CN108874979B - Suspicious vehicle discovery method based on space-time range constraint - Google Patents
Suspicious vehicle discovery method based on space-time range constraint Download PDFInfo
- Publication number
- CN108874979B CN108874979B CN201810589430.XA CN201810589430A CN108874979B CN 108874979 B CN108874979 B CN 108874979B CN 201810589430 A CN201810589430 A CN 201810589430A CN 108874979 B CN108874979 B CN 108874979B
- Authority
- CN
- China
- Prior art keywords
- space
- time
- time constraint
- constraint condition
- constraint
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method based on space and timeA suspicious vehicle discovery algorithm with range constraint relates to the application fields of intelligent transportation, a time-space database, artificial intelligence and the like, in particular to a suspicious vehicle discovery method independent of visual features. The method comprises the following steps: 1) submissionkA spatiotemporal constraint;kis a natural number which is a natural number,kthe value range of (A) is not less than 1k≤10；kA space-time constraint namelykA relatively well-defined spatial and temporal extent; 2) finding matches in turn in a trajectory databasekAnd (5) vehicles under space-time constraint conditions are selected, and a final query result is obtained. The time range and the space range of the movement of the suspicious vehicle are represented as space-time constraint conditions by utilizing the space-time characteristics of the movement of the suspicious vehicle, and the suspicious vehicle is inquired based on the space-time constraint, so that the multi-space-time constraint conditions and a distributed processing mode can be supported to improve the efficiency of discovering the suspicious vehicle.
Description
Technical Field
The invention discloses a suspicious vehicle discovery method based on space-time range constraint, relates to the application fields of intelligent transportation, a space-time database, artificial intelligence and the like, and particularly relates to a suspicious vehicle discovery method independent of visual features.
Background
Suspicious vehicle identification is an important issue in intelligent transportation and aims to help public security agencies track illegal and illegal vehicles. Real life suspicious vehicles generally include two categories: the method comprises the following steps that firstly, a vehicle with typical external characteristics, such as license plate number, brand model number, color and the like, of a suspicious vehicle is recorded by a monitoring system such as a witness or a certain place; the second is a vehicle with unknown external features. The current suspicious vehicle discovery algorithm mainly adopts suspicious vehicles with typical external features, and generally adopts an image processing-based discovery algorithm, which mainly comprises two types of determining the license plate number of the suspicious vehicle and determining the driving route of the suspicious vehicle. The license plate number of the suspicious vehicle is determined, namely, for the characteristics such as the brand, the color and the model of the vehicle provided by the witness, all the vehicle information meeting the characteristics in the accident occurrence time interval near the accident scene is inquired through the monitoring system, and the vehicle information is checked one by one until the license plate number of the suspicious vehicle is determined, so that the identity information of the criminal suspect is determined. The method comprises the steps of determining the driving route of a suspicious vehicle, wherein the driving route refers to a given license plate number and a time interval, and inquiring the motion track of the given vehicle through a monitoring system, namely information such as which road sections the vehicle passes through and how long the vehicle stays in the given time interval. The identification of the suspicious vehicle license plate number and the query of the driving route can be combined, and the efficiency and the precision of license plate number identification can be improved according to the driving route.
The license plate number identification is mainly carried out by an image processing method, the characteristics of each vehicle are extracted from a video shot by a monitoring system and are matched with the vehicle characteristics provided by witnesses on site, and all vehicles meeting the characteristics near an accident site and in a time interval before and after the accident are found. Current traffic monitoring systems have been able to quickly and efficiently identify the license plate numbers of passing vehicles so that the license plate numbers of all vehicles meeting the characteristics can be determined. And then according to the license plate number, combining the driving route of each vehicle to further narrow the range of the suspicious vehicle.
For suspicious vehicles without typical external features, a query mode is usually adopted to search vehicles meeting conditions in a traffic system according to the time and the position presumed by police. The existing suspicious vehicle query method adopts a data storage and query method of a relational database, and comprises two types: the first is the query of determining time and determining position, namely, all vehicles with the time and the position equal to given conditions are searched in a relational database; the second is the query of time range and determined location, i.e. all vehicles passing a specific location within a given time range are queried. Because of the symbolic position usually adopted by the position information in such methods, only the query for determining the position is supported, and the method is hardly suitable for the query of the space range. However, in practical applications, the query of the suspicious vehicle is usually based on the time range and the space range, that is, the vehicle passing through a certain area in a certain time period is the suspicious vehicle. Moreover, to increase the query accuracy, it is usually necessary to set multiple sets of constraints of temporal and spatial extent according to case clues.
The discovery of suspicious vehicles is helpful for police to catch criminal suspects, and the suspicious vehicles usually adopt various modes to hide own traffic paths in order to interfere the catching work of the police, so that the police is difficult to accurately locate the position of the suspicious vehicles at a certain moment, a large amount of manpower and material resources are required to be invested to screen monitoring videos of a traffic system in a manual mode, and the time and space range is gradually enlarged to be checked. Therefore, it is necessary to provide a suspicious vehicle discovery method independent of visual features, which can automatically and efficiently discover suspicious vehicles and provide decision support for police to fight against criminal behaviors.
The existing suspicious vehicle discovery method mainly discovers a suspicious vehicle with certain external characteristics through an image identification technology, or discovers a vehicle passing through a specific place through a relational database query technology, so that the problem of discovering the suspicious vehicle without definite external characteristics under the condition of uncertain time and place cannot be efficiently solved.
Disclosure of Invention
The invention aims to provide a suspicious vehicle discovering method based on space-time range constraint aiming at the defects, which can discover suspicious vehicles without clear external features under the condition of uncertain time and places and solve the query problem that the suspicious vehicles without typical external features and uncertain position information cannot be efficiently discovered by an image recognition-based method and a relational database-based method.
The invention is realized by adopting the following technical scheme:
a suspicious vehicle discovery method based on space-time range constraint comprises the following steps,
1) submissionkIndividual space-time constraint condition
kIs a natural number which is a natural number,kthe value range of (A) is not less than 1k≤10；
kA space-time constraint namelykSet relatively well-defined spatial and temporal ranges;
2) finding matches in turn in a trajectory databasekA vehicle subject to a space-time constraint,and obtain the final query result.
The specific process of the step 2) comprises the following steps:
as described belowiIs a natural number, and is not more than 1i<k；
2-1) if toiIf the query result of each space-time constraint condition is empty, the traffic management system does not have the vehicle which meets all the space-time constraint conditions, if the query result is empty, the step 2-3 is carried out; if it is toiIf the query result of the space-time constraint condition is not empty, entering the step 2-2);
2-2) updating the source database and the current space-time constraint condition;
if it is firstiIf the space-time constraint condition is the last space-time constraint condition, the current query result is the final query result, and the step 2-3 is carried out; otherwise, it will be directed toiThe query result of the space-time constraint condition is used as the firsti+1 data sources queried by space-time constraint conditions, entering step 2-1) to carry out the second stepi+1 query of spatio-temporal constraints;
2-3) the user confirms whether the space-time constraint condition needs to be adjusted, if not, the query result is output, and the algorithm execution process is ended; if the adjustment is needed, the user determines that the current result is more or less, and if the current result is less, the step 2-4) is carried out; if the current result is more than a few, entering the step 2-5);
2-4) deleting the time and space limitation in the least important space-time constraint condition or the space-time constraint condition with the strongest relaxed constraint in the currently calculated space-time constraint condition by the user according to the actual situation; or selecting a space-time constraint condition with the minimum ratio of the query result to the input source data amount in the currently calculated space-time constraint conditions by an algorithm, and relaxing the time and space limitations of the space-time constraint condition; if it is deletedjTaking the source data of the jth space-time constraint condition as the second timej+1 space-time constraint source data, proceeding to step 2-1) toj+1 query of spatio-temporal constraints; if relaxed, it isjA space-time constraint is maintainedjThe source data for the +1 spatio-temporal constraints are unchanged,step 2-1) is carried outjQuerying a space-time constraint condition; the above-mentionedjIs a natural number, and is not more than 1j < k；
2-5) calculated by the user at present according to the actual conditionskStrengthening time and space limitation in the weakest time-space constraint condition or adding new time-space constraint condition in the time-space constraint condition; or selected by an algorithm to be completed at the current calculationkThe space-time constraint condition with the maximum ratio of the query result to the input source data amount in the space-time constraint conditions strengthens the time and space limitations of the space-time constraint conditions, takes the current query result as the source data, and takes the new space-time constraint condition or the space-time constraint condition after strengthening the time and space limitations as the first time of the systemk+1 space-time constraint condition, proceeding to step 2-1) tok+1 queries of spatiotemporal constraints.
The invention has the advantages that: the invention is mainly used for efficiently discovering suspicious vehicles without typical external features under the condition of uncertain time and place. When the suspicious vehicle has no clear external features, the existing image identification-based method is not available; when the position information is uncertain, the query method based on the relational database cannot efficiently find suspicious vehicles. The time range and the space range of the movement of the suspicious vehicle are represented as space-time constraint conditions by utilizing the space-time characteristics of the movement of the suspicious vehicle, and the suspicious vehicle is inquired based on the space-time constraint conditions, so that a plurality of space-time constraint conditions and a distributed processing mode can be supported to improve the efficiency of discovering the suspicious vehicle.
Drawings
The invention will be further explained with reference to the drawings, in which:
FIG. 1 is a logic flow diagram of a suspicious vehicle discovery method of the present invention;
FIG. 2 is a schematic diagram of an embodiment of the constraint space partition of the suspicious vehicle discovery method of the present invention;
FIG. 3 is a flow chart of a distributed method based on data partitioning employed by the present invention;
FIG. 4 is a flow chart of a distributed method for data-based backup as employed by the present invention;
fig. 5 is a flowchart of an implementation of the cloud platform-based distributed algorithm suspicious vehicle discovery method.
Detailed Description
The invention is described below by way of specific embodiments with reference to the accompanying drawings.
Suppose that each vehicleobjDuring movementtTime passing positionpThe timing system will generate a record (obj,t, p) Each space-time constraint is in the form ofC=(T 1,T 2,pos 1,pos 2) WhereinT 1AndT 2represents time, andT 1≤T 2，pos 1=(x 1,y 1) Andpos 2= (x 2,y 2)，x 1andx 2indicating a locationpos 1Andpos 2the longitude of (a) is determined,y 1andy 2indicating a locationpos 1Andpos 2latitude, in positionpos 1Andpos 2the rectangular area formed by the diagonal vertices is denoted asArea(C). If objectobjIn thattThe moment being in the areaArea(C) In, i.e.
Then call the objectobjSatisfy the space-time constraint conditionCIs marked as
。
For a suspect, the probable escape time and route can be deduced, and the escape route is selectedkRelatively well-defined spatial and temporal extent, i.e. givenkA space-time constraint, each space-time constraintC i The format of (A) is:C i =(T i1,T i2,pos i1,pos i2)，1≤i≤kis shown bySuspect vehicle is in time intervalT i1,T i2]Will pass by the positionpos i1Andpos i2the rectangular area formed by the diagonal vertices is denoted asArea(C i ). Each vehicleobjThe path trajectory of (1) is a set of a series of sampling points if the vehicle is int 1Time passing positionp 1In at 2Time passing positionp 2… int m Time passing positionp m Then its path isPath(obj)={(obj,t 1,p 1), (obj,t 2,p 2),…,(obj,t m ,p m )}. If a certain vehicleobjCan satisfy given total possible pathskA space-time constraint, i.e.
Then the vehicle is consideredobjIs a suspect vehicle.
The process of finding out suspicious vehicles is to search for the conforming conditions in the traffic management systemkAll vehicles with space-time constraint condition (1 ≦k≦ 10), the logic flow of the method is shown in FIG. 1. Finding matches in turn in a trajectory databasekVehicles with space-time constraints, in whichi（1≤i<k) The query result of the space-time constraint condition is taken asi+1 data sources for spatio-temporal constraint queries. If the query result is null or the user considers that the current query result is more or less, the setting of the current space-time constraint condition is unreasonable, and the user is required to adjust the given space-time constraint condition. There are two cases of adjustment of the spatiotemporal constraints: firstly, the space-time constraint condition is too strong, so that the query result is empty or the result is less, and at the moment, some space-time constraint conditions need to be deleted or the time and space limitations in some space-time constraint conditions need to be relaxed; secondly, the number of query results is large due to the weak space-time constraint condition, and more space-time constraint conditions or some space-time constraint bars need to be added at the momentThe time and space limitations of the member are enhanced. The specific scheme of the adjustment of the space-time constraint condition is selected by a user according to the practical application.
kA space-time constraint namelykA relatively well-defined spatial and temporal extent.
SubmissionkThe specific steps of the space-time constraint condition comprise:
1-1) for each space-time constraintC i =(T i1,T i2,pos i1,pos i2) Building a rectangular spatial regionArea(C i ) The rectangular space region is used asC i Of which two diagonal vertices arepos i1=(x i1,y i1) Andpos i2=(x i2,y i2) WhereiniIs the serial number of the space-time constraint,T i1、T i2、pos i1、pos i2respectively represent the time interval of the suspicious vehicleT i1,T i2]Will pass throughArea(C i ) Zone, 1 ≦i≤k；
1-2) will constrain spaceArea(C i ) Dividing the GeoHash into a plurality of subspaces according to different granularities of the GeoHash, and specifically dividing the GeoHash into:
1-2-1) utilizing a multi-granularity index structure of GeoHash to find out a subspace covering a constraint space from the coarse granularity;
1-2-2) finding out subspaces covering the constraint space on finer granularity;
1-2-3) searching successively on more detailed level until the constraint space is completely coveredArea(C i )。
FIG. 2 is a schematic diagram of an embodiment of the constraint space partition adopted in the present invention, in which the constraint space is dividedArea(C i ) Can be divided intoarea 1、area 2Andarea 3、area 4~area 66 subspaces of three different granularities.
Finding matches in turn in a trajectory databasekA vehicle subject to space-time constraints, namely:
2-1) in the track databaseTDB i In the time intervalT i1,T i2]Internally passing through each subspacearea j All vehiclesLobj(area j )；
2-2) combining all results generated in the step 2-1) to obtain a product meeting space-time constraint conditionsC i All suspicious vehicles
；
2-3) if all the space-time constraint conditions are processed, outputting a suspicious vehicle setGobj(C i ) (ii) a Otherwise, updating the track databaseTDB i+1=Gobj(C i ) And jumping to the step 2-1).
The suspicious vehicle discovery method can improve the efficiency in a parallel processing mode under a distributed environment.
In the following, for example, a distributed method based on data partitioning and a distributed method based on data backup are mainly used.
1) Distributed method based on data partitioning:
if the original track dataTDBStored in a plurality of nodes in a horizontally divided mannern 1,n 2,…,n q On, i.e.
And is
WhereinqIndicating the number of nodes, each noden i Producing the final result according to the algorithmGobj(n i ). The final output suspicious vehicle set is
WhereinResultThe set of suspicious vehicles that satisfy all the spatiotemporal constraints is shown in FIG. 3.
2) The distributed method based on data backup comprises the following steps:
if the original track dataTDBAt a plurality of nodesn 1,n 2,…,n q Store the original trace data onTDBA complete copy of, i.e. each noden i All store data setTDBWherein 1 is less than or equal toi≤q. At this time, ifq≥kThen selectkA node, each noden i Up processing 1 space-time constraintC i To obtain the resultResult(C i ) Then, intersection of these results can be used to obtain suspicious vehicles satisfying all space-time constraints, i.e., vehicles that are suspicious
As shown in fig. 4. If it isq<kThen each noden i Upper treatment of
Suspicious vehicle query under space-time constraint conditions to obtain resultsResult(n i ) Then, intersection of these results can be used to obtain suspicious vehicles satisfying all space-time constraints, i.e., vehicles that are suspicious
。
The distributed method based on data division is based on a centralized method, and the calculated amount is less; the distributed method based on data backup improves algorithm efficiency by using redundant nodes and can return results in a short time. In a cloud environment, because data is generally divided horizontally, and each data block stores a certain amount of backups, a distributed method based on combination of data division and data backup can be adopted, a distributed method based on data division is adopted among data blocks, and a distributed method based on data backup is adopted on each data backup in the data blocks.
The algorithm of the invention requires that the absolute position of the vehicle be represented by latitude and longitude rather than symbol-based relative position during the acquisition of the vehicle position information. In a traffic system, due to the huge data volume, a cloud platform is generally adopted to store and analyze vehicle position information. The centralized algorithm is deployed on each data node, and the distributed strategy is deployed on a master. When a user submits a space-time constraint condition to inquire suspicious vehicles, the master firstly judges how to utilize the characteristics of the cloud platform to carry out distributed deployment according to metadata, carries out inquiry of a centralized algorithm on each node to generate a local result, and then merges the local results to generate all final suspicious vehicles.
Assume that the existing track database TDB is divided horizontally into two data blocks TDB1 and TDB2, each storing 3 copies, three copies of TDB1 being stored on the N1, N2 and N3 nodes, respectively, and a copy of TDB2 data block being stored on the three nodes N4, N5 and N6. In the distributed processing, a data copy-based distributed method is adopted on the data blocks TDB1 and TDB2, respectively, and the suspicious vehicle discovery result on the TDB is the union of the results on the TDB1 and TDB2, as shown in fig. 5. In solving the three copies of TDB1 for local suspicious vehicles satisfying space-time constraints C1, C2 and C3, respectively, results R11, R21 and R31 are generated, and then suspicious vehicles for data block TDB1 are generated
(ii) a Likewise, a suspect vehicle for data block TDB1 may be obtained
. It is thus possible to obtain a total data set TDB that satisfies the spatio-temporal constraints C1, C2 and C3All suspicious vehicles
。
Through the above embodiments and analysis, compared with the conventional suspicious vehicle discovery method, the method of the invention comprises the following steps:
1. the suspicious vehicle can be found without depending on visual features, namely, without providing external features such as a license plate, a color, a model and the like of the suspicious vehicle;
2. suspicious vehicle discovery in uncertain time and space is supported, namely suspicious vehicle searching can be carried out only by setting a time range and a position range;
3. and distributed deployment is supported to improve efficiency, and the distributed deployment based on data partitioning and the distributed deployment based on data backup are supported.
Claims (3)
1. A suspicious vehicle discovery method based on space-time range constraint is characterized by comprising the following steps:
1) submissionkA spatiotemporal constraint;
kis a natural number which is a natural number,kthe value range of (A) is not less than 1k≤10；
kA space-time constraint namelykA relatively well-defined spatial and temporal extent;
2) finding matches in turn in a trajectory databasekObtaining a final query result by vehicles with space-time constraint conditions;
step 2) searching coincidence in the track database in sequencekThe concrete steps of the vehicle of the space-time constraint condition include:
2-1) if toiIf the query result of each space-time constraint condition is empty, the traffic management system does not have the vehicle which meets all the space-time constraint conditions, if the query result is empty, the step 2-3 is carried out; if it is toiIf the query result of the space-time constraint condition is not empty, entering the step 2-2);
2-2) updating the source database and the current space-time constraint condition;
if it is firstiIndividual space-time constraint conditionIf the query result is the last space-time constraint condition, the current query result is the final query result, and the step 2-3) is carried out; otherwise, it will be directed toiThe query result of the space-time constraint condition is used as the firsti+1 data sources queried by space-time constraint conditions, entering step 2-1) to carry out the second stepi+1 query of spatio-temporal constraints;
2-3) the user confirms whether the space-time constraint condition needs to be adjusted, if not, the query result is output, and the algorithm execution process is ended; if the adjustment is needed, the user determines that the current result is more or less, and if the current result is less, the step 2-4) is carried out; if the current result is more than a few, entering the step 2-5);
2-4) deleting the time and space limitation in the least important space-time constraint condition or the space-time constraint condition with the strongest relaxed constraint in the currently calculated space-time constraint condition by the user according to the actual situation; or selecting a space-time constraint condition with the minimum ratio of the query result to the input source data amount in the currently calculated space-time constraint conditions by an algorithm, and relaxing the time and space limitations of the space-time constraint condition; if it is deletedjTaking the source data of the jth space-time constraint condition as the second timej+1 space-time constraint source data, proceeding to step 2-1) toj+1 query of spatio-temporal constraints; if relaxed, it isjA space-time constraint is maintainedjThe source data of +1 space-time constraint conditions are not changed, and the step 2-1) is carried outjQuerying a space-time constraint condition; the above-mentionedjIs a natural number, and is not more than 1j < k；
2-5) calculated by the user at present according to the actual conditionskStrengthening time and space limitation in the weakest time-space constraint condition or adding new time-space constraint condition in the time-space constraint condition; or selected by an algorithm to be completed at the current calculationkThe space-time constraint condition with the maximum ratio of the query result to the input source data amount in the space-time constraint conditions strengthens the time and space limitations of the space-time constraint conditions, takes the current query result as the source data, and takes the new space-time constraint condition or the space-time constraint condition after strengthening the time and space limitations as the first time of the systemk+1 space-time periodUnder the beam condition, the step 2-1) is carried outk+1 query of spatio-temporal constraints;
saidiIs a natural number, and is not more than 1i<k。
2. The method for suspicious vehicle discovery based on spatiotemporal constraints according to claim 1 wherein step 1) submitskThe specific steps of the space-time constraint condition comprise:
1-1) for each space-time constraintC i =(T i1,T i2,pos i1,pos i2) Building a rectangular spatial regionArea(C i ) The rectangular space region is used asC i Of which two diagonal vertices arepos i1=(x i1,y i1) Andpos i2=(x i2,y i2) WhereiniIs the serial number of the space-time constraint,T i1、T i2、pos i1、pos i2respectively represent the time interval of the suspicious vehicleT i1,T i2]Will pass throughArea(C i ) Zone, 1 ≦i≤k；
1-2) will constrain spaceArea(C i ) Dividing the GeoHash into a plurality of subspaces according to different granularities of the GeoHash, and specifically dividing the GeoHash into:
1-2-1) utilizing a multi-granularity index structure of GeoHash to find out a subspace covering a constraint space from the coarse granularity;
1-2-2) finding out subspaces covering the constraint space on finer granularity;
1-2-3) searching successively on more detailed level until the constraint space is completely coveredArea(C i )。
3. The space-time based range of claim 2A method for finding out suspicious vehicles with surrounding constraint is characterized by sequentially searching the track database for the suspicious vehicles meeting the constraintkA vehicle subject to space-time constraints, namely:
2-1) in the track databaseTDB i In the time intervalT i1,T i2]Internally passing through each subspacearea j All vehiclesLobj(area j )；
2-2) combining all results generated in the step 2-1) to obtain a product meeting space-time constraint conditionsC i All suspicious vehicles
；
2-3) if all the space-time constraint conditions are processed, outputting a suspicious vehicle setGobj(C i ) (ii) a Otherwise, updating the track databaseTDB i+1=Gobj(C i ) And jumping to the step 2-1).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810589430.XA CN108874979B (en) | 2018-06-08 | 2018-06-08 | Suspicious vehicle discovery method based on space-time range constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810589430.XA CN108874979B (en) | 2018-06-08 | 2018-06-08 | Suspicious vehicle discovery method based on space-time range constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108874979A CN108874979A (en) | 2018-11-23 |
| CN108874979B true CN108874979B (en) | 2021-04-30 |
Family
ID=64337573
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810589430.XA Active CN108874979B (en) | 2018-06-08 | 2018-06-08 | Suspicious vehicle discovery method based on space-time range constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108874979B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083599B (en) * | 2019-03-28 | 2023-06-23 | 华东师范大学 | Vehicle track data indexing method based on space-time interpolation |
| CN110334232A (en) * | 2019-06-28 | 2019-10-15 | 深圳市商汤科技有限公司 | Records application method and device, storage medium |
| CN110888884B (en) * | 2019-11-26 | 2022-08-23 | 杭州三汇数字信息技术有限公司 | Vehicle code fitting method and system based on geohash matching |
| CN113742606A (en) * | 2020-05-29 | 2021-12-03 | 京东城市(北京)数字科技有限公司 | Object identification method and device, electronic equipment and readable storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007058608A (en) * | 2005-08-24 | 2007-03-08 | Ntt Communications Kk | Parking lot guiding device |
| CN103294775B (en) * | 2013-05-10 | 2016-12-28 | 苏州祥益网络科技有限公司 | Police service cloud image recognition vehicle administrating system based on geographic space-time constraint |
| US10713686B2 (en) * | 2014-03-22 | 2020-07-14 | Retailmenot, Inc. | Peer-to-peer geotargeting content with ad-hoc mesh networks |
| CN104581803B (en) * | 2015-02-03 | 2018-11-13 | 杨奎武 | A kind of city vehicle-mounted net network mobility model of space-time restriction |
| CN108021561A (en) * | 2016-10-28 | 2018-05-11 | 沈阳建筑大学 | A kind of abnormal mobile object detection method based on track data stream |
| CN107730889B (en) * | 2017-10-30 | 2021-01-05 | 昆明理工大学 | Target vehicle retrieval method based on traffic video |
-
2018
- 2018-06-08 CN CN201810589430.XA patent/CN108874979B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108874979A (en) | 2018-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108874979B (en) | Suspicious vehicle discovery method based on space-time range constraint | |
| Zhang et al. | A novel generative adversarial network for estimation of trip travel time distribution with trajectory data | |
| CN107180530B (en) | A kind of road network trend prediction method based on depth space-time convolution loop network | |
| JP6849898B2 (en) | A method and an apparatus for generating a composite image in which it is impossible to determine whether or not to synthesize a synthetic image by using a generational neural network and a hostile generation network including a discrimination neural network. | |
| CN108197326B (en) | Vehicle retrieval method and device, electronic equipment and storage medium | |
| WO2015098442A1 (en) | Video search system and video search method | |
| Zhang et al. | City brain: practice of large‐scale artificial intelligence in the real world | |
| CN113591380B (en) | Traffic flow prediction method, medium and equipment based on graph Gaussian process | |
| CN105138525B (en) | Traffic video processing unit and method and retrieval device and method | |
| CN103294775A (en) | Police service cloud image recognition vehicle management and control system based on geographic space-time constraint | |
| CN112215427A (en) | Vehicle driving track reconstruction method and system under condition of bayonet data loss | |
| CN112579921B (en) | Track indexing and query method and system based on inverted sorting index and prefix tree | |
| Petrou et al. | Online long-term trajectory prediction based on mined route patterns | |
| TW201426637A (en) | Unsupervised adaptation method and image automatic classification method applying the same | |
| CN114998993A (en) | Combined pedestrian target detection and tracking combined method in automatic driving scene | |
| Choi et al. | Activity recognition in wide aerial video surveillance using entity relationship models | |
| CN113807457A (en) | Method, device and equipment for determining road network characterization information and storage medium | |
| CN114265952B (en) | Target retrieval method and device | |
| CN109492065B (en) | Extraction method of indoor semantic map space-time relationship | |
| Zhu et al. | Instant discovery of moment companion vehicles from big streaming traffic data | |
| KR102426594B1 (en) | System and method for estimating the location of object in crowdsourcing environment | |
| CN109800685A (en) | The determination method and device of object in a kind of video | |
| CN113609317A (en) | Image library construction method and device and electronic equipment | |
| CN109815369B (en) | Filing method and device | |
| Fofana et al. | Optimal Flame Detection of Fires in Videos Based on Deep Learning and the Use of Various Optimizers |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230105 Address after: 221000 7th floor, Building 8, Xuzhou Software Park, No. 6, Software Park Road, Quanshan District, Xuzhou City, Jiangsu Province Patentee after: Jiangsu Tianze Zhilian Information Technology Co.,Ltd. Address before: No. 80, yunlongshan Road, Jianye District, Nanjing City, Jiangsu Province, 210019 Patentee before: TIANZE INFORMATION INDUSTRY Corp. |
|
| TR01 | Transfer of patent right |