CN115719546A - Track abnormity detection method and device, electronic equipment and readable storage medium - Google Patents
Track abnormity detection method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN115719546A CN115719546A CN202110975911.6A CN202110975911A CN115719546A CN 115719546 A CN115719546 A CN 115719546A CN 202110975911 A CN202110975911 A CN 202110975911A CN 115719546 A CN115719546 A CN 115719546A
- Authority
- CN
- China
- Prior art keywords
- time
- real
- target object
- historical
- night
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 claims abstract description 31
- 230000005856 abnormality Effects 0.000 claims abstract description 20
- 206010059013 Nocturnal emission Diseases 0.000 claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims abstract description 14
- 238000012216 screening Methods 0.000 claims description 19
- 238000004458 analytical method Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 8
- 230000014759 maintenance of location Effects 0.000 claims description 8
- 230000000422 nocturnal effect Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 4
- 230000002354 daily effect Effects 0.000 description 21
- 238000004891 communication Methods 0.000 description 6
- 230000035945 sensitivity Effects 0.000 description 4
- 241001622623 Coeliadinae Species 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007667 floating Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Landscapes
- Alarm Systems (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention provides a track abnormity detection method and device, electronic equipment and a readable storage medium, wherein the method comprises the following steps: acquiring historical driving track information and real-time driving track information of a target object; performing density clustering on the historical driving track information, and acquiring the day-time frequent station and the night-time frequent station of the target object based on the density clustering result; obtaining a day-night-time occurrence judgment ratio based on a first historical stay time of the target object at the day-night frequent residence and a second historical stay time of the night frequent residence respectively; based on the real-time running track information, acquiring and based on a first real-time stay time of the target object at the daily frequent site and a second real-time stay time of the target object at the night frequent site respectively to obtain a daytime and nighttime appearance real-time ratio; and detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission judgment ratio and the diurnal and nocturnal emission real-time ratio. The method can effectively remove data redundancy, reduce the time consumption of operation and improve the accuracy of abnormality judgment.
Description
Technical Field
The invention relates to the field of big data analysis and intelligent transportation, in particular to a track abnormity detection method and device, electronic equipment and a readable storage medium.
Background
In the prior art, people with daytime and nighttime can be found by counting the time of each person entering and leaving through security personnel in a community or by periodically touching the police. The judgment is also carried out by counting the total times of night appearance and the total times of day of the same person and calculating the ratio of night appearance and day appearance. In addition, a model detection method of daytime and nighttime arises, and specifically, assuming that vehicles which are daytime and nighttime appear are in a binomial distribution, the nighttime appearance probabilities of the various vehicle checkpoints are independent from each other, and a model of the nighttime appearance probability is established according to actual travel records.
With the increasing of the size of the floating population, the number of people in the community is increased, and the method for finding out people in the daytime and night type can bring greater working pressure to the workers through the scheme that security personnel in the community count the access time of each person or the scheme that a policeman periodically gropes at home, and also can leave omissions to cause incorrect judgment. The method for judging daytime/night travel times by simply counting the daytime/night travel times does not fully dig the value of data, and the accuracy is also influenced by the simple statistics of the data. The night-day occurrence probability model assumes that the night occurrence probabilities of all the vehicle bayonets are mutually independent, and the personnel traveling probabilities are not mutually independent in actual conditions, so that the probability model cannot be simply applied.
In addition, with the progress of electronic technology, more and more devices such as bayonet cameras and the like are used on roads, intersections, shopping malls, residential areas and the like, massive face data can be generated every day, common operations are stored in a relational database, however, the calculation amount is large during analysis, the processing is difficult, and people who go out at night in the daytime and night cannot be effectively and quickly controlled.
When the user is subjected to the permanent location analysis, the movement range of the user is wide, and the resident information is inaccurate in statistics due to a positioning result obtained by using the communication information of the simple mobile terminal. The method comprises the steps of carrying out clustering analysis on one or more track points of a user to be detected to obtain an aggregation region, not considering sequence correlation among the track points and factors of stay time, and simply using the frequency of the regions appearing in a preset time interval to judge whether the user is a permanent station.
Disclosure of Invention
The invention provides a track anomaly detection method and device, electronic equipment and a readable storage medium, which are used for solving the technical defects in the prior art.
The invention provides a track abnormity detection method, which comprises the following steps:
acquiring historical driving track information and real-time driving track information of a target object;
performing density clustering on the historical driving track information, and acquiring the day permanent station and the night permanent station of the target object based on the result of the density clustering; obtaining a daytime and nighttime judgment proportion based on a first historical stay time of the target object at the daily frequent residence and a second historical stay time of the nighttime frequent residence respectively;
based on the real-time running track information, acquiring and based on a first real-time stay time of the target object at a daily frequent site and a second real-time stay time of the target object at a night frequent site respectively, and acquiring a daytime and nighttime emergence real-time ratio;
and detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission judgment ratio and the diurnal and nocturnal emission real-time ratio.
The track anomaly detection method according to the invention, wherein the density clustering is performed on the historical driving track information, and the day-night permanent station and the night permanent station of the target object are obtained based on the density clustering result, and the method comprises the following steps:
performing density clustering on the historical driving track information of the target object according to the similarity in time and space to obtain clustering centers, performing cluster clustering on each clustering center to obtain clustering results, and screening the clustering results based on preset rules to obtain day-time frequent stops and night-time frequent stops of the target object.
The track anomaly detection method according to the invention, wherein the density clustering is performed on the historical driving track information, and the day-night permanent station and the night permanent station of the target object are obtained based on the density clustering result, and the method comprises the following steps:
for a target object i, recording an acquired elapsed time sequence x of historical K time periods im Corresponding sequence of transit points y im And corresponding dwell time sequences z im M =1, \8230;, K; screening sequences with the same data dimension from the K retention time sequences for clustering, and recording a set X of each group of elapsed time sequences ij Corresponding set of sequences of places Y ij And corresponding set of dwell time sequences Z ij J =1, \8230, N, and N groups;
for the j group of the elapsed time series set X ij And a corresponding set of sequences of passing sites Y ij Carrying out group clustering, wherein j =1, \ 8230, and N, the specific steps are as follows:
setting the h-th data X in the time sequence set and the place sequence set ijh And Y ijh Traversing all classes for class centers of a certain class if the I data X in the set ijl And X ijh Poly as one type, Y ijl And Y ijh If the cluster is one type, clustering is successful, and a class center is updated; if the two groups do not form a class, the new class center is formed by the self; otherwise, abandoning the first data X ijl And Y ijl ；
Circularly traversing the time sequence set and the place sequence set to obtain a final elapsed time sequence and a final elapsed place sequence clustering result CX j ，CY j J =1, \8230, N; clustering results and CZ of dwell time series j With CX j ，CY j Corresponding; the number of W classes is obtained,
calculating class centers of the time sequences, the corresponding passing place sequences and the corresponding staying time sequences in the W classes;
performing cluster clustering on the W cluster centers passing through the site sequence to obtain S clustering results, namely the number of the permanent sites,
Location w the number of clusters in the w-th cluster;
mapping the clustering result to a passing time sequence and a staying time sequence to obtain a one-to-one correspondence relationship between the residence place and the passing time and the staying time;
screening S permanent premises, corresponding residence time and elapsed time: reserving a normal station with the stay time larger than a preset time threshold; calculating the distance between every two reserved normal stations, and deleting the normal station with short residence time if the distance is smaller than a preset distance threshold value to obtain a pre-selected normal station;
and pre-selecting the constant station based on a preset day and night division rule to obtain the day constant station and the night constant station.
The track anomaly detection method according to the invention is characterized in that the obtaining of the judgment ratio of the daytime and nighttime eminences based on the first historical stay time of the target object at the daily frequent residence and the second historical stay time of the night frequent residence respectively comprises the following steps:
based on the pre-selected permanent station, the corresponding stay time and the elapsed time, the first historical stay time of the target object at the permanent station in the daytime and the second historical stay time of the permanent station at night are counted;
calculating the daytime and nighttime output judgment ratio r of the target object i i = second history stay length/(first history stay length + second history stay length).
The track anomaly detection method according to the invention, wherein the obtaining and the obtaining of the real-time ratio of the daytime and the night-time exhibitions based on the first real-time staying time of the target object at the daily frequent place and the second real-time staying time of the night frequent place respectively based on the real-time traveling track information comprises:
acquiring a real-time constant station and corresponding residence time and elapsed time based on the real-time driving track information, and counting first real-time residence time of the target object at the day constant station and second real-time residence time of the target object at the night constant station;
calculating the daytime and nighttime output real-time ratio b = second real-time stay duration/(first real-time stay duration + second real-time stay duration) of the target object i.
The track anomaly detection method according to the present invention further includes:
performing density clustering on the real-time driving track information of the target object and the W classes, if the real-time driving track information of the target object is successfully clustered into one class, updating a class center, otherwise, independently clustering into one class to obtain updated classification;
updating the judgment ratio of the current station and the daytime and nighttime of the target object based on the updated classification; and/or the track abnormity detection method further comprises the following steps:
and when the real-time running track abnormality of the target object is detected, giving an alarm.
The track abnormality detection method according to the present invention, wherein the detecting whether the real-time travel track of the target object is abnormal based on a result of comparing the daytime/nighttime output determination ratio and the daytime/nighttime output real-time ratio, includes:
real-time ratio b as the diurnal and nocturnal emissions of the target object<Daytime and nighttime judging ratio r i When, satisfy | b-r i |/b>And p, detecting the real-time running track abnormity of the target object, wherein p represents the set diurnal and nocturnal fluctuation ratio.
The invention also provides a track anomaly detection device, which comprises:
the data acquisition module is used for acquiring historical driving track information and real-time driving track information of the target object;
the historical track analysis module is used for carrying out density clustering on the historical driving track information and obtaining the day-time frequent station and the night-time frequent station of the target object based on the density clustering result; obtaining a day and night emergence judgment proportion based on a first historical stay time of the target object at a day permanent site and a second historical stay time of the target object at a night permanent site respectively;
the real-time track analysis module is used for acquiring and obtaining a real-time daytime and nighttime emergence proportion based on a first real-time stay time of the target object at a daily station and a second real-time stay time of the target object at a night station respectively based on the real-time driving track information;
and the detection module is used for detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission judgment proportion and the diurnal and nocturnal emission real-time proportion.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the steps of any one of the above track abnormity detection methods.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the trajectory anomaly detection method as described in any one of the above.
According to the method, only the stay time and the stay proportion of the real-time data of the target object at the daily and night permanent places need to be counted for judgment, so that data redundancy can be effectively removed, the time consumption of operation is reduced, and a closed-loop system is formed on the basis of the internal relation and consistency between the historical data and the real-time data; clustering and screening the track time sequences of each target object in a plurality of time periods to finally obtain day-time permanent stations and night-time permanent stations, and considering the correlation of the time sequences and the internal relation between track points; the individuation and differentiation of each composite body are fully utilized to respectively obtain the daytime and night output judgment proportion of each target object and the real-time daytime and night output proportion for comparison, so that the accuracy of abnormality judgment is improved.
Drawings
In order to more clearly illustrate the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic flow chart of a track anomaly detection method according to the present invention;
FIG. 2 is a second schematic flow chart of the track anomaly detection method provided by the present invention;
FIG. 3 is a schematic structural diagram of a track anomaly detection device provided by the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
A track anomaly detection method according to the present invention is described below with reference to fig. 1, and the method includes:
s1, acquiring historical driving track information and real-time driving track information of a target object;
the target object can be different personnel, and at this moment, gather the data of different personnel's trip in-process record on equipment such as camera head, bayonet socket, entrance guard, gather the data of two parts respectively: the method comprises the following steps that firstly, historical driving track information of different personnel in a plurality of time periods is obtained; and secondly, real-time running track information of the target object in a time period.
The information of the driving track in a time period comprises identity information of personnel, at least one passing place (longitude and latitude), a passing time and a set of stopping time in the time period. Here, a time period means 24 hours a day and night, and only minutes and seconds are recorded without recording the year, month and day through the time.
S2, performing density clustering on the historical driving track information, and obtaining the day permanent station and the night permanent station of the target object based on the density clustering result; obtaining a daytime and nighttime judgment proportion based on a first historical stay time of the target object at the daily frequent residence and a second historical stay time of the nighttime frequent residence respectively;
the data collected by different personnel in the data collection module 10 is modeled and analyzed based on a density clustering method, specifically, each personnel can be modeled respectively through a DBSCAN density clustering algorithm, the collected data are clustered according to similarity in time and space, if the time and the space meet requirements, the current time sequence is clustered into one class, then each clustering center is clustered, and daily residences and nights of different personnel are obtained through screening by a certain rule.
S3, based on the real-time running track information, obtaining and based on a first real-time stay time of the target object at the daily frequent site and a second real-time stay time of the target object at the night frequent site respectively, and obtaining a real-time ratio of the daytime and the night;
the obtaining mode of the daytime and nighttime output real-time proportion is the same as the daytime and nighttime output judging proportion and is determined based on the stay time of the daily frequent residence and the stay time of the nighttime frequent residence.
And S4, detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the daytime and nighttime output judgment proportion and the real-time daytime and nighttime output proportion.
If the real-time daytime and nighttime ratio is smaller than the daytime and nighttime judgment ratio, the target object stays at the night place for a time ratio smaller than the daytime and nighttime judgment ratio obtained from the historical driving track information, and then the real-time driving track abnormality of the target object is detected; on the contrary, it can be detected that the real-time travel track of the target object is normal.
According to the method, only the stay time and the stay proportion of the real-time data of the target object at the daily and night permanent places need to be counted for judgment, so that data redundancy can be effectively removed, the time consumption of operation is reduced, and a closed-loop system is formed on the basis of the internal relation and consistency between the historical data and the real-time data; clustering and screening the track time sequences of each target object in a plurality of time periods to finally obtain day-time permanent places and night-time permanent places, and considering the correlation of the time sequences and the internal relation between track points; the individuation and differentiation of each combination body are fully utilized to respectively obtain the daytime and nighttime judgment proportion and the daytime and nighttime proportion of each target object for comparison, so that the accuracy of abnormality judgment is improved.
The track anomaly detection method according to the invention, wherein the density clustering is performed on the historical driving track information, and the daytime frequent residence and the night frequent residence of the target object are obtained based on the result of the density clustering, and the method comprises the following steps:
performing density clustering on the historical driving track information of the target object according to the similarity in time and space to obtain clustering centers, performing cluster clustering on each clustering center to obtain clustering results, and screening the clustering results based on preset rules to obtain day-time frequent stops and night-time frequent stops of the target object.
The track anomaly detection method according to the invention, wherein the density clustering is performed on the historical driving track information, and the daytime frequent residence and the night frequent residence of the target object are obtained based on the result of the density clustering, and the method comprises the following steps:
for a target object i, recording an elapsed time sequence x of collected historical K time periods im Corresponding sequence of transit points y im And corresponding dwell time sequences z im M =1, \8230;, K; screening sequences with the same data dimension from the K retention time sequences for clustering, and recording a set X of each group of elapsed time sequences ij Corresponding set of sequences of passing sites Y ij And correspond toSet of dwell time sequences Z ij J =1, \ 8230, N, total N groups;
for the obtained N groups of sequences X with the same data dimension ij ，Y ij ，Z ij J =1, \ 8230, N is clustered in groups. And screening by using a certain rule to obtain an integral clustering result, and specifically comprising the following steps:
1) For the j group of the elapsed time series set X ij And corresponding sets of passing location (latitude and longitude) sequences Y ij Carrying out group clustering, wherein j =1, \ 8230, and N;
2) For the jth group X ij ，Y ij Get the intersection of the subscripts of the members in the cluster to obtain the incompatible W j Cluster, specifically: x ij Clustering members of a class with Y ij And traversing the members of each class of the clustering result to obtain the member combination with the most intersected same subscript, namely if the l, h and k data X ijl 、X ijh And X ijk Poly are one, and Y ijl 、Y ijh And Y ijk Clustering is successful if the cluster is one type; otherwise clustering fails. If the number of elements in the class is less than 2, it means that there is only a single data in the class, and a failure is also indicated. Discarding the data and category that fails clustering. So as to obtain the final clustering result CX of the time sequence and the place sequence j ，CY j J =1, \8230, N; clustering results of retention time series with CZ j With CX j ，CY j Correspondingly, no separate clustering is performed here;
a total of W classes are obtained,
calculating class centers of the time sequences, the corresponding passing place sequences and the corresponding staying time sequences in the W classes;
performing cluster clustering on the W cluster centers passing through the site sequence to obtain S clustering results, namely the number of the permanent sites,
Location w the number of clusters in the w-th cluster;
mapping the clustering result to a passing time sequence and a staying time sequence to obtain a one-to-one correspondence relationship between the residence place and the passing time and the staying time;
screening S permanent sites and corresponding residence time and elapsed time: reserving the ordinary station with the stay time larger than a preset time threshold; calculating the distance between every two reserved ordinary stations, and deleting the ordinary station with shorter residence time if the distance is smaller than a preset distance threshold value, namely if the distance is close to the preset distance threshold value, so as to obtain a pre-selected ordinary station; for example, the preset time threshold may be set to 1h or other time value, and the preset distance threshold may be set to any value within an interval of 10 meters to 500 meters. The step is to further screen the permanent station and select the pre-selected permanent station with reference meaning.
The day permanent premises and the night permanent premises are obtained based on a preset day and night division rule (here, 7 00-19 is day time, 19-00 is night time, 7.
The track anomaly detection method according to the invention is characterized in that the obtaining of the judgment ratio of the daytime and nighttime eminences based on the first historical stay time of the target object at the daily frequent residence and the second historical stay time of the night frequent residence respectively comprises the following steps:
counting a first historical stay time of the target object at the day permanent station and a second historical stay time of the target object at the night permanent station respectively based on the preselected permanent station, the corresponding stay time and the elapsed time;
calculating the daytime and nighttime output judgment ratio r of the target object i i = second historical stay duration/(first historical stay duration + second historical stay duration).
The track anomaly detection method according to the invention, wherein the obtaining and the obtaining of the real-time ratio of the daytime and the night-time exhibitions based on the first real-time staying time of the target object at the daily frequent place and the second real-time staying time of the night frequent place respectively based on the real-time traveling track information comprises:
acquiring a real-time constant station and corresponding residence time and elapsed time based on the real-time driving track information, and counting first real-time residence time of the target object at the day constant station and second real-time residence time of the target object at the night constant station;
and calculating the diurnal and nocturnal real-time proportion b = second real-time stay duration/(first real-time stay duration + second real-time stay duration) of the target object i.
The track anomaly detection method according to the present invention further includes:
performing density clustering on the real-time driving track information of the target object and the W classes, if the real-time driving track information of the target object is successfully clustered into one class, updating a class center, otherwise, independently clustering into one class to obtain updated classification;
and updating the current station and the diurnal and photovoltaic emergence judgment ratio of the target object based on the updated classification for judgment of the diurnal and photovoltaic emergence module of the next round. The invention considers the internal relation and consistency between historical data and real-time data, is not simple data statistics, but forms a closed-loop system, has dynamic correction function, and improves the judgment accuracy of people in daytime and nighttime.
Specifically, a new piece of real-time travel track information can be compared with W classes, the distance is calculated, the distance between the new piece of real-time travel track information and the class center of the W class is minimum, the distance is smaller than a preset threshold value, the new data belongs to the W class, the number of the W class members is +1, and the class center is recalculated; otherwise, the new data becomes a single class, the W +1 th class, and the total number of classes is also +1. Namely, the historical running track information is updated and iterated by utilizing the real-time running track information every time, so that better accuracy is realized.
Further, the track anomaly detection method further includes:
and when the real-time running track abnormity of the target object is detected, alarming. If the target object is determined to be of the daytime and night-time output type, immediately notifying security or policemen to perform further checking.
The track abnormality detection method according to the present invention, wherein the detecting whether the real-time travel track of the target object is abnormal based on a result of comparing the daytime/nighttime output determination ratio and the daytime/nighttime output real-time ratio, includes:
real-time ratio b as the diurnal and nocturnal emissions of the target object<Ratio r for judging daytime and nighttime i While satisfying | b-r i |/b>And p, detecting the real-time running track abnormality of the target object, wherein p represents the set diurnal and nocturnal fluctuation ratio. p is set artificially, and because the sensitivity of different scenes to the abnormality is different, p can be set to 10% or 20%, and the value of p can be freely adjusted. The floating proportion of the daytime and night is increased, the sensitivity of detection can be adjusted, and a critical condition for judging the abnormity is increased, so that the judgment is more accurate.
To further illustrate the track anomaly detection method provided by the present invention, referring to fig. 2, the present invention provides a specific embodiment, which includes:
firstly, gather the data of different personnel's trip in-process record on equipment such as camera head, bayonet socket, entrance guard, gather the data of two parts respectively: the method comprises the following steps that firstly, historical driving track information of different personnel in a plurality of time periods is obtained; secondly, real-time driving track information of the target object in a time period;
the method comprises the steps of carrying out modeling analysis on data acquired by different personnel in a data acquisition module 10 based on a density clustering method, specifically, respectively modeling each personnel through a DBSCAN density clustering algorithm, clustering the acquired data according to similarity in time and space, clustering current time sequences into one class if the time and the space meet requirements, clustering the cluster centers into one class, and screening through a certain rule to obtain day permanent premises and night permanent premises of different personnel.
Suppose a person i has day permanent sites A, B, C and night permanent sites D, E, and the historical stay time is recorded as t l L = A, \8230E, E, the first history staying time is t A +t B +t C The second history staying time is t D +t E ，r i ＝(t D +t E )/(t A +t B +t C +t D +t E ) (ii) a The new real-time track information of the person i is counted, and the staying time at the normal station is T l ，l＝A,…,E， b＝(T D +T E )/(T A +T B +T C +T D +T E ) The diurnal and nocturnal fluctuation ratio is set to p.
For example: setting p =10%, enabling a person i to stay at work place for 12h at night and stay at work place for r at night i Is 8/(8 + 12) =0.4; counting a new real-time track to obtain 4h at home at night, and keeping 8h at a working place in a white day, wherein b is 4/(4 + 8) =0.33; b is obtained by calculation<r i And | b-r i |/b＝ 21％>And p, the person track is considered to be abnormal preliminarily, and at the moment, security or policemen can be immediately informed to further check.
And using the real-time data driving track information of the target object in a time period to update the current station set and the diurnal and photovoltaic judgment proportion of the target object, and using the current station set and the diurnal and photovoltaic judgment proportion to judge the diurnal and photovoltaic module in the next round.
Referring to fig. 3, a description will be given below of a track anomaly detection device according to the present invention, and the track anomaly detection device described below and the track anomaly detection method described above may be referred to in correspondence with each other, and the track anomaly detection device includes:
the data acquisition module 10 is used for acquiring historical driving track information and real-time driving track information of a target object;
the target object can be different personnel, and at this moment, gather the data of different personnel's trip in-process record on equipment such as camera head, bayonet socket, entrance guard, gather the data of two parts respectively: the method comprises the following steps that firstly, historical driving track information of different personnel in a plurality of time periods is obtained; and secondly, real-time running track information of the target object in a time period.
The information of the driving track in a time period comprises identity information of personnel, at least one passing place (longitude and latitude) in the time period, a passing time and a set of stopping time. Here, a time period means 24 hours a day and night, and only minutes and seconds are recorded without recording the year, month and day through the time.
The historical track analysis module 20 is configured to perform density clustering on the historical driving track information, and obtain a day-night permanent station and a night permanent station of the target object based on a result of the density clustering; obtaining a day-night emergence judgment ratio based on a first historical stay time of the target object at a day-night station and a second historical stay time of the target object at a night-night station respectively;
the data collected by different personnel in the data collection module 10 is modeled and analyzed based on a density clustering method, specifically, each personnel can be modeled respectively through a DBSCAN density clustering algorithm, the collected data are clustered according to similarity in time and space, if the time and the space meet requirements, the current time sequence is clustered into one class, then each clustering center is clustered, and daily residences and nights of different personnel are obtained through screening by a certain rule.
The real-time track analysis module 30 is configured to obtain and obtain a real-time daytime/nighttime ratio based on the real-time traveling track information and a first real-time staying time of the target object at a daily frequent site and a second real-time staying time at a nighttime frequent site, respectively;
the mode of acquiring the daytime and nighttime output real-time ratio is the same as the daytime and nighttime output judgment ratio and is determined based on the stay time of the daily and nighttime residences and the stay time of the nighttime residences.
And the detection module 40 is configured to detect whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission determination ratio and the diurnal and nocturnal emission real-time ratio.
If the real-time daytime and nighttime ratio is smaller than the daytime and nighttime judgment ratio, the target object stays at the night place for a time ratio smaller than the daytime and nighttime judgment ratio obtained from the historical driving track information, and then the real-time driving track abnormality of the target object is detected; conversely, it can be detected that the real-time travel trajectory of the target object is normal.
The track anomaly detection system according to the present invention, wherein the historical track analysis module 20 is specifically configured to:
performing density clustering on the historical driving track information of the target object according to the similarity in time and space to obtain clustering centers, performing cluster clustering on each clustering center to obtain clustering results, and screening the clustering results based on preset rules to obtain day-time frequent stops and night-time frequent stops of the target object.
The track anomaly detection system according to the present invention, wherein the historical track analysis module 20 is specifically configured to:
for a target object i, recording an elapsed time sequence x of collected historical K time periods im Corresponding sequence of places y im And corresponding dwell time sequences z im M =1, \8230;, K; screening sequences with the same data dimension from the K retention time sequences for clustering, and recording a set X of each group of elapsed time sequences ij Corresponding set of sequences of passing sites Y ij And corresponding set of dwell time sequences Z ij J =1, \ 8230, N, total N groups;
for the obtained N groups of sequences X with the same data dimension ij ，Y ij ，Z ij J =1, \8230, N carries out intra-group clustering. And screening by using a certain rule to obtain an integral clustering result, which comprises the following specific steps:
for the j group of the elapsed time series set X ij And corresponding sets of passing location (latitude and longitude) sequences Y ij Carrying out group clustering, wherein j =1, \ 8230, and N;
for jth group X ij ，Y ij The subscripts of the members in the cluster are intersected in pairs to obtain incompatible W j Cluster, specifically: x ij Clustering members of a class with Y ij And traversing the members of each class of the clustering result to obtain the member combination with the most intersected same subscript, namely if the l, h and k data X ijl 、X ijh And X ijk Poly is of one type, and Y ijl 、Y ijh And Y ijk Clustering is successful if the cluster is one type; whether or notThe clustering fails. If the number of elements in the class is less than 2, it means that there is only a single piece of data in the class, and this also indicates a failure. Discarding the data and category that fails clustering. So as to obtain the final clustering result CX of the time sequence and the place sequence j ，CY j J =1, \8230, N; clustering results of retention time series with CZ j With CX j ，CY j Correspondingly, no separate clustering is performed here;
a total of W classes are obtained and,
calculating class centers of the time sequences, the corresponding passing place sequences and the corresponding staying time sequences in the W classes;
performing cluster cohesion clustering on W cluster centers passing through the site sequence respectively to obtain S clustering results, namely the number of the permanent sites,
Location w the number of clusters in the w class;
mapping the clustering result to a passing time sequence and a staying time sequence to obtain a one-to-one correspondence relation between the residence place and the passing time and the staying time;
screening S permanent sites and corresponding residence time and elapsed time: reserving a normal station with the stay time larger than a preset time threshold; calculating the distance between every two reserved ordinary stations, and deleting the ordinary station with shorter staying time if the distance is smaller than a preset distance threshold value, namely if the distance is close to the preset distance threshold value, so as to obtain a pre-selected ordinary station; for example, the preset time threshold may be set to 1h or other time value, and the preset distance threshold may be set to any value within an interval of 10 meters to 500 meters. The step is to further screen the permanent station and select the pre-selected permanent station with reference meaning.
Based on a preset day-night division rule (here, 7-19 is day time, 19-the next day, 7 00 is night time), the day ordinary premises and the night ordinary premises are preselected to be obtained.
The track anomaly detection system according to the present invention, wherein the historical track analysis module 20 is specifically configured to:
counting a first historical stay time of the target object at the day permanent station and a second historical stay time of the target object at the night permanent station respectively based on the preselected permanent station, the corresponding stay time and the elapsed time;
calculating the daytime and nighttime output judgment ratio r of the target object i i = second history stay length/(first history stay length + second history stay length).
The track anomaly detection system according to the present invention, wherein the real-time track analysis module 30 is specifically configured to:
acquiring a real-time permanent station, corresponding residence time and elapsed time based on the real-time running track information, and counting first real-time residence time of the target object in the permanent station in the daytime and second real-time residence time of the permanent station at night;
and calculating the diurnal and nocturnal real-time proportion b = second real-time stay duration/(first real-time stay duration + second real-time stay duration) of the target object i.
The track anomaly detection system according to the present invention, wherein said system further comprises:
the updating module is used for carrying out density clustering on the real-time driving track information of the target object and the W classes, if the real-time driving track information of the target object is successfully clustered into one class, a class center is updated, otherwise, the real-time driving track information of the target object is singly clustered into one class, and the updated class is obtained;
and updating the judging proportion of the current station and the diurnal emissions of the target object based on the updated classification for judging by a diurnal emissions module in the next round. The invention considers the internal relation and consistency between the historical data and the real-time data, is not simple data statistics, but forms a closed-loop system, has a dynamic correction function and improves the judgment accuracy of the person who appears at night in the daytime.
Specifically, a new piece of real-time travel track information can be compared with W classes, the distance is calculated, the distance between the new piece of real-time travel track information and the class center of the W class is minimum, the distance is smaller than a preset threshold value, the new data belongs to the W class, the number of the W class members is +1, and the class center is recalculated; otherwise, the new data becomes a single class, the W +1 th class and the total number of classes is also +1. Namely, the historical running track information is updated and iterated by utilizing the real-time running track information every time, so that better accuracy is realized.
Further, the track anomaly detection system further comprises an anomaly alarm module, configured to:
and when the real-time running track abnormality of the target object is detected, giving an alarm. If the target object is determined to be of the daytime and night-time output type, immediately notifying security or policemen to perform further checking.
The track anomaly detection system according to the present invention, wherein the detection module 40 is specifically configured to:
real-time ratio b as the diurnal and nocturnal emissions of the target object<Ratio r for judging daytime and nighttime i While satisfying | b-r i |/b>And when p is obtained, detecting the real-time running track abnormity of the target object, wherein p represents the set diurnal and nocturnal fluctuation ratio. p is set artificially, and can be set to 10% or 20% due to different scenes with different sensitivity degrees to the abnormality, and the value of p can be adjusted automatically. The floating proportion of the daytime and night is increased, the sensitivity of detection can be adjusted, and a critical condition for judging the abnormity is increased, so that the judgment is more accurate.
Fig. 4 illustrates a physical structure diagram of an electronic device, which may include: a processor (processor) 310, a communication Interface (Communications Interface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may call logic instructions in the memory 330 to perform a method of trace anomaly detection, the method comprising:
s1, acquiring historical driving track information and real-time driving track information of a target object;
s2, performing density clustering on the historical driving track information, and obtaining the day permanent station and the night permanent station of the target object based on the result of the density clustering; obtaining a day-night-time judgment proportion based on a first historical stay time of the target object at the day station and a second historical stay time of the target object at the night station respectively;
s3, acquiring and obtaining a real-time ratio of daytime and nighttime based on the first real-time stay time of the target object at the daily frequent place and the second real-time stay time of the nighttime frequent place respectively based on the real-time running track information;
and S4, detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission judgment ratio and the diurnal and nocturnal emission real-time ratio.
In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method for detecting track anomalies provided by the above methods, the method comprising:
s1, acquiring historical driving track information and real-time driving track information of a target object;
s2, performing density clustering on the historical driving track information, and obtaining the day permanent station and the night permanent station of the target object based on the result of the density clustering; obtaining a daytime and nighttime judgment proportion based on a first historical stay time of the target object at the daily frequent residence and a second historical stay time of the nighttime frequent residence respectively;
s3, acquiring and obtaining a real-time ratio of daytime and nighttime based on the first real-time stay time of the target object at the daily frequent place and the second real-time stay time of the nighttime frequent place respectively based on the real-time running track information;
and S4, detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the daytime and nighttime output judgment proportion and the real-time daytime and nighttime output proportion.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor is implemented to perform the trajectory anomaly detection methods provided above, the method comprising:
s1, acquiring historical driving track information and real-time driving track information of a target object;
s2, performing density clustering on the historical driving track information, and obtaining the day permanent station and the night permanent station of the target object based on the density clustering result; obtaining a daytime and nighttime judgment proportion based on a first historical stay time of the target object at the daily frequent residence and a second historical stay time of the nighttime frequent residence respectively;
s3, acquiring and obtaining a real-time ratio of daytime and nighttime based on the first real-time stay time of the target object at the daily frequent place and the second real-time stay time of the nighttime frequent place respectively based on the real-time running track information;
and S4, detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission judgment ratio and the diurnal and nocturnal emission real-time ratio.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the embodiment. One of ordinary skill in the art can understand and implement the present invention without any inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, and not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; such modifications or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A track anomaly detection method is characterized by comprising the following steps:
acquiring historical driving track information and real-time driving track information of a target object;
performing density clustering on the historical driving track information, and acquiring the day permanent station and the night permanent station of the target object based on the density clustering result; obtaining a daytime and nighttime judgment proportion based on a first historical stay time of the target object at the daily frequent residence and a second historical stay time of the nighttime frequent residence respectively;
based on the real-time running track information, acquiring and based on a first real-time stay time of the target object at a day permanent location and a second real-time stay time of the target object at a night permanent location respectively, and acquiring a daytime and nighttime real-time ratio;
and detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal emission judgment ratio and the diurnal emission real-time ratio.
2. The track abnormality detection method according to claim 1, wherein the density clustering of the historical travel track information and the obtaining of the day-night-day and night-day residences of the target object based on the result of the density clustering include:
performing density clustering on the historical driving track information of the target object according to the similarity in time and space to obtain clustering centers, performing cluster clustering on each clustering center to obtain clustering results, and screening the clustering results based on preset rules to obtain day-time frequent stops and night-time frequent stops of the target object.
3. The track abnormality detection method according to claim 2, wherein the density clustering of the historical travel track information and the obtaining of the day-night-day and night-day residences of the target object based on the result of the density clustering include:
for a target object i, recording an acquired elapsed time sequence x of historical K time periods im Corresponding sequence of transit points y im And corresponding dwell time sequences z im M =1, \ 8230;, K; screening sequences with the same data dimension from K retention time sequences for clustering, and recording a set X of each group of elapsed time sequences ij Corresponding set of sequences of passing sites Y ij And corresponding dwell time series set Z ij J =1, \ 8230, N, total N groups;
for the j group of the elapsed time series set X ij And a corresponding set of sequences of passing sites Y ij Respectively carrying out group clustering, wherein j =1, \ 8230, and N, and the specific steps are as follows:
setting the h-th data X in the time sequence set and the place sequence set ijh And Y ijh Traversing all classes for class centers of a certain class if the 1 st data X in the set ijl And X ijh Poly as one type, Y ijl And Y ijh If the cluster is one type, the cluster is successful, and the class center is updated; if the two groups do not form a class, the two groups become a new class center by themselves; otherwise, abandoning the 1 st data X ijl And Y ijl ；
Circularly traversing the time sequence set and the place sequence set to obtain a final elapsed time sequence and a final elapsed place sequence clustering result CX j ，CY j J =1, \ 8230, N; clustering results of retention time series with CZ j With CX j ，CY j Corresponding; the number of W classes is obtained, and,
calculating class centers of the time sequences, the corresponding passing place sequences and the corresponding staying time sequences in the W classes;
performing cluster cohesion clustering on W cluster centers passing through the site sequence respectively to obtain S clustering results, namely the number of the permanent sites,
Location w the number of clusters in the w class;
mapping the clustering result to a passing time sequence and a staying time sequence to obtain a one-to-one correspondence relationship between the ordinary station and the passing time and the staying time;
screening S permanent premises, corresponding residence time and elapsed time: reserving a permanent station with the residence time larger than a preset time threshold; calculating the distance between every two reserved normal stations, and if the distance is smaller than a preset distance threshold, deleting the normal station with shorter retention time to obtain a pre-selected normal station;
and pre-selecting the ordinary station based on a preset day and night division rule to obtain the day ordinary station and the night ordinary station.
4. The track abnormality detection method according to claim 3, wherein obtaining a diurnal emission determination ratio based on a first historical stay time of the target object at a daytime frequent location and a second historical stay time of the target object at a nighttime frequent location, respectively, includes:
counting a first historical stay time of the target object at the day permanent station and a second historical stay time of the target object at the night permanent station respectively based on the preselected permanent station, the corresponding stay time and the elapsed time;
calculating a daytime/nighttime occurrence judgment ratio r of the target object i i = second historical stay duration/(first historical stay duration + second historical stay duration).
5. The track abnormality detection method according to claim 4, wherein obtaining a daytime and nighttime emergence real-time ratio based on obtaining and respectively based on a first real-time stay time of the target object at a daytime regular premises and a second real-time stay time of the target object at a nighttime regular premises based on the real-time travel track information includes:
acquiring a real-time permanent station, corresponding residence time and elapsed time based on the real-time running track information, and counting first real-time residence time of the target object in the permanent station in the daytime and second real-time residence time of the permanent station at night;
and calculating the diurnal and nocturnal real-time proportion b = second real-time stay duration/(first real-time stay duration + second real-time stay duration) of the target object i.
6. The trajectory anomaly detection method according to claim 3, further comprising:
performing density clustering on the real-time driving track information of the target object and the W classes, if the real-time driving track information of the target object is successfully clustered into one class, updating class centers, otherwise, independently clustering into one class to obtain updated classes;
updating the judgment ratio of the current station and the daytime and nighttime of the target object based on the updated classification; and/or, the track abnormity detection method further comprises the following steps:
and when the real-time running track abnormality of the target object is detected, giving an alarm.
7. The track abnormality detection method according to claim 5, wherein the detecting whether the real-time travel track of the target object is abnormal based on a result of comparing the diurnal emission determination ratio and the diurnal emission real-time ratio includes:
real-time ratio b as the diurnal and nocturnal emissions of the target object<Ratio r for judging daytime and nighttime i When, satisfy | b-r i |/b>And p, detecting the real-time running track abnormality of the target object, wherein p represents the set diurnal and nocturnal fluctuation ratio.
8. A trajectory abnormality detection device characterized by comprising:
the data acquisition module is used for acquiring historical driving track information and real-time driving track information of the target object;
the historical track analysis module is used for carrying out density clustering on the historical driving track information and obtaining the day-time frequent station and the night-time frequent station of the target object based on the result of the density clustering; obtaining a daytime and nighttime judgment proportion based on a first historical stay time of the target object at the daily frequent residence and a second historical stay time of the nighttime frequent residence respectively;
the real-time track analysis module is used for acquiring and obtaining a real-time ratio of daytime and nighttime emergence based on a first real-time stay time of the target object at a daily frequent site and a second real-time stay time of the target object at a nighttime frequent site respectively based on the real-time running track information;
and the detection module is used for detecting whether the real-time running track of the target object is abnormal or not based on the comparison result of the diurnal and nocturnal emission judgment proportion and the diurnal and nocturnal emission real-time proportion.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the program.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the trajectory anomaly detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110975911.6A CN115719546B (en) | 2021-08-24 | 2021-08-24 | Track abnormality detection method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110975911.6A CN115719546B (en) | 2021-08-24 | 2021-08-24 | Track abnormality detection method and device, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115719546A true CN115719546A (en) | 2023-02-28 |
| CN115719546B CN115719546B (en) | 2024-10-25 |
Family
ID=85253469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110975911.6A Active CN115719546B (en) | 2021-08-24 | 2021-08-24 | Track abnormality detection method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115719546B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018176162A1 (en) * | 2017-03-31 | 2018-10-04 | Avigilon Corporation | Unusual motion detection method and system |
| CN109446186A (en) * | 2018-09-27 | 2019-03-08 | 江苏大学 | A kind of social relationships judgment method based on motion track |
| CN110751164A (en) * | 2019-03-01 | 2020-02-04 | 西安电子科技大学 | Old man travel abnormity detection method based on location service |
| CN111739325A (en) * | 2020-06-15 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Method, apparatus, device and medium for determining abnormal parking probability and navigation control |
| WO2020210504A1 (en) * | 2019-04-09 | 2020-10-15 | Avigilon Corporation | Anomaly detection method, system and computer readable medium |
| CN111882873A (en) * | 2020-07-22 | 2020-11-03 | 平安国际智慧城市科技股份有限公司 | Track anomaly detection method, device, equipment and medium |
| WO2021077761A1 (en) * | 2019-10-23 | 2021-04-29 | 江苏智通交通科技有限公司 | Intersection abnormal vehicle trajectory identification and analysis method based on hierarchical clustering |
-
2021
- 2021-08-24 CN CN202110975911.6A patent/CN115719546B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018176162A1 (en) * | 2017-03-31 | 2018-10-04 | Avigilon Corporation | Unusual motion detection method and system |
| CN109446186A (en) * | 2018-09-27 | 2019-03-08 | 江苏大学 | A kind of social relationships judgment method based on motion track |
| CN110751164A (en) * | 2019-03-01 | 2020-02-04 | 西安电子科技大学 | Old man travel abnormity detection method based on location service |
| WO2020210504A1 (en) * | 2019-04-09 | 2020-10-15 | Avigilon Corporation | Anomaly detection method, system and computer readable medium |
| WO2021077761A1 (en) * | 2019-10-23 | 2021-04-29 | 江苏智通交通科技有限公司 | Intersection abnormal vehicle trajectory identification and analysis method based on hierarchical clustering |
| CN111739325A (en) * | 2020-06-15 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Method, apparatus, device and medium for determining abnormal parking probability and navigation control |
| CN111882873A (en) * | 2020-07-22 | 2020-11-03 | 平安国际智慧城市科技股份有限公司 | Track anomaly detection method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115719546B (en) | 2024-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lin et al. | Using machine learning to assist crime prevention | |
| Donovan et al. | Using coarse GPS data to quantify city-scale transportation system resilience to extreme events | |
| CN107798876B (en) | Road traffic abnormal jam judging method based on event | |
| CN106022296B (en) | A kind of fake-licensed car detection method of the probability polymerization based on vehicle hot spot region | |
| CN107832364B (en) | Method and device for locking target object based on spatio-temporal data | |
| US20080154829A1 (en) | Temporal-influenced geospatial modeling system and method | |
| Han et al. | Traffic incident detection: A trajectory-based approach | |
| CN105719489A (en) | Fake-licensed vehicle detection method based on bayonet vehicle flow direction probability | |
| CN106558220B (en) | A kind of statistical method and device of traffic volume of highway | |
| CN109377757A (en) | The vehicle driving track extraction method of license plate identification data based on the rough error containing multi-source | |
| CN111445369A (en) | Urban large-scale gathering activity intelligence early warning method and device based on L BS big data | |
| Qi et al. | Vehicle trajectory reconstruction on urban traffic network using automatic license plate recognition data | |
| CN108257385B (en) | Method for discriminating abnormal events based on public transportation | |
| CN111461460B (en) | Police service inspection route planning method and personnel inspection method | |
| CN114706988B (en) | Knowledge graph-based monitoring and early warning method and device | |
| CN110503032A (en) | Individual important place detection method based on monitoring camera track data | |
| CN113470369B (en) | Method and system for judging true number plate of fake-licensed vehicle based on multi-dimensional information | |
| CN112052405B (en) | Passenger searching area recommendation method based on driver experience | |
| CN104778355B (en) | The abnormal track-detecting method of traffic system is distributed based on wide area | |
| CN112434074A (en) | Natural conservation place human activity supervision system based on mobile phone positioning information | |
| CN113516850B (en) | Pipeline traffic flow data acquisition method based on space syntactic analysis | |
| CN115719546A (en) | Track abnormity detection method and device, electronic equipment and readable storage medium | |
| CN114419894B (en) | Method and system for setting and monitoring parking positions in road | |
| CN112309129B (en) | OD matrix forming method for vehicle travel characteristic classification based on checkpoint license plate recognition | |
| CN111369394B (en) | Scenic spot passenger flow volume statistical evaluation system and method based on big 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 |