CN110164132B - Method and system for detecting road traffic abnormity - Google Patents
Method and system for detecting road traffic abnormity Download PDFInfo
- Publication number
- CN110164132B CN110164132B CN201910455487.5A CN201910455487A CN110164132B CN 110164132 B CN110164132 B CN 110164132B CN 201910455487 A CN201910455487 A CN 201910455487A CN 110164132 B CN110164132 B CN 110164132B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- passing
- information
- vehicle information
- travel time
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- 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
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
-
- 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
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a method and a system for detecting road traffic abnormity. The method comprises the following steps: acquiring vehicle passing information identified by an electronic police in the detection area; acquiring the corresponding travel time of each vehicle according to the vehicle passing information; acquiring a vehicle mode in a current time window; acquiring the average value of the vehicle modes in all time windows in the current natural day; calculating the Mahalanobis distance between the vehicle mode in the current time window and the average value of the vehicle modes in all the time windows; judging whether the Mahalanobis distance is smaller than a set threshold value or not, and if so, determining that no abnormal traffic condition occurs in the detection area in the current time window; if not, determining that the traffic abnormal condition occurs in the detection area in the current time window. The invention can improve the accuracy of the traffic abnormity detection and the reliability of the detection result.
Description
Technical Field
The invention relates to the field of road traffic detection, in particular to a method and a system for detecting road traffic abnormity.
Background
Traffic accidents such as sudden vehicle breakdown, traffic accidents, cargo throwing and the like in urban roads can cause the generation of road traffic abnormity, namely the abnormity of traffic flow information such as traffic flow, running speed, travel time and the like. Traffic abnormity is like sudden small wound in an urban road network, if the information of the generation time, the location and the like of the small wound cannot be effectively identified and positioned in time, targeted treatment is carried out, the wound is liable to spread and spread to peripheral areas, and large wound which seriously affects the urban road network is generated, such as regional traffic jam, traffic capacity reduction and the like. Therefore, the economic loss, the travel cost, the travel satisfaction degree and other consequences are brought, and further influences are caused. At present, the urban traffic management department detects abnormal events in a manual or artificial intelligent mode through video detection equipment arranged in road sections, so that the cost is high, and the missed detection is easy. Meanwhile, the time and the place generated by the emergency traffic incident have high randomness, and for a macroscopic urban road network, the objective problems of long road mileage, complex network and the like exist, so that the timeliness and the effectiveness of traffic anomaly detection are poor.
In the prior art, the detection of traffic abnormality includes the following modes:
(1) dividing a normal traffic video image sequence into video block sequences, detecting the number of lenses in the video block sequences, establishing a Gaussian model of the number of lenses in the video block sequences, and performing anomaly detection on a test traffic video image by using the Gaussian model. The method is mainly characterized in that the video image data is utilized to detect the abnormal traffic events in the video acquisition range. The method essentially belongs to pattern recognition based on images, has high requirements on the quality of the images, requires a large number of images of normal traffic and abnormal traffic as training samples, and has high implementation difficulty. Meanwhile, the method can only identify traffic abnormality in the video acquisition range, and if the urban road network is to be comprehensively monitored, the conventional video detection equipment cannot meet the requirement, a large number of video detectors with good shooting angles need to be arranged, and the cost is huge.
(2) Based on a microwave detector arranged in a road network, the detection of the road section abnormity is realized by the following steps: 1) cleaning vehicle speed and flow data, fusing sampling data to form a sample space, 2) performing dynamic inspection on the vehicle speed and flow data sources, 3) calculating the mean value and variance of the vehicle speed and flow of each microwave point, 4) calculating vehicle speed abnormal indexes and flow abnormal indexes, 5) outputting early warning by descending order of the abnormal indexes, traversing and calculating road abnormal indexes D in all microwave points in the whole road network in the current time slot, and outputting the first K most abnormal road segment early warnings by arranging the calculated abnormal index results from large to small. The method is characterized in that a microwave detector arranged in a road section is utilized to carry out deep analysis and excavation on the speed and flow data of the road section, so that the traffic abnormity detection of the road section is realized. The microwave detectors are arranged in the existing cities on fewer roads, the detection of traffic abnormity of the urban whole road network is realized, the microwave detectors of all road sections are required to be widely added, the detection cost is undoubtedly increased, and the popularization and the application are difficult.
(3) The traffic anomaly detection method based on the route travel time calculation comprises the following steps: establishing a historical database of path travel time and road speed; step two: detecting an abnormal path; step three: the influence condition of the abnormal path covering road section is measured, and the method can filter out local road section abnormality caused by accidental factors (such as single-pass time increase caused by short-time parking of an individual taxi, abnormality which can be quickly dissipated automatically caused by short-time illegal behaviors of an individual taxi and the like); if the road sections are connected and mutually influenced, the synergistic influence of the multiple road sections can be superposed and amplified, so that a better abnormity detection effect is obtained. The method is characterized in that a continuous track model is established by utilizing taxi GPS data, the time difference between the initial track point and the final track point of each GPS track is taken as the travel time of the GPS track, a density-based clustering algorithm is used for clustering the travel time, one travel time record is taken as a clustering example, all passage records form a clustering data set, after clustering, the examples in the clustering data set are divided into a plurality of clusters, the cluster with the most examples is taken as a central cluster, the maximum value in the central cluster is taken as an initial abnormal threshold value, in the clustering data set, all the examples smaller than the initial abnormal threshold value are preliminarily defined as normal examples, and the travel time exceeding the abnormal threshold value is taken as an abnormal value. The sampling rate of taxi GPS data is usually 30 seconds to 1 minute, a plurality of roads are often spanned between two adjacent positioning points, the number of urban taxis is only about 5% -10% of that of all motor vehicles, and the taxi GPS data has disadvantages in the aspects of accuracy and distribution. Meanwhile, due to the functional particularity of taxis, the taxis have behaviors of walking, stopping, directionally searching passengers and the like, and the traffic flow characteristics of the taxis cannot reflect the real traffic flow characteristics.
Disclosure of Invention
The invention aims to provide a method and a system for detecting road traffic abnormity, which are used for carrying out abnormity detection through full sample data acquired by electric alarm equipment so as to improve the accuracy of traffic abnormity detection and the reliability of a detection result.
In order to achieve the purpose, the invention provides the following scheme:
a method of detecting a road traffic anomaly, comprising:
acquiring vehicle passing information identified by an electronic police in the detection area; the detection area is a road area between an upstream intersection and a downstream intersection of the road; the vehicle passing information comprises first vehicle information and second vehicle information; the first vehicle information is vehicle information identified by an electronic police at the upstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the upstream intersection; the second vehicle information is vehicle information identified by an electronic police at the downstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the downstream intersection;
acquiring travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information; the travel time is the time from an upstream intersection to a downstream intersection of the vehicle;
acquiring a vehicle mode in a current time window; the vehicle mode is a two-dimensional vector consisting of the number of vehicles arriving at the downstream intersection in the current time window and the average value of the travel time of all vehicles arriving at the downstream intersection in the current time window;
acquiring the average value of the vehicle modes in all time windows in the current natural day;
calculating the Mahalanobis distance between the vehicle mode in the current time window and the average value of the vehicle modes in all the time windows;
judging whether the Mahalanobis distance is smaller than a set threshold value or not to obtain a first judgment result;
when the first judgment result shows that the Mahalanobis distance is smaller than a set threshold value, determining that no abnormal traffic condition occurs in the detection area in the current time window;
and when the first judgment result shows that the Mahalanobis distance is not smaller than a set threshold value, determining that the traffic abnormal condition occurs in the detection area in the current time window.
Optionally, the obtaining of the vehicle passing information identified by the electronic police in the detection area further includes performing noise processing on the vehicle passing information, and the method specifically includes:
judging whether each piece of passing information comprises a license plate and a passing timestamp to obtain a second judgment result;
when the second judgment result shows that the passing information does not include the license plate or the passing timestamp, deleting the passing information;
judging whether license plates of any plurality of pieces of first vehicle information are the same or not, wherein the serial numbers are different from the passing timestamps, and obtaining a third judgment result;
when the third judgment result shows that the license plates of the vehicles are the same among the first pieces of vehicle information and the numbers are different from the vehicle-passing timestamps, determining the license plate, the number and the vehicle-passing timestamp corresponding to the maximum value of the vehicle-passing timestamp as the final first vehicle information of the vehicle, and deleting the rest first vehicle information of the vehicle;
judging whether the license plates of any plurality of pieces of second vehicle information are the same or not, wherein the serial numbers are different from the passing timestamps, and obtaining a fourth judgment result;
and when the fourth judgment result shows that the license plates of the vehicles are the same among the second pieces of vehicle information and the numbers are different from the vehicle-passing timestamps, determining the license plate, the number and the vehicle-passing timestamp corresponding to the maximum value of the vehicle-passing timestamp as the final second vehicle information of the vehicle, and deleting the rest second vehicle information of the vehicle.
Optionally, the obtaining, according to the first vehicle information and the second vehicle information, a travel time corresponding to each vehicle specifically includes:
matching license plates in the first vehicle information and the second vehicle information to obtain a first time stamp and a second time stamp corresponding to the same license plate; the first timestamp is a vehicle-passing timestamp in the first vehicle information, and the second timestamp is a vehicle-passing timestamp in the second vehicle information;
the first time stamp and the second time stamp are subjected to difference to obtain initial travel time;
judging whether the initial travel time is within a set time range or not to obtain a fifth judgment result;
and when the fifth judgment result shows that the initial travel time is in a set time range, determining the initial travel time as the travel time corresponding to the vehicle.
Optionally, the obtaining, according to the first vehicle information and the second vehicle information, a travel time corresponding to each vehicle, and then performing noise processing on the travel times corresponding to all the vehicles further includes:
determining a sampling window and a sampling step length of noise processing;
noise processing is carried out on the travel time in each sampling window according to a box diagram principle, abnormal value data outside an upper boundary and a lower boundary are removed, and the travel time after noise processing of the sampling windows is obtained;
and sequentially obtaining the travel time of each sampling window after noise processing according to the sampling step length.
Optionally, the mahalanobis distance between the vehicle mode in the current time window and the average value of the vehicle modes in all the time windows is
Wherein i is more than or equal to 2, STMiSTM for vehicle mode in current time window1～iIs the average value of the vehicle modes in all time windows of the current natural day,
s is STMiAnd STM1～iThe covariance matrix of the constructed full sample.
Optionally, the set threshold is
Delta is a set coefficient, and j has a value range of [1, i]，STMjTo slave STM1To STMiThe traffic pattern of the jth time window.
The invention also provides a system for detecting road traffic abnormality, comprising:
the vehicle passing information acquisition module is used for acquiring vehicle passing information identified by an electronic police in the detection area; the detection area is a road area between an upstream intersection and a downstream intersection of the road; the vehicle passing information comprises first vehicle information and second vehicle information; the first vehicle information is vehicle information identified by an electronic police at the upstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the upstream intersection; the second vehicle information is vehicle information identified by an electronic police at the downstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the downstream intersection;
the travel time acquisition module is used for acquiring the travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information; the travel time is the time from an upstream intersection to a downstream intersection of the vehicle;
the vehicle mode acquisition module is used for acquiring a vehicle mode in a current time window; the vehicle mode is a two-dimensional vector consisting of the number of vehicles arriving at the downstream intersection in the current time window and the average value of the travel time of all vehicles arriving at the downstream intersection in the current time window;
the vehicle mode average value acquisition module is used for acquiring vehicle mode average values in all time windows in the current natural day;
the Mahalanobis distance calculation module is used for calculating the Mahalanobis distances between the vehicle modes in the current time window and the average value of the vehicle modes in all the time windows;
the first judgment module is used for judging whether the Mahalanobis distance is smaller than a set threshold value or not to obtain a first judgment result;
the traffic abnormity determining module is used for determining that no traffic abnormity occurs in the detection area in the current time window when the first judgment result shows that the Mahalanobis distance is smaller than a set threshold value; and when the first judgment result shows that the Mahalanobis distance is not smaller than a set threshold value, determining that the traffic abnormal condition occurs in the detection area in the current time window.
Optionally, the system further includes a first noise processing module, configured to perform noise processing on the vehicle passing information after obtaining the vehicle passing information identified by the electronic police in the detection area, where the noise processing module specifically includes:
the second judgment unit is used for judging whether each piece of passing information comprises a license plate and a passing timestamp to obtain a second judgment result;
a deleting unit, configured to delete the vehicle passing information when the second determination result indicates that the vehicle passing information does not include the license plate or the vehicle passing timestamp;
the third judging unit is used for judging whether the license plates of any plurality of pieces of first vehicle information are the same or not, and the serial numbers are different from the passing timestamp to obtain a third judging result;
a first determining unit, configured to determine, when the third determination result indicates that license plates of vehicles are the same among the plurality of pieces of first vehicle information and that the numbers are different from the vehicle-passing timestamps, the license plate corresponding to the maximum value of the vehicle-passing timestamp, the number, and the vehicle-passing timestamp as final first vehicle information of the vehicle, and delete the remaining first vehicle information of the vehicle;
the fourth judging unit is used for judging whether the license plates of any plurality of pieces of second vehicle information are the same or not, and the serial numbers are different from the passing timestamps to obtain a fourth judging result;
and a second determining unit, configured to determine, when the fourth determination result indicates that the license plates of the vehicles are the same among the plurality of pieces of second vehicle information and the numbers are different from the passing timestamps, the license plate corresponding to the maximum value of the passing timestamp, the number, and the passing timestamp as final second vehicle information of the vehicle, and delete the remaining second vehicle information of the vehicle.
Optionally, the journey time obtaining module specifically includes:
the matching unit is used for matching license plates in the first vehicle information and the second vehicle information to obtain a first time stamp and a second time stamp corresponding to the same license plate; the first timestamp is a vehicle-passing timestamp in the first vehicle information, and the second timestamp is a vehicle-passing timestamp in the second vehicle information;
the difference making unit is used for making a difference between the first time stamp and the second time stamp to obtain an initial travel time;
the fifth judging unit is used for judging whether the initial travel time is within a set time range or not to obtain a fifth judging result;
and a third determining unit, configured to determine the initial travel time as a travel time corresponding to the vehicle when the fifth determination result indicates that the initial travel time is within a set time range.
Optionally, the system further includes a second noise processing module, configured to perform noise processing on the travel time corresponding to all vehicles after obtaining the travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information, where the method specifically includes:
the sampling parameter determining unit is used for determining a sampling window and a sampling step length of noise processing;
the eliminating unit is used for carrying out noise processing on the travel time in each sampling window according to the box diagram principle, eliminating abnormal value data outside an upper boundary and a lower boundary and obtaining the travel time after the noise processing of the sampling windows;
and the travel time determining unit is used for sequentially obtaining the travel time after the noise processing of each sampling window according to the sampling step length.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention carries out real-time matching calculation on the travel time of the vehicle based on the vehicle passing information such as license plates, timestamps and the like detected by electronic police equipment arranged at the upper and lower intersections of the road section in real time, screens and retains reasonable data through a box diagram, constructs the road section traffic mode of the road section traffic flow and the travel time on the basis, and achieves the purpose of identifying the road section traffic abnormity through identifying the abnormal value of the road section traffic mode. On one hand, the electronic police equipment commonly arranged in the existing urban road network is utilized, so that multiplexing can be realized to the greatest extent, the laying and maintenance cost brought by a special detector is reduced, and the cost minimization of traffic anomaly detection is realized; on the other hand, traffic flow data acquired by the electric police is full-sample data relative to floating car data such as a taxi GPS (global positioning system), individual running state characteristics of various vehicles can be comprehensively covered, changes of real traffic flow can be more reasonably and accurately reflected, and the accuracy rate of abnormal recognition is higher. Meanwhile, the invention has very important application value for the aspects of urban intelligent traffic control, emergency response of public security traffic control departments, improvement of smooth travel safety of residents and the like.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a method for detecting road traffic anomalies according to the present invention;
FIG. 2 is a schematic diagram of a detection area in the method for detecting road traffic abnormality according to the present invention;
fig. 3 is a schematic structural diagram of the detection system for road traffic abnormality according to 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 make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a schematic flow chart of a method for detecting road traffic abnormality according to the present invention. As shown in fig. 1, the detection method includes the following steps:
step 100: and acquiring vehicle passing information identified by an electronic police in the detection area. Fig. 2 is a schematic diagram of a detection area in the method for detecting road traffic abnormality, as shown in fig. 2, an electronic police is arranged at about 20 meters upstream of a stop line of an exit lane of an intersection to detect the vehicle running condition in the direction of an entrance lane, mainly monitors illegal behaviors of vehicles such as violation lane change and red light running, and collects vehicle passing information such as license plates and time stamps of vehicles passing through the stop line of the intersection. Since the subsequent travel time is matched from the detection area of the upstream intersection to the detection area of the downstream intersection. The travel time of the vehicle within the upstream intersection is negligible compared to the travel time of the vehicle on the road segment. Therefore, the detection range of the traffic abnormality is a road area between the upstream intersection and the downstream intersection of the road. The vehicle passing information comprises first vehicle information and second vehicle information; the first vehicle information is vehicle information identified by an electronic police at the upstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the upstream intersection; the second vehicle information is the vehicle information identified by the electronic police at the downstream intersection, and comprises the serial number, the license plate and the passing timestamp of the vehicle at the downstream intersection.
Step 200: and acquiring the travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information. The travel time is the time from an upstream intersection to a downstream intersection of the vehicle, and the specific process is as follows: the vehicle passing information (namely, the second vehicle information) identified by the electronic police at the junction is compared with the vehicle passing information (namely, the second vehicle information) identified by the electronic police at the junction at the upstream, vehicles with the same license plates are matched, the vehicle passing timestamp (namely, the first timestamp) at the upstream junction and the vehicle passing timestamp (namely, the second timestamp) at the downstream junction of the same license plates are obtained, the first timestamp and the second timestamp are differentiated, the initial travel time is obtained, whether the initial travel time is within a set time range or not is further judged, and if the initial travel time is within the set time range, the initial travel time is the travel time corresponding to the vehicle. The time range is set here in the sense that the calculated travel time after matching does not exceed a reasonable range.
As another implementation manner, the range of the first timestamp to be matched can be determined according to the second timestamp and the set time range, then whether the vehicle passing information with the same license plate is included is searched in the range of the first timestamp, and then the travel time of the vehicle corresponding to the two vehicle passing information is calculated. Suppose that the electric police at the downstream intersection recognizes that the timestamp is ts when a certain license plate passes1If the timestamp range of the retrieval of the electric warning license plate data at the upstream intersection is within
L is the length of the road section and v is the speed per hour of the road design
Searching whether the license plate is included in the time range, and if so, displaying the license plate in the time range
Time stamps and ts within range1And obtaining the final travel time of the license plate by making a difference.
Because the electronic police vehicle license plate identification data has a certain proportion of noise data, as a preferred embodiment, the vehicle passing information identified by the electronic police can be further subjected to noise processing, and the method mainly comprises the following steps: (1) the unidentified vehicle passing information, namely only vehicle passing time and vehicle passing number, but lack of license plate number data, is subjected to pre-deletion operation before matching aiming at the noise data; (2) the method comprises the following steps that (1) repeated information of vehicle passing at a downstream intersection is obtained, more than one same license plate data appears in the vehicle passing information within a short time (5min), the vehicle passing number and the vehicle passing time are different, only the data with the maximum timestamp is reserved aiming at the noise data, and the vehicle passing information corresponding to the timestamp is deleted; (3) the upstream cross traffic information is repeated in the same way as the downstream cross traffic information, and only the data with the maximum timestamp is reserved.
The electronic police collects full sample data, and inevitably counts some random and special conditions when matching and calculating the travel time. For example, when a taxi stops on a road due to passenger carrying and unloading, the travel time is increased, and when there is a demand point for travel generation and attraction on the road, the vehicle stops, and the travel time is increased. The travel time obtained by the vehicle matching calculation is greatly deviated from the travel time of other vehicles, so that as another embodiment, the data is further eliminated to obtain more objective and real travel time data. In the embodiment, a method for eliminating abnormal data in the box diagram is adopted for noise processing. The specific process is as follows:
first, a sampling window and a sampling step size for noise processing are determined. The change of the traffic flow often has certain time sequence characteristics, the characteristic difference of the traffic flow is large in different time periods in a day, and if the noise processing operation is carried out on the full-sample travel time data in the day, a large number of normal values are eliminated. Traffic flow tends to advance along a time axis and presents slowly, gradually and continuously changing characteristics. Therefore, the present invention divides the noise data processing into a low peak period and a high peak period, the low peak being between 0-7 and 22-24 points, and the periods excluding these are the high peak period. In the low peak period, 30min is taken as a travel time sample acquisition window, 1min is taken as a step length, noise data are processed in a rolling mode, for example, noise elimination operation is performed once on travel time samples from 0:00 to 0:30 from 0:00, then operation is performed again from 0:01 to 0:31, and rolling is performed for multiple times, wherein the total number of the operations is 31. Noise data was processed in a rolling fashion with 5 minutes as the sampling window and 1 minute as the step size during the peak period.
Then, noise processing is carried out on the travel time in each sampling window according to a box diagram principle, abnormal value data outside an upper boundary and a lower boundary are removed, and the travel time after noise processing of the sampling windows is obtained. Noise data culling generally adopts a method of abnormal data culling in a box chart. And respectively calculating the maximum value (namely the upper boundary), the minimum value (namely the lower boundary), the separation difference, the 25% quantile and the 75% quantile of the group of data aiming at a specific data sample, generating a box body diagram by utilizing the data, wherein most of normal data are contained in the box body diagram, abnormal values are out of the upper boundary and the lower boundary of the box body, and the abnormal values are removed, namely the noise processing process in the sampling window is completed. The calculation method is as follows:
potential difference division: iQR 75% -quantile-25% quantile
An upper boundary: uplimit 75% quantile + IQR 1.5
Lower bound: lowlimit 25% quantile-IQR 1.5
And sequentially obtaining the travel time of each sampling window after noise processing according to the sampling step length to finish the noise processing process.
Step 300: and acquiring the vehicle mode in the current time window. The vehicle mode is a two-dimensional vector STM consisting of the number of vehicles arriving at the downstream intersection in the current time window and the average value of the travel time of all vehicles arriving at the downstream intersection in the current time window<N,T>Wherein N refers to the number of vehicles arriving at the downstream intersection within the current time window; t is the average of the travel time of all vehicles arriving at the downstream intersection on the road section in the time windowThe average value of the average value is calculated,
wherein, tiThe travel time of the ith vehicle reaching the downstream intersection on the road section is determined from all the N vehicles.
Step 400: and acquiring the average value of the vehicle modes in all time windows in the current natural day. Is calculated by the formula
STM1～iRefers to the first time window tw1To the current time window twiAverage of all vehicle modes.
Step 500: the mahalanobis distance between the vehicle pattern in the current time window and the average of the vehicle patterns in all time windows is calculated. In the road section traffic mode, the number N of vehicles arriving at a downstream intersection and the average travel time T of the vehicles have an internal association relation (namely in a normal traffic flow state, the travel time is increased due to the increase of the number of vehicles, and the travel time is reduced corresponding to the reduction of the number of vehicles), and the two dimensions are different.
The mahalanobis distance between the vehicle mode in the current time window and the average of the vehicle modes in all time windows is
Wherein i is more than or equal to 2, STMiSTM for vehicle mode in current time window1～iThe average value of the vehicle modes in all time windows of the current natural day is S is STMiAnd STM1～iThe covariance matrix of the constructed full sample.
Step 600: and judging whether the Mahalanobis distance is smaller than a set threshold value or not. If yes, go to step 700; if not, step 800 is performed. In the present invention, the threshold is set to
Delta is a set coefficient, the smaller delta, the more sensitive the detection method is, and the capability of the detection method isDetecting less influential anomalies, j having a value in the range [1, i]，STMjTo slave STM1To STMiThe traffic pattern of the jth time window.
Step 700: and determining that no traffic abnormal condition occurs in the detection area in the current time window.
Step 800: and determining the abnormal traffic condition of the detection area in the current time window.
Road segment traffic patterns typically produce four anomalies in the data, described as follows:
1) the number of vehicles arriving at the downstream intersection is reduced, the travel time of the vehicles is increased, and the traffic is interfered by the abnormal traffic events on the road section;
2) the number of vehicles arriving at the downstream intersection is reduced, the travel time of the vehicles is shortened, the traffic volume of the road section is reduced, and the vehicles reach a free-running state;
3) the number of vehicles arriving at the downstream intersection is increased, the travel time of the vehicles is increased, the traffic volume of the road section is increased, and the vehicles queue at the intersection and the like;
4) the number of vehicles arriving at the downstream intersection is increased, the travel time of the vehicles is shortened, the road traffic flow is increased, but the running speed is increased, and the situation does not exist in reality and can be eliminated.
Therefore, the first abnormal situation is the road section abnormality to be detected by the invention. Within a single natural day, there are a total of n time windows (i.e., tw)1,tw2,tw3,...,twn) The corresponding road section traffic modes are STM respectively1,STM2,STM3,...,STMn. The logic for anomaly detection is that when the time window is scrolled to twiJudging STM by using the traffic mode of the road section in the preorder time windowiWhether an exception is generated. When an emergency occurs in a road segment, traffic flow from upstream to downstream of the road segment may be abnormal in that the number of vehicles traveling from an upstream intersection to a downstream intersection is reduced and the travel time of the road segment is increased. Therefore, the invention firstly establishes the road section traffic mode of the vehicle number and the travel time, and then identifies the road section traffic modeAnd further completing the detection of the traffic abnormality of the identified road section.
In the road traffic abnormality detection, the time window tw and the abnormality determination threshold δ are important parameters, and calibration is required to increase the accuracy of the abnormality detection. Generally speaking, the larger the time window is, the larger the data volume of the number of vehicles and the travel time of the vehicles on the road section in the time window is, that is, the more accurate the traffic mode of the road section is. The road section traffic mode caused by an overlarge time window cannot reflect the continuous change condition of the traffic flow in time sequence. The larger the abnormality determination threshold value is, the lower the sensitivity of the abnormality detection model is. Therefore, model parameters with different sensitivities are given to adapt to different demand scenarios. As shown in table 1.
TABLE 1 threshold parameters for different sensitivities
The invention provides a road section traffic abnormity detection method based on electric warning license plate identification data. Firstly, based on electronic police equipment arranged at an upstream intersection and a downstream intersection of a road section, utilizing recognized license plate data to match the upstream intersection and the downstream intersection, and calculating the travel time of the road section; secondly, preprocessing the travel time data such as abnormal deletion, duplicate redundancy removal and the like to ensure the quality and the validity of the data; finally, the traffic abnormality in the identified road section is detected by using the pattern change of the travel time. The invention has the following two advantages: (1) the electronic police are used as data sources, the electronic police are used as law enforcement equipment, and entrance roads at main intersections of cities are widely arranged, so that illegal snapshot is realized, a large amount of license plate data and vehicle passing information can be recognized and collected, and the cost of traffic abnormity detection can be reduced to the maximum extent by multiplexing the electronic police equipment. (2) The method can quickly and effectively identify the traffic abnormality of the road section by mining the sudden change of the travel time mode of the road section, and timely give an alarm to a traffic management department so as to quickly respond and dispose the traffic abnormality, control the influence range of the abnormality to be minimum, and ensure the safety and the order of the urban road network.
Corresponding to the detection method shown in fig. 1, the present invention further provides a system for detecting road traffic abnormality, and fig. 3 is a schematic structural diagram of the system for detecting road traffic abnormality according to the present invention. As shown in fig. 3, the structure includes the following:
the vehicle passing information acquisition module 301 is used for acquiring vehicle passing information identified by an electronic police in the detection area; the detection area is a road area between an upstream intersection and a downstream intersection of the road; the vehicle passing information comprises first vehicle information and second vehicle information; the first vehicle information is vehicle information identified by an electronic police at the upstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the upstream intersection; the second vehicle information is vehicle information identified by an electronic police at the downstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the downstream intersection;
a travel time obtaining module 302, configured to obtain a travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information; the travel time is the time from an upstream intersection to a downstream intersection of the vehicle;
a vehicle mode obtaining module 303, configured to obtain a vehicle mode in a current time window; the vehicle mode is a two-dimensional vector consisting of the number of vehicles arriving at the downstream intersection in the current time window and the average value of the travel time of all vehicles arriving at the downstream intersection in the current time window;
a vehicle mode average value obtaining module 304, configured to obtain vehicle mode average values in all time windows in a current natural day;
a mahalanobis distance calculation module 305 for calculating mahalanobis distances between the vehicle mode in the current time window and the average value of the vehicle modes in all the time windows;
the first judging module 306 is configured to judge whether the mahalanobis distance is smaller than a set threshold, so as to obtain a first judgment result;
a traffic anomaly determination module 307, configured to determine that no traffic anomaly occurs in the detection area within the current time window when the first determination result indicates that the mahalanobis distance is smaller than a set threshold; and when the first judgment result shows that the Mahalanobis distance is not smaller than a set threshold value, determining that the traffic abnormal condition occurs in the detection area in the current time window.
As another embodiment, the detection system further includes a first noise processing module, configured to perform noise processing on the vehicle passing information after obtaining the vehicle passing information identified by the electronic police in the detection area, where the noise processing module specifically includes:
the second judgment unit is used for judging whether each piece of passing information comprises a license plate and a passing timestamp to obtain a second judgment result;
a deleting unit, configured to delete the vehicle passing information when the second determination result indicates that the vehicle passing information does not include the license plate or the vehicle passing timestamp;
the third judging unit is used for judging whether the license plates of any plurality of pieces of first vehicle information are the same or not, and the serial numbers are different from the passing timestamp to obtain a third judging result;
a first determining unit, configured to determine, when the third determination result indicates that license plates of vehicles are the same among the plurality of pieces of first vehicle information and that the numbers are different from the vehicle-passing timestamps, the license plate corresponding to the maximum value of the vehicle-passing timestamp, the number, and the vehicle-passing timestamp as final first vehicle information of the vehicle, and delete the remaining first vehicle information of the vehicle;
the fourth judging unit is used for judging whether the license plates of any plurality of pieces of second vehicle information are the same or not, and the serial numbers are different from the passing timestamps to obtain a fourth judging result;
and a second determining unit, configured to determine, when the fourth determination result indicates that the license plates of the vehicles are the same among the plurality of pieces of second vehicle information and the numbers are different from the passing timestamps, the license plate corresponding to the maximum value of the passing timestamp, the number, and the passing timestamp as final second vehicle information of the vehicle, and delete the remaining second vehicle information of the vehicle.
As another embodiment, the journey time obtaining module 302 specifically includes:
the matching unit is used for matching license plates in the first vehicle information and the second vehicle information to obtain a first time stamp and a second time stamp corresponding to the same license plate; the first timestamp is a vehicle-passing timestamp in the first vehicle information, and the second timestamp is a vehicle-passing timestamp in the second vehicle information;
the difference making unit is used for making a difference between the first time stamp and the second time stamp to obtain an initial travel time;
the fifth judging unit is used for judging whether the initial travel time is within a set time range or not to obtain a fifth judging result;
and a third determining unit, configured to determine the initial travel time as a travel time corresponding to the vehicle when the fifth determination result indicates that the initial travel time is within a set time range.
As another embodiment, the detection system further includes a second noise processing module, configured to perform noise processing on travel times corresponding to all vehicles after obtaining a travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information, and specifically includes:
the sampling parameter determining unit is used for determining a sampling window and a sampling step length of noise processing;
the eliminating unit is used for carrying out noise processing on the travel time in each sampling window according to the box diagram principle, eliminating abnormal value data outside an upper boundary and a lower boundary and obtaining the travel time after the noise processing of the sampling windows;
and the travel time determining unit is used for sequentially obtaining the travel time after the noise processing of each sampling window according to the sampling step length.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. A method for detecting road traffic abnormality, comprising:
acquiring vehicle passing information identified by an electronic police in the detection area; the detection area is a road area between an upstream intersection and a downstream intersection of the road; the vehicle passing information comprises first vehicle information and second vehicle information; the first vehicle information is vehicle information identified by an electronic police at the upstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the upstream intersection; the second vehicle information is vehicle information identified by an electronic police at the downstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the downstream intersection;
acquiring travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information; the travel time is the time from an upstream intersection to a downstream intersection of the vehicle;
acquiring a vehicle mode in a current time window; the vehicle mode is a two-dimensional vector consisting of the number of vehicles arriving at the downstream intersection in the current time window and the average value of the travel time of all vehicles arriving at the downstream intersection in the current time window;
acquiring the average value of the vehicle modes in all time windows in the current natural day;
calculating the Mahalanobis distance between the vehicle mode in the current time window and the average value of the vehicle modes in all the time windows;
judging whether the Mahalanobis distance is smaller than a set threshold value or not to obtain a first judgment result;
when the first judgment result shows that the Mahalanobis distance is smaller than a set threshold value, determining that no abnormal traffic condition occurs in the detection area in the current time window;
and when the first judgment result shows that the Mahalanobis distance is not smaller than a set threshold value, determining that the traffic abnormal condition occurs in the detection area in the current time window.
2. The method according to claim 1, wherein the step of obtaining the passing information identified by the electronic police in the detection area further comprises a step of performing noise processing on the passing information, and specifically comprises:
judging whether each piece of passing information comprises a license plate and a passing timestamp to obtain a second judgment result;
when the second judgment result shows that the passing information does not include the license plate or the passing timestamp, deleting the passing information;
judging whether license plates of any plurality of pieces of first vehicle information are the same or not, wherein the serial numbers are different from the passing timestamps, and obtaining a third judgment result;
when the third judgment result shows that the license plates of the vehicles are the same among the first pieces of vehicle information and the numbers are different from the vehicle-passing timestamps, determining the license plate, the number and the vehicle-passing timestamp corresponding to the maximum value of the vehicle-passing timestamp as the final first vehicle information of the vehicle, and deleting the rest first vehicle information of the vehicle;
judging whether the license plates of any plurality of pieces of second vehicle information are the same or not, wherein the serial numbers are different from the passing timestamps, and obtaining a fourth judgment result;
and when the fourth judgment result shows that the license plates of the vehicles are the same among the second pieces of vehicle information and the numbers are different from the vehicle-passing timestamps, determining the license plate, the number and the vehicle-passing timestamp corresponding to the maximum value of the vehicle-passing timestamp as the final second vehicle information of the vehicle, and deleting the rest second vehicle information of the vehicle.
3. The method for detecting a road traffic abnormality according to claim 1, wherein the obtaining of the travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information specifically includes:
matching license plates in the first vehicle information and the second vehicle information to obtain a first time stamp and a second time stamp corresponding to the same license plate; the first timestamp is a vehicle-passing timestamp in the first vehicle information, and the second timestamp is a vehicle-passing timestamp in the second vehicle information;
the first time stamp and the second time stamp are subjected to difference to obtain initial travel time;
judging whether the initial travel time is within a set time range or not to obtain a fifth judgment result;
and when the fifth judgment result shows that the initial travel time is in a set time range, determining the initial travel time as the travel time corresponding to the vehicle.
4. The method according to claim 1, wherein the step of obtaining the travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information and then performing noise processing on the travel times corresponding to all vehicles comprises:
determining a sampling window and a sampling step length of noise processing;
noise processing is carried out on the travel time in each sampling window according to a box diagram principle, abnormal value data outside an upper boundary and a lower boundary are removed, and the travel time after noise processing of the sampling windows is obtained;
and sequentially obtaining the travel time of each sampling window after noise processing according to the sampling step length.
5. The method of claim 1, wherein the mahalanobis distance between the vehicle pattern in the current time window and the average of the vehicle patterns in all time windows is
Wherein i is more than or equal to 2, STMiSTM for vehicle mode in current time window1～iIs the average value of the vehicle modes in all time windows of the current natural day,
s is STMiAnd STM1～iThe covariance matrix of the constructed full sample.
6. The method of detecting a road traffic abnormality according to claim 5, characterized in that the set threshold value is
Delta is a set coefficient, and j has a value range of [1, i]，STMjTo slave STM1To STMiThe vehicle mode of the jth time window.
7. A system for detecting a road traffic anomaly, comprising:
the vehicle passing information acquisition module is used for acquiring vehicle passing information identified by an electronic police in the detection area; the detection area is a road area between an upstream intersection and a downstream intersection of the road; the vehicle passing information comprises first vehicle information and second vehicle information; the first vehicle information is vehicle information identified by an electronic police at the upstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the upstream intersection; the second vehicle information is vehicle information identified by an electronic police at the downstream intersection, and comprises the serial number, the license plate and a passing timestamp of the vehicle at the downstream intersection;
the travel time acquisition module is used for acquiring the travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information; the travel time is the time from an upstream intersection to a downstream intersection of the vehicle;
the vehicle mode acquisition module is used for acquiring a vehicle mode in a current time window; the vehicle mode is a two-dimensional vector consisting of the number of vehicles arriving at the downstream intersection in the current time window and the average value of the travel time of all vehicles arriving at the downstream intersection in the current time window;
the vehicle mode average value acquisition module is used for acquiring vehicle mode average values in all time windows in the current natural day;
the Mahalanobis distance calculation module is used for calculating the Mahalanobis distances between the vehicle modes in the current time window and the average value of the vehicle modes in all the time windows;
the first judgment module is used for judging whether the Mahalanobis distance is smaller than a set threshold value or not to obtain a first judgment result;
the traffic abnormity determining module is used for determining that no traffic abnormity occurs in the detection area in the current time window when the first judgment result shows that the Mahalanobis distance is smaller than a set threshold value; and when the first judgment result shows that the Mahalanobis distance is not smaller than a set threshold value, determining that the traffic abnormal condition occurs in the detection area in the current time window.
8. The system for detecting road traffic abnormality according to claim 7, further comprising a first noise processing module, configured to perform noise processing on the vehicle passing information after obtaining the vehicle passing information identified by the electronic police in the detection area, specifically comprising:
the second judgment unit is used for judging whether each piece of passing information comprises a license plate and a passing timestamp to obtain a second judgment result;
a deleting unit, configured to delete the vehicle passing information when the second determination result indicates that the vehicle passing information does not include the license plate or the vehicle passing timestamp;
the third judging unit is used for judging whether the license plates of any plurality of pieces of first vehicle information are the same or not, and the serial numbers are different from the passing timestamp to obtain a third judging result;
a first determining unit, configured to determine, when the third determination result indicates that license plates of vehicles are the same among the plurality of pieces of first vehicle information and that the numbers are different from the vehicle-passing timestamps, the license plate corresponding to the maximum value of the vehicle-passing timestamp, the number, and the vehicle-passing timestamp as final first vehicle information of the vehicle, and delete the remaining first vehicle information of the vehicle;
the fourth judging unit is used for judging whether the license plates of any plurality of pieces of second vehicle information are the same or not, and the serial numbers are different from the passing timestamps to obtain a fourth judging result;
and a second determining unit, configured to determine, when the fourth determination result indicates that the license plates of the vehicles are the same among the plurality of pieces of second vehicle information and the numbers are different from the passing timestamps, the license plate corresponding to the maximum value of the passing timestamp, the number, and the passing timestamp as final second vehicle information of the vehicle, and delete the remaining second vehicle information of the vehicle.
9. The system for detecting road traffic abnormality according to claim 7, wherein the travel time acquisition module specifically includes:
the matching unit is used for matching license plates in the first vehicle information and the second vehicle information to obtain a first time stamp and a second time stamp corresponding to the same license plate; the first timestamp is a vehicle-passing timestamp in the first vehicle information, and the second timestamp is a vehicle-passing timestamp in the second vehicle information;
the difference making unit is used for making a difference between the first time stamp and the second time stamp to obtain an initial travel time;
the fifth judging unit is used for judging whether the initial travel time is within a set time range or not to obtain a fifth judging result;
and a third determining unit, configured to determine the initial travel time as a travel time corresponding to the vehicle when the fifth determination result indicates that the initial travel time is within a set time range.
10. The system according to claim 7, further comprising a second noise processing module, configured to perform noise processing on travel times corresponding to all vehicles after obtaining a travel time corresponding to each vehicle according to the first vehicle information and the second vehicle information, specifically including:
the sampling parameter determining unit is used for determining a sampling window and a sampling step length of noise processing;
the eliminating unit is used for carrying out noise processing on the travel time in each sampling window according to the box diagram principle, eliminating abnormal value data outside an upper boundary and a lower boundary and obtaining the travel time after the noise processing of the sampling windows;
and the travel time determining unit is used for sequentially obtaining the travel time after the noise processing of each sampling window according to the sampling step length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910455487.5A CN110164132B (en) | 2019-05-29 | 2019-05-29 | Method and system for detecting road traffic abnormity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910455487.5A CN110164132B (en) | 2019-05-29 | 2019-05-29 | Method and system for detecting road traffic abnormity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110164132A CN110164132A (en) | 2019-08-23 |
| CN110164132B true CN110164132B (en) | 2020-03-06 |
Family
ID=67629855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910455487.5A Active CN110164132B (en) | 2019-05-29 | 2019-05-29 | Method and system for detecting road traffic abnormity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110164132B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI774984B (en) * | 2019-09-05 | 2022-08-21 | 中華電信股份有限公司 | Traffic incident detection system and method |
| CN110570664B (en) * | 2019-09-23 | 2023-04-07 | 山东科技大学 | Automatic detection system for highway traffic incident |
| CN111681430B (en) * | 2020-04-30 | 2022-03-29 | 安徽科力信息产业有限责任公司 | Method for predicting number of stop lines of signal lamp intersection in future in real time |
| CN111986483A (en) * | 2020-08-28 | 2020-11-24 | 上海宝康电子控制工程有限公司 | Method and device for studying and judging road congestion state based on electric alarm data collision and storage medium |
| CN112669608B (en) * | 2020-12-25 | 2022-04-22 | 浙江大学 | License plate recognition data re-matching method based on maximum travel time probability |
| CN113192326B (en) * | 2021-04-23 | 2022-04-29 | 长安大学 | Cascade type road operation risk prevention and control system and single-point fault isolation method |
| CN113554264B (en) * | 2021-06-07 | 2023-11-17 | 华北电力科学研究院有限责任公司 | Alarm method and device for power generation equipment |
| CN113936454B (en) * | 2021-09-22 | 2023-02-21 | 上海宝康电子控制工程有限公司 | Information processing method and system based on track fusion |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100582555B1 (en) * | 2004-11-10 | 2006-05-23 | 한국전자통신연구원 | Apparatus for detectiong and visualizing anomalies of network traffic and method therof |
| CN100444210C (en) * | 2007-04-20 | 2008-12-17 | 东南大学 | Mixed controlling method of single dot signal controlling crossing |
| CN103383816B (en) * | 2013-07-01 | 2015-09-02 | 青岛海信网络科技股份有限公司 | The traffic signal control method that multiplexing electronic police mixed traffic flow detects and device |
| JP6620056B2 (en) * | 2016-03-31 | 2019-12-11 | 三菱日立パワーシステムズ株式会社 | Device abnormality diagnosis method and device abnormality diagnosis device |
| KR101703058B1 (en) * | 2016-08-30 | 2017-02-06 | 주식회사 블루시그널 | System for predicting traffic state pattern by analysis of traffic data and predicting method thereof |
| CN108922170B (en) * | 2018-06-13 | 2020-11-27 | 同济大学 | Intersection safety evaluation method based on electronic police snapshot data |
| CN109191872B (en) * | 2018-10-09 | 2021-03-19 | 东南大学 | Intersection traffic flow characteristic parameter extraction method based on number plate data |
-
2019
- 2019-05-29 CN CN201910455487.5A patent/CN110164132B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110164132A (en) | 2019-08-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110164132B (en) | Method and system for detecting road traffic abnormity | |
| CN110197588B (en) | Method and device for evaluating driving behavior of large truck based on GPS track data | |
| CN109448369B (en) | Real-time operation risk calculation method for expressway | |
| CN102136190B (en) | Dispatching management system and method for event emergency response of urban bus passenger transport | |
| CN104809878B (en) | Method for detecting abnormal condition of urban road traffic by utilizing GPS (Global Positioning System) data of public buses | |
| CN107590999B (en) | Traffic state discrimination method based on checkpoint data | |
| CN104751642B (en) | A kind of advanced road real-time predictor method of traffic flow operation risk | |
| CN110176139A (en) | A kind of congestion in road identification method for visualizing based on DBSCAN+ | |
| CN112053558A (en) | Traffic jam state identification method, device and equipment | |
| CN107204115B (en) | Method for estimating parking rate of main line traffic flow | |
| CN110379169B (en) | Saturated headway calculation method for lane to be transferred based on video number plate equipment | |
| CN108922168B (en) | A kind of mid-scale view Frequent Accidents road sentences method for distinguishing | |
| CN113436432A (en) | Method for predicting short-term traffic risk of road section by using road side observation data | |
| CN111768619A (en) | Express way vehicle OD point determining method based on checkpoint data | |
| CN112767684A (en) | Highway traffic jam detection method based on charging data | |
| CN115691120A (en) | Congestion identification method and system based on highway running water data | |
| CN115527364B (en) | Traffic accident tracing method and system based on radar data fusion | |
| CN108492563B (en) | Overspeed event detection method based on average speed | |
| CN106530714A (en) | Secondary traffic accident time prediction method based on traffic flow data | |
| CN102360524B (en) | Automatic detection and confirmation method of dangerous traffic flow characteristics of highway | |
| CN111899517A (en) | Expressway fatigue driving illegal behavior determination method | |
| US20220319323A1 (en) | Method for identifying road risk based on networked vehicle-mounted adas | |
| WO2017107790A1 (en) | Method and apparatus for predicting road conditions using big data | |
| Cohen et al. | Travel time estimation between loop detectors and FCD: A compatibility study on the Lille network, France | |
| Wu et al. | Analysis of drivers’ stop-or-run behavior at signalized intersections with high-resolution traffic and signal event data |
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 |