CN114594437A - Automatic labeling method, system and storage medium for radar trace data - Google Patents

Automatic labeling method, system and storage medium for radar trace data Download PDF

Info

Publication number
CN114594437A
CN114594437A CN202210086638.6A CN202210086638A CN114594437A CN 114594437 A CN114594437 A CN 114594437A CN 202210086638 A CN202210086638 A CN 202210086638A CN 114594437 A CN114594437 A CN 114594437A
Authority
CN
China
Prior art keywords
radar
clusters
condition
merging
trace data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210086638.6A
Other languages
Chinese (zh)
Inventor
张靓
李荣锋
池姗姗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aerospace Nanhu Electronic Information Technology Co ltd
Original Assignee
Aerospace Nanhu Electronic Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Aerospace Nanhu Electronic Information Technology Co ltd filed Critical Aerospace Nanhu Electronic Information Technology Co ltd
Priority to CN202210086638.6A priority Critical patent/CN114594437A/en
Publication of CN114594437A publication Critical patent/CN114594437A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/40—Means for monitoring or calibrating
    • G01S7/4004—Means for monitoring or calibrating of parts of a radar system
    • G01S7/4021—Means for monitoring or calibrating of parts of a radar system of receivers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a method, a system and a storage medium for automatically labeling radar trace data, which comprise the following steps: acquiring radar trace data; performing characteristic sensitivity analysis on radar trace point data, establishing a trace point classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace data are clustered to obtain a plurality of clusters; merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and when the merging result meets the integral DBI condition, outputting the merging result as a final class. The improved CFDP algorithm divides the trace data, better covers the original category of the radar trace data, can meet the requirement on the accuracy rate of samples in the automatic marking of the radar trace data, and can stably and accurately carry out the automatic marking of the radar trace data.

Description

Automatic labeling method, system and storage medium for radar trace data
Technical Field
The invention relates to the technical field of radar trace data marking, in particular to an automatic marking method, system and storage medium for radar trace data.
Background
In the radar data processing, as a data basis for the trace point quality evaluation, the objective degree of trace point data samples to different types of trace point descriptions directly determines the accuracy of the trace point quality evaluation. The existing trace point data labeling method basically takes manual work as a main part, radar track data are compared with answering equipment data such as broadcast type automatic correlation monitoring, secondary radar or a friend or foe identifier, radar tracks close to the truth values of the answering equipment are selected in space, time and two dimensions, real tracks are considered to be satisfied, and manual labeling is carried out. The method can obtain more accurate marking data, but has higher requirements on the experience of marking personnel. In the radar trace quality evaluation, no matter a likelihood estimation method or a machine learning method is adopted, the prior information of historical data needs to be extracted by a statistical method, the more the effective historical data is, the more accurate a classification model is, but the manual comparison is undoubtedly not friendly, the accuracy of manual comparison is obviously reduced along with the increase of data volume, and the time cost is increased, so that a stable and accurate automatic labeling method of radar trace data is needed to label massive radar trace data.
Disclosure of Invention
The invention aims to overcome the technical defects, provides an automatic marking method, system and storage medium for radar trace data, and solves the technical problem that trace data is difficult to mark in radar trace quality evaluation in the prior art.
In order to achieve the above technical object, in a first aspect, a technical solution of the present invention provides an automatic radar trace data labeling method, including the following steps:
acquiring radar trace point data;
performing characteristic sensitivity analysis on the radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value;
taking the sensitive characteristics as the input of a CFDP (clustering by fast search and find of diversity peaks, based on improved fast density peak clustering) algorithm, and clustering the radar trace data to obtain a plurality of clusters;
merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI (Davies-Bouldin) condition or not;
and when the merging result meets the integral DBI condition, outputting the merging result as a final class.
Compared with the prior art, the method for automatically labeling radar trace point data has the beneficial effects that:
the automatic labeling method of the radar trace point data comprises the following steps: firstly, acquiring radar trace data; then, performing characteristic sensitivity analysis on the radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, the sensitive features are used as input of a CFDP algorithm, and clustering processing is carried out on the radar trace data to obtain a plurality of clusters; secondly, merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class.
The original category of the radar trace data is better covered by the partition of the improved CFDP algorithm on the trace data, the requirement on the sample accuracy in the automatic marking of the radar trace data can be met, an effective data base can be provided for the quality evaluation of the radar trace, the automatic marking of the radar trace data can be stably and accurately carried out, and the method has a very good practical value.
According to some embodiments of the invention, the performing the feature sensitivity analysis on the radar trace data comprises:
and for the radar trace characteristics of each piece of radar trace data, carrying out qualitative analysis and quantitative analysis on the contribution of the radar trace characteristics to radar true and false trace classification by using a true and false trace scatter diagram and a frequency distribution statistical diagram corresponding to the radar trace characteristics, and sequencing the separability of all the radar trace characteristics to the radar true and false traces.
According to some embodiments of the invention, said taking said sensitive feature as an input to a CFDP algorithm comprises the steps of:
and taking the number of traces, the azimuth extension, the distance extension, the amplitude variance, the nearest trace distance and the trace duration as the input of the CFDP algorithm.
According to some embodiments of the present invention, the clustering the radar trace data to obtain a plurality of clusters includes:
and (3) clustering the radar trace data in a high-dimensional space to obtain a plurality of clusters by iteration of local density and distance between the radar trace data and a higher density point.
According to some embodiments of the present invention, the merging the clusters satisfying the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result includes:
When two clusters have similar intra-cluster average distances and the minimum distance of the two clusters is similar to the average distance of the two clusters, marking the two clusters as clusters to be merged;
and if the overall DBI value of the clusters to be merged is reduced after merging, merging the clusters to be merged to obtain a merging result.
According to some embodiments of the invention, the two clusters are clustered with Ci、CjIs represented by Ci、CjHaving similar intra-cluster average distances should satisfy the following condition:
d(Ci,Cj)＜3min(dist(Ci),dist(Cj))
the average distance within a cluster is the average of the distances from all objects within the cluster to its nearest neighbor object;
Figure BDA0003488245710000031
neigh(xiand 2) is a neighbor function representing the distance x except itselfiThe most recent object;
the minimum distance between the two clusters and the average distance between the two clusters are similar to satisfy the following condition:
2min(dist(Ci),dist(Cj))＞max(dist(Ci),dist(Cj))
the inter-cluster distance is represented by the minimum distance.
Figure BDA0003488245710000032
According to some embodiments of the present invention, the clustering the radar trace data to obtain a plurality of clusters includes:
in the select outlier stage, the outlier condition is relaxed so that the CFDP algorithm tends to produce more classes.
In a second aspect, the present invention provides an automatic labeling system for radar trace data, including:
The characteristic sensitivity analysis module is used for carrying out characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space and screening out a plurality of sensitive characteristics with separability above a preset value;
the CFDP algorithm clustering processing module is in communication connection with the characteristic sensitivity analysis module and is used for clustering the radar trace data to obtain a plurality of clusters;
the merging module is in communication connection with the CFDP algorithm clustering processing module and is used for merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merging result;
the DBI condition judgment module is in communication connection with the merging module and is used for judging whether the merging result meets the integral DBI condition;
and the output module is in communication connection with the DBI condition judgment module and is used for outputting the merging result meeting the integral DBI condition as a final class.
In a third aspect, the present invention provides an automatic labeling system for radar trace data, including: the system 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 computer program to realize the automatic labeling method of the radar trace data according to any one of the first aspect.
In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are configured to cause a computer to execute the method for automatically labeling radar trace data according to any one of the first aspect.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which the abstract is to be fully consistent with one of the figures of the specification:
fig. 1 is a flowchart of an automatic radar trace data labeling method according to an embodiment of the present invention;
fig. 2 is a flowchart of an automatic labeling method for radar trace data according to another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
It is noted that while a division of functional blocks is depicted in the system diagram, and logical order is depicted in the flowchart, in some cases the steps depicted and described may be performed in a different order than the division of blocks in the system or the flowchart. The terms first, second and the like in the description and in the claims, as well as in the drawings described above, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
The invention provides an automatic labeling method for radar trace data, which better covers the original category of the radar trace data by dividing the trace data through an improved CFDP algorithm, can meet the requirement on the sample accuracy in the automatic labeling of the radar trace data, can provide an effective data basis for radar trace quality evaluation, can stably and accurately perform the automatic labeling of the radar trace data, and has very good practical value.
The embodiments of the present invention will be further explained with reference to the drawings.
Referring to fig. 1, fig. 1 is a flowchart of an automatic labeling method for radar trace data according to an embodiment of the present invention, where the automatic labeling method for radar trace data includes, but is not limited to, steps S110 to S150.
Step S110, acquiring radar trace data;
step S120, performing characteristic sensitivity analysis on radar trace point data, establishing a trace point classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value;
step S130, taking the sensitive characteristics as the input of a CFDP algorithm, and clustering radar trace point data to obtain a plurality of clusters;
step S140, merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not;
and S150, outputting the merging result as a final class when the merging result meets the integral DBI condition.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace point data; secondly, performing characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace point data are clustered to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. The improved CFDP algorithm divides the trace data, so that the original category of the radar trace data is better covered, the requirement on the sample accuracy in the automatic marking of the radar trace data can be met, an effective data base can be provided for the quality evaluation of the radar trace, the automatic marking of the radar trace data can be stably and accurately carried out, and the method has a very good practical value.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace point data; secondly, performing characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace point data are clustered to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. The characteristic sensitivity analysis is carried out on the radar trace data, and the method comprises the following steps: and for the radar trace characteristics of each radar trace data, carrying out qualitative analysis and quantitative analysis on the contribution of the radar trace characteristics to radar true and false trace classification by using a true and false trace scatter diagram and a frequency distribution statistical diagram corresponding to the radar trace characteristics, and sequencing the separability of all the radar trace characteristics to the radar true and false traces.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace data; secondly, performing characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace point data are clustered to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. Taking the sensitive characteristic as an input of a CFDP algorithm, and comprising the following steps: and taking the number of traces, the azimuth extension, the distance extension, the amplitude variance, the nearest trace distance and the trace duration as the input of the CFDP algorithm.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace data; secondly, performing characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace point data are clustered to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. Clustering radar trace point data to obtain a plurality of clusters, comprising the following steps: and (3) clustering the radar trace data in a high-dimensional space to obtain a plurality of clusters by iteration of the local density and the distance between the radar trace data and a higher density point.
Referring to fig. 2, fig. 2 is a flowchart of an automatic labeling method for radar trace data according to another embodiment of the present invention, where the automatic labeling method for radar trace data includes, but is not limited to, steps S210 to S220.
Step S210, when the two clusters have similar intra-cluster average distances and the minimum distance of the two clusters is similar to the average distance of the two clusters, marking the two clusters as clusters to be merged;
and step S220, if the overall DBI value of the clusters to be merged is reduced after merging, merging the clusters to be merged to obtain a merging result.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace data; secondly, performing characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace point data are clustered to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. Merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, wherein the merging step comprises the following steps of: when the two clusters have similar intra-cluster average distances and the minimum distance of the two clusters is similar to the average distance of the two clusters, marking the two clusters as clusters to be merged; and if the overall DBI value of the clusters to be merged is reduced after merging, merging the clusters to be merged to obtain a merging result.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace point data; secondly, performing characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value; secondly, clustering radar trace point data by taking the sensitive characteristics as the input of a CFDP algorithm to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. Merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, wherein the merging step comprises the following steps of: when the two clusters have similar intra-cluster average distances and the minimum distance of the two clusters is similar to the average distance of the two clusters, marking the two clusters as clusters to be merged; and if the overall DBI value of the clusters to be merged is reduced after merging, merging the clusters to be merged to obtain a merging result.
For the CFDP clustering result C, considering any two clusters in C, if two clusters Ci and Cj have similar intra-cluster average distances and the minimum distance between the two clusters is close to the average distance between the two clusters, then Ci and Cj may be two different clusters that are relatively close to each other, or may be two sub-clusters in a multi-peak cluster, which is called a cluster to be merged. At this time, the DBI index can be used for judgment, and if the combined Ci and Cj can reduce the overall DBI value (the smaller the DBI value, the better the clustering result), the two clusters are considered to be two sub-clusters in the multi-peak cluster, and should be combined.
Two clusters using Ci、CjIs represented by Ci、CjHaving similar intra-cluster average distances should satisfy the following condition:
d(Ci,Cj)＜3min(dist(Ci),dist(Cj))
the average distance within a cluster is the average of the distances from all objects within the cluster to its nearest neighbor object;
Figure BDA0003488245710000081
neigh(xiand 2) is a neighbor function representing the distance x except itselfiThe most recent object;
the minimum distance between two clusters and the average distance between two clusters should be similar to satisfy the following condition:
2min(dist(Ci),dist(Cj))＞max(dist(Ci),dist(Cj))
the inter-cluster distance is represented by the minimum distance.
Figure BDA0003488245710000082
DBI is defined as follows:
Figure BDA0003488245710000083
Figure BDA0003488245710000091
Figure BDA0003488245710000092
Figure BDA0003488245710000093
Figure BDA0003488245710000094
the improved CFDP clustering algorithm pseudo-code is as follows:
Figure BDA0003488245710000095
the input to the algorithm is an m X n dimensional dataset X and the output is a cluster structure C. The algorithm is divided into two stages of CFDP clustering and merging. Firstly, a CFDP algorithm is used for a data set X to obtain a cluster structure C of a clustering result, and a merging list is initialized and used for storing class marks corresponding to two clusters to be merged. For result C of CFDP, consider any two clusters Ci、CjAnd if the closeness degree of the cluster-to-cluster distance and the cluster-to-cluster average distance of the two clusters meets the merging condition, the two clusters are considered as the clusters to be merged. Then trying to merge the clusters to be merged, if the DBI of the merged whole is reduced, the two clusters can be merged, and the class pairs (p) corresponding to the two clusters are markedi、pj) Put into merge list mergePairs. After all the cluster pairs are traversed once, actually merging the C, sequentially taking out the cluster pairs from the mergePairs list, merging, and finally returning to the C.
In one embodiment, the automatic labeling method for radar trace data comprises the following steps: firstly, acquiring radar trace point data; then, performing feature sensitivity analysis on radar trace point data, establishing a trace point classification sensitive feature space, and screening out a plurality of sensitive features with separability above a preset value; secondly, the sensitive characteristics are used as the input of a CFDP algorithm, and radar trace point data are clustered to obtain a plurality of clusters; secondly, merging the clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not; and finally, when the merging result meets the integral DBI condition, outputting the merging result as a final class. Clustering radar trace point data to obtain a plurality of clusters, comprising the following steps: in the select outlier phase, the outlier condition is relaxed so that the CFDP algorithm tends to produce more classes.
The invention also provides an automatic marking system for radar trace data, which comprises: the characteristic sensitivity analysis module is used for carrying out characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space and screening out a plurality of sensitive characteristics with separability above a preset value; the CFDP algorithm clustering processing module is in communication connection with the characteristic sensitivity analysis module and is used for clustering radar trace data to obtain a plurality of clusters; the merging module is in communication connection with the CFDP algorithm clustering processing module and is used for merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merging result; the DBI condition judgment module is in communication connection with the merging module and is used for judging whether the merging result meets the integral DBI condition; and the output module is in communication connection with the DBI condition judgment module and is used for outputting the merging result meeting the integral DBI condition as a final class.
The invention also provides an automatic marking system for radar trace data, which comprises: the system comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the automatic marking method of the radar trace data when executing the computer program.
The processor and memory may be connected by a bus or other means.
The memory, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs as well as non-transitory computer executable programs. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the processor, and these remote memories may be connected to the processor through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
It should be noted that the system for automatically labeling radar trace data in this embodiment may include a service processing module, an edge database, a server version information register, and a data synchronization module, and when a processor executes a computer program, the method for automatically labeling radar trace data applied to the system for automatically labeling radar trace data is implemented.
The above-described embodiments of the apparatus are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, i.e. may be located in one place, or may also be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
Furthermore, an embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by a processor or a controller, for example, by a processor in the terminal embodiment, so that the processor executes the method for automatically labeling radar trace data in the above embodiment.
One of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those of ordinary skill in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.
While the preferred embodiments of the present invention have been described in detail, it will be understood by those skilled in the art that the foregoing and various other changes, omissions and deviations in the form and detail thereof may be made without departing from the scope of this invention.
The above-described embodiments of the present invention should not be construed as limiting the scope of the present invention. Any other corresponding changes and modifications made according to the technical idea of the present invention should be included in the protection scope of the claims of the present invention.

Claims (10)

1. An automatic labeling method for radar trace data is characterized by comprising the following steps:
acquiring radar trace data;
performing characteristic sensitivity analysis on the radar trace data, establishing a trace classification sensitive characteristic space, and screening out a plurality of sensitive characteristics with separability above a preset value;
taking the sensitive characteristics as the input of a CFDP algorithm, and clustering the radar trace data to obtain a plurality of clusters;
merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result, and judging whether the merged result meets the integral DBI condition or not;
And when the merging result meets the integral DBI condition, outputting the merging result as a final class.
2. The method for automatically labeling radar locus data according to claim 1, wherein the characteristic sensitivity analysis of the radar locus data comprises the following steps:
and for the radar trace characteristics of each piece of radar trace data, carrying out qualitative analysis and quantitative analysis on the contribution of the radar trace characteristics to radar true and false trace classification by using a true and false trace scatter diagram and a frequency distribution statistical diagram corresponding to the radar trace characteristics, and sequencing the separability of all the radar trace characteristics to the radar true and false traces.
3. The method for automatically labeling radar locus data as claimed in claim 1, wherein the step of taking the sensitive feature as an input of a CFDP algorithm comprises the steps of:
and taking the number of traces, the azimuth extension, the distance extension, the amplitude variance, the nearest trace distance and the trace duration as the input of the CFDP algorithm.
4. The method for automatically labeling radar locus data according to claim 1, wherein the clustering processing of the radar locus data to obtain a plurality of clusters comprises the steps of:
And (3) performing clustering processing on the radar trace data in a high-dimensional space to obtain a plurality of clusters through iteration of local density and distance between the radar trace data and a higher density point.
5. The method for automatically labeling radar trace point data as claimed in claim 1, wherein the step of merging clusters satisfying the intra-cluster distance condition and the inter-cluster distance condition to obtain a merged result comprises the steps of:
when two clusters have similar intra-cluster average distances and the minimum distance of the two clusters is similar to the average distance of the two clusters, marking the two clusters as clusters to be merged;
and if the overall DBI value of the clusters to be merged is reduced after merging, merging the clusters to be merged to obtain a merging result.
6. The method as claimed in claim 5, wherein the two clusters are marked with Ci、CjIs represented by Ci、CjHaving similar intra-cluster average distances should satisfy the following condition:
d(Ci,Cj)＜3min(dist(Ci),dist(Cj))
the average distance within a cluster is the average of the distances from all objects within the cluster to its nearest neighbor object;
Figure FDA0003488245700000021
neigh(xiand 2) is a neighbor function representing the distance x except itselfiThe most recent object;
the minimum distance between the two clusters and the average distance between the two clusters are similar to satisfy the following condition:
2min(dist(Ci),dist(Cj))＞max(dist(Ci),dist(Cj))
The inter-cluster distance is represented by the minimum distance.
Figure FDA0003488245700000022
7. The method for automatically labeling radar locus data according to claim 1, wherein the clustering processing of the radar locus data to obtain a plurality of clusters comprises the steps of:
in the select outliers stage, the outliers condition is relaxed so that the CFDP algorithm tends to produce more classes.
8. An automatic labeling system for radar trace data is characterized by comprising:
the characteristic sensitivity analysis module is used for carrying out characteristic sensitivity analysis on radar trace data, establishing a trace classification sensitive characteristic space and screening out a plurality of sensitive characteristics with separability above a preset value;
the CFDP algorithm clustering processing module is in communication connection with the characteristic sensitivity analysis module and is used for clustering the radar trace data to obtain a plurality of clusters;
the merging module is in communication connection with the CFDP algorithm clustering processing module and is used for merging clusters meeting the intra-cluster distance condition and the inter-cluster distance condition to obtain a merging result;
the DBI condition judgment module is in communication connection with the merging module and is used for judging whether the merging result meets the integral DBI condition;
And the output module is in communication connection with the DBI condition judgment module and is used for outputting the merging result meeting the integral DBI condition as a final class.
9. An automatic labeling system for radar trace data is characterized by comprising: the device 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 computer program to realize the automatic radar locus data labeling method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing computer-executable instructions for causing a computer to execute the method for automatically labeling radar locus data according to any one of claims 1 to 7.
CN202210086638.6A 2022-01-25 2022-01-25 Automatic labeling method, system and storage medium for radar trace data Pending CN114594437A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210086638.6A CN114594437A (en) 2022-01-25 2022-01-25 Automatic labeling method, system and storage medium for radar trace data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210086638.6A CN114594437A (en) 2022-01-25 2022-01-25 Automatic labeling method, system and storage medium for radar trace data

Publications (1)

Publication Number Publication Date
CN114594437A true CN114594437A (en) 2022-06-07

Family

ID=81807018

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210086638.6A Pending CN114594437A (en) 2022-01-25 2022-01-25 Automatic labeling method, system and storage medium for radar trace data

Country Status (1)

Country Link
CN (1) CN114594437A (en)

Similar Documents

Publication Publication Date Title
US8737739B2 (en) Active segmentation for groups of images
US10176246B2 (en) Fast grouping of time series
CN112348081B (en) Migration learning method for image classification, related device and storage medium
CN112529204A (en) Model training method, device and system
CN112329847A (en) Abnormity detection method and device, electronic equipment and storage medium
CN113742387A (en) Data processing method, device and computer readable storage medium
CN112766170B (en) Self-adaptive segmentation detection method and device based on cluster unmanned aerial vehicle image
CN111914932A (en) Model selection and parameter estimation for anomaly detection
CN111723815A (en) Model training method, image processing method, device, computer system, and medium
CN111291867A (en) Data prediction model generation method and device and data prediction method and device
CN112329617A (en) New scene face recognition model construction method and system based on single source domain sample
CN113191432A (en) Outlier factor-based virtual machine cluster anomaly detection method, device and medium
CN112597871A (en) Unsupervised vehicle re-identification method and system based on two-stage clustering and storage medium
CN114594437A (en) Automatic labeling method, system and storage medium for radar trace data
CN113012088A (en) Circuit board fault detection and twin network training method, device and equipment
CN109767457A (en) Online multi-instance learning method for tracking target, terminal device and storage medium
CN117523218A (en) Label generation, training of image classification model and image classification method and device
CN112115996B (en) Image data processing method, device, equipment and storage medium
CN113255601B (en) Training method and system for vehicle weight recognition model and related equipment
CN112884866B (en) Coloring method, device, equipment and storage medium for black-and-white video
CN115292361A (en) Method and system for screening distributed energy abnormal data
CN114898182A (en) Image data screening method and system based on target detection learning algorithm
CN117171141B (en) Data model modeling method based on relational graph
CN116541252B (en) Computer room fault log data processing method and device
US20240061867A1 (en) Cluster-based dataset evolution

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