CN104931960A - Trend message and radar target state information whole-track data correlation method - Google Patents

Trend message and radar target state information whole-track data correlation method Download PDF

Info

Publication number
CN104931960A
CN104931960A CN201510270745.4A CN201510270745A CN104931960A CN 104931960 A CN104931960 A CN 104931960A CN 201510270745 A CN201510270745 A CN 201510270745A CN 104931960 A CN104931960 A CN 104931960A
Authority
CN
China
Prior art keywords
flight path
association
fuzzy
track
target
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
Application number
CN201510270745.4A
Other languages
Chinese (zh)
Other versions
CN104931960B (en
Inventor
俞鸿波
谢卫
陈怀新
张宇
李绪维
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CETC 10 Research Institute
Original Assignee
CETC 10 Research Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CETC 10 Research Institute filed Critical CETC 10 Research Institute
Priority to CN201510270745.4A priority Critical patent/CN104931960B/en
Publication of CN104931960A publication Critical patent/CN104931960A/en
Application granted granted Critical
Publication of CN104931960B publication Critical patent/CN104931960B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/86—Combinations of radar systems with non-radar systems, e.g. sonar, direction finder
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Automation & Control Theory (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a trend message and radar target state information whole-track data association method comprising the following steps: extracting track information from trend message and radar target state information; using a double-association threshold rule including a fuzzy association threshold and a shape association threshold to make coarse association judgment on whole-track data, forming a target association track pair of the trend message and radar target state information, and determining the mode of subsequent track data association processing according to the result of coarse judgment; for track pair data in line with the fuzzy association threshold, constructing a fuzzy factor according to information corresponding to the target track point location, and then allocating a weight according to the degree of importance of the fuzzy factor, and carrying out fuzzy comprehensive treatment to realize whole-track association; for track pair data in line with the shape association threshold, using a chain code to describe the track shape, and completing whole-track shape association through chain code matching; and finally, carrying out comprehensive association track processing to realize association of trend message and radar target state information whole-track data.

Description

Trend message and radar target situation information full flight path segment data correlating method
Technical field
The invention belongs to target intelligence process and analysis field, particularly relate to the targetpath correlating method of a kind of trend message and radar target situation information.
Background technology
Along with the development of associating air-sea feelings pre-alarming system, the importance of multi-sensor information fusion is fully reflected.Wherein Radar Target Track and trend message to be associated in target identification aspect significant.Radar Target Track mainly refers to the data such as position, direction, speed of the air-sea target obtained by radar detection, includes some track points positions information of air-sea target in trend message.By calculating the target full flight path section feature that two kinds of means obtain, the incidence relation of two class target informations can being set up, distinguishing that target provides foundation for further identifying.
Track association research both at home and abroad about sensing data is hotter, and main method comprises: arest neighbors data correlation (NN), probabilistic data association (PDA), JPDA (JPDA) etc. are based on the data correlation method of probability statistics; In addition, the method such as data correlation, FCM data correlation, fuzzy synthesis association based on fuzzy logic is also had.Associate different from general sensing data, trend message originally has with the principal feature that associates of radar situation information:
(1) object difference is associated: the object of sensing data association is the duplicate removal complementation in order to complete many means track data, and generates unified target situation, thus the tracking of realize target and supervision.And trend message basis being the relation obtaining targetpath in order to set up means of different with associating of radar situation information, realizing the mutual confirmation of information, comprehensively sentencing the excavation of card or realize target relevant information;
(2) association utilizes information gap: sensing data associates the target metric data (comprising: position, speed, acceleration, identity attribute etc.) mainly utilizing the characterisitic parameter of sensor and sensor to obtain in real time, substantially or seldom need not use complete targetpath information.And trend message this be associate for whole targetpath section with associating of radar situation information, utilization be information and target signature, the trajectory shape feature of the full flight path section of target;
(3) difference of data transfer rate and real-time: the data output period of sensing data association was determined by the Data Update cycle of sensor, and be generally second or Millisecond, real-time is higher and turnover rate is basically identical.And trend message this and the association results of radar situation information are disposablely to export in batch, real-time is low and data transfer rate difference large, is typically used as treatment and analyses afterwards;
(4) difference of aftertreatment: sensing data has associated and generally will do Track Fusion afterwards.And trend message originally carries out target information confirmation and comprehensive distinguishing afterwards usually with associating of radar situation information, do not do follow-up Track Fusion.
In sum, trend message basis is associating for the full flight path section of target with the targetpath feature association of radar situation information, and belong to target reconnaissance information post analysis process, its Major Difficulties comprises:
(1) in trend message basis, targetpath is counted few, has openness, and comparatively large with the data transfer rate difference of radar situation information, even occur the situation of the corresponding complete flight path section of minority track points, traditional target association algorithm cannot process;
Targetpath point time, the positional precision difference in (2) two sources are large, and between the precision of data acquisition and system, deviation is unknown, be therefore difficult to adopt the method for accurate quantification to carry out space-time aligning when association, parameter calculates or error correction.
Summary of the invention
The Data Association that the present invention is directed to existing sensing data is not suitable for trend message and associates weak point with the full flight path segment data of radar target situation information, there is provided that a kind of implementation is simple, automaticity is high, the trend message of good stability and radar target situation information full flight path segment data correlating method, to solve multiple track association problems that targetpath point is sparse and position measurement error difference is larger, thus improve intelligence analysis work efficiency.
The technical solution adopted for the present invention to solve the technical problems is: a kind of trend message and radar target situation information full flight path segment data correlating method, is characterized in that comprising the steps:
Step S1, obtain sensor raw data and extract targetpath information: will be stored in the radar target situation information of database, adopt the mode directly extracting target item, from radar target situation information, extract target full flight path section flight path information, adopt the method for regular expression from trend message, obtain the full flight path raw data of target;
Step S2, two-stage association is slightly sentenced: automatically calculate Fuzzy Correlation thresholding and shape thresholding by computing machine, utilize double connected thresholding rule to carry out two-stage to the full flight path segment data of trend message and the target in radar target situation information associate and slightly sentence, the target association flight path pair of formation trend message and radar target situation information;
Step S3, full flight path section fuzzy matching association: to meeting the flight path of fuzzy matching thresholding to data, according to the corresponding informance structure fuzzy factor of targetpath point position, to data, Fuzzy processing is carried out to the flight path meeting fuzzy matching thresholding, then assign weight and the fuzzy synthesis process of multiple feature according to the importance of fuzzy factor, to realize full flight path section fuzzy matching association;
Step S4, full flight path section shape correlation: to not meeting fuzzy matching thresholding, but meet the flight path of shape correlation thresholding to data, calculate region, common point position, for meeting the flight path of shape correlation thresholding to data, in public domain, adopt chain code to describe flight path shape, then by the method for chain codes table, in conjunction with full flight path section shape facility COMPREHENSIVE CALCULATING targetpath between the degree of association;
Step S5, the overall treatment of association flight path: comprise polysemy process and sequence output, carry out the overall treatment associating flight path, using the diversity factor of similarity vectors norm and difference chain code as associate weighing criteria, drawing the targetpath feature association result of final trend message and radar situation information.
By above-mentioned steps, finally realize associating of trend message and the full flight path segment data of radar target situation information.
Compared with prior art, good effect of the present invention is:
The present invention, by obtaining the Fuzzy processing of multiple features of target full flight path section to radar and trend message two kinds of means, avoids the interference such as noise, can be good at the uncertainty representing the multiple feature of sensor, react the essence of problem better;
The present invention makes full use of the shape facility of the full flight path section of target, for meeting the flight path of shape correlation thresholding to data, in public domain, adopt chain code to describe flight path shape, solve multiple track association problems that targetpath point is sparse and position measurement error difference is larger;
The present invention is by the feature such as distance, orientation, motion, objective attribute target attribute of flight path in complete flight path section, in conjunction with full flight path section shape facility COMPREHENSIVE CALCULATING targetpath between the degree of association, fuzzy membership is utilized to calculate track points characteristic type and the inconsistent flight path similarity of observation interval, draw final trend message basis and the targetpath feature association result of radar situation information, solve means of different and obtain flight path precision and data rate differential problem.
Accompanying drawing explanation
In order to more clearly understand the present invention, now by embodiment of the present invention, simultaneously with reference to accompanying drawing, the present invention will be described, wherein:
Fig. 1 is trend message of the present invention and radar target situation information full flight path segment data correlating method process flow diagram.
Fig. 2 is trend message and radar target situation information full flight path segment data example.
Fig. 3 is the process flow diagram that the present invention utilizes double connected thresholding to carry out two-stage slightly to sentence.
Fig. 4 is full flight path section fuzzy matching association process flow diagram.
Fig. 5 is target full flight path section shape facility association process flow diagram.
Fig. 6 is the direction value of the Freeman chain code describing flight path shape.
Embodiment
For making the object of the application, technical scheme and advantage clearly, below in conjunction with drawings and the specific embodiments, the application is described in further detail.
Consult Fig. 1, Fig. 2.In embodiment described below, Fig. 2 show schematically in the full flight path section of the target extracted in trend text and radar target situation information, solid line is the full flight path segment data of radar situation, pentagram is the targetpath point extracted in trend message basis, and dotted line is the full flight path segment data of text of track points composition.The flow process of the targetpath correlating method of the trend message that Fig. 1 describes and radar target situation information, specifically can comprise the following steps:
Step S1, extracts flight path information from trend message and radar target situation information, to obtain the full flight path segment data of multiple target, wherein, adopts the method for regular expression from trend message, extract the full track data of target; Radar target situation information is the three-dimensional radar targetpath stored in a database, adopts the mode directly extracting target item to obtain full flight path segment data;
Step S2, arranges double connected thresholding and carries out two-stage to trend message and radar target situation targetpath and slightly sentence to obtain multiple preliminary targetpath pair associated;
Step S3, to meeting the flight path of fuzzy matching thresholding to data, according to the corresponding informance structure fuzzy factor of targetpath point position, to data, Fuzzy processing is carried out to the flight path meeting fuzzy matching thresholding, then assign weight according to the importance of fuzzy factor, fuzzy synthesis process is to realize full flight path section fuzzy matching association;
Step S4, to do not meet fuzzy matching thresholding but the flight path meeting shape correlation thresholding to data, calculate region, common point position, then by the method for chain codes table, in conjunction with full flight path section shape facility COMPREHENSIVE CALCULATING targetpath between the degree of association;
Step S5, the overall treatment of association flight path, comprise polysemy process and sequence output, carry out the overall treatment associating flight path, using the diversity factor of similarity vectors norm and difference chain code as associate weighing criteria, drawing the targetpath feature association result of final trend message and radar situation information.
In trend message, targetpath adopts some position mode to describe, and some bit format comprises:
(1) time: adopt the mode of date Hour Minute Second to describe, wherein, the date can adopt text description, such as: during 9 days 12 October in 2014 20 points 30 seconds, also can remove word, as: 20141009122030.
(2) position: adopt the mode of longitude and latitude to describe, such as: 120 degree 22 points 33 seconds; Latitude: 25 degree 40 points 56 seconds.
(3) highly: adopt numerical value to describe, unit is rice.
(4) direction of motion: adopt angular values to describe, with positive north for zero degree line, clockwise direction, XX degree YY divide ZZ second.
(5) speed: adopt numerical value to describe, to aircraft, unit is meter per second; To naval vessel, unit is in the sea/hour.
For above-mentioned form, the mode of available regular expression extracts the information such as time and position of some positions from trend message, and adopts coordinate conversion technology target longitude and latitude to be changed into x and the y coordinate of Descartes's rectangular coordinate system.Because the data abstraction techniques of regular expression and coordinate conversion technology are mature technologies, directly can be adopted, be not specifically described in the present invention.
Combine in chronological order for after the some bit sequence extraction of same target designation, form the full flight path segment data of text of this target.
In radar target situation information, main target flight path is three-dimensional radar flight path, therefrom can obtain the x, y, z three-dimensional coordinate of corresponding track points, and the information such as speed and position angle in rectangular coordinate system.Track points information for same target designation combines in order, forms the full flight path segment data of radar situation of this target.
Consult Fig. 3.Method that two-stage slightly sentences is to utilize double connected thresholding to carry out:
First double connected thresholding is determined: Fuzzy Correlation thresholding and shape correlation thresholding.Wherein, Fuzzy Correlation distance threshold Δ d mwith shape correlation distance threshold Δ d svalue be:
Δd M＝V max×Δt M
Δd S＝V max×Δt S
When target be aircraft or naval vessel time, V maxrepresent the maximum headway of corresponding target respectively.
Δ t mwith Δ t srepresent Fuzzy Correlation time threshold and shape correlation time threshold respectively.When target is aircraft, Δ t m=1min, Δ t s=3min; When target is naval vessel, Δ t m=10min, Δ t s=60min.
Track association one-level is slightly sentenced:
For a track points p2 of flight path j in the track points p1 of flight path i in trend text and radar target situation information, if p1 writing time tp1 and p2 tp2 writing time there is following relation:
|tp1-tp2|＜Δt S
And the position (x of p1 point 1, y 1) with the position (x of p2 point 2, y 2) meet following relation:
( ( x 1 - x 2 ) 2 + ( y 1 - y 2 ) 2 ) / 2 2 < &Delta;d S ,
Then think that above-mentioned two track points meet thick Correlation Criteria.
Pointwise judgement is carried out to track points whole in flight path i and flight path j, when existence is for the track points pair meeting thick Correlation Criteria, then carries out counting and add up.As counter number N sbe greater than N 1time (value is 8), namely there is the track points of certain the objective flight path in more than 8 track points and radar situation (track points of track points centering does not all repeat) to meet thick Correlation Criteria in certain text flight path, then think that two flight paths meet one-level and slightly sentence condition (that is: meeting shape correlation thresholding).
Track association secondary is slightly sentenced:
For a track points p2 of flight path j in the track points p1 of flight path i in trend text and radar target situation information, if p1 writing time tp1 and p2 tp2 writing time there is following relation:
|tp1-tp2|＜Δt M
And the position (x of p1 point 1, y 1) with the position (x of p2 point 2, y 2) meet following relation:
( ( x 1 - x 2 ) 2 + ( y 1 - y 2 ) 2 ) / 2 2 < &Delta;d M ,
Then think that above-mentioned two track points meet thick Correlation Criteria.
Pointwise judgement is carried out to two flight path local nodes, counting is carried out for the track points meeting thick Correlation Criteria and adds up, as counter values N mbe greater than N 2time (value is 6), think that two flight path meeting tier 2s slightly sentence condition (that is: meeting Fuzzy Correlation thresholding), then this flight path is sent into follow-up link to (i, j) and carry out full flight path segment information Fuzzy Correlation; Otherwise, by flight path to carrying out shape facility association process.
In step s3, to meeting the flight path of fuzzy matching thresholding to data, the character pair information structuring fuzzy factor according to targetpath point position carries out Fuzzy processing.Obfuscation refers to according to certain obfuscation rule, an original class or a few category feature variable is divided into multiple fuzzy variable, makes each fuzzy variable express the local characteristics of former feature.Original feature is replaced to carry out subsequent treatment by these new fuzzy characteristics.The object original a kind of feature being become some fuzzy characteristics is to make new feature react the essence of problem better.In addition, because the existence of the disturbing factors such as noise makes the target signature to extracting have certain uncertainty.Compared with statistical method, fuzzy mathematics has the superperformance of process uncertain problem.Therefore, track association problem is based on fuzzy mathematics, the information of the targetpath point position information first using trend report to provide and the corresponding track points position of radar target situation, calculate the driftage factor, the time difference factor, direction factor, height factors, velocity factor etc., then the quantitative target of the track points degree of correlation of a trend report text objects full flight path section and the full flight path section of radar target is constructed, this targetpath of trend message and Radar Target Track are carried out whole matching, calculate the Fuzzy Relationship Degree of full flight path section, and judge whether association according to this index.
Consult Fig. 4.Full flight path section fuzzy matching association flow process is specifically implemented as follows:
Sub-step S31: determine fuzzy factor collection
In complete flight path section, carry out full flight path segment information Fuzzy Correlation, first construct fuzzy factor collection U={ μ 1, μ 2..., μ n, wherein each fuzzy factor computing method are:
Alternate position spike fuzzy factor: reacted in complete flight path section, namely trend text objects flight path i and radar situation datum target flight path j meets the track points position departure degree that track association secondary slightly sentences condition, represent the association moment, represent in radar situation datum target flight path j, with the moment the time value of immediate track points;
Difference in height fuzzy factor: react complete flight path section and meet track association secondary and slightly sentence the track points highly deviated degree of condition;
Velocity contrast fuzzy factor: react complete flight path section and meet track association secondary and slightly sentence the track points speed deviations degree of condition;
Direction difference fuzzy factor: react complete flight path section and meet track association secondary and slightly sentence the track points course drift degree of condition.
If two track association results are divided into m rank, then the set be made up of these results is called as passes judgment on collection, is designated as V={v 1, v 2..., v m, wherein, v 1, l=1,2 ..., m is l court verdict.To the court verdict of any two flight paths, be actually a fuzzy subset on V.When evaluating association, owing to only whether associating interested to targetpath, therefore, from the consideration of the simplify processes of the angle of practical application and problem, the rank m=2 evaluating collection is selected, wherein, v 1represent association, v 2represent and do not associate.
Sub-step S32: structure fuzzy factor power collection
According to the set of factors U of structure, be each factor right of distribution coefficient, form fuzzy set A={a 1, a 2..., a n, a in formula kfor a kth factor μ kcorresponding power, general provision a kselection need according to a kth factor to adjudicate importance or influence degree decide.According to the diversity of objective attribute target attribute and mode of motion, a kvalue to embody the importance of each factor and actual environment as far as possible to the impact of sensor, and Wrong, missing association should be reduced as far as possible.
Consider that the positional precision of each flight path observation station is higher than height, velocity accuracy, weight vector selects a 1=0.6, a 2=0.15, a 4=0.15, a 4=0.1.
Sub-step S33: determine fuzzy membership functions, calculates comprehensive similarity and Fuzzy Correlation matrix
Similar degree of membership based on two flight paths of above-mentioned four fuzzy factors is:
r k＝exp{-τ k(μ k/σ k)}，k=1，2，3，4
In formula, σ kthe latitude of emulsion of factor in fuzzy set, and τ kit is adjustment degree;
After the degree of membership calculating each factor respectively, weighted mean just can be adopted to carry out comprehensive evaluation, so comprehensive similarity is:
For the n from text gobjective flight path i and the n from radar situation hobjective flight path j just constitutes Fuzzy Correlation matrix:
Sub-step S34: track association is checked
Maximum comprehensive similarity and threshold value discrimination principle is adopted to carry out track association inspection.Process is: first in matrix F, find out greatest member f ijif, f ij> ε, then judge that flight path i and j is test association, after then deleting corresponding row column element from matrix,
Obtain new depression of order Fuzzy Correlation matrix, repeat said process until all elements is all less than ε in matrix; Parameter ε is threshold value, and value is 0.5.
In order to control finishing and termination of track association inspection, introduce trace association quality.Based on double threshold criterion, get positive integer and N m, in above-mentioned checkout procedure, if test is successfully associated, then associate quality
Its numerical value represents moment, flight path i and j was judged to the number of times meeting Testing Association.
If test association is unsuccessful, flight path departs from quality:
Do not associate if to be that is judged to integrated track j at l moment desired track i, then it leaks association mass number and adds 1, represents flight path i and j and is judged to the number of times do not associated in the l moment.
At N mafter secondary Testing Association completes, have:
m ij(N M)≥I，i∈U g，j∈U h
I is flight path quality thresholding, then adjudicate flight path i with j for having associated.
If for some i, more than one of the j that above formula is set up, then want follow-up and carry out polysemy process.To given N mwith if moment exists then stop Fuzzy Correlation inspection.
Step S4, full flight path section shape correlation flow process as shown in Figure 5, comprises following sub-step:
The pre-service of sub-step S41 flight path:
The length extracting targetpath and radar situation targetpath due to message is inconsistent, therefore needs to carry out flight path resampling, and both are regular to same length.Wherein, initial trace is processed into equal length by sampling by resampling.
The shape analysis of sub-step S42 flight path:
Analyze for flight path shape, judge whether flight path shape meets singularity and be applicable to subsequent shapes feature association.Judge according to comprising flight path knowledge corresponding to the some figure place of flight path, complexity and objective attribute target attribute.
When calculating flight path complexity, the main feature utilized comprises:
Linear complexity: use track common point position end to end 2 set up total slope, the diversity factor of the slope between all the other tracing points and total slope is less than given thresholding, then think that whole piece flight path does not meet linear complexity requirement; Otherwise think that it meets linear complexity requirement;
Direction complexity: on flight path direction, corner point occurs more than twice, thinks that flight path meets direction complexity requirement;
In conjunction with user experience, objective attribute target attribute degree of conformity: according to the target type attribute of Text Feature Extraction, judges whether targetpath has complicacy, as: early warning plane and reconnaissance plane etc.;
Flight path shape singularity measure: when flight path meet above-mentioned three require time, think flight path shape meet singularity tolerance.
The method of discrimination of corner point recited above is as follows:
Definition difference accumulated value: d (i)=d 1(i)+d 2(i)
Wherein
d 1 ( i ) = | &xi; ( i + 1 ) - &xi; ( i ) | , | &xi; ( i + 1 ) - &xi; ( i ) | < 4 8 - | &xi; ( i + 1 ) - &xi; ( i ) | , | &xi; ( i + 1 ) - &xi; ( i ) | > 4 4 , | &xi; ( i + 1 ) - &xi; ( i ) | = 4
d 2 ( i ) = | &xi; ( i + 2 ) - &xi; ( i - 1 ) | , | &xi; ( i + 2 ) - &xi; ( i ) - 1 | < 4 8 - | &xi; ( i + 2 ) - &xi; ( i - 1 ) | , | &xi; ( i + 2 ) - &xi; ( i - 1 ) | > 4 4 , | &xi; ( i + 2 ) - &xi; ( i - 1 ) | = 4
In formula, i is the index value of track points, the chain code value that ξ (i) is corresponding track points, d 1i () is difference value, d 2i () is secondary difference value.After obtaining d (i), utilize the corner point that criterion below judges in track points:
(1) if d (i) >=3, then think that this point is corner point;
(2) otherwise, think that this point is non-corner point;
Sub-step S43 signature of flight path extracts:
Extract the shape facility of targetpath, be convenient to follow-up form fit.The flight path shape facility extracted adopts normalization chain code.
Normalization chain code: chain code is for representing by the boundary line formed along the line segment with assigned direction connected, usually with 8 sections of chain representation boundary lines.
Because chain code only has Pan and Zoom unchangeability, to the rotation between the targetpath of the same name existed because of detecting error, not there is adaptive faculty, simultaneously, in order to overcome the chain code problem that difference is brought with rising point selection different, adopt normalization first order difference Freeman chain code to describe flight path shape, namely there is rotational invariance.From all directions to Freeman chain code value as shown in Figure 6.
Freeman chain code takes full advantage of the spatial relation between pixel, with eight, the point of shape is represented that the numeral in direction represents, thus greatly save storage space and computing time.
Sub-step S44 flight path form fit:
According to the targetpath feature extracted, calculate the degree of correlation between flight path by chain codes table method.Flight path form fit does not do hard decision, only the result of coupling is preserved in the mode of numerical value, to carry out follow-up association comprehensive assessment.
The method of chain codes table is as follows:
Above formula represents a text kth normalization difference chain code with radar individual normalization difference chain code diversity factor, to element ξ each in chain code ask difference rear chain code element sum be normalized, γ represents the weighting to the corner point factor.
Finally set matching threshold T sM, value is 0.2, if think that two flight paths meet shape correlation, otherwise not think and associate.
Step S5, the overall treatment of association flight path, comprises polysemy process and sequence exports.Wherein, polysemy process refers to when in the one group of flight path meeting association requirement, process during more than one of association flight path number.
First, determine that polysemy process criterion is as follows:
When polysemy appears in full flight path section whole matching association, use similarity vectors norm carry out association weighing criteria;
When polysemy appears in flight path shape correlation, use the diversity factor of normalization difference chain code as association weighing criteria.
The criterion that final sequence exports is that the result that full flight path section whole matching associates when single interrelational form exists polysemy, then sorts according to the value size of R1 and R2 before coming flight path shape correlation result respectively.

Claims (10)

1. trend message and a radar target situation information full flight path segment data correlating method, is characterized in that comprising the steps:
Step S1, obtain sensor raw data and extract targetpath information: will be stored in the radar target situation information of database, adopt the mode directly extracting target item, from radar target situation information, extract target full flight path section flight path information, adopt the method for regular expression from trend message, obtain the full flight path raw data of target;
Step S2, two-stage association is slightly sentenced: automatically calculate Fuzzy Correlation thresholding and shape thresholding by computing machine, utilize double connected thresholding rule to carry out two-stage to the full flight path segment data of trend message and the target in radar target situation information associate and slightly sentence, the target association flight path pair of formation trend message and radar target situation information;
Step S3, full flight path section fuzzy matching association: to meeting the flight path of fuzzy matching thresholding to data, according to the corresponding informance structure fuzzy factor of targetpath point position, to data, Fuzzy processing is carried out to the flight path meeting fuzzy matching thresholding, then assign weight and the fuzzy synthesis process of multiple feature according to the importance of fuzzy factor, to realize full flight path section fuzzy matching association;
Step S4, full flight path section shape correlation: to not meeting fuzzy matching thresholding, but meet the flight path of shape correlation thresholding to data, calculate region, common point position, for meeting the flight path of shape correlation thresholding to data, in public domain, adopt chain code to describe flight path shape, then by the method for chain codes table, in conjunction with full flight path section shape facility COMPREHENSIVE CALCULATING targetpath between the degree of association;
Step S5, the overall treatment of association flight path: comprise polysemy process and sequence output, carry out the overall treatment associating flight path, using the diversity factor of similarity vectors norm and difference chain code as associate weighing criteria, drawing the targetpath feature association result of final trend message and radar situation information.
2. trend message according to claim 1 and radar target situation information full flight path segment data correlating method, it is characterized in that, the method that in described step S2, two-stage is slightly sentenced is:
First double threshold is determined: Fuzzy Correlation thresholding and shape correlation thresholding, wherein, Fuzzy Correlation distance threshold Δ d mwith shape correlation distance threshold Δ d svalue be:
Δd M＝V max×Δt M
Δd S＝V max×Δt S
When target be aircraft or naval vessel time, V maxrepresent the maximum headway of corresponding target respectively, Δ t mwith Δ t srepresent Fuzzy Correlation time threshold and shape correlation time threshold respectively;
In track association one-level is slightly sentenced: for a track points p2 of flight path j in the track points p1 of flight path i in trend text and radar target situation information, if p1 writing time tp1 and p2 tp2 writing time there is following relation:
|tp1-tp2|＜Δt S
And the position (x of p1 point 1, y 1) with the position (x of p2 point 2, y 2) meet following relation:
( ( x 1 - x 2 ) 2 + ( y 1 - y 2 ) 2 ) / 2 2 < &Delta;d S ,
Then above-mentioned two track points meet thick Correlation Criteria;
In track association secondary is slightly sentenced: for a track points p2 of flight path j in the track points p1 of flight path i in trend text and radar target situation information, if p1 writing time tp1 and p2 tp2 writing time there is following relation:
|tp1-tp2|＜Δt M
And the position of p1 point with the position of p2 point meet following relation:
( ( x 1 - x 2 ) 2 + ( y 1 - y 2 ) 2 ) / 2 2 < &Delta;d M ,
Then above-mentioned two track points meet thick Correlation Criteria.In formula, Δ t mrepresent Fuzzy Correlation time threshold, occurrence is specified by user.
3. trend message according to claim 2 and radar target situation information full flight path segment data correlating method, it is characterized in that, computer program carries out pointwise judgement to track points whole in flight path i and flight path j, when existence is for the track points pair meeting thick Correlation Criteria, then carries out counting and adds up.As counter number N sbe greater than N 1time, namely have the track points of certain the objective flight path in more than 8 track points and radar situation in certain text flight path, and the track points of track points centering does not all repeat, meet thick Correlation Criteria, then think that two flight paths meet one-level and slightly sentence condition, namely meet shape correlation thresholding, wherein N 1value is 8.
4. trend message according to claim 3 and radar target situation information full flight path segment data correlating method, it is characterized in that, computer program carries out pointwise judgement to two flight path local nodes, carries out counting add up, as counter values N for the track points meeting thick Correlation Criteria mbe greater than N 2time, think that two flight path meeting tier 2s slightly sentence condition, namely meet Fuzzy Correlation thresholding, then this flight path is sent into follow-up link to (i, j) and carry out full flight path segment information Fuzzy Correlation; Otherwise, by flight path to carrying out shape facility association process, wherein N 2value is 6.
5. trend message according to claim 2 and radar target situation information full flight path segment data correlating method, it is characterized in that, for the flight path pair meeting thick Correlation Criteria in described step S3, in complete flight path section, carry out full flight path segment information Fuzzy Correlation, first construct fuzzy factor collection U={ μ 1, μ 2..., μ n, on wherein each fuzzy factor computing method be:
Alternate position spike fuzzy factor: &mu; 1 = ( x gi ( l ) - x hj ( l * ) ) 2 + ( y gi ( l ) - y hj ( l * ) ) 2 , Having reacted in complete flight path section is that trend text objects flight path i and radar situation datum target flight path j meets the track points position departure degree that track association secondary slightly sentences condition, and l represents the association moment, l *represent in radar situation datum target flight path j, with the time value of the immediate track points of moment l, x gi(l) and y gil () represents x and the y coordinate of g track points in association moment l text objects flight path i respectively, x hj(l *) and y hj(l *) represent moment l respectively *x and the y coordinate of h1 track points in radar situation datum target flight path j;
Difference in height fuzzy factor: react complete flight path section and meet track association secondary and slightly sentence the track points highly deviated degree of condition, wherein z gil () represents the z coordinate of g track points in association moment l text objects flight path i, z hj(l *) represent moment l *the z coordinate of h1 track points in radar situation datum target flight path j;
Velocity contrast fuzzy factor: μ 3=| v gi(l)-v hj(l *) |, react complete flight path section and meet track association secondary and slightly sentence the track points speed deviations degree of condition, wherein v gil () represents the target velocity of g track points in association moment l text objects flight path i, v hj(l *) represent moment l *the target velocity of h1 track points in radar situation datum target flight path j;
Direction difference fuzzy factor: μ 4=| θ gi(l)-θ hj(l *) |, react complete flight path section and meet track association secondary and slightly sentence the track points course drift degree of condition, wherein θ gil () represents the target course of g track points in association moment l text objects flight path i, θ hj(l *) represent moment l *the target course of h1 track points in radar situation datum target flight path j.
Similar degree of membership based on two flight paths of above-mentioned four fuzzy factors is:
r k＝exp{-τ k(μ k/σ k)}，k＝1，２，３，４
In formula, σ kthe latitude of emulsion of factor in fuzzy set, and τ kadjustment degree, r krepresent the similar degree of membership of a two flight path kth factor;
After the degree of membership calculating each factor respectively, weighted mean just can be adopted to carry out comprehensive evaluation, so comprehensive similarity is:
f ij ( l ) = &Sigma; k = 1 n a k ( l ) r k
For the n from text gobjective flight path i and the n from radar situation hobjective flight path j just constitutes Fuzzy Correlation matrix:
F (l) represents the similarity of l moment text and radar situation.The maximum comprehensive similarity of follow-up employing and threshold value discrimination principle can carry out track association.
6. trend message according to claim 1 and radar target situation information full flight path segment data correlating method, it is characterized in that, in described step S4, the step of flight path shape correlation is:
The pre-service of sub-step S41 flight path: the targetpath that message is extracted and radar situation targetpath regular to same length.Wherein, initial trace is processed into equal length by sampling by flight path resampling.
Sub-step S42 flight path shape analysis: analyze for flight path shape, judges whether flight path shape meets singularity and be applicable to subsequent shapes feature association; Judge according to comprising flight path knowledge corresponding to the some figure place of flight path, complexity and objective attribute target attribute.
Sub-step S43 signature of flight path extracts: the shape facility extracting targetpath, is convenient to follow-up form fit.The flight path shape facility extracted adopts normalization first order difference Freeman chain code to describe, and namely has translation and rotates convergent-divergent unchangeability;
Sub-step S44 flight path form fit: according to the targetpath feature extracted, calculates the degree of correlation between flight path by chain codes table method.Flight path form fit does not do hard decision, only the result of coupling is preserved in the mode of numerical value, to carry out follow-up association comprehensive assessment.
7. trend message according to claim 1 and radar target situation information full flight path segment data correlating method, it is characterized in that, the method for chain codes table is as follows:
T ( X k g , X l h ) = 1 N S &Sigma; i = 1 n &gamma; i | &xi; k , i g - &xi; l . i h |
Above formula represents a message targetpath kth normalization difference chain code with radar target situation targetpath l normalization difference chain code diversity factor, to element ξ each in chain code ask difference rear chain code element sum be normalized, γ represents the weighting to the corner point factor, N srepresent the chain data code sum in the public domain of two flight paths.
8. trend message according to claim 1 and radar target situation information full flight path segment data correlating method, it is characterized in that, if two track association results are divided into m rank, then the set be made up of these results is called as passes judgment on collection, is designated as V=(v 1, v 2..., v m, wherein, v 1, 1=1,2 ..., m is the 1st court verdict; To the court verdict of any two flight paths, being actually a fuzzy subset on V, when evaluating association, selecting the rank m=2 evaluating collection, wherein, v 1represent association, v 2represent and do not associate.
9. trend message according to claim 1 and radar target situation information full flight path segment data correlating method, is characterized in that, for the n from text gobjective flight path i and the n from radar situation hobjective flight path j just constitutes Fuzzy Correlation matrix:
In formula, represent the n from text respectively gobjective flight path and the n from radar situation hthe comprehensive similarity of objective flight path.
10. trend message according to claim 1 and radar target situation information full flight path segment data correlating method, is characterized in that, adopts maximum comprehensive similarity and threshold value discrimination principle to carry out track association inspection: first in matrix F, to find out greatest member f ijif, f ij> ε, then judge that flight path i and j is test association, after then deleting corresponding row column element from matrix, obtain new depression of order Fuzzy Correlation matrix, repeat said process until all elements is all less than ε in matrix; Parameter ε is threshold value, and value is 0.5.
CN201510270745.4A 2015-05-25 2015-05-25 Trend message and radar target situation information full flight path segment data correlating method Active CN104931960B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510270745.4A CN104931960B (en) 2015-05-25 2015-05-25 Trend message and radar target situation information full flight path segment data correlating method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510270745.4A CN104931960B (en) 2015-05-25 2015-05-25 Trend message and radar target situation information full flight path segment data correlating method

Publications (2)

Publication Number Publication Date
CN104931960A true CN104931960A (en) 2015-09-23
CN104931960B CN104931960B (en) 2017-03-08

Family

ID=54119213

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510270745.4A Active CN104931960B (en) 2015-05-25 2015-05-25 Trend message and radar target situation information full flight path segment data correlating method

Country Status (1)

Country Link
CN (1) CN104931960B (en)

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105469153A (en) * 2015-11-25 2016-04-06 中国电子科技集团公司第二十八研究所 Air situation sharing apparatus based on system ordering
CN107063259A (en) * 2017-03-08 2017-08-18 四川九洲电器集团有限责任公司 A kind of Data Association and electronic equipment
CN108896973A (en) * 2018-07-13 2018-11-27 北京无线电测量研究所 Scaling method, point the mark method for evaluating quality, storage medium of radar data
CN109143224A (en) * 2018-08-28 2019-01-04 中国电子科技集团公司第三十六研究所 A kind of multiple target correlating method and device
CN109190647A (en) * 2018-06-26 2019-01-11 中国电子科技集团公司第二十九研究所 A kind of active passive data fusion method
CN109656916A (en) * 2018-12-14 2019-04-19 北京无线电测量研究所 A kind of acquisition methods of radar target data
CN110033017A (en) * 2019-02-27 2019-07-19 中国人民解放军空军工程大学 A kind of more radar track substep Interconnected Fuzzy clustering algorithms
CN110145966A (en) * 2019-02-27 2019-08-20 中国人民解放军空军工程大学 A kind of finite time convergence control Global robust sliding formwork method of guidance
CN110907903A (en) * 2019-10-30 2020-03-24 中电科技（合肥）博微信息发展有限责任公司 Self-adaptive tracking processing method based on multiple sensors
CN110991539A (en) * 2019-09-05 2020-04-10 北京无线电测量研究所 Spatial target high-frequency repetitive behavior identification method
CN111860688A (en) * 2020-07-31 2020-10-30 北京无线电测量研究所 Target type identification method and system and electronic equipment
CN113064164A (en) * 2021-03-16 2021-07-02 北京润科通用技术有限公司 Vehicle-mounted radar data association method and device
CN113447027A (en) * 2021-07-16 2021-09-28 电信科学技术第五研究所有限公司 Target fusion filtering method based on geometric figure
CN113702964A (en) * 2021-08-23 2021-11-26 中国北方工业有限公司 Radar adaptive area aggregation method based on track information
CN114967751A (en) * 2022-06-21 2022-08-30 深圳华创电科技术有限公司 Aircraft track tracking method, device, equipment and storage medium

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1417591A (en) * 2001-11-09 2003-05-14 财团法人资讯工业策进会 Fusion processing method of flight path information in combined radar/automatic echo monitoring
US20030191610A1 (en) * 2002-03-26 2003-10-09 Hai-Wen Chen Method and system for multi-sensor data fusion using a modified dempster-shafer theory
EP2343672A1 (en) * 2009-12-17 2011-07-13 BAE Systems PLC Producing data describing states of a plurality of targets.
CN102162847A (en) * 2011-01-14 2011-08-24 中国人民解放军理工大学 High-efficiency track correlation method based on singular value decomposition
CN102707276A (en) * 2012-05-23 2012-10-03 中国人民解放军海军航空工程学院 Automatic identification system (AIS) and radar track robust association algorithm based on target topological information
CN103869281A (en) * 2014-03-13 2014-06-18 中国民用航空总局第二研究所 Passive multipoint positioning system-based double-threshold association method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1417591A (en) * 2001-11-09 2003-05-14 财团法人资讯工业策进会 Fusion processing method of flight path information in combined radar/automatic echo monitoring
US20030191610A1 (en) * 2002-03-26 2003-10-09 Hai-Wen Chen Method and system for multi-sensor data fusion using a modified dempster-shafer theory
EP2343672A1 (en) * 2009-12-17 2011-07-13 BAE Systems PLC Producing data describing states of a plurality of targets.
CN102162847A (en) * 2011-01-14 2011-08-24 中国人民解放军理工大学 High-efficiency track correlation method based on singular value decomposition
CN102707276A (en) * 2012-05-23 2012-10-03 中国人民解放军海军航空工程学院 Automatic identification system (AIS) and radar track robust association algorithm based on target topological information
CN103869281A (en) * 2014-03-13 2014-06-18 中国民用航空总局第二研究所 Passive multipoint positioning system-based double-threshold association method

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105469153B (en) * 2015-11-25 2019-05-21 中国电子科技集团公司第二十八研究所 A kind of air situation sharing means ordered based on system
CN105469153A (en) * 2015-11-25 2016-04-06 中国电子科技集团公司第二十八研究所 Air situation sharing apparatus based on system ordering
CN107063259A (en) * 2017-03-08 2017-08-18 四川九洲电器集团有限责任公司 A kind of Data Association and electronic equipment
CN107063259B (en) * 2017-03-08 2020-06-09 四川九洲电器集团有限责任公司 Track association method and electronic equipment
CN109190647A (en) * 2018-06-26 2019-01-11 中国电子科技集团公司第二十九研究所 A kind of active passive data fusion method
CN109190647B (en) * 2018-06-26 2022-03-08 中国电子科技集团公司第二十九研究所 Active and passive data fusion method
CN108896973A (en) * 2018-07-13 2018-11-27 北京无线电测量研究所 Scaling method, point the mark method for evaluating quality, storage medium of radar data
CN109143224A (en) * 2018-08-28 2019-01-04 中国电子科技集团公司第三十六研究所 A kind of multiple target correlating method and device
CN109143224B (en) * 2018-08-28 2023-01-20 中国电子科技集团公司第三十六研究所 Multi-target association method and device
CN109656916A (en) * 2018-12-14 2019-04-19 北京无线电测量研究所 A kind of acquisition methods of radar target data
CN110033017A (en) * 2019-02-27 2019-07-19 中国人民解放军空军工程大学 A kind of more radar track substep Interconnected Fuzzy clustering algorithms
CN110145966A (en) * 2019-02-27 2019-08-20 中国人民解放军空军工程大学 A kind of finite time convergence control Global robust sliding formwork method of guidance
CN110991539A (en) * 2019-09-05 2020-04-10 北京无线电测量研究所 Spatial target high-frequency repetitive behavior identification method
CN110991539B (en) * 2019-09-05 2023-10-31 北京无线电测量研究所 Space target high-frequency repetitive behavior recognition method
CN110907903A (en) * 2019-10-30 2020-03-24 中电科技（合肥）博微信息发展有限责任公司 Self-adaptive tracking processing method based on multiple sensors
CN110907903B (en) * 2019-10-30 2023-08-01 中电科技（合肥）博微信息发展有限责任公司 Self-adaptive tracking processing method based on multiple sensors
CN111860688A (en) * 2020-07-31 2020-10-30 北京无线电测量研究所 Target type identification method and system and electronic equipment
CN111860688B (en) * 2020-07-31 2023-10-31 北京无线电测量研究所 Target type identification method and system and electronic equipment
CN113064164B (en) * 2021-03-16 2022-11-22 北京润科通用技术有限公司 Vehicle-mounted radar data association method and device
CN113064164A (en) * 2021-03-16 2021-07-02 北京润科通用技术有限公司 Vehicle-mounted radar data association method and device
CN113447027B (en) * 2021-07-16 2022-03-08 电信科学技术第五研究所有限公司 Target fusion filtering method based on geometric figure
CN113447027A (en) * 2021-07-16 2021-09-28 电信科学技术第五研究所有限公司 Target fusion filtering method based on geometric figure
CN113702964A (en) * 2021-08-23 2021-11-26 中国北方工业有限公司 Radar adaptive area aggregation method based on track information
CN113702964B (en) * 2021-08-23 2023-09-26 中国北方工业有限公司 Radar self-adaptive region aggregation method based on track information
CN114967751A (en) * 2022-06-21 2022-08-30 深圳华创电科技术有限公司 Aircraft track tracking method, device, equipment and storage medium

Also Published As

Publication number Publication date
CN104931960B (en) 2017-03-08

Similar Documents

Publication Publication Date Title
CN104931960A (en) Trend message and radar target state information whole-track data correlation method
Li et al. Localizing and quantifying damage in social media images
Zhang et al. AIS data driven general vessel destination prediction: A random forest based approach
CN109002835A (en) A kind of particle filter data correlation method based on maximum entropy fuzzy clustering
Shen et al. Visual interpretation of recurrent neural network on multi-dimensional time-series forecast
CN110348437A (en) It is a kind of based on Weakly supervised study with block the object detection method of perception
CN112508105A (en) Method for detecting and retrieving faults of oil extraction machine
CN109753874A (en) A kind of low slow small classification of radar targets method based on machine learning
CN104715154A (en) Nuclear K-mean value track correlation method based on KMDL criteria
CN107832778A (en) A kind of same target recognition methods based on spatial synthesis similarity
Zhao et al. Abnormal trajectory detection based on a sparse subgraph
Ji et al. A divisive hierarchical clustering approach to hyperspectral band selection
CN113284369B (en) Prediction method for actually measured airway data based on ADS-B
Zhang et al. A novel extreme learning machine using privileged information
CN104537392B (en) A kind of method for checking object based on the semantic part study of identification
Lin et al. An Unsupervised Transformer-based Multivariate Alteration Detection Approach for Change Detection in VHR Remote Sensing Images
CN111563532B (en) Unknown target identification method based on attribute weight fusion
Armah et al. A deep analysis of the precision formula for imbalanced class distribution
Daudt et al. Learning to understand earth observation images with weak and unreliable ground truth
CN104574345B (en) False change detecting method in a kind of Land_use change/cover diverse vector figure based on SYMMETRY THEORY
Sasaki et al. Road segment interpolation for incomplete road data
CN112906746B (en) Multi-source track fusion evaluation method based on structural equation model
CN104462826A (en) Method for detecting and measuring multi-sensor evidence conflicts on basis of matrix singular value decomposition
CN103983269A (en) Asynchronous track association algorithm based on interval-real number mixed sequence
CN113947309A (en) Shield tunnel construction standard working hour measuring and calculating and scoring method based on big construction data

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant