CN106570330A - Shape estimated performance evaluation method for extended target tracing - Google Patents
Shape estimated performance evaluation method for extended target tracing Download PDFInfo
- Publication number
- CN106570330A CN106570330A CN201610977631.8A CN201610977631A CN106570330A CN 106570330 A CN106570330 A CN 106570330A CN 201610977631 A CN201610977631 A CN 201610977631A CN 106570330 A CN106570330 A CN 106570330A
- Authority
- CN
- China
- Prior art keywords
- target
- morphology
- estimated
- extension
- shape
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a shape estimated performance evaluation method for extended target tracing and belongs to the field of sensors and information signal processing. The method comprises a step that according to a complicate extension shape feature of the extended target, in particular for an extended target model built based on a supporting function, various shape parameter description ways are fully considered; a Hausdorff distance is discrete-sampled and then Monte Carlo-averaged, so overall matching degree between a real target shape and an estimated target shape can be measured; and an aim of extended target shape estimated performance evaluation can be achieved. The method can evaluate advantages and disadvantages of shape estimated performance in the extended target tracing algorithms, and can be realized in an engineering way; and the method has strong application value and promotion prospect.
Description
Technical field
The invention belongs to sensor and information signal process field, are related to extend form and the true shape that target state estimator goes out
Matching problem between state, i.e., it is a kind of for extending Target Tracking System in form estimate performance estimating method, can be effective
To evaluate the quality of extension target tracking algorism.
Background technology
The continuous development of modern high-precision sensor resolution techniques so that it is not only able to provide motion state measurement, also
Measurement in terms of the portion forms characteristic information such as width, size of target can be provided.In this case, movable body is typically recognized
To be the extension target with certain form, point target is no longer considered.
In order to verify the validity of certain track algorithm, generally require that its property estimated is contrasted and assessed with other algorithms
Can, and estimate that the quality of performance is then by calculating true target state and estimating the evaluated error size between dbjective state
To embody.By taking point target tracking as an example, root-mean-square error can be come to the target state property estimated as a kind of measurement criterion
Can be estimated.But for the estimation Performance Evaluation of form, the mean square error of target morphology parameter can not objective measure expansion
Form and the similarity degree of true form that exhibition target state estimator goes out.This is because for the extension target with different shape, making
Different root-mean-square error results can be produced with different morphological parameters description forms.Therefore, in order to extending target following calculation
Form estimates that performance carries out objective evaluation in method, inventor by the form for extending target estimate assessment be regarded as target true form with
The whole matching problem of estimated target morphology.During the present invention is realized, particular for being built based on support function
The extension object module for erecting, fully takes into account the describing mode of its different shape parameter having, by Hausdorff distance
Discrete sampling and the whole matching journey for carrying out Monte Carlo averagely to objectively respond real goal form with estimate target morphology
Degree, so as to reach the purpose that extension target morphology estimates Performance Evaluation.
The content of the invention
It is an object of the invention to provide a kind of form for extending target following estimates performance estimating method, described expansion
Exhibition target morphology estimates performance estimating method, is not only able to objectively respond between the estimated form of extension target and true form
Similarity degree, and can according in invention propose Hausdorff distance discrete sampling after obtain Monte Carlo average result come
Effectively weigh the quality of extension target tracking algorism.
The technical solution adopted in the present invention is：A kind of form for extending target following estimates performance estimating method,
Comprise the following steps：
Step one, hypothesis extension target, according to the complicated expanded configuration feature that extension target has, its morphologic description is adopted
The mathematical form of support functionTo describe,, wherein
N is the number for extending the sub-goal form that target morphology decomposes, along the angle of sightThe support function of sub-goal is expressed as on direction,,；Then extend targetSupport function be expressed as, matrix point
AmountTo extend the expanded configuration parameter of targetComponent,For vectorial transposition；
Step 2, hypothesis
It is to be calculated using extension target following
The target morphology parameter that method is estimated, then the description form for estimating the support function of target morphology is
；
Step 3, according to the estimated target morphology of step 2The real goal form calculated with step oneBetween discrete sampling Hausdorff distanceTo evaluate the complexity extension target
Estimate performance, wherein,,,,, a is target morphologyMiddle basisThe certain point of discrete sampling,It is expressed as point a and targetBetween
Euclidean distance；
Step 4, adopt formulaAdopt to discrete
Sample Hausdorff distanceCarry out Monte Carlo average, according to's
Result of calculation being estimated to the estimation performance of target morphology, wherein,,It is special to cover
Carlow simulation times,Represent theThe discrete sampling Hao Siduofu that secondary Monte Carlo averagely obtains away from
From.
Beneficial effects of the present invention：Using the present invention, be not only able to the objective reaction estimated form of extension target with it is true
Similarity degree between real form estimates performance to assess it, and can effectively weigh the quality for extending target tracking algorism,
It is easy to Project Realization, with stronger using value and promotion prospect.
Description of the drawings
Fig. 1 is the flow chart of embodiment of the present invention；
Fig. 2 is extension target following trajectory diagram；
Fig. 3 is extension target following track partial approach figure；
Fig. 4 is that the extension target morphology under the influence of different measurement noises estimates performance evaluation result figure.
Specific embodiment
Understandable to enable the above objects, features and advantages of the present invention to become apparent from, below in conjunction with the accompanying drawings 1 is real with concrete
Apply mode to be described in further detail the present invention.
A kind of form for extending target following estimates performance estimating method, comprises the following steps：
Step one, hypothesis extension target, according to the complicated expanded configuration feature that extension target has, its morphologic description is adopted
With the mathematical form of support functionTo describe,,
Wherein n is the number for extending the sub-goal form that target morphology decomposes, along the angle of sightThe support function table of sub-goal on direction
It is shown as,,；Then extend targetSupport function table
It is shown as, square
The component of battle arrayTo extend the expanded configuration parameter of targetComponent,For vectorial transposition；
Step 2, hypothesis
It is to be calculated using extension target following
The target morphology parameter that method is estimated, then the description form for estimating the support function of target morphology is
；
Step 3, according to the estimated target morphology of step 2The real goal form calculated with step oneBetween discrete sampling Hausdorff distanceTo evaluate the complexity extension target
Estimate performance, wherein,,,,, a is target morphologyMiddle basisThe certain point of discrete sampling,It is expressed as point a and targetIt
Between Euclidean distance；
Step 4, adopt formulaAdopt to discrete
Sample Hausdorff distanceCarry out Monte Carlo average, according to
Result of calculation being estimated to the estimation performance of target morphology, wherein,,To cover
Special Carlow simulation times,Represent theDiscrete sampling that secondary Monte Carlo averagely obtains person of outstanding talent this
Many husband's distances.
Embodiment 1
By taking a complicated extension target as an example, its target morphology is decomposed into two oval sub-goal forms.So complex target
Support function be the simple of two oval sub-goal support functions plus and：
Wherein along the angle of sightThe support function of two oval sub-goals is expressed as on direction
Accordingly, the support function representation of complicated extension target morphology is：
So component of matrixThe expanded configuration parameter of complex target can be considered to be。
IfIt is the target estimated using extension target tracking algorism
Morphological parameters, then the support function description form for estimating target morphology is：
Estimated target morphology can be passed through based on the complicated extension target state estimator performance of support functionWith it is true
FormBetween discrete sampling Hausdorff distanceTo evaluate, this is becauseIt is that a continuum cannot be used directly, if carry out Performance Evaluation according to it be not obviously inconsistent with actual conditions.
So by the thought of even angle sampling, by continuumUniform sampling is into discrete set,
WhereinFor discrete sampling number.Therefore be used for assessing extension target morphology estimate performance discrete sampling Hao Siduofu away from
FromFor
,
Wherein
For target morphologyMiddle basisThe certain point of discrete employing, thenIt is expressed as a littleAnd targetBetween Euclidean distance.Therefore person of outstanding talent Si using discrete sampling is more
Husband's distanceCome to the target morphology for estimating
With real goal form
Carry out similarity degree contrast and estimate performance to assess it.Usually, the validity of checking extension target tracking algorism, generally
Using the method for Monte Carlo simulation.So in order to carry out overall merit to the form performance for estimating, need to discrete sampling
Hausdorff distanceCarry out Monte Carlo average, i.e.,
,
WhereinFor Monte Carlo simulation number of times,Represent theWhat secondary Monte Carlo averagely obtained
Discrete sampling Hausdorff distance.Obviously,Size reflect
The whole matching degree of estimated target morphology and real goal form, so as to according to its result of calculation come to target morphology
Estimation performance be estimated.
The present invention estimates that the effect of Performance Evaluation can be real by following emulation for extending form in target tracking algorism
Test and further illustrate：
1. simulating scenes and parameter
Consider following simulating scenes, an extension target with complicated form makees approximate along the track shown in accompanying drawing 2
Linear uniform motion,For its target initial motion state, high score
Resolution radar observation point is located at the origin of cartesian coordinate plane, the sampling period。
2. emulation content and interpretation of result
Accompanying drawing 2 gives the complicated extension target following track in this scene, and wherein Fig. 3 is the partial enlarged drawing of Fig. 2.In order to survey
Can discrete sampling Hausdorff distance estimate that performance is effectively commented to the form of complicated morphological dilation target in the examination present invention
Estimate, by carrying out for the extension target following under the influence of high measurement noise and low measurement noise different situations in this scene
100 Monte Carlo simulations.Obviously, as shown in figure 4, the discrete sampling Hao Siduofu obtained by Monte Carlo simulation average computation
DistanceIt is less, estimated form just closer to true form, so as to reach extension target with
Form estimates the purpose of Performance Evaluation in track.Generally speaking, the form for extending target following in the present invention estimates performance
Appraisal procedure, can pass through comparison object true form and estimate the illiteracy of form discrete sampling Hausdorff distance therebetween
Special Carlow mean value weighs the quality for extending target tracking algorism with effective.
Claims (1)
1. a kind of form for extending target following estimates performance estimating method, it is characterised in that：Comprise the following steps：
Step one, hypothesis extension target, according to the complicated expanded configuration feature that extension target has, its morphologic description is adopted
The mathematical form of support functionTo describe,, wherein n is to expand
The number of the sub-goal form that exhibition target morphology decomposes, along the angle of sightThe support function of sub-goal is expressed as on direction,,；Then extend targetSupport function be expressed as, matrix
ComponentTo extend the expanded configuration ginseng of target
NumberComponent,For vectorial transposition；
Step 2, hypothesis
It is to be estimated using extension target tracking algorism
Target morphology parameter, then the description form for estimating the support function of target morphology is
；
Step 3, according to the estimated target morphology of step 2The real goal form calculated with step oneBetween discrete sampling Hausdorff distanceTo evaluate the complexity extension target
Estimate performance, wherein,,,,
, a is target morphologyMiddle basisThe certain point of discrete sampling,It is expressed as point a and targetIt
Between Euclidean distance；
Step 4, adopt formulaTo discrete sampling
Hausdorff distanceCarry out Monte Carlo average, according toMeter
Calculate result to be estimated the estimation performance of target morphology, wherein,,For Monte Carlo
Simulation times,Represent theThe discrete sampling Hausdorff distance that secondary Monte Carlo averagely obtains.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610977631.8A CN106570330A (en) | 2016-11-08 | 2016-11-08 | Shape estimated performance evaluation method for extended target tracing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610977631.8A CN106570330A (en) | 2016-11-08 | 2016-11-08 | Shape estimated performance evaluation method for extended target tracing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106570330A true CN106570330A (en) | 2017-04-19 |
Family
ID=58540209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610977631.8A Pending CN106570330A (en) | 2016-11-08 | 2016-11-08 | Shape estimated performance evaluation method for extended target tracing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106570330A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111724417B (en) * | 2020-06-15 | 2022-08-02 | 中国电子科技集团公司第二十九研究所 | Fourier transform-based multi-target tracking evaluation method considering shape difference |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5842156A (en) * | 1996-11-12 | 1998-11-24 | The United States Of America As Represented By The Secretary Of The Air Force | Multirate multiresolution target tracking |
| CN103886605A (en) * | 2014-03-31 | 2014-06-25 | 江南大学 | Method for predicting and tracking moving object based on center of curvature |
| CN104075710A (en) * | 2014-04-28 | 2014-10-01 | 中国科学院光电技术研究所 | Real-time mobile extended target axial posture estimation method based on track prediction |
| CN104777465A (en) * | 2014-01-09 | 2015-07-15 | 江南大学 | Random extended object shape and state estimation method based on B spline function |
| CN105913080A (en) * | 2016-04-08 | 2016-08-31 | 西安电子科技大学昆山创新研究院 | Random matrix-based maneuvering non-ellipse expanding object combined tracking and classifying method |
-
2016
- 2016-11-08 CN CN201610977631.8A patent/CN106570330A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5842156A (en) * | 1996-11-12 | 1998-11-24 | The United States Of America As Represented By The Secretary Of The Air Force | Multirate multiresolution target tracking |
| CN104777465A (en) * | 2014-01-09 | 2015-07-15 | 江南大学 | Random extended object shape and state estimation method based on B spline function |
| CN103886605A (en) * | 2014-03-31 | 2014-06-25 | 江南大学 | Method for predicting and tracking moving object based on center of curvature |
| CN104075710A (en) * | 2014-04-28 | 2014-10-01 | 中国科学院光电技术研究所 | Real-time mobile extended target axial posture estimation method based on track prediction |
| CN105913080A (en) * | 2016-04-08 | 2016-08-31 | 西安电子科技大学昆山创新研究院 | Random matrix-based maneuvering non-ellipse expanding object combined tracking and classifying method |
Non-Patent Citations (1)
| Title |
|---|
| 李淼 等: "采用多伯努利滤波器的过采样点目标检测前跟踪", 《光学精密工程》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111724417B (en) * | 2020-06-15 | 2022-08-02 | 中国电子科技集团公司第二十九研究所 | Fourier transform-based multi-target tracking evaluation method considering shape difference |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106950562B (en) | State fusion target tracking method based on predicted value measurement conversion | |
| CN106443661B (en) | Motor-driven extension method for tracking target based on Unscented kalman filtering | |
| CN103345577B (en) | Variation Bayesian probability assumed density multi-object tracking method | |
| CN104155650B (en) | A kind of method for tracking target based on entropy weight method point mark quality evaluation | |
| CN104035083B (en) | A kind of radar target tracking method based on measurement conversion | |
| CN104730537B (en) | Infrared/laser radar data fusion target tracking method based on multi-scale model | |
| KR101914550B1 (en) | Method for tracking target position of radar | |
| CN103996171B (en) | Fluid motion vector estimation method based on space-time image | |
| EP2169422A1 (en) | System and method for acoustic tracking an underwater vehicle trajectory | |
| CN110503071A (en) | Multi-object tracking method based on the more Bernoulli Jacob's Additive Models of variation Bayes's label | |
| CN109284677A (en) | A kind of Bayesian filter target tracking algorism | |
| CN104330768B (en) | A kind of motor-driven sound bearing based on acoustic vector sensors method of estimation | |
| US9213100B1 (en) | Bearing-only tracking for horizontal linear arrays with rapid, accurate initiation and a robust track accuracy threshold | |
| CN108761387B (en) | Double-station time difference and frequency difference combined positioning method for fixed radiation source | |
| CN104504728B (en) | Multiple maneuver target tracking methods, system and broad sense JPDA device thereof | |
| CN111722214A (en) | Radar multi-target tracking PHD implementation method | |
| CN105974412A (en) | Target feature extraction method used for synthetic aperture radar | |
| CN108871365B (en) | State estimation method and system under course constraint | |
| CN105353351A (en) | Improved positioning method based on multi-beacon arrival time differences | |
| CN111623703A (en) | Novel Kalman filtering-based Beidou deformation monitoring real-time processing method | |
| CN110231620A (en) | A kind of noise correlation system tracking filter method | |
| CN114757241A (en) | Doppler parameter coupling line extraction method | |
| Kolouri et al. | Acoustic tomography of the atmosphere using unscented Kalman filter | |
| CN106707234B (en) | A kind of sensor network target localization method for combining delay inequality and angle measurement | |
| CN106168662A (en) | The error registration method of passive sensor based on Maximum-likelihood estimation and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170419 |