holographic sensing - Google Patents
Intersection event detection system based on road side edge holographic sensing Download PDFInfo
- Publication number
- CN114333330B CN114333330B CN202210097755.2A CN202210097755A CN114333330B CN 114333330 B CN114333330 B CN 114333330B CN 202210097755 A CN202210097755 A CN 202210097755A CN 114333330 B CN114333330 B CN 114333330B
- Authority
- CN
- China
- Prior art keywords
- intersection
- track
- road side
- real
- motor vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention provides an intersection event detection system based on road side edge holographic sensing, which comprises road side sensing equipment, an edge computing unit and a cloud control platform, and comprises the following specific steps: s1, the road side sensing equipment acquires crossed sensing data and transmits the crossed sensing data to an edge computing unit; s2, the edge computing unit acquires a real-time data stream of the road side sensing equipment, and carries out real-time track extraction through a deep learning algorithm; s3, the edge computing unit identifies road traffic events based on the running track characteristics of the motor vehicle, and reports the occurrence range of the traffic events to the cloud control platform; and S4, the cloud control platform receives traffic event data, feeds back the occurrence position of the traffic event and the on-site real-time video based on the visual map, and performs early warning and secondary confirmation on the traffic event. The invention improves the accuracy of detecting the traffic event at the intersection, improves the response capability of traffic management departments to the traffic event at the intersection area, and can effectively ensure the running efficiency of the urban traffic system.
Description
Technical Field
The invention belongs to the technical field of road traffic detection, and particularly relates to an intersection event detection system based on road side edge holographic perception.
Background
At present, with the rapid increase of the quantity of motor vehicles in China, the urban road traffic flow rapidly rises, and the problems of traffic jam and the like are increasingly serious. The crossing is used as a main node of urban traffic, is influenced by signal control, and has lower traffic capacity than a common road section. When vehicles go out in the early and late peaks, road intersections are frequent points of congestion, and the areas are also main concern points of urban traffic management. Particularly, when events such as retrograde traffic and traffic accidents occur at the road, whether the events can be found and responded in real time directly influences whether the traffic events can be properly handled in time or not, and whether secondary events occur or expansion and spreading of traffic jams are formed or not can be avoided. Therefore, timeliness and effectiveness of traffic event detection is particularly important at road intersections.
The existing urban traffic supervision infrastructure mainly comprises an electric police, a bayonet, millimeter wave radars, coils and the like, and traffic events such as sprinkles and abnormal vehicles on roads are generally detected in a mode of recognizing road states through videos. But the traffic event such as accident, throwing thing etc. that the regional emergence of intersection image feature is various, and the vehicle is also comparatively complicated at the regional travel feature of intersection, if rely on the mode of pure vision degree of depth to learn to detect traffic event, the recognition rate is lower, the false detection rate is higher, the practicality is not strong. Along with the construction of intelligent networking cities, the construction of holographic intersections based on multi-sensor fusion of radars, cameras and the like is steadily promoted, and on the basis of multi-sensor fusion, the target refinement real-time perception data of each traffic participant at the intersections can be obtained, so that the track refinement depiction of the traffic participants is realized. The intersection event identification based on the road side edge holographic perception can effectively avoid the problems existing in pure video detection, greatly improve the response speed of the intersection event, timely link traffic management facilities such as signal lamps and the like, and avoid large-scale traffic jam of cities caused by the intersection traffic event.
The traffic events such as abnormal vehicles at the intersection, sprinkles, traffic accidents and the like often bring about track changes of other vehicles at the intersection, so that the invention determines whether the traffic events occur by detecting and comparing the track changes of the vehicles in real time.
Disclosure of Invention
The invention aims to solve the problems, and provides an intersection event detection system based on road side edge holographic perception, which can identify the occurrence of an intersection traffic event and realize the precise real-time traffic event detection of an intersection region.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
the intersection event detection system based on road side edge holographic perception comprises road side perception equipment, an edge calculation unit and a cloud control platform, and specifically comprises the following steps:
s1, the road side sensing equipment acquires crossed sensing data and transmits the crossed sensing data to an edge computing unit;
s2, the edge computing unit acquires real-time data streams of the road side sensing equipment, performs time stamp unification and sensor space coordinate calibration, and performs real-time track extraction through a deep learning algorithm;
s3, the edge computing unit identifies road traffic events based on the characteristics of the motor vehicle running track on the basis of acquiring the target track of the motor vehicle at the intersection, and reports the occurrence range of the traffic events to the cloud control platform;
and S4, the cloud control platform receives traffic event data reported by the edge computing unit, feeds back the occurrence position of the traffic event and the on-site real-time video based on the visual map, and performs early warning and secondary confirmation on the traffic event. According to the invention, the real-time track data of the motor vehicle targets at the intersection is obtained through the road side sensing equipment, the passing rule of the motor vehicle at the intersection is introduced, and whether the intersection area sends traffic events such as abnormal parking, object throwing and traffic accidents is judged by comparing the motor vehicle track with the passing rule of the intersection in real time.
Further, the road side sensing device comprises a plurality of sensor devices which are deployed on the road side and are used for acquiring the sensing data of the intersection in real time. The road side sensing equipment can acquire the environmental data of the intersection road in real time based on the sensing characteristics of different sensors.
Further, the sensor device includes a lidar that acquires real-time point cloud data and a video camera that acquires video image data. The sensor device can be used for acquiring the cross sensing data, such as laser point cloud, video images and the like, in real time to obtain the original data stream of each sensor at the intersection.
Further, the edge computing unit is deployed in the road side floor cabinet, performs space-time synchronization, extraction of targets of road traffic participants, target fusion and tracking on the original data stream obtained by the road side sensor, performs traffic event analysis based on real-time track data, and uploads the data result to the cloud control platform.
Further, the real-time trajectory includes a target type, a target position, and a target heading angle.
Further, the deep learning algorithm is an algorithm for identifying traffic events based on the characteristics of the running track of the motor vehicle, and specifically comprises the following steps:
s21, acquiring a track data set (timestamp, longitude, latitude and course angle) of each motor vehicle target fixed time period T seconds at an intersection, wherein the corresponding set is { (T) i1 ,x i1 ,y i1 ,h i1 ),……,(t i10T ,x i10T ,y i10T ,h i10T ) Track data frequency 10Hz;
s22, acquiring intersection map data, and extracting a connection track of each entrance road and each exit road and each lane attribute of the entrance road from the map;
s23, endowing lane attributes and corresponding lane numbers with each motor vehicle target ID acquired by the road side sensing equipment by combining a high-precision map; the motor vehicle target lane attribute determines the lane number and the driving direction of the corresponding ID motor vehicle entering the intersection by comparing the longitude and latitude of the motor vehicle entering the intersection with the longitude and latitude of lane lines of each lane;
s24, determining reasonable track routes of the motor vehicle in all driving directions of the intersection based on the map;
s25, matching reasonable track points corresponding to the actual track of the motor vehicle;
s26, comparing the tracks in real time, and alarming the traffic event when the difference exceeds a threshold value.
Further, lane attributes of the entrance lane include straight, left turn, right turn, straight left, straight right, and straight left and right.
Further, in step S23, each motor vehicle structured data set is { (timestamp, longitude, latitude, heading angle, lane to which the motor vehicle structured data set belongs, direction of travel) }, and the motor vehicle structured data set corresponding to ID number i is
{(t i1 ,x i1 ,y i1 ,h i1 ,n i1 ,l i1 ),……,(t i10T ,x i10T ,y i10T ,h i10T ,n i10T ,l i10T )}。
Further, in step S25, a distance calculation is performed by connecting each point of the real-time track points of the vehicle running with a reasonable track point, and the reasonable track corresponding to the vehicle is obtained by using the distance sum as the matching track.
Further, the real-time comparison track in step S26 specifically includes: comparing the real-time position of the motor vehicle with a reasonable track, and calculating the vertical distance D between the real-time position and the reasonable track it When D it >When delta, outputting event alarm information outwards, wherein delta is a minimum threshold value; the vertical distance D between the real-time position and the reasonable track it Is the distance between the motor vehicle i and a reasonable track line at the t moment.
Compared with the prior art, the invention has the advantages that:
1. according to the intersection event detection system based on the road side edge holographic perception, real-time tracks of motor vehicles at an intersection are identified and extracted through the multidimensional sensor, traffic rule features of vehicles at the intersection are introduced based on track data, traffic events in the intersection are identified based on the traffic rule features, and an intersection area refined real-time traffic event detection function is realized; compared with the traditional mode of identifying the traffic event of the intersection based on the video picture, the method can effectively avoid the problems that the detection rate is low and the false detection rate is high because the video image cannot be fully covered due to the algorithm training data and is difficult to deal with the identification of different image features presented by different traffic events, different shape types of sprinkles and the like;
2. the invention focuses on the intersection of the urban traffic treatment core area, avoids the influence of different illumination and the like on the perception effect based on the vehicle running track characteristics, effectively detects the traffic event of the intersection, provides real-time early warning service for urban traffic managers, ensures smooth urban roads, and has great popularization and application prospect and value.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of the system components of the present invention;
FIG. 3 is a trace line drawing of the intersection entrance and exit tracks of the present invention.
Detailed Description
In order to enable those skilled in the art to better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.
As shown in fig. 1 and fig. 2, the intersection event detection system based on road side edge holographic sensing in this embodiment includes road side sensing equipment, an edge computing unit and a cloud control platform, and specifically includes the following steps:
s1, the road side sensing equipment acquires crossed sensing data and transmits the crossed sensing data to an edge computing unit;
s2, the edge computing unit acquires real-time data flow of road side sensing equipment, performs time stamp unification and sensor space coordinate calibration, runs a deep learning algorithm to realize intersection traffic target identification, multi-sensor fusion and target tracking, and realizes real-time track extraction of intersection traffic participant targets, namely motor vehicles, wherein the real-time track extraction comprises target types, target positions (longitude and latitude) and target course angles;
s3, in the edge computing unit, on the basis of acquiring a target track of the motor vehicle at the intersection, identifying road traffic events such as road casting objects, abnormal vehicles and the like based on the characteristics of the running track of the motor vehicle, and reporting the approximate range of the traffic events to a cloud control platform;
and S4, the cloud control platform receives traffic event data reported by the edge computing unit, and feeds back the occurrence position of the traffic event and the on-site real-time video to a traffic manager based on a visual map, so as to perform early warning and manual secondary confirmation. The road side multisource sensing device, namely the drive test sensing device, can acquire the road environment data of the intersection in real time based on different sensor sensing characteristics. The edge computing unit is used for realizing space-time unification, target recognition and fusion of different sensors, so that real-time track data of motor vehicles in an intersection area are obtained, and intersection traffic rules are introduced. And judging whether the vehicle track is abnormal or not by comparing the errors of the vehicle track and the reasonable track points, and further judging whether an intersection traffic event occurs or not.
The road side sensing equipment is a plurality of sensor equipment deployed on the road side, and the road side sensing equipment can acquire the road environment data of the intersection in real time based on different sensor sensing characteristics. The road side sensor comprises a laser radar, a video camera and the like, wherein the laser radar can acquire real-time point cloud data, and the video camera can acquire video image data; the sensor device of the embodiment includes, but is not limited to, sensing devices such as a road side radar (laser radar/millimeter wave radar), a video camera and the like, and acquires cross sensing data such as laser point cloud, video image and the like through the sensor device in real time to obtain original data streams of each sensor at an intersection. The laser radar of the intersection equipment can be arranged in a mechanical laser radar diagonal line arrangement mode, and the video camera is arranged on the electric warning rod piece of the intersection, so that holographic perception of the intersection area is realized.
The edge computing unit of the embodiment is a hardware device with computing power deployed in a road side floor cabinet, performs space-time synchronization, target extraction, target fusion and tracking on the original data stream obtained by the road side sensor, analyzes traffic events such as road casting objects, vehicle anomalies and the like based on real-time track data, and reports the data result to the cloud control platform.
According to the embodiment, the edge computing unit acquires real-time data flow of the road side sensing device, the time from the transmission of sensor data to the edge computing unit is taken as unified time, and the coordinates of each calibration point in the sensor are subjected to coordinate conversion so as to be unified to a WGS84 coordinate system, so that the sensor data coordinate unification is realized. And identifying the motor vehicle target perceived by the sensor by using a deep learning method, tracking the identified structured data so as to ensure that the motor vehicle is continuously tracked in an intersection area and the target ID is unique, and outputting the target ID, the target type, the target position (longitude and latitude) and the target course angle according to the frequency of 10 Hz.
In the edge calculation unit of the embodiment, on the basis of the obtained motor vehicle target ID, target type, target position (longitude and latitude) and target course angle, the motor vehicle passing rule of the intersection is introduced, for example, the motor vehicle entering the intersection of the straight entrance road must travel straight to the opposite exit road, the motor vehicle of the left-turn entrance road must travel left to the intersection, the motor vehicle of the right-turn entrance road must travel right to the intersection, and meanwhile, the fluctuation of the vehicle running track in a certain range does not have larger change under the general condition. And analyzing the running track of the motor vehicle based on the motor vehicle passing rule, and outputting an intersection traffic event early warning outwards and reporting the traffic event occurrence approximate range to the cloud control platform when the running track exceeds a set threshold value.
The cloud control platform is application software deployed in the background and can be deployed in the traffic police intranet. The edge computing unit reports traffic event analysis data such as road sprinkles, vehicle anomalies and the like to the cloud control platform and is used for displaying early warning to traffic managers. The cloud control platform of the embodiment receives traffic event data reported by the road side edge calculation unit, feeds back traffic event occurrence positions and on-site real-time videos to a traffic manager based on a visual map, and the traffic manager carries out secondary confirmation on the traffic events of the intersection area through the videos to complete an event acquisition process.
The deep learning algorithm of the embodiment is an algorithm for identifying traffic events based on the characteristics of the running track of the motor vehicle, and specifically comprises the following steps:
s21, acquiring a track data set (timestamp, longitude, latitude and course angle) of each motor vehicle target in the latest fixed time period T seconds at an intersection, wherein the corresponding set is { (T) i1 ,x i1 ,y i1 ,h i1 ),……,(t i10T ,x i10T ,y i10T ,h i10T ) Track data frequency 10Hz; in this embodiment, the T second data is mainly used to fit a section of historical track with a reasonable track point line described below, and generally 5 seconds of data are selected according to experience, and the passing speed of 20km/h at a common intersection is about 25 meters.
S22, as shown in FIG. 3, acquiring intersection map data, extracting a connection track of each entrance road and each exit road and each lane attribute of the entrance road from the map, including straight running, left turning, right turning, straight left, straight right and straight left and right, and corresponding the entrance road number with the running direction of the entrance road;
s23, combining a high-precision map to endow lane attributes and corresponding lane numbers with each motor vehicle target ID acquired by the road side sensing equipment, wherein the lane attributes and the corresponding lane numbers can be respectively set to be 1,2 and 3 … … from left to right; the motor vehicle target lane attribute determines the lane number and the driving direction of the corresponding ID motor vehicle entering the intersection by comparing the longitude and latitude of the motor vehicle entering the intersection with the longitude and latitude of lane lines of each lane; each motor vehicle target structured data set is { (timestamp, longitude, latitude, course angle, lane, driving direction) }, and the motor vehicle data with corresponding ID number i is
{(t i1 ,x i1 ,y i1 ,h i1 ,n i1 ,l i1 ),……,(t i10T ,x i10T ,y i10T ,h i10T ,n i10T ,l i10T )},
Wherein n is the lane number 1,2,3, … … to which it belongs; l is the running direction, which is one of straight running, left turning, right turning, straight left, straight right and straight left and right respectively.
S24, determining reasonable track routes of the motor vehicle in all driving directions of the intersection based on the map; the reasonable track route is based on the connection line of the center point of the entrance lane and the center point of the exit lane, as shown in figure 3, the straight lane P 2 、P 3 Center point and S 1 、S 2 Center point, S 2 、S 3 The central points are connected;
s25, matching theoretical reasonable track points corresponding to the actual track of the motor vehicle. Because more than one reasonable lane may exist in a certain running direction of the vehicle, a reasonable track point corresponding to the actual track of the motor vehicle needs to be matched. Calculating the vertical distance connected with reasonable track points according to the longitude and latitude of each point of the real-time track points of the vehicle running, obtaining the square of the error value of each point, and summing the squares sigma theta 2 The minimum is the matching track, and the corresponding reasonable track of the vehicle is obtained;
s26, comparing the tracks in real time, and alarming traffic events when the difference exceeds a threshold value.
The real-time comparison track in step S26 specifically includes: comparing the real-time position of the motor vehicle with a reasonable track, and calculating the vertical distance D between the real-time position and the reasonable track it When D it >When delta, outputting event alarm information outwards, wherein delta is a minimum threshold value; the vertical distance D between the real-time position and the reasonable track it Is the distance between the motor vehicle i and a reasonable track line at the t moment. According to the embodiment, the value of the general detour behavior characteristic delta is 4 meters after the occurrence of an event of the vehicle, and at least the fact that a plurality of vehicles deviate at the same time is detected, the occurrence of the traffic event is determined, and the event alarm information is output.
According to the method, the real-time track data of the motor vehicle targets at the intersection are obtained through the holographic perception of the edges of the road sides, the traffic rules of the motor vehicles at the intersection are introduced, whether the intersection area sends traffic events such as abnormal parking, object throwing and traffic accidents or not is judged through real-time comparison and analysis of the motor vehicle tracks and the traffic rules of the intersection, compared with a traditional video method, the accuracy of detecting the traffic events at the intersection is improved, the response capability of traffic management departments to the traffic events at the intersection area is improved, and the running efficiency of an urban traffic system can be effectively guaranteed. Therefore, the method has more popularization and application values and prospects in the aspect of urban refined traffic control.
The specific embodiments described herein are offered by way of example only to illustrate the spirit of the invention. Those skilled in the art may make various modifications or additions to the described embodiments or substitutions thereof without departing from the spirit of the invention or exceeding the scope of the invention as defined in the accompanying claims.
Claims (8)
1. The intersection event detection system based on road side edge holographic perception comprises road side perception equipment, an edge calculation unit and a cloud control platform, and is characterized by comprising the following specific steps:
s1, the road side sensing equipment acquires crossed sensing data and transmits the crossed sensing data to an edge computing unit;
s2, the edge computing unit acquires real-time data streams of the road side sensing equipment, performs time stamp unification and sensor space coordinate calibration, and performs real-time track extraction through a deep learning algorithm;
s3, the edge computing unit identifies road traffic events based on the characteristics of the motor vehicle running track on the basis of acquiring the target track of the motor vehicle at the intersection, and reports the occurrence range of the traffic events to the cloud control platform;
s4, the cloud control platform receives traffic event data reported by the edge computing unit, feeds back the occurrence position of the traffic event and the on-site real-time video based on the visual map, and performs early warning and secondary confirmation on the traffic event;
the deep learning algorithm is an algorithm for identifying traffic events based on the characteristics of the running track of the motor vehicle, and specifically comprises the following steps:
s21, acquiring a track data set (timestamp, longitude, latitude and course angle) of each motor vehicle target fixed time period T seconds at an intersection, wherein the corresponding set is { (T) i1 ,x i1 ,y i1 ,h i1 ),……,(t i10T ,x i10T ,y i10T ,h i10T ) Track data frequency 10Hz;
s22, acquiring intersection map data, and extracting a connection track of each entrance road and each exit road and each lane attribute of the entrance road from the map;
s23, endowing lane attributes and corresponding lane numbers with each motor vehicle target ID acquired by the road side sensing equipment by combining a high-precision map; the motor vehicle target lane attribute determines the lane number and the driving direction of the corresponding ID motor vehicle entering the intersection by comparing the longitude and latitude of the motor vehicle entering the intersection with the longitude and latitude of lane lines of each lane;
s24, determining reasonable track routes of the motor vehicle in all driving directions of the intersection based on the map;
s25, matching reasonable track points corresponding to the actual track of the motor vehicle;
s26, comparing the tracks in real time, and alarming traffic events when the difference exceeds a threshold value;
and step S25, calculating the vertical distance connected with the reasonable track points according to the longitude and latitude of each point of the real-time track points of the vehicle running, obtaining the square of the error value of each point, and obtaining the corresponding reasonable track of the vehicle by taking the minimum sum of squares as the matching track.
2. The intersection event detection system based on road side edge holographic sensing as claimed in claim 1, wherein said road side sensing device comprises a plurality of sensor devices deployed at the road side for acquiring intersection sensing data in real time.
3. The intersection event detection system based on road side edge holographic sensing of claim 2, wherein the sensor device comprises a lidar and a video camera, the lidar acquiring real-time point cloud data, the video camera acquiring video image data.
4. The intersection event detection system based on road side edge holographic sensing according to claim 1, wherein the edge calculation unit is deployed in a road side floor cabinet, performs space-time synchronization, target extraction, target fusion and tracking on an original data stream obtained by a road side sensor, performs traffic event analysis based on real-time track data, and uploads a data result to a cloud control platform.
5. The intersection event detection system based on road side edge holographic awareness of claim 4, wherein the real-time trajectory comprises a target type, a target position and a target heading angle.
6. The intersection event detection system based on road side edge holographic awareness of claim 1, wherein the lane attributes of the entrance lane include straight-going, left-turn, right-turn, straight-left, straight-right, and straight-left-right.
7. The intersection event detection system based on road side edge holographic sensing as claimed in claim 1, wherein each of the motor vehicle structured data sets in the step S23 is { (time stamp, longitude, latitude, heading angle, lane to which the vehicle belongs, traveling direction) }, and the motor vehicle data corresponding to ID number i is { (t) i1 ,x i1 ,y i1 ,h i1 ,n i1 ,l i1 ),……,(t i10T ,x i10T ,y i10T ,h i10T ,n i10T ,l i10T )}。
8. The intersection event detection system based on road side edge holographic sensing according to claim 1, wherein the real-time comparison track in step S26 is specifically: comparing the real-time position of the motor vehicle with a reasonable track, and calculating the real-time position and the reasonable trackIs a vertical distance D of (2) it When D it >When delta, outputting event alarm information outwards, wherein delta is a minimum threshold value; the vertical distance D between the real-time position and the reasonable track it Is the distance between the motor vehicle i and a reasonable track line at the t moment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210097755.2A CN114333330B (en) | 2022-01-27 | 2022-01-27 | Intersection event detection system based on road side edge holographic sensing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210097755.2A CN114333330B (en) | 2022-01-27 | 2022-01-27 | Intersection event detection system based on road side edge holographic sensing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114333330A CN114333330A (en) | 2022-04-12 |
CN114333330B true CN114333330B (en) | 2023-04-25 |
Family
ID=81030251
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210097755.2A Active CN114333330B (en) | 2022-01-27 | 2022-01-27 | Intersection event detection system based on road side edge holographic sensing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114333330B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115294757A (en) * | 2022-04-26 | 2022-11-04 | 江苏未来都市出行科技集团有限公司 | Recognition and release system for lane-level traffic flow and traffic incident |
CN115063969A (en) * | 2022-04-26 | 2022-09-16 | 阿里云计算有限公司 | Data processing method, device, medium, roadside cooperative device and system |
CN114973663B (en) * | 2022-05-16 | 2023-08-29 | 浙江机电职业技术学院 | Intelligent road side unit device based on edge calculation |
CN115063973A (en) * | 2022-05-25 | 2022-09-16 | 上海交通大学 | Road intersection vehicle track extraction system based on computer vision |
WO2023241536A1 (en) * | 2022-06-13 | 2023-12-21 | Anlyvis Limited | A system and method for detecting traffic events |
CN115019512A (en) * | 2022-07-05 | 2022-09-06 | 北京动视元科技有限公司 | Road event detection system based on radar video fusion |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105513349A (en) * | 2014-12-03 | 2016-04-20 | 中咨泰克交通工程集团有限公司 | Double-perspective learning-based mountainous area highway vehicle event detection method |
CN105788274A (en) * | 2016-05-18 | 2016-07-20 | 武汉大学 | Urban intersection lane-level structure extraction method based on time-space trajectory big data |
JP2018010506A (en) * | 2016-07-14 | 2018-01-18 | オムロン株式会社 | Traffic information acquisition device, traffic information acquisition method and traffic information acquisition program |
CN110473402A (en) * | 2019-08-20 | 2019-11-19 | 河北德冠隆电子科技有限公司 | A kind of accident detection early warning system based on target abnormal behaviour trajectory analysis |
CN111402612A (en) * | 2019-01-03 | 2020-07-10 | 北京嘀嘀无限科技发展有限公司 | Traffic incident notification method and device |
CN111540237A (en) * | 2020-05-19 | 2020-08-14 | 河北德冠隆电子科技有限公司 | Method for automatically generating vehicle safety driving guarantee scheme based on multi-data fusion |
WO2020177767A1 (en) * | 2019-03-06 | 2020-09-10 | 路特迩科技(杭州)有限公司 | Electronic map-based traffic and travel information service system and method |
CN112382093A (en) * | 2020-11-13 | 2021-02-19 | 浙江交投高速公路建设管理有限公司 | Highway ramp auxiliary early warning method based on vehicle-road cooperation |
CN113419244A (en) * | 2021-05-28 | 2021-09-21 | 同济大学 | Vehicle track splicing method based on millimeter wave radar data |
WO2021184841A1 (en) * | 2020-03-19 | 2021-09-23 | 中移(上海)信息通信科技有限公司 | Internet of vehicles method and apparatus, device, storage medium, and system |
WO2021190067A1 (en) * | 2020-03-24 | 2021-09-30 | 北京京东乾石科技有限公司 | Method for determining passing by self-driving vehicle and related device |
CN113947907A (en) * | 2021-10-28 | 2022-01-18 | 高新兴科技集团股份有限公司 | Vehicle traffic accident early warning method, device, medium and equipment based on V2X |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6760061B1 (en) * | 1997-04-14 | 2004-07-06 | Nestor Traffic Systems, Inc. | Traffic sensor |
CN102956104A (en) * | 2012-10-19 | 2013-03-06 | 上海交通大学无锡研究院 | Method for detecting traffic incident based on vehicle trajectory description |
US10121370B2 (en) * | 2014-09-20 | 2018-11-06 | Mohamed Roshdy Elsheemy | Comprehensive traffic control system |
JP2017207315A (en) * | 2016-05-17 | 2017-11-24 | 富士通株式会社 | Movement assist system, movement assist method and movement assist program |
KR101685900B1 (en) * | 2016-07-06 | 2016-12-13 | 이용호 | Monitoring system for Intersection detection accident |
CN106952473A (en) * | 2017-04-01 | 2017-07-14 | 深圳市元征科技股份有限公司 | Road service system detection method and device |
US10489976B2 (en) * | 2017-08-11 | 2019-11-26 | Jing Jin | Incident site investigation and management support system based on unmanned aerial vehicles |
JP6969962B2 (en) * | 2017-10-05 | 2021-11-24 | トヨタ自動車株式会社 | Map information providing system for vehicle driving support and / or driving control |
US20200219396A1 (en) * | 2019-01-08 | 2020-07-09 | Continental Automotive Systems, Inc. | Automated detection of traffic incidents via intelligent infrastructure sensors |
CN110570664B (en) * | 2019-09-23 | 2023-04-07 | 山东科技大学 | Automatic detection system for highway traffic incident |
CN110728842B (en) * | 2019-10-23 | 2021-10-08 | 江苏智通交通科技有限公司 | Abnormal driving early warning method based on reasonable driving range of vehicles at intersection |
CN113326719A (en) * | 2020-02-28 | 2021-08-31 | 华为技术有限公司 | Method, equipment and system for target tracking |
CN111578964B (en) * | 2020-04-13 | 2022-01-07 | 河北德冠隆电子科技有限公司 | High-precision map road information rapid generation system and method based on space-time trajectory reconstruction |
CN111583675B (en) * | 2020-05-14 | 2021-05-14 | 吴钢 | Regional road network traffic signal lamp coordination control system and method |
CN112816954B (en) * | 2021-02-09 | 2024-03-26 | 中国信息通信研究院 | Road side perception system evaluation method and system based on true value |
CN113744564B (en) * | 2021-08-23 | 2023-11-28 | 上海智能新能源汽车科创功能平台有限公司 | Intelligent networking bus road cooperative control system based on edge calculation |
CN113888871B (en) * | 2021-10-20 | 2023-05-05 | 上海电科智能***股份有限公司 | Automatic handling linkage system and method for expressway traffic incidents |
-
2022
- 2022-01-27 CN CN202210097755.2A patent/CN114333330B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105513349A (en) * | 2014-12-03 | 2016-04-20 | 中咨泰克交通工程集团有限公司 | Double-perspective learning-based mountainous area highway vehicle event detection method |
CN105788274A (en) * | 2016-05-18 | 2016-07-20 | 武汉大学 | Urban intersection lane-level structure extraction method based on time-space trajectory big data |
JP2018010506A (en) * | 2016-07-14 | 2018-01-18 | オムロン株式会社 | Traffic information acquisition device, traffic information acquisition method and traffic information acquisition program |
CN111402612A (en) * | 2019-01-03 | 2020-07-10 | 北京嘀嘀无限科技发展有限公司 | Traffic incident notification method and device |
WO2020177767A1 (en) * | 2019-03-06 | 2020-09-10 | 路特迩科技(杭州)有限公司 | Electronic map-based traffic and travel information service system and method |
CN110473402A (en) * | 2019-08-20 | 2019-11-19 | 河北德冠隆电子科技有限公司 | A kind of accident detection early warning system based on target abnormal behaviour trajectory analysis |
WO2021184841A1 (en) * | 2020-03-19 | 2021-09-23 | 中移(上海)信息通信科技有限公司 | Internet of vehicles method and apparatus, device, storage medium, and system |
WO2021190067A1 (en) * | 2020-03-24 | 2021-09-30 | 北京京东乾石科技有限公司 | Method for determining passing by self-driving vehicle and related device |
CN111540237A (en) * | 2020-05-19 | 2020-08-14 | 河北德冠隆电子科技有限公司 | Method for automatically generating vehicle safety driving guarantee scheme based on multi-data fusion |
CN112382093A (en) * | 2020-11-13 | 2021-02-19 | 浙江交投高速公路建设管理有限公司 | Highway ramp auxiliary early warning method based on vehicle-road cooperation |
CN113419244A (en) * | 2021-05-28 | 2021-09-21 | 同济大学 | Vehicle track splicing method based on millimeter wave radar data |
CN113947907A (en) * | 2021-10-28 | 2022-01-18 | 高新兴科技集团股份有限公司 | Vehicle traffic accident early warning method, device, medium and equipment based on V2X |
Also Published As
Publication number | Publication date |
---|---|
CN114333330A (en) | 2022-04-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114333330B (en) | Intersection event detection system based on road side edge holographic sensing | |
US11990036B2 (en) | Driver behavior monitoring | |
CN112700470B (en) | Target detection and track extraction method based on traffic video stream | |
CN111223302B (en) | External coordinate real-time three-dimensional road condition auxiliary device for mobile carrier and system | |
US11836985B2 (en) | Identifying suspicious entities using autonomous vehicles | |
US10713500B2 (en) | Identification and classification of traffic conflicts using live video images | |
CN110717433A (en) | Deep learning-based traffic violation analysis method and device | |
CN103176185A (en) | Method and system for detecting road barrier | |
CN109101939B (en) | Method, system, terminal and readable storage medium for determining vehicle motion state | |
CN111477011A (en) | Detection device and detection method for road intersection early warning | |
CN106327880B (en) | A kind of speed recognition methods and its system based on monitor video | |
CN115019512A (en) | Road event detection system based on radar video fusion | |
WO2022213542A1 (en) | Method and system for clearing information-controlled intersection on basis of lidar and trajectory prediction | |
CN105405297B (en) | A kind of automatic detection method for traffic accident based on monitor video | |
CN113658427A (en) | Road condition monitoring method, system and equipment based on vision and radar | |
CN115188187A (en) | Roadside perception data quality monitoring system and method based on vehicle-road cooperation | |
Zheng | Developing a traffic safety diagnostics system for unmanned aerial vehicles usingdeep learning algorithms | |
CN117238145A (en) | Intelligent traffic management method and system based on big data | |
CN217426263U (en) | Holographic road network road monitoring system | |
CN104408942A (en) | Intelligent vehicle speed measuring device and method | |
CN113128847A (en) | Entrance ramp real-time risk early warning system and method based on laser radar | |
CN117494029B (en) | Road casting event identification method and device | |
CN112927514B (en) | Motor vehicle yellow light running behavior prediction method and system based on 3D laser radar | |
Liu et al. | Ubiquitous sensing for smart cities with autonomous vehicles | |
Tageldin | The development of behavior-based traffic conflict indicators through automated traffic safety analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |