CN111077494A - Tracking method and device - Google Patents

Tracking method and device Download PDF

Info

Publication number
CN111077494A
CN111077494A CN201911182982.XA CN201911182982A CN111077494A CN 111077494 A CN111077494 A CN 111077494A CN 201911182982 A CN201911182982 A CN 201911182982A CN 111077494 A CN111077494 A CN 111077494A
Authority
CN
China
Prior art keywords
parameters
initialization
target
predicting
initialization parameters
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
Application number
CN201911182982.XA
Other languages
Chinese (zh)
Inventor
李鹏
房艺伟
余水
王文慧
邱俊达
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.)
Jiangsu University of Technology
Original Assignee
Jiangsu University of Technology
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 Jiangsu University of Technology filed Critical Jiangsu University of Technology
Priority to CN201911182982.XA priority Critical patent/CN111077494A/en
Publication of CN111077494A publication Critical patent/CN111077494A/en
Pending legal-status Critical Current

Links

Images

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
    • G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/0294—Trajectory determination or predictive filtering, e.g. target tracking or Kalman filtering

Abstract

The invention discloses a tracking method, which comprises the following steps: acquiring an initialization parameter; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters; updating the initialization parameters of the corresponding categories to obtain updated parameters; and performing mixed extraction on the updated parameters so as to observe. The method can track a single point target or an extended target, can deal with the processing of a mixed target, and can correctly track the target under the condition that the number of targets is not preset and the target is a single point target or an extended target.

Description

Tracking method and device
Technical Field
The present invention relates to a tracking technology, and in particular, to a tracking method and apparatus.
Background
In conventional object tracking, when the detected object is far from the detector, the object is processed as a point, because the object is small relative to the detector and occupies only one resolution cell of the detector, which is called a single-point object. When an object approaches the detector, the echo signal of the object left on the detector occupies a plurality of resolution units of the detector, and cannot be equivalent to a point, and the object can generate a plurality of measurements, and the object is called an extended object.
The conventional GM-PHD, when used to describe the cluster state, treats the measurements generated by one target as multiple targets, and is not accurate. When the ET-GM-PHD tracks the single-point target, the measurement number generated by the single-point target is too small, so that the single-point target cannot be divided into a measurement set, and the target is lost due to the fact that the ET-GM-PHD cannot track the single-point target.
Disclosure of Invention
In order to achieve the above purpose, the technical solution of the embodiment of the present invention is realized as follows:
according to an aspect of the embodiments of the present invention, a monitoring method and apparatus are provided, the method including: acquiring an initialization parameter; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters; updating the initialization parameters of the corresponding categories to obtain updated parameters; and performing mixed extraction on the updated parameters so as to observe.
In the foregoing solution, predicting the initialization parameter and obtaining a corresponding category of the initialization parameter includes: predicting the state of the initialization parameter to obtain a prediction result; if the prediction result meets a first preset condition, determining the prediction result to be a first category; and if the prediction result meets a second preset condition, determining the prediction result to be in a second category.
In the above scheme, the method further comprises: adopting GM-PHD as the initialization parameter of the first category; and the initialization parameter of the second category adopts ET-GM-PHD.
In the above scheme, after the observation, the method further comprises: acquiring an observation result within a preset time; and if the observation result does not meet the observation condition, predicting the initialization parameters again.
In the above scheme, acquiring the initialization parameters includes setting an initial target state ξ0＝{m0,P0In which m is0Is the position of the target, P0Is the target motion noise covariance. And setting parameters assuming that Q and R are the covariance of the state noise and the covariance of the measurement noise respectively
Figure BDA0002291752460000025
Wherein
Figure BDA0002291752460000026
The maximum distance between measurements is generated for the same extended target.
In the foregoing solution, predicting the initialization parameter, and obtaining the corresponding category of the initialization parameter includes: when k is more than or equal to 1, the measurement set Z is measuredkCarrying out pretreatment; for metrology set Z generated at time kkGrouping measurements of proximity into one class; if a certain measurement class
Figure BDA0002291752460000021
Satisfy the requirement of
Figure BDA0002291752460000022
Then
Figure BDA0002291752460000023
Otherwise
Figure BDA0002291752460000024
Where the number of elements of the set, Zk,TSet of measurements, Z, representing possible point targetsk,ETA set of metrics representing possible extended targets.
In the foregoing solution, performing mixed extraction on the updated parameters to perform observation further includes: for each Gaussian component in the state sets of the single-point target and the extended target, if the weight is smaller than a given threshold value, the weight is discarded, and if the Mahalanobis distance between different Gaussian components is smaller than the given threshold value, the states are combined by a weighted average rule.
In the foregoing solution, performing mixed extraction on the updated parameters to perform observation further includes: if the weight is more than 0.5, the state extraction is carried out, and the system output is the detected target
According to another aspect of the embodiments of the present invention, there is provided a tracking apparatus, the apparatus including: an acquisition unit, configured to acquire an initialization parameter; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters; the updating unit is used for updating the initialization parameters of the corresponding categories and acquiring the updated parameters; and the extraction unit is used for carrying out mixed extraction on the updated parameters so as to observe.
According to another aspect of the embodiments of the present invention, there is provided a tracking apparatus, the apparatus including: a memory, a processor and a responsive program stored in the memory for movement by the processor, the processor being responsive to the steps of any of the above-described tracking methods when executing the responsive program.
The invention provides a tracking method and a tracking device, which are used for acquiring initialization parameters; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters; updating the initialization parameters of the corresponding categories to obtain updated parameters; and performing mixed extraction on the updated parameters so as to observe. The method can track a single point target or an extended target, can deal with the processing of a mixed target, and can correctly track the target under the condition that the target number is not preset and the target is a single point target or an extended target.
Drawings
Fig. 1 is a schematic flow chart illustrating an implementation of a monitoring method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of another implementation provided by the embodiment of the present invention;
FIG. 3 is a schematic flow chart of another implementation provided by the embodiment of the present invention;
FIG. 4 is a schematic diagram of another implementation flow provided in the embodiment of the present invention
FIG. 5 is a schematic structural diagram of a monitoring device according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating the effect of C-GM-PHD on hybrid target tracking in accordance with an embodiment of the present invention;
FIG. 7 is a comparison of the number of mixed target trails by C-GM-PHD in accordance with an embodiment of the present invention.
Detailed Description
So that the manner in which the features and aspects of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the appended drawings.
Fig. 1 is a schematic flow chart of an implementation of a credit tracking method according to an embodiment of the present invention, as shown in fig. 1, the method includes:
step S101, acquiring initialization parameters;
in the present application, let initial time k be 0, and single-point target initialization parameter m0,T,w0,T,P0,TExpanding the target initialization parameter m0,ET,w0,ET,P0,ET。
Step S102, predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters;
in the application, when k is more than or equal to 1, the state is predicted in parallel;
(2.1) predicting the next time PHD of the corresponding component of the single-point target:
Figure BDA0002291752460000041
wherein the content of the first and second substances,
Figure BDA0002291752460000042
Figure BDA0002291752460000043
wherein, Jk-1Representing the number of predicted objects, Fk-1Representing a state transition matrix.
(2.2) predicting the next time PHD of the extension target corresponding component:
Figure RE-GDA0002417639690000051
wherein the content of the first and second substances,
Figure BDA0002291752460000045
Figure BDA0002291752460000046
where d is the physical space dimension, I is the d-dimensional identity matrix, the symbol
Figure BDA0002291752460000049
Representing the new matrix obtained by multiplying all the elements of the left matrix by the right matrix.
Preprocessing the measurement set to generate a single-point target measurement set Zk,TAnd extending the target metrology set Zk,ET；
(3.1) for the metrology set Z generated at time kkTaking two measurements from any of the two measurements
Figure BDA0002291752460000047
And
Figure BDA0002291752460000048
calculate the distance between them if
Figure BDA0002291752460000051
Then it is grouped into one
Figure BDA0002291752460000052
If there is Z for any measurement of this typekIf other measurements satisfy the relationship, the clustering process is continued until all measurements are clustered at the moment;
(3.2) if a certain measurement is similar
Figure BDA0002291752460000053
Satisfy the requirement of
Figure BDA0002291752460000054
Then
Figure BDA0002291752460000055
Otherwise
Figure BDA0002291752460000056
Where | is the number of elements of the set, Zk,TSet of measurements, Z, representing possible point targetsk,ETA set of metrics representing possible extended targets.
Step S103, updating the initialization parameters of the corresponding category, and acquiring the updated parameters.
In this application, (4.1) the state of the single point object is updated,
Figure BDA0002291752460000057
wherein the content of the first and second substances,
Figure BDA0002291752460000058
Figure BDA0002291752460000059
Figure BDA00022917524600000510
Figure BDA00022917524600000511
the weight is updated to
Figure BDA00022917524600000512
Wherein, ckIs the probability of a clutter of the scene,
Figure BDA00022917524600000513
is a Gaussian probability density function, pDIs the sensor detection probability.
(4.2) updating the state of the extension point target;
(4.2a) to Zk,ETDividing, given a set of distance thresholds
Figure BDA00022917524600000514
Clustering process and procedure for each threshold3(3.1) same, the partition subset is denoted W;
(4.2b) PHD of extended target update is:
Figure BDA0002291752460000061
wherein the content of the first and second substances,
Figure BDA0002291752460000062
Figure BDA0002291752460000063
wherein the content of the first and second substances,
Figure BDA0002291752460000064
Figure BDA0002291752460000065
Figure BDA0002291752460000066
Figure BDA0002291752460000067
Figure BDA0002291752460000068
the weight is updated to
Figure BDA0002291752460000069
Figure BDA00022917524600000610
Figure BDA00022917524600000611
Figure BDA00022917524600000612
Wherein, γ(j)Is the measurement rate of the extended target.
And step S104, performing mixed extraction on the updated parameters so as to observe.
In the present application, component pruning, state extraction:
(5.1) for each Gaussian component in the state sets of the single-point target and the extended target, if the weight is smaller than a given threshold value, the weight is discarded, and if the Mahalanobis distance between different Gaussian components is smaller than the given threshold value, the states are combined by a weighted average rule; and if the weight is more than 0.5, performing state extraction, and outputting the system as the detected target.
And (5.2) for each Gaussian component in the state set of the single-point target and the extended target, if the weight is more than 0.5, performing state extraction, and outputting the system as the detected target.
In another embodiment, as shown in fig. 2, predicting the initialization parameters and obtaining the corresponding category of the initialization parameters includes:
predicting the state of the initialization parameter to obtain a prediction result; if the prediction result meets a first preset condition, determining the prediction result to be a first category; and if the prediction result meets a second preset condition, determining the prediction result to be in a second category.
In another embodiment, further comprising: adopting GM-PHD as the initialization parameter of the first category; and the initialization parameter of the second category adopts ET-GM-PHD.
In another embodiment, as shown in fig. 3, the observing further includes: acquiring an observation result within a preset time; and if the observation result does not meet the observation condition, predicting the initialization parameter again.
In another embodiment, obtaining initialization parameters comprises:
set initial target state ξ0＝{m0,P0In which m is0Is the position of the target, P0Is the target motion noise covariance. And assuming Q and R are covariance of state noise and covariance of measurement noise, respectively, setting parameters
Figure BDA0002291752460000071
Wherein
Figure BDA0002291752460000072
The maximum distance between measurements is generated for the same extended target.
In another embodiment, predicting the initialization parameters, and obtaining the corresponding category of the initialization parameters includes:
when k is more than or equal to 1, the measurement set Z is measuredkCarrying out pretreatment; for metrology set Z generated at time kkGrouping close measurements into one class; if a certain measurement class
Figure BDA0002291752460000081
Satisfy the requirement of
Figure BDA0002291752460000082
Then
Figure BDA0002291752460000083
Otherwise
Figure BDA0002291752460000084
Where the number of elements of the set, Zk,TSet of measurements, Z, representing possible point targetsk,ETA set of metrics representing possible extended targets.
In another embodiment, performing hybrid extraction on the updated parameters to perform observation further includes:
for each Gaussian component in the state sets of the single-point target and the extended target, if the weight is smaller than a given threshold value, the weight is discarded, and if the Mahalanobis distance between different Gaussian components is smaller than the given threshold value, the states are combined by a weighted average rule.
In another embodiment, performing hybrid extraction on the updated parameters to perform observation further includes:
and if the weight is more than 0.5, performing state extraction, and outputting the system as the detected target.
In another embodiment, as shown in fig. 4, it is a detailed flow chart of the whole method, i.e. the GM-PHD tracks single point and extended target hybrid scenarios.
In another embodiment, the apparatus comprises: an acquisition unit configured to acquire an initialization parameter; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters; the updating unit is used for updating the initialization parameters of the corresponding categories and acquiring the updated parameters; and the extraction unit is used for carrying out mixed extraction on the updated parameters so as to observe.
In another embodiment, the apparatus comprises: a memory, a processor, and a responsive program stored in the memory for movement by the processor, wherein the processor is responsive to the steps of the tracking method when executing the responsive program.
It should be noted that: in the data processing apparatus provided in the above embodiment, when the program is developed, only the division of each program module is exemplified, and in practical applications, the above processing may be distributed to different program modules according to needs, that is, the internal structure of the data processing apparatus may be divided into different program modules to complete all or part of the above-described processing. In addition, the data processing apparatus provided in the above embodiment and the data processing method embodiment belong to the same concept, and specific implementation processes thereof are detailed in the method embodiment and are not described herein again.
Fig. 5 is a schematic structural diagram of a data processing device in an embodiment of the present invention, and as shown in fig. 5, the data processing device 500 may be a handle, a mouse, a trackball, a mobile phone, a smart pen, a smart watch, a smart ring, a smart bracelet, a smart glove, or the like. The data processing apparatus 500 shown in fig. 3 includes: at least one processor 501, memory 502, at least one network interface 504, and a user interface 503. The various components in the data processing device 500 are coupled together by a bus system 505. It is understood that the bus system 505 is used to enable connection communications between these components. The bus system 505 includes a power bus, a control bus, and a status signal bus in addition to a data bus. For clarity of illustration, however, the various buses are labeled as bus system 505 in FIG. 5.
The user interface 503 may include a display, a keyboard, a mouse, a trackball, a click wheel, a key, a button, a touch pad, a touch screen, or the like, among others.
It will be appreciated that the memory 502 can be either volatile memory or nonvolatile memory, and can include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic random access Memory (FRAM), a Flash Memory (Flash Memory), a magnetic surface Memory, an optical Disc, or a Compact Disc Read-Only Memory (CD-ROM); the magnetic surface memory may beMagnetic disk memoryOrMagnetic tape memory. The volatile Memory may be a Random Access Memory (RAM) which serves as an external cache. By way of illustration, and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM, Enhanced Synchronous Dynamic Random Access Memory (DRDRAM), Synchronous linked Dynamic Random Access Memory (SLS DRAM, direct DRAM),direct Rambus Random Access Memory). The memory 302 described in connection with the embodiments of the invention is intended to comprise, without being limited to, these and any other suitable types of memory.
The memory 502 in embodiments of the present invention is used to store various types of data to support the operation of the data processing apparatus 500. Examples of such data include: any computer programs for operating on the data processing apparatus 500, such as an operating system 5021 and application programs 5022; music data; animation data; book information; video, drawing information, etc. The operating system 5021 includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, and is used for implementing various basic services and processing hardware-based tasks. The application 5022 may contain various applications such as a media player (MediaPlayer), a Browser (Browser), etc., for implementing various application services. The program for implementing the method according to the embodiment of the present invention may be included in the application program 5022.
The method disclosed by the above-mentioned embodiments of the present invention may be applied to the processor 501, or implemented by the processor 501. The processor 501 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be implemented by integrated logic circuits of hardware or instructions in the form of software in the processor 501. The Processor 501 may be a general purpose Processor, a Digital Signal Processor (DSP), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc. Processor 501 may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed in the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium located in the memory 502 and the processor 501 reads the information in the memory 302 and in combination with its hardware performs the steps of the method described above.
In an exemplary embodiment, the data processing apparatus 500 may be implemented by one or more Application Specific Integrated Circuits (ASICs), DSPs, Programmable Logic Devices (PLDs), Complex Programmable Logic Devices (CPLDs), Field Programmable Gate Arrays (FPGAs), general purpose processors, controllers, Micro Controllers (MCUs), microprocessors (microprocessors), or other electronic components for performing the foregoing methods.
Specifically, when the processor 501 runs the computer program, it executes: acquiring an initialization parameter; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters; updating the initialization parameters of the corresponding categories to obtain updated parameters; and performing mixed extraction on the updated parameters so as to observe.
When the processor 501 runs the computer program, it further executes: predicting the initialization parameters and obtaining corresponding categories of the initialization parameters, wherein the steps comprise: predicting the state of the initialization parameter to obtain a prediction result; if the prediction result meets a first preset condition, determining the prediction result to be a first category; and if the prediction result meets a second preset condition, determining the prediction result to be a second category.
When the processor 501 runs the computer program, it further executes: further comprising: adopting GM-PHD as the initialization parameter of the first category; and the initialization parameter of the second category adopts ET-GM-PHD.
When the processor 501 runs the computer program, it further executes: after the observation, the method further comprises the following steps: acquiring an observation result within a preset time; and if the observation result does not meet the observation condition, predicting the initialization parameter again.
When the processor 501 runs the computer program, it further executes that acquiring the initialization parameter includes setting an initial target state ξ0＝{m0,P0In which m is0Is the position of the target, P0The target motion noise covariance is taken. And assuming Q and R as state noise respectivelyCovariance of covariance and measured noise, setting parameters
Figure BDA0002291752460000125
Wherein
Figure BDA0002291752460000126
The maximum distance between measurements is generated for the same extended target.
When the processor 501 runs the computer program, it further executes: predicting the initialization parameters, and acquiring corresponding categories of the initialization parameters comprises: when k is more than or equal to 1, the measurement set Z is measuredkCarrying out pretreatment; for metrology set Z generated at time kkGrouping close measurements into one class; if a certain measurement class
Figure BDA0002291752460000121
Satisfy the requirement of
Figure BDA0002291752460000122
Then
Figure BDA0002291752460000123
Otherwise
Figure BDA0002291752460000124
Where the number of elements of the set, Zk,TSet of measurements, Z, representing possible point targetsk,ETA set of metrics representing possible extended targets.
When the processor 501 runs the computer program, it further executes: performing mixed extraction on the updated parameters so as to perform observation, further comprising: for each Gaussian component in the state sets of the single-point target and the extended target, if the weight is smaller than a given threshold value, the weight is discarded, and if the Mahalanobis distance between different Gaussian components is smaller than the given threshold value, the states are combined by a weighted average rule.
When the processor 501 runs the computer program, it further executes: performing mixed extraction on the updated parameters so as to perform observation, further comprising: and if the weight is more than 0.5, performing state extraction, and outputting the system as the detected target.
In an exemplary embodiment, the present invention further provides a computer readable storage medium, such as a memory 502, comprising a computer program, which is executable by a processor 501 of a data processing apparatus 500 to perform the steps of the aforementioned method. The computer readable storage medium may be memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flashmemory, magnetic surface memory, optical disk, or CD-ROM; or a variety of devices, such as mobile phones, computers, tablet devices, personal digital assistants, etc., that include one or any combination of the above-mentioned memories.
A computer-readable storage medium, having stored thereon a computer program which, when executed by a processor, performs: acquiring an initialization parameter; predicting the initialization parameters and acquiring the corresponding categories of the initialization parameters; updating the initialization parameters of the corresponding categories to obtain updated parameters; and performing mixed extraction on the updated parameters so as to observe.
The computer program, when executed by the processor, further performs: predicting the initialization parameters and obtaining corresponding categories of the initialization parameters, wherein the corresponding categories comprise: predicting the state of the initialization parameter to obtain a prediction result; if the prediction result meets a first preset condition, determining the prediction result to be a first category; and if the prediction result meets a second preset condition, determining the prediction result to be in a second category.
The computer program, when executed by the processor, further performs: further comprising: adopting GM-PHD as the initialization parameter of the first category; and the initialization parameter of the second category adopts ET-GM-PHD.
The computer program, when executed by the processor, further performs: after the observation, the method further comprises the following steps: acquiring an observation result within a preset time; and if the observation result does not meet the observation condition, predicting the initialization parameters again.
The computer program, when executed by the processor, further performs obtaining initialization parameters including setting an initial target state ξ0＝{m0,P0In which m is0Is the position of the object and is,P0is the target motion noise covariance. And setting parameters assuming that Q and R are the covariance of the state noise and the covariance of the measurement noise respectively
Figure BDA0002291752460000135
Wherein
Figure BDA0002291752460000137
The maximum distance between measurements is generated for the same extended target.
The computer program, when executed by the processor, further performs: predicting the initialization parameters, and acquiring corresponding categories of the initialization parameters comprises: when k is more than or equal to 1, the measurement set Z is measuredkCarrying out pretreatment; for metrology set Z generated at time kkGrouping close measurements into one class; if a certain measurement class
Figure BDA0002291752460000131
Satisfy the requirement of
Figure BDA0002291752460000132
Then
Figure BDA0002291752460000133
Otherwise
Figure BDA0002291752460000134
Where the number of elements of the set, Zk,TSet of measurements, Z, representing possible point targetsk,ETA set of metrics representing possible extended targets.
The computer program, when executed by the processor, further performs: performing mixed extraction on the updated parameters so as to perform observation, further comprising: for each Gaussian component in the state sets of the single-point target and the extended target, if the weight is smaller than a given threshold value, the weight is discarded, and if the Mahalanobis distance between different Gaussian components is smaller than the given threshold value, the states are combined by a weighted average rule.
The computer program, when executed by the processor, further performs: performing mixed extraction on the updated parameters so as to perform observation, further comprising: and if the weight is more than 0.5, performing state extraction, and outputting the system as the detected target.
In one embodiment, as shown in FIG. 6, the effect of C-GM-PHD on hybrid target tracking is shown. Where the red line represents the true trajectory, the blue circle represents the estimated target coordinates, and the black point represents the measurement. It can be seen that when the C-GM-PHD is used for tracking the mixed target, the algorithm can better track the scene target, and the performance is better than that of the GM-PHD and ET-GM-PHD algorithms which are used independently.
In one embodiment, as shown in FIG. 7, a comparison of the number of C-GM-PHD versus hybrid target tracking is shown. The black solid line represents the real number of the targets, and the circle represents the estimated number of the targets, so that the estimation performance of the number of the targets of the proposed algorithm is better than that of the GM-PHD algorithm and the ET-GM-PHD algorithm which are used independently.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims (10)

1. A tracking method, characterized in that the method comprises:
acquiring an initialization parameter;
predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters;
updating the initialization parameters of the corresponding categories to obtain updated parameters;
and performing mixed extraction on the updated parameters so as to observe.
2. The method of claim 1, wherein predicting the initialization parameters and obtaining corresponding categories of initialization parameters comprises:
predicting the state of the initialization parameter to obtain a prediction result;
if the prediction result meets a first preset condition, determining the prediction result to be a first category;
and if the prediction result meets a second preset condition, determining the prediction result to be in a second category.
3. The method of claim 2, further comprising:
adopting GM-PHD as the initialization parameter of the first category;
and the initialization parameter of the second category adopts ET-GM-PHD.
4. The method of claim 3, further comprising, after the observing:
acquiring an observation result within a preset time;
and if the observation result does not meet the observation condition, predicting the initialization parameter again.
5. The method of claim 1, wherein obtaining initialization parameters comprises:
set initial target state ξ0＝{m0,P0In which m is0Is the position of the target, P0Is the target motion noise covariance. And setting parameters assuming that Q and R are the covariance of the state noise and the covariance of the measurement noise respectively
Figure FDA0002291752450000011
Wherein
Figure FDA0002291752450000012
The maximum distance between measurements is generated for the same extended target.
6. The method of claim 1, wherein predicting the initialization parameters and obtaining corresponding categories of initialization parameters comprises:
when k is more than or equal to 1, the measurement set Z is measuredkCarrying out pretreatment; for metrology set Z generated at time kkWill be close toMeasuring the aggregation as one type; if a certain measurement class
Figure FDA0002291752450000021
Satisfy the requirement of
Figure FDA0002291752450000022
Then
Figure FDA0002291752450000023
Otherwise
Figure FDA0002291752450000024
Where the number of elements of the set, Zk,TSet of measurements, Z, representing possible point targetsk,ETA set of metrics representing possible extended targets.
7. The method of claim 1, wherein performing hybrid extraction of the updated parameters for observation further comprises:
for each Gaussian component in the state sets of the single-point target and the extended target, if the weight is smaller than a given threshold value, the weight is discarded, and if the Mahalanobis distance between different Gaussian components is smaller than the given threshold value, the states are combined by a weighted average rule.
8. The method of claim 1, wherein performing hybrid extraction of the updated parameters for observation further comprises:
and if the weight is more than 0.5, performing state extraction, and outputting the system as the detected target.
9. A tracking apparatus, characterized in that the apparatus comprises:
an acquisition unit configured to acquire an initialization parameter; predicting the initialization parameters and obtaining the corresponding categories of the initialization parameters;
the updating unit is used for updating the initialization parameters of the corresponding categories and acquiring the updated parameters;
and the extraction unit is used for carrying out mixed extraction on the updated parameters so as to observe.
10. A tracking apparatus, characterized in that the apparatus comprises: memory, processor and a responsive program stored in the memory for movement by the processor, wherein the processor is responsive to the steps of the tracking method of any of claims 1 to 8 when running the responsive program.
CN201911182982.XA 2019-11-27 2019-11-27 Tracking method and device Pending CN111077494A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911182982.XA CN111077494A (en) 2019-11-27 2019-11-27 Tracking method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911182982.XA CN111077494A (en) 2019-11-27 2019-11-27 Tracking method and device

Publications (1)

Publication Number Publication Date
CN111077494A true CN111077494A (en) 2020-04-28

Family

ID=70311855

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911182982.XA Pending CN111077494A (en) 2019-11-27 2019-11-27 Tracking method and device

Country Status (1)

Country Link
CN (1) CN111077494A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160109566A1 (en) * 2014-10-21 2016-04-21 Texas Instruments Incorporated Camera Assisted Tracking of Objects in a Radar System
CN108445480A (en) * 2018-02-02 2018-08-24 重庆邮电大学 Mobile platform based on laser radar adaptively extends Target Tracking System and method
CN109509207A (en) * 2018-11-09 2019-03-22 电子科技大学 The method that a kind of pair of point target and extension target carry out seamless tracking
CN109671096A (en) * 2017-10-13 2019-04-23 南京航空航天大学 A kind of space-time neighbour target detection and Grid Clustering measure more extension method for tracking target under dividing
CN109917372A (en) * 2018-12-28 2019-06-21 江苏理工学院 Extension target based on target prediction measures collection division and tracking

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160109566A1 (en) * 2014-10-21 2016-04-21 Texas Instruments Incorporated Camera Assisted Tracking of Objects in a Radar System
CN109671096A (en) * 2017-10-13 2019-04-23 南京航空航天大学 A kind of space-time neighbour target detection and Grid Clustering measure more extension method for tracking target under dividing
CN108445480A (en) * 2018-02-02 2018-08-24 重庆邮电大学 Mobile platform based on laser radar adaptively extends Target Tracking System and method
CN109509207A (en) * 2018-11-09 2019-03-22 电子科技大学 The method that a kind of pair of point target and extension target carry out seamless tracking
CN109917372A (en) * 2018-12-28 2019-06-21 江苏理工学院 Extension target based on target prediction measures collection division and tracking

Similar Documents

Publication Publication Date Title
CN110910422A (en) Target tracking method and device, electronic equipment and readable storage medium
CN111159413A (en) Log clustering method, device, equipment and storage medium
KR102346815B1 (en) A system to mitigate against adversarial samples for ml and ai models
CN115510982A (en) Clustering method, device, equipment and computer storage medium
CN108108769B (en) Data classification method and device and storage medium
CN116822651A (en) Large model parameter fine adjustment method, device, equipment and medium based on incremental learning
TWI713019B (en) Data label generation, model training, event recognition method and device
CN116797973A (en) Data mining method and system applied to sanitation intelligent management platform
CN111077494A (en) Tracking method and device
CN114155388B (en) Image recognition method and device, computer equipment and storage medium
CN115827918A (en) Method, device, storage medium and electronic equipment for executing service
KR102242042B1 (en) Method, apparatus and computer program for data labeling
CN111951791A (en) Voiceprint recognition model training method, recognition method, electronic device and storage medium
CN113159100B (en) Circuit fault diagnosis method, circuit fault diagnosis device, electronic equipment and storage medium
CN114792256B (en) Crowd expansion method and device based on model selection
CN116580063B (en) Target tracking method, target tracking device, electronic equipment and storage medium
CN112052861A (en) Method for calculating effective receptive field of deep convolutional neural network and storage medium
CN115408449B (en) User behavior processing method, device and equipment
CN113360356B (en) Method for identifying reading cheating behaviors, computing device and computer storage medium
US10983892B2 (en) Method and system for estimating efficiency of machine vision
CN110991543B (en) Image region of interest clustering method and device, computing device and storage medium
EP3955165A1 (en) Method and apparatus with convolution operation processing based on redundancy reduction
CN117520737A (en) Time sequence data detection method and device, electronic equipment and storage medium
CN113827981A (en) Game loss user prediction method and system based on naive Bayes
CN116912760A (en) Internet of things data processing method and device, electronic equipment and storage medium

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

Application publication date: 20200428

RJ01 Rejection of invention patent application after publication