CN109800279A - A kind of algorithm based on space-time trajectory Fast Collision - Google Patents
A kind of algorithm based on space-time trajectory Fast Collision Download PDFInfo
- Publication number
- CN109800279A CN109800279A CN201910024972.7A CN201910024972A CN109800279A CN 109800279 A CN109800279 A CN 109800279A CN 201910024972 A CN201910024972 A CN 201910024972A CN 109800279 A CN109800279 A CN 109800279A
- Authority
- CN
- China
- Prior art keywords
- data
- time
- collision
- record
- list
- 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.)
- Withdrawn
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of algorithms based on space-time trajectory Fast Collision, are related to algorithm field.This algorithm is: 1. data prediction (10), A, initial data storage (11)；B, data temporally ascending sort (12)；C, input time and apart from variable (13)；D, data grouping (14) are carried out according to input time；E, packet data (15) are obtained；2. multi-trace collides (20), a, traversal raw data set (21)1；B, packet data (22) are obtained；C, judge time offset (23)；D, judge ranging offset amount (24)；E, collision result (25) is saved；F, analysis collision result (26).The present invention has following advantages and good effect: 1. originality supports the space-time trajectory collision of big data quantity to be maintained at certain computational complexity, reduces operation time, promotes user experience；2. algorithm has scalability.
Description
Technical field
The present invention relates to algorithm field more particularly to a kind of algorithms based on space-time trajectory Fast Collision.
Background technique
As information-based is popularized, the action message of people is all recorded by electronic data, is passed through and is analyzed track activity data
The spatio-temporal activity track that can analyze out specific people obtains clue by information-based means in major criminal case
Case is more and more, is increasingly becoming an important clue source.
Traditional algorithm is to [S0,S1,S2...SN] data acquisition system by personage is divided into M set, and successively traverses M and collect
It closes and obtains different data combinations, when M is 1, time complexity is O (n)；When M is n, time complexity O
(n！).It is traversed and is compared by full dose datacycle, to infer in a case whether multiple people have contact or meet, Yi Jitong
Contact or the number met are counted to infer the relationship of more people in case.It increases substantially, violates in retrievable original data volume now
Case part is related to that number is more and more, and under the background that relationship becomes increasingly complex, the operand of traditional analysis significantly increases,
Analysis efficiency is low, and the data analysis being unable to satisfy under big data quantity, more personage's application scenarios requires.
Summary of the invention
The object of the invention is that solve existing algorithm data volume is big, more than personage in the case where inefficiency ask
Topic, provides a kind of algorithm based on space-time trajectory Fast Collision, computational complexity is maintained at 0 (n), effectively promotes space-time rail
The operation efficiency of mark collision, reduces operation time, promotes user experience.
The object of the invention purpose is achieved in that specifically, this method includes the following steps:
1. data prediction；
A, initial data is put in storage
Multi-trace crash analysis starts；
B, data temporally ascending sort
Initial data temporally ascending sort obtains data acquisition system So (Sorigin)；
C, input time and apart from variable
Input multi-trace impact conditions time range Δ t and distance range Δ d；
D, data grouping is carried out according to input time
Data acquisition system So is further grouped according to Δ t；
E, packet data is obtained
Data acquisition system Sg (S after being groupedgroup), make first record S in each group of datafirstWith last
Item records SlastTime difference be less than or equal to Δ t, i.e. all records in every group of data all meet condition Δ t；
2. multi-trace collides
A, raw data set is traversed
Ergodic data set So obtains record S1；
B, packet data is obtained
S is obtained from Sg1The grouping S of place time rangeg1And next grouping S of the groupingg2；
C, judge time offset
Compare the record in S1 and Sg1, Sg2, judges whether time interval is less than Δ t, be to then follow the steps d, otherwise jump
Go to step b；
D, judge ranging offset amount
According to S1 and compared the longitude and latitude of record, obtains the distance between the place of two records, whether judge distance
Less than Δ d, it is to then follow the steps e, otherwise jumps to step b；
E, collision result is saved
Collision result is saved in the ID of S1 as KEY, LIST collection is combined into the MAP of value, other collision knots of subsequent S1
Fruit is equally saved in LIST set；It after the completion of traversal, can be obtained with Record ID as KEY, the collision result of the record
LIST is the MAP of value；
F, analysis collision result
The identification card number of all persons in each LIST is extracted, piece identity is obtained after deduplication and demonstrate,proves number combination
The data group of like combinations KEY is put into the same LIST by KEY.
The present invention has following advantages and good effect:
1. originality supports the space-time trajectory collision of big data quantity to be maintained at certain computational complexity, when reducing operation
Between, promote user experience；
2. algorithm has scalability.
Detailed description of the invention:
Fig. 1 is the flow chart of this method.
Wherein:
10-data predictions；
11-initial data storage；
12-data temporally ascending sort；
13-input times and apart from variable；
14-carry out data grouping according to input time；
15-obtain packet data；
The collision of 20-multi-traces；
21-traversal raw data sets；
22-obtain packet data；
23-judge time offset；
24-judge ranging offset amount；
25-save collision result；
26-analysis collision results.
Fig. 2 is the final result data structure diagram of this system；
Fig. 3 is the data structure diagram for colliding result map；
Wherein KEY is personage's identity card combining characters string, and VALUE is the data acquisition system that result is collided comprising multiple groups.
English to Chinese
1, MAP: key-value pair container；
2, KEY: key；
3, VALUE: value
4, LIST: data acquisition system；
4, StartTime: time started；
6, EndTime: end time.
Specific embodiment
With reference to the accompanying drawing with real case to further instruction.
One, algorithm
Such as Fig. 1, this algorithm is:
1. data prediction -10,
A, initial data storage -11；
B, data temporally ascending sort -12；
C, input time and apart from variable -13；
D, data grouping -14 is carried out according to input time；
E, packet data -15 is obtained；
2. multi-trace collision -20,
A, raw data set -21 is traversed1；
B, packet data -22 is obtained；
C, judge time offset -23；
D, judge ranging offset amount -24；
E, collision result -25 is saved；
F, analysis collision result -26.
Two, embodiment
Application scenarios carry out crash analysis to the track data in a period of time [StartTime, EndTime], search out
In certain time error range Δ t minutes, the track within d kilometers of certain distance range delta is overlapped number, searches out common
Clique.(Δ t is set as 60min in case, and Δ d is set as 1km.)
1. data load, data within the scope of a period of time are extracted, by initial data according to time ascending sort,
Form [S0, S1, S2…SN] data list set S.(the initial data S after representing sequenceorigin)-12。
Table 1: initial data Sorigin
| Record number | Recorder | Time | Other information |
| S0 | A | 2018-01-01 08:00 | Other |
| S1 | B | 2018-01-01 08:30 | Other |
| S2 | A | 2018-02-01 09:30 | Other |
| S3 | C | 2018-02-01 09:50 | Other |
| S4 | B | 2018-03-01 08:00 | Other |
| S5 | C | 2018-03-01 08:10 | Other |
| S6 | C | 2018-03-01 08:40 | Other |
| S7 | B | 2018-03-01 09:10 | Other |
| S8 | A | 2018-03-01 09:20 | Other |
| S9 | C | 2018-03-01 09:50 | Other |
2. Δ t, Δ d-13 are inputted, by [S0, S1, S2…SN] set be grouped according to time range, obtain [S0,
S1..Sm],[Sm+1,Sm+2..Sm+k] ... data group Sg(represent the data S after being temporally groupedgroup), each data
The time difference of all data between any two in group is less than or equal to Δ t, such as time difference≤△ t of S1-S0, such as following table -14.
Table 2: packet data Sgroup
3. traversing the data of set So, the collision record that each record is likely to occur centainly is in the time where it
Grouping or future time be grouped in two groups of data (if the time position rearward being compared data and being in grouping, it is likely that
There is the data S collided as a result, in such as table 2 in next grouping1With data S2, therefore every record needs and place group and next
Group data are compared), it is only necessary to -22 compared with being compared time and distance with the data in the two data groups.
4. obtaining data S from So1, S is obtained from Sg1The letters group data S at placeg1With next letters group data Sg2,
Compare S1 and set Sg1、Sg2Middle data, acquisition time difference t continue distance and compare (step 5., if t≤△ t
Otherwise the secondary comparison is terminated, return step 3. -23.
5. obtaining the longitude and latitude data of two records in relatively, two are obtained according to longitude and latitude using HAVERSINE formula
The distance between point d, compares d and distance range △ d, if d≤△ d, obtains a collision record.HAVERSINE formula
As follows -24:
R is earth radius, can be averaged 6371km；
Indicate the latitude of two o'clock；
The difference of Δ λ expression two o'clock longitude.
6. creation result set MAP is for saving collision as a result, being saved in collision record with S1ID be KEY, LIST is
In the MAP of value, subsequent S1When obtaining other collision results again, will also it be saved in -25 in the List.
⑦S1With Sg1、Sg2In all data it is completeer after, by S1And its collision result is removed from So, avoids repeating
3. data return to step, remove a record, and two data groups for needing to collide, repeat above step.
After 8. all record has collided, result MAP data structure such as Fig. 3 of acquisition.
9. continuing to analyze result Map, the ID card No. of all persons is extracted from VALUE, generates identity after deduplication
Number combination idnums is demonstrate,proved, the data set of identical idnums is put into idnums as KEY, LIST is data in the set of value
Structure such as Fig. 2.
Case collision the results are shown in Table 3.
Table 3: case collides result
In conclusion the present invention first pre-processes initial data, according to the time error range delta t of input by data
Grouping, then initial data is traversed, to each data to be analyzed, the data and next packet data only organized with data place
It compares, the full dose of abandoning tradition traverses analysis method, is changed to draw period analysis method, operand is greatly decreased, in big data
It measures, under more personage's scenes, remains to quickly analyze with superior performance, obtain meeting between personage or frequency of exposure, Position/Point Of Contact
Information.
Three, it applies
This patent algorithm is applied to public security and handles a case system, passes through case related personnel's data of analysis acquisition, analyzes personnel
Between contact relation, give a clue support for case.Applying step is as follows:
1. acquiring the data such as trip, lodging and the ticket of case related personnel, input system raw data base；
2. obtaining spot and the time of origin of data, more people's track data crash analysis are quickly carried out by this algorithm,
Obtain frequency of exposure, time of contact and the Position/Point Of Contact between case related personnel.
Claims (1)
1. a kind of algorithm based on space-time trajectory Fast Collision, it is characterised in that:
1. data prediction (10)；
A, initial data storage (11)
Multi-trace crash analysis starts；
B, data temporally ascending sort (12)
Initial data temporally ascending sort obtains data acquisition system So(Sorigin)；
C, input time and apart from variable (13)
Input multi-trace impact conditions time range Δ t and distance range Δ d；
D, data grouping (14) are carried out according to input time
Data acquisition system So is further grouped according to Δ t；
E, packet data (15) are obtained
Data acquisition system Sg(S after being groupedgroup), make first record S in each group of datafirstIt is recorded with the last item
SlastTime difference be less than or equal to Δ t, i.e. all records in every group of data all meet condition Δ t；
2. multi-trace collides (20)
A, raw data set (21) are traversed
Ergodic data set So obtains record S1；
B, packet data (22) are obtained
S is obtained from Sg1The grouping S of place time rangeg1And next grouping S of the groupingg2；
C, judge time offset (23)
Compare the record in S1 and Sg1, Sg2, judges whether time interval is less than Δ t, be to then follow the steps d, otherwise jump to
Step b；
D, judge ranging offset amount (24)
According to S1 and compared the longitude and latitude of record, obtains the distance between the place of two records, judge whether distance is less than
Δ d is to then follow the steps e, otherwise jumps to step b；
E, collision result (25) is saved
Collision result is saved in the ID of S1 as KEY, LIST collection is combined into the MAP of value, and other collision results of subsequent S1 are same
Sample is saved in LIST set；It after the completion of traversal, can be obtained with Record ID as KEY, the collision result LIST of the record is
The MAP of value；
F, analysis collision result (26)
The identification card number of all persons in each LIST is extracted, piece identity is obtained after deduplication and demonstrate,proves number combination KEY, it will
The data group of like combinations KEY is put into the same LIST.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910024972.7A CN109800279A (en) | 2019-01-11 | 2019-01-11 | A kind of algorithm based on space-time trajectory Fast Collision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910024972.7A CN109800279A (en) | 2019-01-11 | 2019-01-11 | A kind of algorithm based on space-time trajectory Fast Collision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109800279A true CN109800279A (en) | 2019-05-24 |
Family
ID=66557132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910024972.7A Withdrawn CN109800279A (en) | 2019-01-11 | 2019-01-11 | A kind of algorithm based on space-time trajectory Fast Collision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109800279A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633279A (en) * | 2019-09-09 | 2019-12-31 | 四川东方网力科技有限公司 | Multi-source space-time trajectory data processing method, device, equipment and storage medium |
| CN110990722A (en) * | 2019-12-19 | 2020-04-10 | 南京柏跃软件有限公司 | Fuzzy co-station analysis algorithm model based on big data mining and analysis system thereof |
| CN111460074A (en) * | 2020-04-09 | 2020-07-28 | 深圳云天励飞技术有限公司 | Trajectory collision analysis method and device and electronic equipment |
| CN113536083A (en) * | 2021-05-31 | 2021-10-22 | 中国人民公安大学 | Target person track collision analysis method based on case event space-time coordinates |
| CN113704342A (en) * | 2021-07-30 | 2021-11-26 | 济南浪潮数据技术有限公司 | Method, system, equipment and storage medium for trace accompanying analysis |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590250A (en) * | 2017-09-18 | 2018-01-16 | 广州汇智通信技术有限公司 | A kind of space-time orbit generation method and device |
| CN107832364A (en) * | 2017-10-26 | 2018-03-23 | 浙江宇视科技有限公司 | A kind of method and device based on space-time data lock onto target object |
| CN108197140A (en) * | 2017-11-24 | 2018-06-22 | 中国电子科技集团公司电子科学研究院 | Method for digging, server and the computer readable storage medium of space-time co-occurrence |
| CN108920499A (en) * | 2018-05-24 | 2018-11-30 | 河海大学 | It is a kind of towards periodically retrieve space-time trajectory index and search method |
-
2019
- 2019-01-11 CN CN201910024972.7A patent/CN109800279A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590250A (en) * | 2017-09-18 | 2018-01-16 | 广州汇智通信技术有限公司 | A kind of space-time orbit generation method and device |
| CN107832364A (en) * | 2017-10-26 | 2018-03-23 | 浙江宇视科技有限公司 | A kind of method and device based on space-time data lock onto target object |
| CN108197140A (en) * | 2017-11-24 | 2018-06-22 | 中国电子科技集团公司电子科学研究院 | Method for digging, server and the computer readable storage medium of space-time co-occurrence |
| CN108920499A (en) * | 2018-05-24 | 2018-11-30 | 河海大学 | It is a kind of towards periodically retrieve space-time trajectory index and search method |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633279A (en) * | 2019-09-09 | 2019-12-31 | 四川东方网力科技有限公司 | Multi-source space-time trajectory data processing method, device, equipment and storage medium |
| CN110990722A (en) * | 2019-12-19 | 2020-04-10 | 南京柏跃软件有限公司 | Fuzzy co-station analysis algorithm model based on big data mining and analysis system thereof |
| CN110990722B (en) * | 2019-12-19 | 2020-11-06 | 南京柏跃软件有限公司 | Fuzzy co-site analysis method and system based on big data mining |
| CN111460074A (en) * | 2020-04-09 | 2020-07-28 | 深圳云天励飞技术有限公司 | Trajectory collision analysis method and device and electronic equipment |
| CN113536083A (en) * | 2021-05-31 | 2021-10-22 | 中国人民公安大学 | Target person track collision analysis method based on case event space-time coordinates |
| CN113536083B (en) * | 2021-05-31 | 2023-11-24 | 中国人民公安大学 | Target person track collision analysis method based on event space-time coordinates |
| CN113704342A (en) * | 2021-07-30 | 2021-11-26 | 济南浪潮数据技术有限公司 | Method, system, equipment and storage medium for trace accompanying analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109800279A (en) | A kind of algorithm based on space-time trajectory Fast Collision | |
| Wang et al. | SMOTETomek-based resampling for personality recognition | |
| CN106874432B (en) | A kind of public transport passenger trip space-time trajectory extracting method | |
| US10810870B2 (en) | Method of processing passage record and device | |
| CN103617234B (en) | Active video enrichment facility and method | |
| CN104794171B (en) | Mark the method and device of picture geographical location information | |
| CN106778876A (en) | User classification method and system based on mobile subscriber track similitude | |
| CN107665289B (en) | Operator data processing method and system | |
| CN106126751A (en) | A kind of sorting technique with time availability and device | |
| CN106934275A (en) | A kind of password intensity evaluating method based on personal information | |
| CN107993444B (en) | Suspected vehicle identification method based on bayonet vehicle-passing big data analysis | |
| CN106649849A (en) | Text information base building method and device and searching method, device and system | |
| CN103793447A (en) | Method and system for estimating semantic similarity among music and images | |
| CN111523577A (en) | Mass trajectory similarity calculation method based on improved LCSS algorithm | |
| CN103177436A (en) | Method, system, computer program product and recording medium for object tracking | |
| CN110110325A (en) | It is a kind of to repeat case lookup method and device, computer readable storage medium | |
| CN109872541A (en) | A kind of information of vehicles analysis method and device | |
| CN108241856A (en) | Information generation method and equipment | |
| CN110853672A (en) | Data expansion method and device for audio scene classification | |
| CN109615573A (en) | Field staff based on big data flees about to commit crimes analysis method and system | |
| CN115798022A (en) | Artificial intelligence identification method based on feature extraction | |
| Moosavi et al. | Discovery of driving patterns by trajectory segmentation | |
| Pacheco et al. | Interactive concept learning for uncovering latent themes in large text collections | |
| CN110851450A (en) | Accompanying vehicle instant discovery method based on incremental calculation | |
| CN105844204A (en) | Method and device for recognizing behavior of human body |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190524 |