CN108010066A - Multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information - Google Patents

Multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information Download PDF

Info

Publication number
CN108010066A
CN108010066A CN201711177478.1A CN201711177478A CN108010066A CN 108010066 A CN108010066 A CN 108010066A CN 201711177478 A CN201711177478 A CN 201711177478A CN 108010066 A CN108010066 A CN 108010066A
Authority
CN
China
Prior art keywords
mtd
mrow
flight path
mtr
msup
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
CN201711177478.1A
Other languages
Chinese (zh)
Other versions
CN108010066B (en
Inventor
夏庆
施明绚
赵凯生
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.)
Luoyang Institute of Electro Optical Equipment AVIC
Original Assignee
Luoyang Institute of Electro Optical Equipment AVIC
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 Luoyang Institute of Electro Optical Equipment AVIC filed Critical Luoyang Institute of Electro Optical Equipment AVIC
Priority to CN201711177478.1A priority Critical patent/CN108010066B/en
Publication of CN108010066A publication Critical patent/CN108010066A/en
Application granted granted Critical
Publication of CN108010066B publication Critical patent/CN108010066B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/277—Analysis of motion involving stochastic approaches, e.g. using Kalman filters
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10048—Infrared image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30241—Trajectory

Abstract

The present invention relates to a kind of multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information, utilize infrared target gray scale cross-correlation characteristic information and angle information, flight path confidence evaluation model in multiple hypotheis tracking method is rebuild, establish the flight path confidence evaluation model based on gray scale cross-correlation and angle information, flight path confidence calculations are completed using the model, realize that the quick of multi-target traces confirms and quickly delete, reduce follow-up flight path and assume quantity, effectively improve flight path settling time and reduce algorithm complex.Therefore preferable effect can be reached by the flight path confidence evaluation model of infrared target gray scale cross-correlation and angle information, better than the multiple hypotheis tracking method using angle information route evaluation model.

Description

Multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information
Technical field
The invention belongs to technical field of data processing, is related to a kind of method for multiple target tracking, and in particular to a kind of Multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information.
Background technology
Since the retrievable information content of infrared warning system is less, general only target angle information, and obtained target There are a large amount of false-alarms and clutter in information, multiple target tracking is there are the contradiction of real-time and accuracy, it is necessary to go deep into it Research.At present, summing up multiple target tracking algorithm mainly includes global nearest-neighbor method (GNN), Joint Probabilistic Data Association Method (JPDA) and multiple hypotheis tracking method (MHT).Global nearest-neighbor method and Joint Probabilistic Data Association method calculation amount is small, real-time Height, but the target easy to be lost under highly dense clutter environment；Multiple hypotheis tracking algorithm is to solve Multi-target Data under high clutter environment The highest method of accuracy is associated, but the algorithm calculation amount is larger, and real-time is poor.
The content of the invention
Technical problems to be solved
In order to avoid the shortcomings of the prior art, the present invention proposes that one kind is based on infrared target gray scale cross-correlation and angle The multiple hypotheis tracking method of information, using infrared target gray scale cross-correlation feature and angle information, rebuilds multiple hypotheis tracking Middle flight path confidence evaluation model, realizes that the quick of multi-target traces confirms and quickly delete, effectively improves flight path settling time With reduction algorithm complex.
Technical solution
A kind of multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information, it is characterised in that：First To initial track, then step is as follows：
Step 1：Data correlation is carried out, k filtering of Kalman, the residual error of wave filter are carried out to the flight path after prediction Remaining covariance matrix S and residual norm d2It is as follows：
S=HP (k/k) HT+R(k)
In formula, H is measurement matrix,Exported for wave filter, P (k/k) is error co-variance matrix, and R (k) is Measurement noise covariance matrix, c is correlation threshold, using adaptive threshold；
Step 2, route evaluation：Flight path confidence level is calculated, is accumulated and produced by recursion, flight path confidence level is as follows：
L (k)=L (k-1)+Δ Lk
Wherein：PdFor the detection probability of target, βfFor false-alarm space exploration density, M is to measure space dimensionality, d2It is residual error Norm, S are remaining covariance matrixes, and Cov (k | k-1) it is flight path Track (k-1) associated objects 5*5 neighborhoods and flight path Track (k) cross-correlation coefficient of associated objects 5*5 neighborhoods；
It is described
Step 3, flight path are deleted：
As L (k) ＜ TL, delete flight path；
As L (k) ＞ TU, confirm flight path；
Wherein, TLConfidence threshold value, T are deleted for flight pathUFor track confirmation confidence threshold value；
Work as TL＜ L (k) ＜ TU：
1. flight path clusters：Flight path clusters generating criteria：In Different Flight class without interaction；
2. flight path is assumed：Flight path assumes generating criteria：Travel through all potential tracks in flight path cluster；An optional flight path is made For the flight path of hypothesis；Exclude and the incompatible all flight paths of this flight path；
Calculate and assume flight path confidence level；Repeat above-mentioned flight path and cluster the process assumed to flight path；
Step 4, Trajectory Prediction：Kalman prediction is carried out to the flight path after confirmation using even acceleration model CA, is completed Track filtering predicts that formula is as follows：
In formula：
Wherein, X (k-1) is dbjective state vector, and A (k-1), E (k-1) represent k-1 moment azimuth of target, pitching respectively Angular data,K-1 moment target bearings, elevating movement speed are represented respectively, Respectively Represent k-1 moment target bearings, elevating movement acceleration；F (k-1) is state-transition matrix, and G (k-1) is input matrix；T is The infrared warning system detector sampling time；W (k-1) is state-noise, is zero mean Gaussian white noise, its covariance matrix is Q(k-1)；
For next frame target data, 1~step 4 of repeat step.
Beneficial effect
A kind of multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information proposed by the present invention, utilizes Infrared target gray scale cross-correlation characteristic information and angle information, to the flight path confidence evaluation model in multiple hypotheis tracking method into Row is rebuild, and establishes the flight path confidence evaluation model based on gray scale cross-correlation and angle information, completes to navigate using the model Mark confidence calculations, realize that the quick of multi-target traces confirms and quickly delete, reduce follow-up flight path and assume quantity, effectively improve Flight path settling time and reduction algorithm complex.Therefore by infrared target gray scale cross-correlation and the flight path confidence level of angle information Evaluation model can reach preferable effect, better than the multiple hypotheis tracking method using angle information route evaluation model.
Brief description of the drawings
Fig. 1 is the method flow schematic diagram of the present invention
Embodiment
In conjunction with embodiment, attached drawing, the invention will be further described：
The experimental data source of embodiment includes 200 frame image sequences (640*512) and corresponding detection target position data.
Specific implementation step of the present invention is as follows：
1) data correlation
Data correlation have passed through the filtering of k times using the flight path after prediction.Wave filter is calculated according to Kalman filter theory Residual errorRemaining covariance matrix S and residual norm d2：
S=HP (k/k) HT+R(k) (2)
In formula, H is measurement matrix,Exported for wave filter, P (k/k) is error co-variance matrix, and R (k) is Measurement noise covariance matrix, c is correlation threshold, using adaptive threshold.
2) route evaluation
First, flight path Track (k-1) is taken in the 5*5 neighborhoods x of k-1 secondary association targetsi,j(k-1) (1≤i≤5,1≤j≤ 5) flight path Track (k), is taken in the 5*5 neighborhoods x of k secondary association targetsi,j(k) (1≤i≤5,1≤j≤5), then gray scale cross correlation Number Cov (k | k-1) it is as follows：
In formula,
Flight path confidence level is calculated, is accumulated and produced by recursion, flight path confidence level is as follows：
L (k)=L (k-1)+Δ Lk (5)
In formula：PdFor the detection probability of target, βfFor false-alarm space exploration density, M is to measure space dimensionality, d2It is residual error Norm, S are remaining covariance matrixes, and Cov (k | k-1) it is flight path Track (k-1) associated objects 5*5 neighborhoods and flight path Track (k) cross-correlation coefficient of associated objects 5*5 neighborhoods.
3) flight path is deleted
Flight path confidence level L (k) is calculated according to formula (6), according to L (k) ＜ TL, delete flight path；L (k) ＞ TU, confirm flight path；TL ＜ L (k) ＜ TU, wait more data more criterion to carry out flight path and delete.Wherein, TLConfidence threshold value, T are deleted for flight pathUFor boat Mark confirms confidence threshold value.
TL＜ L (k) ＜ TU:
1. flight path clusters
Flight path clusters generating criteria：In Different Flight class without interaction.
2. flight path is assumed
Flight path assumes generating criteria：Travel through all potential tracks in flight path cluster；An optional flight path is as the boat assumed Mark；Exclude and the incompatible all flight paths of this flight path；
Calculated using formula (5) and assume flight path confidence level；Repeat above-mentioned 1. flight path and cluster~2. flight paths hypothesis.
4) Trajectory Prediction
Kalman prediction is carried out to the flight path after confirmation using even acceleration model (CA), flight path is completed using formula (7) Filter forecasting, formula are as follows：
In formula：
Wherein, X (k-1) is dbjective state vector, and A (k-1), E (k-1) represent k-1 moment azimuth of target, pitching respectively Angular data,K-1 moment target bearings, elevating movement speed are represented respectively, Respectively Represent k-1 moment target bearings, elevating movement acceleration；F (k-1) is state-transition matrix, and G (k-1) is input matrix；T is The infrared warning system detector sampling time；W (k-1) is state-noise, is zero mean Gaussian white noise, its covariance matrix is Q(k-1)。
Receive fresh target data, repeat step 1) -5), complete multiple target tracking.

Claims (1)

  1. A kind of 1. multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information, it is characterised in that：It is right first Initial track, then step is as follows：
    Step 1：Data correlation is carried out, k filtering of Kalman, the residual error of wave filter are carried out to the flight path after predictionIt is remaining Covariance matrix S and residual norm d2It is as follows：
    <mrow> <mover> <mi>Y</mi> <mo>~</mo> </mover> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>Y</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>-</mo> <mi>H</mi> <mover> <mi>X</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>k</mi> <mo>/</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
    S=HP (k/k) HT+R(k)
    <mrow> <msup> <mi>d</mi> <mn>2</mn> </msup> <mo>=</mo> <mover> <mi>Y</mi> <mo>~</mo> </mover> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <msup> <mi>S</mi> <mrow> <mo>-</mo> <mn>1</mn> </mrow> </msup> <mover> <mi>Y</mi> <mo>~</mo> </mover> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> </mrow>
    In formula, H is measurement matrix,Exported for wave filter, P (k/k) is error co-variance matrix, and R (k) is measurement Noise covariance matrix, c is correlation threshold, using adaptive threshold；
    Step 2, route evaluation：Flight path confidence level is calculated, is accumulated and produced by recursion, flight path confidence level is as follows：
    L (k)=L (k-1)+Δ Lk
    Wherein：PdFor the detection probability of target, βfFor false-alarm space exploration density, M is to measure space dimensionality, d2It is residual norm, S is remaining covariance matrix, and Cov (k | k-1) it is that flight path Track (k-1) associated objects 5*5 neighborhoods are associated with flight path Track (k) The cross-correlation coefficient of target 5*5 neighborhoods；
    It is described
    <mrow> <msub> <mi>X</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>-</mo> <mfrac> <mn>1</mn> <mn>25</mn> </mfrac> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>5</mn> </munderover> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>5</mn> </munderover> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>X</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>-</mo> <mfrac> <mn>1</mn> <mn>25</mn> </mfrac> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>5</mn> </munderover> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>5</mn> </munderover> <msub> <mi>x</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>j</mi> </mrow> </msub> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
    Step 3, flight path are deleted：
    As L (k) ＜ TL, delete flight path；
    As L (k) ＞ TU, confirm flight path；
    Wherein, TLConfidence threshold value, T are deleted for flight pathUFor track confirmation confidence threshold value；
    Work as TL＜ L (k) ＜ TU：
    1. flight path clusters：Flight path clusters generating criteria：In Different Flight class without interaction；
    2. flight path is assumed：Flight path assumes generating criteria：Travel through all potential tracks in flight path cluster；An optional flight path is as false If flight path；Exclude and the incompatible all flight paths of this flight path；
    Calculate and assume flight path confidence level；Repeat above-mentioned flight path and cluster the process assumed to flight path；
    Step 4, Trajectory Prediction：Kalman prediction is carried out to the flight path after confirmation using even acceleration model CA, completes flight path Filter forecasting, formula are as follows：
    <mrow> <mover> <mi>X</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mi>k</mi> <mo>/</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mi>F</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mi>X</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>+</mo> <mi>G</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mi>W</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
    In formula：
    <mrow> <mi>F</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <mn>1</mn> </mtd> <mtd> <mi>T</mi> </mtd> <mtd> <mrow> <msup> <mi>T</mi> <mn>2</mn> </msup> <mo>/</mo> <mn>2</mn> </mrow> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>1</mn> </mtd> <mtd> <mi>T</mi> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>1</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>1</mn> </mtd> <mtd> <mi>T</mi> </mtd> <mtd> <mrow> <msup> <mi>T</mi> <mn>2</mn> </msup> <mo>/</mo> <mn>2</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>1</mn> </mtd> <mtd> <mi>T</mi> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>0</mn> </mtd> <mtd> <mn>1</mn> </mtd> </mtr> </mtable> </mfenced> </mrow>
    <mrow> <mi>G</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open = "[" close = "]"> <mtable> <mtr> <mtd> <mrow> <msup> <mi>T</mi> <mn>2</mn> </msup> <mo>/</mo> <mn>2</mn> </mrow> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mi>T</mi> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mn>1</mn> </mtd> <mtd> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mrow> <msup> <mi>T</mi> <mn>2</mn> </msup> <mo>/</mo> <mn>2</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mi>T</mi> </mtd> </mtr> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mn>1</mn> </mtd> </mtr> </mtable> </mfenced> </mrow>
    Wherein, X (k-1) is dbjective state vector, and A (k-1), E (k-1) represent k-1 moment azimuth of target, pitch angle number respectively According to,K-1 moment target bearings, elevating movement speed are represented respectively, Represent respectively K-1 moment target bearings, elevating movement acceleration；F (k-1) is state-transition matrix, and G (k-1) is input matrix；T is infrared The warning system detector sampling time；W (k-1) is state-noise, is zero mean Gaussian white noise, its covariance matrix is Q (k- 1)；
    For next frame target data, 1~step 4 of repeat step.
CN201711177478.1A 2017-11-23 2017-11-23 Multi-hypothesis tracking method based on infrared target gray level cross-correlation and angle information Active CN108010066B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711177478.1A CN108010066B (en) 2017-11-23 2017-11-23 Multi-hypothesis tracking method based on infrared target gray level cross-correlation and angle information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711177478.1A CN108010066B (en) 2017-11-23 2017-11-23 Multi-hypothesis tracking method based on infrared target gray level cross-correlation and angle information

Publications (2)

Publication Number Publication Date
CN108010066A true CN108010066A (en) 2018-05-08
CN108010066B CN108010066B (en) 2021-10-22

Family

ID=62053216

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711177478.1A Active CN108010066B (en) 2017-11-23 2017-11-23 Multi-hypothesis tracking method based on infrared target gray level cross-correlation and angle information

Country Status (1)

Country Link
CN (1) CN108010066B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109509214A (en) * 2018-10-15 2019-03-22 杭州电子科技大学 A kind of ship target tracking based on deep learning
CN109525220A (en) * 2018-12-10 2019-03-26 中国人民解放军国防科技大学 Gaussian mixture CPHD filtering method with track association and extraction capability
CN116228018A (en) * 2023-03-02 2023-06-06 中国人民解放军63921部队 Early warning equipment efficiency evaluation method based on non-target theoretical truth value
CN117471423A (en) * 2023-12-28 2024-01-30 安徽隼波科技有限公司 Track quality real-time assessment method based on millimeter wave radar trace information

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030014692A1 (en) * 2001-03-08 2003-01-16 California Institute Of Technology Exception analysis for multimissions
US20050147278A1 (en) * 2001-12-03 2005-07-07 Mircosoft Corporation Automatic detection and tracking of multiple individuals using multiple cues
CN103759732A (en) * 2014-01-14 2014-04-30 北京航空航天大学 Angle information assisted centralized multi-sensor multi-hypothesis tracking method
CN104965199A (en) * 2015-07-28 2015-10-07 中国人民解放军海军航空工程学院 Radar video moving object feature fusion determination method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030014692A1 (en) * 2001-03-08 2003-01-16 California Institute Of Technology Exception analysis for multimissions
US20050147278A1 (en) * 2001-12-03 2005-07-07 Mircosoft Corporation Automatic detection and tracking of multiple individuals using multiple cues
CN103759732A (en) * 2014-01-14 2014-04-30 北京航空航天大学 Angle information assisted centralized multi-sensor multi-hypothesis tracking method
CN104965199A (en) * 2015-07-28 2015-10-07 中国人民解放军海军航空工程学院 Radar video moving object feature fusion determination method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
MILOS VASIC: ""A Collaborative Sensor Fusion Algorithm for Multi-object Tracking Using a Gaussian Mixture Probability Hypothesis Density Filter"", 《2015 IEEE 18TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS》 *
S.J.DAVEY: ""A Markov model for initiating tracks with the probabilistic multi-hypothesis tracker"", 《PROCEEDINGS OF THE FIFTH INTERNATIONAL CONFERENCE ON INFORMATION FUSION. FUSION 2002. (IEEE CAT.NO.02EX5997)》 *
李范鸣等: ""引入目标灰度信息的多假设跟踪方法"", 《系统工程与电子技术》 *
汪圣利等: ""引入目标径向速度信息的多假设跟踪方法"", 《现代雷达》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109509214A (en) * 2018-10-15 2019-03-22 杭州电子科技大学 A kind of ship target tracking based on deep learning
CN109525220A (en) * 2018-12-10 2019-03-26 中国人民解放军国防科技大学 Gaussian mixture CPHD filtering method with track association and extraction capability
CN116228018A (en) * 2023-03-02 2023-06-06 中国人民解放军63921部队 Early warning equipment efficiency evaluation method based on non-target theoretical truth value
CN116228018B (en) * 2023-03-02 2024-03-01 中国人民解放军63921部队 Early warning equipment efficiency evaluation method based on non-target theoretical truth value
CN117471423A (en) * 2023-12-28 2024-01-30 安徽隼波科技有限公司 Track quality real-time assessment method based on millimeter wave radar trace information
CN117471423B (en) * 2023-12-28 2024-04-12 安徽隼波科技有限公司 Track quality real-time assessment method based on millimeter wave radar trace information

Also Published As

Publication number Publication date
CN108010066B (en) 2021-10-22

Similar Documents

Publication Publication Date Title
US10706285B2 (en) Automatic ship tracking method and system based on deep learning network and mean shift
CN101276468B (en) Computer implemented method for tracking object in sequence of frames of video
CN108010066A (en) Multiple hypotheis tracking method based on infrared target gray scale cross-correlation and angle information
CN107193012A (en) Intelligent vehicle laser radar multiple-moving target tracking method based on IMM MHT algorithms
CN106772352B (en) It is a kind of that Weak target detecting method is extended based on the PD radar of Hough and particle filter
Lu et al. Heterogeneous multi-task learning for multiple pseudo-measurement estimation to bridge GPS outages
CN110298865B (en) Space-based starry sky background weak small point target tracking method based on threshold separation clustering device
CN107193011A (en) A kind of method for being used to quickly calculate car speed in automatic driving car area-of-interest
CN103885057A (en) Self-adaptation variable-sliding-window multi-target tracking method
Qiu et al. A survey of motion-based multitarget tracking methods
CN114998276B (en) Robot dynamic obstacle real-time detection method based on three-dimensional point cloud
CN106447696A (en) Bidirectional SIFT (scale invariant feature transformation) flow motion evaluation-based large-displacement target sparse tracking method
CN103759732A (en) Angle information assisted centralized multi-sensor multi-hypothesis tracking method
CN114090718A (en) Bi-LSTM prediction and fuzzy analysis based interrupted track correlation method
Saetchnikov et al. Efficient objects tracking from an unmanned aerial vehicle
Hao et al. Asynchronous information fusion in intelligent driving systems for target tracking using cameras and radars
CN107169299B (en) Tracking method for formation target in decentralized maneuvering mode
CN111739066B (en) Visual positioning method, system and storage medium based on Gaussian process
Leung et al. Evaluating set measurement likelihoods in random-finite-set slam
CN116758153A (en) Multi-factor graph-based back-end optimization method for accurate pose acquisition of robot
Jiang et al. Underwater loop-closure detection for mechanical scanning imaging sonar by filtering the similarity matrix with probability hypothesis density filter
Neloy et al. Alpha-N-V2: Shortest path finder automated delivery robot with obstacle detection and avoiding system
CN115909086A (en) SAR target detection and identification method based on multistage enhanced network
CN113989327B (en) Single UUV target state estimation method based on convolutional neural network particle filter algorithm
CN105353352A (en) MM-PPHDF maneuvering multi-target tracking method through improved search strategy

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant