CN113514824B - Multi-target tracking method and device for safety and lightning protection - Google Patents
Multi-target tracking method and device for safety and lightning protection Download PDFInfo
- Publication number
- CN113514824B CN113514824B CN202110762057.5A CN202110762057A CN113514824B CN 113514824 B CN113514824 B CN 113514824B CN 202110762057 A CN202110762057 A CN 202110762057A CN 113514824 B CN113514824 B CN 113514824B
- Authority
- CN
- China
- Prior art keywords
- target
- state information
- measurement
- objective function
- bayesian
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
- G01S13/72—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar
- G01S13/723—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar by using numerical data
- G01S13/726—Multiple target tracking
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a multi-target tracking method and a device for security radar, wherein the method comprises the following steps: obtaining a plurality of target state information and measurement state information in a current scanning period; generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable; determining the message sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent by each leaf node to the leaf nodes of the adjacent layers; and carrying out security radar multi-target tracking according to the Bayesian free energy objective function. The invention can carry out multi-target tracking of safety radar, effectively improve the operation speed while guaranteeing the anti-interference capability, ensure the real-time performance of the radar during tracking, and avoid the problem of track merging in the process of tracking adjacent targets.
Description
Technical Field
The invention relates to the technical field of radar target analysis, in particular to a radar-safe multi-target tracking method and device.
Background
In recent years, with the complex and changeable application environments, security radar is required to have multi-target tracking capability and simultaneously realize multi-target tracking.
The prior art generally adopts a joint probability data association algorithm to carry out multi-target tracking of safety and lightning protection. The data association aims at determining the association relation between measurement and targets, and is a critical problem in target tracking. Representative association algorithms include nearest neighbor data association (NN), joint probability data association (multi-hypothesis tracking algorithm), multi-hypothesis data association (MHT), track splitting, and the like. Compared with other association algorithms, the joint probability data association (multi-hypothesis tracking algorithm) is more suitable for multi-target tracking in false alarm and clutter environments, has stronger anti-interference capability, but when the targets are more complex, the calculated amount has an exponential multiplication trend along with the linear increase of the targets, and has the problems of large calculated amount and poor instantaneity, and the problem of track merging can occur in the process of tracking adjacent targets.
Thus, there is a need for a secure radar multi-target tracking scheme that overcomes the above problems.
Disclosure of Invention
The embodiment of the invention provides a multi-target tracking method of a security radar, which is used for carrying out multi-target tracking of the security radar, effectively improving the operation speed while guaranteeing the anti-interference capability, guaranteeing the real-time performance of the radar during tracking and avoiding the problem of track merging in the process of tracking adjacent targets, and comprises the following steps:
obtaining a plurality of target state information and measurement state information in a current scanning period;
generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable;
determining the message sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; the message comprises measurement association indicating variable information and edge probability distribution of measurement association indicating variables;
establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent by each leaf node to the leaf nodes of the adjacent layers;
according to the Bayesian free energy target function, carrying out security radar multi-target tracking;
according to the Bayesian free energy objective function, performing multi-target tracking of safety radar, comprising:
optimizing and solving a Bayesian free energy objective function, wherein the objective function meets the convex optimization condition, so that the objective function converges to the maximum value, and solving the approximate solution is as follows:
obtaining an approximate solution of the objective function by optimizing and solving the objective function, and determining an objective x according to a calculation result i And latest measurementThe probability of correlation between the two, and determining the target x through the probability result of correlation i Corresponding measurement data in the s-th scan, thereby updating the target x i Is a track of (a);
generating a probability map model according to the plurality of target state information and the measurement state information, wherein the probability map model comprises the following components:
determining corresponding target association indicating variables and measurement association indicating variables according to the plurality of target state information and measurement state information;
and establishing a probability graph model among the target state, the target association indicating variable and the measurement association indicating variable, wherein the target state, the target association indicating variable and the measurement association indicating variable are respectively used as leaf nodes of a first layer, a second layer and a third layer of the probability graph model.
The embodiment of the invention provides a multi-target tracking device of a security radar, which is used for carrying out multi-target tracking of the security radar, effectively improving operation speed while guaranteeing anti-interference capability, guaranteeing real-time performance of the radar during tracking, and avoiding the problem of track merging in the process of tracking adjacent targets, and comprises the following components:
the information acquisition module is used for acquiring a plurality of target state information and measurement state information in the current scanning period;
the model generation module is used for generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable;
the message determining module is used for determining messages sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; the message comprises measurement association indicating variable information and edge probability distribution of measurement association indicating variables;
the function building module is used for building a Bayesian free energy objective function by utilizing a variation inference algorithm according to the information sent to the leaf nodes of the adjacent layers by the leaf nodes;
the target tracking module is used for carrying out multi-target tracking of safety protection according to the Bayesian free energy target function;
the target tracking module is specifically used for:
optimizing and solving a Bayesian free energy objective function, wherein the objective function meets the convex optimization condition, so that the objective function converges to the maximum value, and solving the approximate solution is as follows:
obtaining an approximate solution of the objective function by optimizing and solving the objective function, and determining an objective x according to a calculation result i And latest measurementThe probability of correlation between the two, and determining the target x through the probability result of correlation i Corresponding measurement data in the s-th scan, thereby updating the target x i Is a track of (a);
the model generation module is used for determining corresponding target association indicating variables and measurement association indicating variables according to the plurality of target state information and measurement state information; and establishing a probability graph model among the target state, the target association indicating variable and the measurement association indicating variable, wherein the target state, the target association indicating variable and the measurement association indicating variable are respectively used as leaf nodes of a first layer, a second layer and a third layer of the probability graph model.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the multi-target tracking method of the security radar.
The embodiment of the invention also provides a computer readable storage medium, which stores a computer program for executing the multi-target tracking method of the security radar.
Compared with the scheme of carrying out multi-target tracking of safety and lightning protection through a joint probability data association algorithm in the prior art, the embodiment of the invention obtains a plurality of target state information and measurement state information in the current scanning period; generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable; determining the message sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent by each leaf node to the leaf nodes of the adjacent layers; and carrying out security radar multi-target tracking according to the Bayesian free energy objective function. According to the embodiment of the invention, a probability graph model is generated according to a plurality of target state information and measurement state information, and the message updating rule is utilized to determine the message sent to the leaf nodes of the adjacent layer by each leaf node in the probability graph model, so that a Bayesian free energy target function is established by using a variation inference algorithm, multiple target tracking of safety radar is performed, the anti-interference capability is ensured, the operation speed is effectively improved, the real-time performance of the radar during tracking is ensured, and the problem of track merging in the process of tracking adjacent targets is avoided.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 is a schematic diagram of a multi-target tracking method of a security radar in an embodiment of the invention;
FIGS. 2-6 are schematic diagrams of multi-target tracking of a security radar in accordance with embodiments of the present invention;
FIG. 7 is a diagram of a multi-target tracking device of a security radar in an embodiment of the present invention;
FIG. 8 is a schematic diagram of a computer device according to an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.
In order to perform multi-target tracking of security radar, effectively improve operation speed while ensuring anti-interference capability, ensure real-time performance of the radar during tracking, and avoid the problem of track merging in the process of tracking adjacent targets, the embodiment of the invention provides a multi-target tracking method of security radar, as shown in fig. 1, which can comprise the following steps:
step 101, obtaining a plurality of target state information and measurement state information in a current scanning period;
102, generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable;
step 103, determining the message sent to the adjacent layer leaf node by each leaf node in the probability graph model by using the message updating rule;
104, establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent to the leaf nodes of the adjacent layers by each leaf node;
and 105, carrying out security radar multi-target tracking according to the Bayesian free energy target function.
As can be seen from fig. 1, the embodiment of the present invention obtains a plurality of target state information and measurement state information in the current scanning period; generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable; determining the message sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent by each leaf node to the leaf nodes of the adjacent layers; and carrying out security radar multi-target tracking according to the Bayesian free energy objective function. According to the embodiment of the invention, a probability graph model is generated according to a plurality of target state information and measurement state information, and the message updating rule is utilized to determine the message sent to the leaf nodes of the adjacent layer by each leaf node in the probability graph model, so that a Bayesian free energy target function is established by using a variation inference algorithm, multiple target tracking of safety radar is performed, the anti-interference capability is ensured, the operation speed is effectively improved, the real-time performance of the radar during tracking is ensured, and the problem of track merging in the process of tracking adjacent targets is avoided.
The probability map model (graphic models) is a generalized uncertain knowledge representation and processing method. It uses a graph-based approach to represent probability distributions based on a probability model. The Bayesian network is also called a belief network or a belief network, is a mathematical model based on probabilistic reasoning, and is based on Bayesian formulas to perform reasoning treatment on uncertainty and incompleteness problems in combination with a belief propagation algorithm (belief propagation). Variability inference was applied in 2003 in the data association problem to solve the target tracking problem with a distributed wireless sensor network.
In an embodiment, a plurality of target state information and measurement state information in a current scanning period are obtained; and generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable.
In the implementation, the state information of the track in the radar target library, namely the target state information and the measurement state information, is updated in real time.
In this embodiment, generating the probability map model according to the plurality of target state information and the measurement state information includes: determining a plurality of corresponding associated variables according to the target state information and the measurement state information; and establishing a probability map model between the target and the associated variable.
In an embodiment, a message update rule is used to determine a message sent by each leaf node to an adjacent layer leaf node in the probability map model.
In this embodiment, the message update rule is a sum-product algorithm.
In an embodiment, a Bayesian free energy objective function is established by using a variation inference algorithm according to the message sent by each leaf node to the leaf nodes of the adjacent layers; and carrying out security radar multi-target tracking according to the Bayesian free energy objective function.
In this embodiment, the method for tracking multiple targets by using a radar further includes: setting constraint conditions for the Bayesian free energy objective function;
according to the Bayesian free energy objective function, performing multi-target tracking of safety radar, comprising: and carrying out multi-target tracking of the security radar according to the Bayesian free energy target function and constraint conditions.
A specific embodiment is given below to illustrate a specific application of the multi-target tracking method of the security radar in the embodiment of the invention. In a specific example, the target and the measured state information in the current scanning period are updated according to the radar real-time detection data.
Taking a single scan as an example, define x= { X 1 ,x 2 ,x i ,.....,x n The state information of the object, s∈s, s= {1,2,.. s i ＝{Z s 1 ,Z s 2 .....Z s m And the status information measured in the s-th scanning period is shown. For each target i e {1, 2..n }, setting a target association indicator variableTo indicate that in the s-th scan period, the measurement is assumed to correlate with the target; similarly, for each measurement j e {1, 2..once, n }, a measurement association indicator variable +.>Representing the s-th scan period, it is assumed that the target correlates with the metrology. A probability map model is then created for the target state, the target-associated indicated variable, and the metrology-associated indicated variable, specifically by using factorization to efficiently represent and calculate joint probability distributions between the plurality of variables, and a probability map model is then created for the target state, the target-associated indicated variable, and the metrology-associated indicated variable. First, the state information x of the target n Target association indicating variable +.>Measurement associated indicator variable +.>Respectively serving as first, second and third leaf nodes; leaf nodes of different levels are then connected to represent implicit relationships between physical quantities of each level. A specific single scan probability map model is shown in fig. 2. In the constructed probability graph modularity, redundant connection exists between the target association indicating variable and the measurement association indicating variable, and the redundancy ensures that each target is associated with at least one measurement.
In the tree-structured probability map, accurate inference is made in conjunction with message update rules to represent information transfer between neighboring vertices. Definition of the definitionRepresenting the information passed between vertex i e V and vertex j e V, and (i, j) e epsilon. The update equation is:
wherein:
the above is also called sum-product algorithm, when the BP algorithm converges, the edge distribution of any vertex j is:
the problem of variation in a single scan is mainly to set a minimization objective function, which is also called bayesian free energy (Bethe free energy, BFE):
constraint conditions:
this constraint is called a consistency constraint because they are necessary to solve the corresponding valid federated association event distribution.
Wherein the method comprises the steps ofConfidence between target i and measurement j;
representing the association between the node factors.
Selecting a target state variable and a neighbor layer node corresponding to the target state variable according to the sequence number of the target in the track library, calculating information sent to all measurement related indicating variable nodes by the node according to an information updating rule, marking the nodes which participate in calculation after the calculation is finished, repeating the process, and finishing one iteration after all the information is updated once. The probability map model of the radar multi-scan data is shown in fig. 3.
The Bayes free energy function is optimized and solved, and the objective function meets the convex optimal condition, so that the objective function can be converged to the maximum value. Solving for its approximate solution is:
obtaining an approximate solution of the objective function by optimizing and solving the objective function, and determining an objective x according to a calculation result i And latest measurementThe probability of correlation between the two, and determining the target x through the probability result of correlation i Corresponding measurement data in the s-th scan, thereby updating the target x i Is a track of the car. In addition, for any target x i For example, if no accurate measurement information is matched in the continuous multi-frames, it indicates that the track corresponding to the target is terminated, and thus the corresponding track number is cancelled. In practice, track merging may occur between a plurality of targets, and by obtaining an optimal correlation solution between the targets and the tracks, track merging can be greatly avoided.
Fig. 4 is a tracking effect diagram of an embodiment of the present invention, which shows a real motion track of a target, a classical data correlation algorithm, and a tracking effect of the present algorithm, and it can be clearly seen from fig. 4 that the tracking effect of the present algorithm is significantly better than that of the classical data correlation algorithm. Position Error (PE) is used as a measure of the accuracy of the algorithm. The experimental results of the algorithm and the classical data correlation algorithm are shown in fig. 5, and from the results, it can be found that the algorithm has smaller average position errors for the target a and the target B, i.e. the accuracy of the algorithm is higher. In addition, the CDF results of the two algorithms in FIG. 6 show that for targets A and B, the algorithm has a PE value of approximately 60% between 20cm and 25.81cm, and 60% PE is superior to the classical joint probability data correlation algorithm, which shows that the algorithm provided herein has higher tracking accuracy in the same tracking scene. The cumulative distribution function (Cumulative Distribution Function, CDF), also called the distribution function, is the integral of the probability density function and fully describes the probability distribution of a real random variable X.
The embodiment of the invention combines the traditional joint probability data association algorithm with variation inference, is not only suitable for single-scanning period, but also suitable for data association during multiple scanning periods, so as to improve the accuracy of track association. According to the method, a probability graph model is introduced on the basis of a traditional joint probability data association algorithm, the target and measurement state information in each scanning period are stored in leaf nodes, a traditional method for constructing a confirmation matrix is replaced, and the maximum posterior probability is calculated by combining a belief propagation algorithm and utilizing the information transmitted between nodes. Therefore, compared with the traditional joint probability data association algorithm, the method has greatly reduced time complexity. Firstly, updating state information of tracks in a radar target library in real time, respectively storing target and measurement related information in leaf nodes of a probability map in single scanning, and initializing likelihood functions of all hidden nodes and potential energy and information between each pair of neighbor nodes; then randomly finding a certain point and its neighbor node, calculating all messages sent by the node to its neighbor node by using message updating rule, then randomly finding a certain node, repeating this process, and finishing an iteration after all messages are updated once; introducing variation inference, constructing a Bayesian free energy function, and iteratively solving an optimal solution; if the algorithm converges, solving a variable which enables the edge probability distribution of each node to be maximum by the confidence coefficient; and updating the track state according to the probability calculation result.
Based on the same inventive concept, the embodiment of the invention also provides a multi-target tracking device of the security radar, as described in the following embodiment. Since the principle of solving the problems is similar to that of the multi-target tracking method for safety and lightning protection, the implementation of the device can be referred to the implementation of the method, and the repetition is omitted.
Fig. 7 is a block diagram of a multi-target tracking device of a security radar according to an embodiment of the present invention, and as shown in fig. 7, the device includes:
an information obtaining module 701, configured to obtain a plurality of target state information and measurement state information in a current scanning period;
the model generating module 702 is configured to generate a probability map model according to the target state information and the measurement state information, where the probability map model includes a plurality of leaf nodes, and each leaf node corresponds to an associated variable;
a message determining module 703, configured to determine a message sent by each leaf node to an adjacent layer of leaf nodes in the probability map model by using a message updating rule;
the function building module 704 is configured to build a bayesian free energy objective function by using a variation inference algorithm according to the message sent by each leaf node to the leaf nodes of the adjacent layer;
and the target tracking module 705 is used for performing security radar multi-target tracking according to the Bayesian free energy target function.
In one embodiment, the model generation module 702 is further configured to:
determining a plurality of corresponding associated variables according to the target state information and the measurement state information;
and establishing a probability map model between the target and the associated variable.
In one embodiment, the message update rule is a sum product algorithm.
In one embodiment, the multi-target tracking device of the security radar further comprises: the constraint setting module is used for setting constraint conditions for the Bayesian free energy objective function;
the target tracking module 705 is further configured to: and carrying out multi-target tracking of the security radar according to the Bayesian free energy target function and constraint conditions.
In summary, the embodiment of the present invention obtains the plurality of target state information and the measurement state information in the current scanning period; generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable; determining the message sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent by each leaf node to the leaf nodes of the adjacent layers; and carrying out security radar multi-target tracking according to the Bayesian free energy objective function. According to the embodiment of the invention, a probability graph model is generated according to a plurality of target state information and measurement state information, and the message updating rule is utilized to determine the message sent to the leaf nodes of the adjacent layer by each leaf node in the probability graph model, so that a Bayesian free energy target function is established by using a variation inference algorithm, multiple target tracking of safety radar is performed, the anti-interference capability is ensured, the operation speed is effectively improved, the real-time performance of the radar during tracking is ensured, and the problem of track merging in the process of tracking adjacent targets is avoided.
Based on the foregoing inventive concept, as shown in fig. 8, the present invention further proposes a computer device 800, including a memory 810, a processor 820, and a computer program 830 stored in the memory 810 and capable of running on the processor 820, where the processor 820 implements the multi-target tracking method of the security radar when executing the computer program 830.
Based on the foregoing inventive concept, the present invention proposes a computer-readable storage medium storing a computer program which, when executed by a processor, implements the multi-target tracking method of the aforementioned security radar.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. A method of multi-target tracking for safety and lightning protection, comprising:
obtaining a plurality of target state information and measurement state information in a current scanning period;
generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable;
determining the message sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; the message comprises measurement association indicating variable information and edge probability distribution of measurement association indicating variables;
establishing a Bayesian free energy objective function by using a variation inference algorithm according to the information sent by each leaf node to the leaf nodes of the adjacent layers;
according to the Bayesian free energy target function, carrying out security radar multi-target tracking;
according to the Bayesian free energy objective function, performing multi-target tracking of safety radar, comprising:
optimizing and solving a Bayesian free energy objective function, wherein the objective function meets the convex optimization condition, so that the objective function converges to the maximum value, and solving the approximate solution is as follows:
；
obtaining an approximate solution of the objective function by optimizing and solving the objective function, and determining an objective according to a calculation resultAnd latest measurement->The probability of association between the two, and the target +.>In->Corresponding measurement data in each scan, thereby updating the targetIs a track of (a);
generating a probability map model according to the plurality of target state information and the measurement state information, wherein the probability map model comprises the following components:
determining corresponding target association indicating variables and measurement association indicating variables according to the plurality of target state information and measurement state information;
and establishing a probability graph model among the target state, the target association indicating variable and the measurement association indicating variable, wherein the target state, the target association indicating variable and the measurement association indicating variable are respectively used as leaf nodes of a first layer, a second layer and a third layer of the probability graph model.
2. The method of claim 1, wherein the message update rule is a sum-product algorithm.
3. The method for multi-target tracking of a security radar of claim 1, further comprising: setting constraint conditions for the Bayesian free energy objective function;
according to the Bayesian free energy objective function, performing multi-target tracking of safety radar, comprising: and carrying out multi-target tracking of the security radar according to the Bayesian free energy target function and constraint conditions.
4. A multi-target tracking device for a security radar, comprising:
the information acquisition module is used for acquiring a plurality of target state information and measurement state information in the current scanning period;
the model generation module is used for generating a probability map model according to the target state information and the measurement state information, wherein the probability map model comprises a plurality of leaf nodes, and each leaf node corresponds to an associated variable;
the message determining module is used for determining messages sent to the adjacent layer leaf nodes by each leaf node in the probability graph model by using the message updating rule; the message comprises measurement association indicating variable information and edge probability distribution of measurement association indicating variables;
the function building module is used for building a Bayesian free energy objective function by utilizing a variation inference algorithm according to the information sent to the leaf nodes of the adjacent layers by the leaf nodes;
the target tracking module is used for carrying out multi-target tracking of safety protection according to the Bayesian free energy target function;
the target tracking module is specifically used for:
optimizing and solving a Bayesian free energy objective function, wherein the objective function meets the convex optimization condition, so that the objective function converges to the maximum value, and solving the approximate solution is as follows:
；
obtaining an approximate solution of the objective function by optimizing and solving the objective function, and determining an objective according to a calculation resultAnd latest measurement->The probability of association between the two, and the target +.>In->Corresponding measurement data in each scan, thereby updating the target +.>Is a track of (a);
the model generation module is used for determining corresponding target association indicating variables and measurement association indicating variables according to the plurality of target state information and measurement state information; and establishing a probability graph model among the target state, the target association indicating variable and the measurement association indicating variable, wherein the target state, the target association indicating variable and the measurement association indicating variable are respectively used as leaf nodes of a first layer, a second layer and a third layer of the probability graph model.
5. The security radar multi-target tracking device of claim 4, wherein the message update rule is a sum-product algorithm.
6. The security radar multi-target tracking device of claim 4, further comprising: the constraint setting module is used for setting constraint conditions for the Bayesian free energy objective function;
the target tracking module is further to: and carrying out multi-target tracking of the security radar according to the Bayesian free energy target function and constraint conditions.
7. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 3 when executing the computer program.
8. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program for executing the method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110762057.5A CN113514824B (en) | 2021-07-06 | 2021-07-06 | Multi-target tracking method and device for safety and lightning protection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110762057.5A CN113514824B (en) | 2021-07-06 | 2021-07-06 | Multi-target tracking method and device for safety and lightning protection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113514824A CN113514824A (en) | 2021-10-19 |
| CN113514824B true CN113514824B (en) | 2023-09-08 |
Family
ID=78066850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110762057.5A Active CN113514824B (en) | 2021-07-06 | 2021-07-06 | Multi-target tracking method and device for safety and lightning protection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113514824B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114217303B (en) * | 2021-11-12 | 2024-07-02 | 鹏城实验室 | Target positioning and tracking method and device, underwater robot and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103472445A (en) * | 2013-09-18 | 2013-12-25 | 电子科技大学 | Detecting tracking integrated method for multi-target scene |
| CN105717505A (en) * | 2016-02-17 | 2016-06-29 | 国家电网公司 | Data association method for utilizing sensing network to carry out multi-target tracking |
| CN106683122A (en) * | 2016-12-16 | 2017-05-17 | 华南理工大学 | Particle filtering method based on Gaussian mixture model and variational Bayes |
| CN106872955A (en) * | 2017-01-24 | 2017-06-20 | 西安电子科技大学 | Radar Multi Target tracking optimization method based on Joint Probabilistic Data Association algorithm |
| CN106980114A (en) * | 2017-03-31 | 2017-07-25 | 电子科技大学 | Target Track of Passive Radar method |
| CN109444505A (en) * | 2018-05-03 | 2019-03-08 | 国网江苏省电力公司盐城供电公司 | A kind of electric automobile charging station harmonic current detection based on variation Bayes's parametric learning method |
| CN109633589A (en) * | 2019-01-08 | 2019-04-16 | 沈阳理工大学 | The Multi-target Data Associations assumed are optimized based on multi-model more in target following |
| CN110542885A (en) * | 2019-08-13 | 2019-12-06 | 北京理工大学 | Millimeter wave radar target tracking method in complex traffic environment |
| CN111178385A (en) * | 2019-12-02 | 2020-05-19 | 江苏大学 | Target tracking method for robust online multi-sensor fusion |
| CN111325776A (en) * | 2020-02-19 | 2020-06-23 | 江苏理工学院 | PHD multi-target tracking method based on variational Bayesian T distribution Kalman filtering |
| CN111988306A (en) * | 2020-08-17 | 2020-11-24 | 北京邮电大学 | Method and system for detecting DDoS attack traffic in network based on variational Bayes |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11315152B2 (en) * | 2019-02-13 | 2022-04-26 | Beijing Jingdong Shangke Information Technology Co., Ltd. | System and method for personalized product recommendation using hierarchical Bayesian |
-
2021
- 2021-07-06 CN CN202110762057.5A patent/CN113514824B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103472445A (en) * | 2013-09-18 | 2013-12-25 | 电子科技大学 | Detecting tracking integrated method for multi-target scene |
| CN105717505A (en) * | 2016-02-17 | 2016-06-29 | 国家电网公司 | Data association method for utilizing sensing network to carry out multi-target tracking |
| CN106683122A (en) * | 2016-12-16 | 2017-05-17 | 华南理工大学 | Particle filtering method based on Gaussian mixture model and variational Bayes |
| CN106872955A (en) * | 2017-01-24 | 2017-06-20 | 西安电子科技大学 | Radar Multi Target tracking optimization method based on Joint Probabilistic Data Association algorithm |
| CN106980114A (en) * | 2017-03-31 | 2017-07-25 | 电子科技大学 | Target Track of Passive Radar method |
| CN109444505A (en) * | 2018-05-03 | 2019-03-08 | 国网江苏省电力公司盐城供电公司 | A kind of electric automobile charging station harmonic current detection based on variation Bayes's parametric learning method |
| CN109633589A (en) * | 2019-01-08 | 2019-04-16 | 沈阳理工大学 | The Multi-target Data Associations assumed are optimized based on multi-model more in target following |
| CN110542885A (en) * | 2019-08-13 | 2019-12-06 | 北京理工大学 | Millimeter wave radar target tracking method in complex traffic environment |
| CN111178385A (en) * | 2019-12-02 | 2020-05-19 | 江苏大学 | Target tracking method for robust online multi-sensor fusion |
| CN111325776A (en) * | 2020-02-19 | 2020-06-23 | 江苏理工学院 | PHD multi-target tracking method based on variational Bayesian T distribution Kalman filtering |
| CN111988306A (en) * | 2020-08-17 | 2020-11-24 | 北京邮电大学 | Method and system for detecting DDoS attack traffic in network based on variational Bayes |
Non-Patent Citations (1)
| Title |
|---|
| 基于非线性滤波的IRST系统弱小目标检测与跟踪方法的研究;田岳鑫;《中国博士学位论文全文数据库 信息科技辑》(第3期);I138-85 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113514824A (en) | 2021-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Oh et al. | Markov chain Monte Carlo data association for multi-target tracking | |
| Colburn et al. | State estimation in wall-bounded flow systems. Part 3. The ensemble Kalman filter | |
| AU2009289008B2 (en) | Estimating a state of at least one target | |
| CN110320512A (en) | A kind of GM-PHD smothing filtering multi-object tracking method based on tape label | |
| Beard et al. | A generalised labelled multi-Bernoulli filter for extended multi-target tracking | |
| CN104156984A (en) | PHD (Probability Hypothesis Density) method for multi-target tracking in uneven clutter environment | |
| Pan et al. | Video tracking based on sequential particle filtering on graphs | |
| Lan et al. | Joint target detection and tracking in multipath environment: A variational Bayesian approach | |
| CN113514824B (en) | Multi-target tracking method and device for safety and lightning protection | |
| CN111830501B (en) | HRRP history feature assisted signal fuzzy data association method and system | |
| Davarynejad et al. | A novel general framework for evolutionary optimization: Adaptive fuzzy fitness granulation | |
| Lan et al. | A survey on joint tracking using expectation–maximization based techniques | |
| Meyer et al. | Scalable multitarget tracking using multiple sensors: A belief propagation approach | |
| CN111862046A (en) | System and method for distinguishing position of catheter in cardiac coronary silhouette | |
| CN106054171A (en) | Information entropy-based multi-radar node adaptive selection and tracking method | |
| Kropfreiter et al. | Track coalescence and repulsion: MHT, JPDA, and BP | |
| Meyer et al. | Joint navigation and multitarget tracking in networks | |
| CN116520281A (en) | DDPG-based extended target tracking optimization method and device | |
| CN111798494A (en) | Maneuvering target robust tracking method under generalized correlation entropy criterion | |
| CN109190787B (en) | Dual particle swarm multi-monitoring point access path planning method for underwater vehicle | |
| Seifzadeh et al. | Soft-data-constrained multi-model particle filter for agile target tracking | |
| Zhang et al. | A novel framework of fuzzy rule interpolation for takagi-sugeno-kang inference systems | |
| Rahmathullah et al. | A batch algorithm for estimating trajectories of point targets using expectation maximization | |
| Cormack et al. | Message passing for joint registration and tracking in multistatic radar | |
| CN113628254A (en) | Target track determination method based on mobile platform and related equipment |
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 |