CN116304757A - Multi-expansion target measurement dividing method, device and storage medium based on OPTICS-FCM - Google Patents
Multi-expansion target measurement dividing method, device and storage medium based on OPTICS-FCM Download PDFInfo
- Publication number
- CN116304757A CN116304757A CN202310213041.8A CN202310213041A CN116304757A CN 116304757 A CN116304757 A CN 116304757A CN 202310213041 A CN202310213041 A CN 202310213041A CN 116304757 A CN116304757 A CN 116304757A
- Authority
- CN
- China
- Prior art keywords
- measurement
- distance
- sample
- optics
- reachable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005259 measurement Methods 0.000 title claims abstract description 232
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000000638 solvent extraction Methods 0.000 claims description 14
- 230000001174 ascending effect Effects 0.000 claims description 6
- 230000001960 triggered effect Effects 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 238000007621 cluster analysis Methods 0.000 description 2
- 238000002592 echocardiography Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The invention provides a multi-expansion target measurement dividing method, a device and a storage medium based on OPTICS-FCM, which relate to the technical field of multi-target tracking and are used for solving the problems that in the existing multi-expansion target tracking process, clutter cannot be effectively removed, measurement secondary division is triggered, measurement division errors and neighbor target measurement division fails, and finally, the target quantity estimation errors are caused, and the realization steps comprise: the target measurement data is expanded for preprocessing, parameters are obtained and used for OPTICS initialization parameter values, clutter measurement is removed, and an reachable distance and a primary measurement subset are output; and estimating the measurement generation rate, and performing FCM measurement subdivision according to the measurement rate to finish final expansion target measurement subdivision. According to the invention, the OPTICS algorithm and FCM subdivision are optimized according to the measurement data of the extended measurement target, and the method can be used for the measurement division link of radar multi-extended target tracking.
Description
Technical Field
The invention relates to the technical field of multi-target tracking, in particular to a multi-expansion target measurement dividing method, a multi-expansion target measurement dividing device and a storage medium based on OPTICS-FCM.
Background
With the rapid development of radar technology, the radar resolution is continuously improved, target echoes are no longer in a single resolution unit, and the target echoes gradually expand to a plurality of resolution units, so that the targets are expanded targets, and a standard measurement model assumed by a conventional multi-target tracking method is not suitable for tracking the expanded targets. The method for firstly dividing the measurement of a plurality of extended targets and then tracking the same is a mainstream multi-extended target tracking method at present. Therefore, the measurement divides an important link in the process of expanding target tracking.
The current classical multi-expansion target measurement set partitioning algorithm comprises: distance division, K-means++ division, prediction division and expected maximum (Expectation maximization, EM) division methods cannot effectively eliminate clutter, particularly in a dense ocean clutter environment, strong ocean clutter can cause interference to measurement division, and in a box particle filtering implementation algorithm, a target amount measurement box body can be enlarged, so that accuracy of tracking method target amount estimation can be seriously affected. In addition, as the measurement number of the expanded targets is uncertain, the measurement subdivision with a fixed threshold value is triggered, the measurement subdivision error is easy to be caused, and the problem of target number estimation error occurs.
The Density-based partitioning method is also a common multi-expansion target measurement set partitioning algorithm, such as a clustering algorithm of DBSCAN (Density-Based Spatial Clustering of Applications with Noise, density-based spatial clustering with noise application), an initial parameter epsilon neighborhood and MinPts need to be manually set, and the difference of parameter values can lead to huge measurement partitioning results, in this case, the clutter environment which changes in real time cannot be adapted, so that clutter cannot be effectively removed, and the following expansion target tracking effect is affected.
Disclosure of Invention
In view of this, the invention provides a multi-expansion target measurement dividing method, device and storage medium based on OPTICS (Ordering Points to identify the clustering structure, identifying cluster structure through point ordering) -FCM (Fuzzy C-Means, fuzzy C-Means clustering), which solves the problems that in the existing multi-expansion target tracking process, the measurement dividing cannot effectively remove clutter, trigger measurement secondary dividing, measurement dividing error and neighbor target measurement dividing failure, and finally result in target quantity estimation error in the dense clutter environment.
For this purpose, the invention provides the following technical scheme:
the invention discloses a multi-expansion target measurement dividing method based on OPTICS-FCM clustering, which comprises the following steps:
acquiring a multi-expansion target measurement set at a preset moment, preprocessing the multi-expansion target measurement set, and calculating the distance between every two measurement samples to obtain a distance maximum value;
setting a measurement sample to be the neighborhood minimum sample number of the core point;
inputting measurement data, a distance maximum value and a neighborhood minimum sample number in the multi-expansion target measurement set into an OPTICS clustering algorithm, and initializing reachable distances by taking the distance maximum value as all measurement samples to obtain an orderly output result and corresponding reachable distances;
traversing the reachable distance sequence of the measurement sample, and judging the threshold value of the reachable distance data of the measurement sample to finish clutter rejection;
taking the average number of the samples of each subset as measurement rate estimation, and completing measurement rate estimation;
performing sub-division judgment on each measurement subset based on the measurement rate estimation, performing measurement sub-division by adopting FCM (fuzzy c-means) in accordance with measurement sub-division conditions, and then outputting a final measurement division subset; the measurement subdivision conditions include: the subset of measurements is made of intersecting or adjacent targets.
Further, the measurement data, the distance maximum value and the neighborhood minimum sample number in the multi-expansion target measurement set are input into an OPTICS clustering algorithm, the distance maximum value is used as all measurement samples to initialize the reachable distance, and an orderly output result and a corresponding reachable distance are obtained, including:
step 2, creating two queues, namely a to-be-processed queue QP and a result queue order; the queue to be processed is used for storing samples in the neighborhood of the core sample and the reachable distances, and is arranged according to the reachable distances in an ascending order; the result queue is used for storing the output order of the sample points and is processed data;
step 3, if all points in the multi-expansion target measurement set are processed or core points do not exist, ending the algorithm; otherwise, selecting a sample point o which is not processed and is a core point, firstly putting o into a result queue order, and deleting o from QP; then find the measurement set Z k All densities of the middle o directly reach the sample point x, the reachable distance from x to o is calculated, if x is not in the QP of the queue to be processed, the x and the reachable distance are put into the QP, if x is in the QP, if the new reachable distance of x is smaller, the reachable distance of x is updated, and finally, the data in the QP are reordered from small to large according to the reachable distance;
step 4, if the queue QP to be processed is empty, returning to step 3, otherwise, taking out the first sample point y in the QP, putting into an order, and measuring the set Z k The marking method is that the corresponding position of the sequence to be processed of the o position of the object is marked as empty, and o is pressed into a result queue order;
step 5, if y is not a core point, repeating the step 4, namely finding a sample point with the minimum reachable distance of the residual data in the QP; if y is a core point, finding all density direct sample points of y in the measurement set, calculating the reachable distance to y, and updating all density direct sample points into QP according to step 3;
step 6, repeating the step 3 and the step 4 until all the measurement samples are processed, and stopping iteration when the queue QP to be processed is empty; finally, an orderly output result and a corresponding reachable distance are obtained.
Further, traversing the measurement sample reachable distance sequence, and performing threshold judgment on the measurement sample reachable distance data, wherein the method comprises the following steps:
setting the number of subsets count, setting the initial value to be 0, traversing the reachable distance RD sequence of the measurement samples, judging the threshold value of RD (order (i)), if RD (order (i)) is not less than RDTH and RD (order (i+1)) is less than RDTH, increasing the count value by 1, and outputting the samples corresponding to RD before the next qualified RD value as measurement division subsets Zp k (1)，Zpk(1)＝Z k (order (i)) and outputting each measurement subset Zp after RD traversal is completed k And (p) measuring the number of the subsets, namely estimating the number of the extended targets, and measuring the number of the subsets without clutter samples in the subset samples to finish clutter rejection.
Further, performing sub-division judgment on each measurement subset based on the measurement rate estimation, performing measurement sub-division by using FCM in accordance with measurement sub-division conditions, including:
judging the measurement number of each measurement subset according to the estimated measurement rate, if the measurement data column number of the measurement subset is greater than 1.5L k And performing FCM measurement subdivision on the measurement subset.
Further, the distance is euclidean distance.
Further, the neighborhood minimum number of samples is set to 5.
The invention also provides a multi-expansion target measurement dividing device based on OPTICS-FCM clustering, which comprises:
the preprocessing unit is used for acquiring a multi-expansion target measurement set at a preset moment, preprocessing the multi-expansion target measurement set, and calculating the distance between every two measurement samples to obtain a distance maximum value;
the parameter setting unit is used for setting the measurement sample to be the neighborhood minimum sample number of the core point;
the OPTICS clustering unit is used for inputting the measurement data in the multi-expansion target measurement set obtained by the preprocessing unit, the distance maximum value and the neighborhood minimum sample number obtained by the parameter setting unit into an OPTICS clustering algorithm, and initializing the reachable distances by taking the distance maximum value as all measurement samples to obtain an ordered output result and corresponding reachable distances;
the clutter removing unit is used for traversing the measuring sample reachable distance sequence obtained by the OPTICS clustering unit, and performing threshold judgment on the measuring sample reachable distance data to complete clutter removing;
the measurement rate estimation unit is used for taking the average number of the samples of each subset after the clutter removal unit removes the clutter as measurement rate estimation and finishing measurement rate estimation;
and the FCM dividing unit is used for carrying out sub-division judgment on each measuring subset based on the measuring rate estimation obtained by the measuring rate estimation unit, adopting FCM to carry out measuring sub-division according with the measuring sub-division condition, and then outputting the final measuring division subset.
The invention also provides a computer readable storage medium, wherein a computer instruction set is stored in the computer readable storage medium, and the multi-expansion target measurement dividing method based on OPTICS-FCM clustering is realized when the computer instruction set is executed by a processor.
The invention has the advantages and positive effects that:
in the invention, the measurement data is preprocessed, so that the operability data processing condition is generated, the condition setting of the OPTICS algorithm can be adaptively adjusted, and the accuracy of measurement division is improved. The method can reject the measurement data marked as clutter, prevents error triggering of measurement subdivision on one hand, and avoids tracking errors caused by overlarge measurement box division in a tracking method based on a box particle filtering technology on the other hand. In addition, after the impurities are removed, the estimation of the measurement generation rate can be easily carried out, the measurement subdivision during the adjacent extended target tracking is convenient, and the accuracy of the estimation of the number of the multiple extended target tracking targets is improved.
Drawings
In order to more clearly illustrate the embodiments of the present 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, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to the drawings without inventive effort to a person skilled in the art.
FIG. 1 is a flowchart of a multi-expansion target measurement partitioning method based on OPTICS-FCM in an embodiment of the invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Aiming at the problem of inaccurate measurement division in the existing multi-expansion target tracking and inaccurate target quantity estimation, the invention provides a multi-expansion target measurement division method based on OPTICS-FCM clustering, which can carry out measurement division on measurement samples with different densities, which are generated by a plurality of expansion targets, in a dense sea clutter scene, output the measurement samples so as to reduce the sensitivity to input parameters, optimize the OPTICS clustering method, ensure that the OPTICS clustering method can not only output cluster structures, but also reject clutter measurement, output the division quantity of measurement samples of the expansion targets according to the multi-expansion target tracking requirement, estimate the measurement generation rate of each division subset sample after each clustering, and input the measurement samples to an FCM algorithm as initial cluster quantity data so as to finish measurement subdivision according to the estimated measurement rate.
The existing OPTICS cluster analysis method is an improved algorithm of DBSCAN cluster, is insensitive to input parameters, and does not influence the clustering result as long as the value of MinPts is determined and the radius eps is slightly changed. The OPTICS does not explicitly generate a cluster, but generates an augmented cluster sequence (for example, a graph with the reachable distance as a vertical axis and the sample point output order as a horizontal axis), the sequence represents a density-based cluster structure of each sample point, and the clustering result of DBSCAN algorithm based on any parameters eps and MinPts can be obtained from the sequence, so that the defect of using a group of global parameters in cluster analysis is overcome, and the problem of poor clustering effect caused by different densities is effectively solved.
According to the invention, the OPTICS method is improved according to the measurement data characteristics of radar multi-expansion target tracking, the RD initial value of each sample point is not set to be a maximum value, but is set to be RDTH according to the maximum Euclidean distance between measurement sample points, and the estimated measurement rate is introduced into the FCM in the measurement sub-division stage, so that clutter is effectively removed, and the problem of inaccurate measurement division quantity is solved.
As shown in FIG. 1, the multi-expansion target measurement dividing method based on OPTICS-FCM clustering in the embodiment of the invention comprises the following steps:
s1, acquiring a k-moment multi-expansion target measurement set Z k For Z k Preprocessing, calculating Euclidean distance between every two measurement samples, and taking the maximum value as RDTH.
The measurement data is preprocessed, and then the operability data processing conditions are generated, so that the condition setting can be adjusted in a self-adaptive mode, and the measurement division accuracy is improved.
S2, setting a parameter MinPts, namely measuring the minimum number of samples of which the neighborhood of the sample becomes a core point.
The parameter only plays an auxiliary role, is generally set to be 5, and has little influence on the whole effect;
s3, inputting the measurement data, RDTH and MinPts into an OPTICS clustering algorithm, initializing the reachable distances by taking RDTH as all measurement samples, and finally obtaining an orderly output result and corresponding reachable distances, namely an order sequence and a reachable distance RD table of each measurement sample.
Wherein, S3 specifically includes the following steps:
s31, calculating the distances between all measured sample points, if the number of sample points (N (x)) I in the neighborhood radius is greater than or equal to MinPts, then the sample points are called core points, calculating the core distance cd (x) of the core points, and the distances d (x, N) between the sample points at the MinPts position and the core point o are arranged in the distance ascending order in the neighborhood of the current core point o MinPts (x) The core distance cd (x) as o (the number of samples included in the actual o-neighborhood o is minpts+1).
S32, creating two queues, namely a to-be-processed queue QP and a result queue order;
the queue to be processed is used for storing samples in the neighborhood of the core sample and the reachable distances, and is arranged according to the reachable distances in an ascending order; the result queue is used to store the output order of the sample points, which is the already processed data.
S33, if multiple extended target measurement set Z k If all the points in the list are processed or no core point exists, the algorithm is ended. Otherwise, selecting a sample point o which is not processed and is a core point, firstly putting o into a result queue order, and deleting o from QP; then find the measurement set Z k And (3) directly reaching the sample points x by all densities of the m-o, calculating the reachable distance of x to o, if x is not in the queue QP to be processed, putting x and the reachable distance thereof into the QP, if x is in the QP, if the new reachable distance of x is smaller, updating the reachable distance of x, and finally reordering the data in the QP from small to large according to the reachable distance.
S34, if the queue QP to be processed is empty, returning to the step S33, otherwise, taking out the first sample point y (i.e. the sample point with the smallest reachable distance) in the QP, putting into an order, and measuring the sample point Z k The marking method is that the corresponding position of the to-be-processed sequence of the o position of the object is marked as empty, and o is pressed into the result queue order.
S35, if y is not a core point, repeating the step S34, namely finding a sample point with the minimum reachable distance of the residual data in the QP; if y is the core point, find all density direct sample points of y in the measurement set, and calculate the reachable distance to y, then update all density direct sample points into QP according to S33.
S36, repeating the steps S33 and S34 until all the measurement samples are processed, and stopping iteration when the queue QP to be processed is empty; finally, an orderly output result and corresponding reachable distances, namely order and RD table, can be obtained.
S4, traversing the RD sequence, and judging the threshold value of the RD data.
Step S3, outputting an reachable distance RD sequence and an order sequence, setting the number of subsets count, setting the initial value to be 0, traversing the RD sequence, performing threshold judgment on RD (order (i)), if RD (order (i)) is not smaller than RDTH and RD (order (i+1)) is smaller than RDTH, increasing the count value by 1, and outputting samples corresponding to RD before the next qualified RD value as measurement division subsets Zp k (1)，Zpk(1)＝Z k After RD traversal is completed, outputting each measurement subset Zpk (p), and the number of subsets count, namely the estimated number of expansion targets, measuring the number of clutter samples in the subset samples, and completing clutter rejection.
If RD (order (i)) is not less than RDTH and RD (order (i+1)) is less than RDTH, then Z is k The first order (i) sample is used as a demarcation point, the number of RD data grooves is searched by taking the demarcation point as a demarcation point, namely the number of sub-sets after measurement and division, the unconditional samples are marked as clutter and are not processed, the conditional measurement samples are stored according to groove data, and after RD traversal is completed, each RD data groove is outputAnd measuring the subset Zpk (p), wherein the number of the subset is the estimated number of the expansion targets, and measuring the sample of the subset that no clutter sample exists, thereby completing clutter rejection.
Step S4 can reject the measurement data marked as clutter, so that on one hand, false triggering of measurement subdivision is prevented, and on the other hand, in the tracking method based on the box particle filtering technology, tracking errors caused by overlarge measurement box body subdivision are avoided.
S5, target measurement rate L k And (3) estimating, namely taking the average number of samples in each subset as a measurement rate estimation, and completing the measurement rate estimation.
Because each measurement subset has completed clutter rejection, the average number of the sample numbers of each measurement subset can be used as the measurement rate estimation according to the sample number relation of each measurement subset, and the sample number of the measurement subset is Zp k And (p) the number of columns to complete the measurement rate estimation.
After removing the impurities, the estimation of the measurement generation rate can be easily carried out, the measurement subdivision during the adjacent extended target tracking is convenient, and the accuracy of the estimation of the number of the multiple extended target tracking targets is improved.
S6, sub-division judgment is carried out on each measurement subset, FCM is adopted to carry out measurement sub-division according with measurement sub-division conditions, and then a final measurement division subset Zp is output k (p)。
Estimating the measurement rate L according to the k moment k Judging the measurement number of each measurement subset, if the measurement data column number of the measurement subset is greater than 1.5L k Indicating that the measurement subset is composed of crossing or adjacent targets, performing FCM measurement sub-division on the measurement subset, and outputting final measurement division result, i.e. final measurement subsets Zp k (p)。
Corresponding to the multi-expansion target measurement dividing method based on OPTICS-FCM clustering in the application, the application also provides a multi-expansion target measurement dividing device based on OPTICS-FCM clustering, which comprises the following steps:
the preprocessing unit is used for acquiring a multi-expansion target measurement set at a preset moment, preprocessing the multi-expansion target measurement set, and calculating the distance between every two measurement samples to obtain a distance maximum value;
the parameter setting unit is used for setting the measurement sample to be the neighborhood minimum sample number of the core point;
the OPTICS clustering unit is used for inputting the measurement data in the multi-expansion target measurement set obtained by the preprocessing unit, the distance maximum value and the neighborhood minimum sample number obtained by the parameter setting unit into an OPTICS clustering algorithm, and initializing the reachable distances by taking the distance maximum value as all measurement samples to obtain an ordered output result and corresponding reachable distances;
the clutter removing unit is used for traversing the measuring sample reachable distance sequence obtained by the OPTICS clustering unit, and performing threshold judgment on the measuring sample reachable distance data to complete clutter removing;
the measurement rate estimation unit is used for taking the average number of the samples of each subset after the clutter removal unit removes the clutter as measurement rate estimation and finishing measurement rate estimation;
and the FCM dividing unit is used for carrying out sub-division judgment on each measuring subset based on the measuring rate estimation obtained by the measuring rate estimation unit, adopting FCM to carry out measuring sub-division according with the measuring sub-division condition, and then outputting the final measuring division subset.
For the multi-expanded target measurement dividing device based on the OPTICS-FCM cluster in the embodiment of the invention, the description is relatively simple because the multi-expanded target measurement dividing device based on the OPTICS-FCM cluster in the above embodiment corresponds to the multi-expanded target measurement dividing method based on the OPTICS-FCM cluster, and the relevant similarities are described in the part of the multi-expanded target measurement dividing method based on the OPTICS-FCM cluster in the above embodiment, which is not described in detail herein.
The embodiment of the application also discloses a computer readable storage medium, wherein a computer instruction set is stored in the computer readable storage medium, and when the computer instruction set is executed by a processor, the multi-expansion target measurement dividing method based on OPTICS-FCM clustering is realized.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (8)
1. The multi-expansion target measurement dividing method based on OPTICS-FCM clustering is characterized by comprising the following steps of:
acquiring a multi-expansion target measurement set at a preset moment, preprocessing the multi-expansion target measurement set, and calculating the distance between every two measurement samples to obtain a distance maximum value;
setting a measurement sample to be the neighborhood minimum sample number of the core point;
inputting measurement data, a distance maximum value and a neighborhood minimum sample number in the multi-expansion target measurement set into an OPTICS clustering algorithm, and initializing reachable distances by taking the distance maximum value as all measurement samples to obtain an orderly output result and corresponding reachable distances;
traversing the reachable distance sequence of the measurement sample, and judging the threshold value of the reachable distance data of the measurement sample to finish clutter rejection;
taking the average number of the samples of each subset as measurement rate estimation, and completing measurement rate estimation;
performing sub-division judgment on each measurement subset based on the measurement rate estimation, performing measurement sub-division by adopting FCM (fuzzy c-means) in accordance with measurement sub-division conditions, and then outputting a final measurement division subset; the measurement subdivision conditions include: the subset of measurements is made of intersecting or adjacent targets.
2. The method for partitioning multi-expansion target measurement based on OPTICS-FCM clustering according to claim 1, wherein the steps of inputting measurement data, a distance maximum value and a neighborhood minimum sample number in the multi-expansion target measurement set into an OPTICS clustering algorithm, initializing reachable distances by using the distance maximum value as all measurement samples, and obtaining an ordered output result and corresponding reachable distances comprise:
step 1, calculating the core distance of a core point, namely, in the neighborhood of the current core point o, arranging sample points with the smallest sample digits in the neighborhood in ascending order from the distance between the current core point o and the core point o, wherein the distance between the current core point o and the current core point o is used as the core distance of o;
step 2, creating two queues, namely a to-be-processed queue QP and a result queue order; the queue to be processed is used for storing samples in the neighborhood of the core sample and the reachable distances, and is arranged according to the reachable distances in an ascending order; the result queue is used for storing the output order of the sample points and is processed data;
step 3, if all points in the multi-expansion target measurement set are processed or core points do not exist, ending the algorithm; otherwise, selecting a sample point o which is not processed and is a core point, firstly putting o into a result queue order, and deleting o from QP; then find the measurement set Z k All densities of the middle o directly reach the sample point x, the reachable distance from x to o is calculated, if x is not in the QP of the queue to be processed, the x and the reachable distance are put into the QP, if x is in the QP, if the new reachable distance of x is smaller, the reachable distance of x is updated, and finally, the data in the QP are reordered from small to large according to the reachable distance;
step 4, if the queue QP to be processed is empty, returning to step 3, otherwise, taking out the first sample point y in the QP, putting into an order, and measuring the set Z k The marking method is that the corresponding position of the sequence to be processed of the o position of the object is marked as empty, and o is pressed into a result queue order;
step 5, if y is not a core point, repeating the step 4, namely finding a sample point with the minimum reachable distance of the residual data in the QP; if y is a core point, finding all density direct sample points of y in the measurement set, calculating the reachable distance to y, and updating all density direct sample points into QP according to step 3;
step 6, repeating the step 3 and the step 4 until all the measurement samples are processed, and stopping iteration when the queue QP to be processed is empty; finally, an orderly output result and a corresponding reachable distance are obtained.
3. The multi-expansion target measurement partitioning method based on OPTICS-FCM clustering of claim 1, wherein traversing the measurement sample reachable distance sequence, performing threshold judgment on measurement sample reachable distance data, comprises:
setting the number of subsets count, setting the initial value to be 0, traversing the reachable distance RD sequence of the measurement samples, judging the threshold value of RD (order (i)), if RD (order (i)) is not less than RDTH and RD (order (i+1)) is less than RDTH, increasing the count value by 1, and outputting the samples corresponding to RD before the next qualified RD value as measurement division subsets Zp k (1)，Zpk(1)＝Z k (order (i)) and outputting each measurement subset Zp after RD traversal is completed k And (p) measuring the number of the subsets, namely estimating the number of the extended targets, and measuring the number of the subsets without clutter samples in the subset samples to finish clutter rejection.
4. The multi-expansion target measurement partitioning method based on the OPTICS-FCM clustering of claim 1, wherein the sub-partitioning judgment is performed on each measurement subset based on the measurement rate estimation, and the measurement sub-partitioning by using FCM according to the measurement sub-partitioning condition comprises:
judging the measurement number of each measurement subset according to the estimated measurement rate, if the measurement data column number of the measurement subset is greater than 1.5L k And performing FCM measurement subdivision on the measurement subset.
5. The multi-expansion target measurement partitioning method based on OPTICS-FCM clustering of claim 1, wherein said distance is Euclidean distance.
6. The multi-expansion target measurement partitioning method based on OPTICS-FCM clustering of claim 1, wherein the neighborhood minimum number of samples is set to 5.
7. The utility model provides a many extension target measurement division device based on OPTICS-FCM cluster which characterized in that includes:
the preprocessing unit is used for acquiring a multi-expansion target measurement set at a preset moment, preprocessing the multi-expansion target measurement set, and calculating the distance between every two measurement samples to obtain a distance maximum value;
the parameter setting unit is used for setting the measurement sample to be the neighborhood minimum sample number of the core point;
the OPTICS clustering unit is used for inputting the measurement data in the multi-expansion target measurement set obtained by the preprocessing unit, the distance maximum value and the neighborhood minimum sample number obtained by the parameter setting unit into an OPTICS clustering algorithm, and initializing the reachable distances by taking the distance maximum value as all measurement samples to obtain an ordered output result and corresponding reachable distances;
the clutter removing unit is used for traversing the measuring sample reachable distance sequence obtained by the OPTICS clustering unit, and performing threshold judgment on the measuring sample reachable distance data to complete clutter removing;
the measurement rate estimation unit is used for taking the average number of the samples of each subset after the clutter removal unit removes the clutter as measurement rate estimation and finishing measurement rate estimation;
and the FCM dividing unit is used for carrying out sub-division judgment on each measuring subset based on the measuring rate estimation obtained by the measuring rate estimation unit, adopting FCM to carry out measuring sub-division according with the measuring sub-division condition, and then outputting the final measuring division subset.
8. A computer readable storage medium, wherein a computer instruction set is stored in the computer readable storage medium, and when the computer instruction set is executed by a processor, the method for partitioning multi-expansion target measurement based on OPTICS-FCM cluster according to any one of claims 1 to 6 is implemented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310213041.8A CN116304757A (en) | 2023-03-07 | 2023-03-07 | Multi-expansion target measurement dividing method, device and storage medium based on OPTICS-FCM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310213041.8A CN116304757A (en) | 2023-03-07 | 2023-03-07 | Multi-expansion target measurement dividing method, device and storage medium based on OPTICS-FCM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116304757A true CN116304757A (en) | 2023-06-23 |
Family
ID=86791945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310213041.8A Pending CN116304757A (en) | 2023-03-07 | 2023-03-07 | Multi-expansion target measurement dividing method, device and storage medium based on OPTICS-FCM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116304757A (en) |
-
2023
- 2023-03-07 CN CN202310213041.8A patent/CN116304757A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6743934B2 (en) | Method, apparatus and system for estimating causal relationship between observed variables | |
| CN110991311B (en) | Target detection method based on dense connection deep network | |
| CN111476302A (en) | fast-RCNN target object detection method based on deep reinforcement learning | |
| CN112526513A (en) | Millimeter wave radar environment map construction method and device based on clustering algorithm | |
| CN111445108B (en) | Data-driven power distribution network line variation relation diagnosis method, device and system | |
| CN112101430A (en) | Anchor frame generation method for image target detection processing and lightweight target detection method | |
| CN110766719A (en) | Target tracking method, device and storage medium | |
| CN111444294B (en) | Track complement method and device and electronic equipment | |
| CN110992396A (en) | K-means + + clustering algorithm-based multi-extension target tracking method | |
| Horzyk et al. | YOLOv3 precision improvement by the weighted centers of confidence selection | |
| CN112906816A (en) | Target detection method and device based on optical differential and two-channel neural network | |
| CN109902641B (en) | Semantic alignment-based face key point detection method, system and device | |
| CN108469609B (en) | Detection information filtering method for radar target tracking | |
| CN117075097B (en) | Maritime radar target tracking method and system based on expanded target cluster division | |
| CN117170294B (en) | Intelligent control method of satellite thermal control system based on space thermal environment prediction | |
| CN111142101B (en) | Data association method | |
| CN110231602B (en) | Large-batch sea surface target point trace extraction algorithm based on two-dimensional partition processing | |
| Ha et al. | Leveraging bayesian optimization to speed up automatic precision tuning | |
| CN116304757A (en) | Multi-expansion target measurement dividing method, device and storage medium based on OPTICS-FCM | |
| CN116958809A (en) | Remote sensing small sample target detection method for feature library migration | |
| CN108469627B (en) | Ground same-frequency multiple-static radiation source positioning method and system based on time-frequency difference | |
| CN115292971B (en) | Bayes-based crack attribute analysis method and device and storage medium | |
| CN113109761B (en) | Trajectory-oriented calculation time reduction method based on multi-hypothesis tracking algorithm | |
| US20220189806A1 (en) | Estimating heights of defects in a wafer | |
| CN113689526A (en) | Method and device for dividing invalid area in map and electronic 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 |