CN112710996A - Data set expansion method and system for radar micro-motion target identification - Google Patents

Data set expansion method and system for radar micro-motion target identification Download PDF

Info

Publication number
CN112710996A
CN112710996A CN202011441343.3A CN202011441343A CN112710996A CN 112710996 A CN112710996 A CN 112710996A CN 202011441343 A CN202011441343 A CN 202011441343A CN 112710996 A CN112710996 A CN 112710996A
Authority
CN
China
Prior art keywords
diagram
time
pulse
distance
frequency
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011441343.3A
Other languages
Chinese (zh)
Other versions
CN112710996B (en
Inventor
理解放
陈小龙
丁昊
吴幸
张海
刘宁波
黄勇
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.)
East China Normal University
Naval Aeronautical University
Original Assignee
East China Normal University
Naval Aeronautical University
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 East China Normal University, Naval Aeronautical University filed Critical East China Normal University
Priority to CN202011441343.3A priority Critical patent/CN112710996B/en
Publication of CN112710996A publication Critical patent/CN112710996A/en
Application granted granted Critical
Publication of CN112710996B publication Critical patent/CN112710996B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • 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
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention relates to a data set expansion method for radar micro-motion target identification. The method comprises the following steps: acquiring echo data of a target radar; determining an echo distance-pulse diagram from echo data, and extracting the micro Doppler characteristics of the distance-pulse diagram by adopting a time-frequency analysis method to obtain a time-frequency diagram; respectively adjusting the color display ranges of the distance-pulse diagram and the time-frequency diagram to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram; respectively cutting the target area of the adjusted distance-pulse diagram and the target area of the adjusted time-frequency diagram to obtain a cut distance-pulse diagram and a cut time-frequency diagram; and performing feature fusion on the cut distance-pulse graph and the cut time-frequency graph to obtain an expanded data set. The method solves the problem of small image sample data amount in the existing radar target identification, and can enhance the characteristic information of the micro-motion target and realize the high-precision identification of the radar target.

Description

Data set expansion method and system for radar micro-motion target identification
Technical Field
The invention relates to the technical field of radar target identification, in particular to a data set expansion method and system for radar micro-motion target identification.
Background
The essence of the radar micro-motion target identification method is that information capable of reflecting target characteristics is extracted from radar echo signals of targets, and the target information is subjected to feature learning by using a machine learning and pattern identification method, wherein the key point is feature selection. The characteristics of the target radar echo signal are selected from a high-resolution range profile characteristic and a micro Doppler characteristic, the high-resolution range profile characteristic has good discrimination on different targets, and the time-varying information of the distance of the target can be acquired; the micro Doppler characteristic is a unique characteristic caused by micro motion of a target, and can acquire characteristic information of micro motion such as vibration, rotation and the like of the target and components thereof.
At present, the identification and classification of radar micro-motion targets use either high-resolution range profiles as target micro-motion identification data sets or time-frequency graphs processed by a time-frequency analysis method as target micro-motion identification data sets, so that only limited characteristic information can be obtained by independently using respective target characteristics as the identified data sets, the optimal accuracy cannot be obtained by reflecting the characteristic information on the target identification precision, and the acquired radar data are generally limited by radar technology and labor cost. The data set is expanded by using a traditional image expansion mode, for example, the data set can be expanded to a certain extent by methods such as image rotation, shearing, blurring and noise addition, but the change of effective characteristics cannot be fundamentally realized, only an image extremely similar to an original image is generated, and along with the increase of the expansion data amount, more and more data in the data set can also cause network overfitting, and the generalization capability is poor.
Disclosure of Invention
The invention aims to provide a data set expansion method and a data set expansion system for radar micro-motion target identification, which are used for solving the problem of small image sample data amount in the existing radar target identification, can enhance the characteristic information of a micro-motion target and provide effective support for realizing high-precision identification of the radar target.
In order to achieve the purpose, the invention provides the following scheme:
a data set expansion method for radar micro-motion target recognition comprises the following steps:
acquiring echo data of a target radar;
determining an echo range-pulse map from the echo data, the echo range-pulse map representing a relationship between a range gate count and a pulse count;
performing micro Doppler feature extraction on the distance-pulse diagram by adopting a time-frequency analysis method to obtain a time-frequency diagram;
respectively adjusting the color display ranges of the distance-pulse diagram and the time-frequency diagram to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram;
respectively cutting the target area of the adjusted distance-pulse diagram and the target area of the adjusted time-frequency diagram according to a preset cutting size to obtain a cut distance-pulse diagram and a cut time-frequency diagram, wherein the target area is an area containing micro-motion characteristics;
and performing feature fusion on the cut distance-pulse graph and the cut time-frequency graph to obtain an expanded data set.
Optionally, the performing micro doppler feature extraction on the distance-pulse diagram by using a time-frequency analysis method further includes:
and performing inter-pulse cancellation on the echo distance-pulse diagram by adopting a moving target display technology to obtain a processed echo distance-pulse diagram.
Optionally, the distance-pulse diagram is subjected to micro doppler feature extraction by using a time-frequency analysis method to obtain a time-frequency diagram, which specifically includes:
selecting a distance-pulse graph segment corresponding to a target distance unit from the distance-pulse graph as a target distance-pulse graph;
and extracting the target distance-pulse diagram by adopting short-time Fourier transform to obtain a time-frequency diagram.
Optionally, the color display ranges of the distance-pulse diagram and the time-frequency diagram are respectively adjusted to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram, and specifically:
using imagesc (C)1,clims1) Adjusting the color display range of the distance-pulse diagram by the function to obtain an adjusted distance-pulse diagram, wherein clims1＝[cmin1 cmax1]，C1Is the distance-pulse diagram, clims1Color display range, cmin, to be adjusted for the distance-pulse diagram1For the minimum value, cmax, of the color display range to be adjusted for the distance-pulse diagram1Maximum value of color display range to be adjusted for the distance-pulse diagram
Using imagesc (C)2,clims2) Adjusting the color display range of the time-frequency diagram by a function to obtain an adjusted time-frequency diagram, wherein clims2＝[cmin2 cmax2]，C2Is said time-frequency diagram, clims2Color display Range, cmin, to be adjusted for said time-frequency diagram2For the minimum value, cmax, of the color display range to be adjusted for the time-frequency diagram2And the maximum value of the color display range to be adjusted of the time-frequency graph is obtained.
Optionally, the extracting the target distance-pulse diagram by using short-time fourier transform to obtain a time-frequency diagram specifically includes:
applying a formula to the target distance-pulse diagram
Figure BDA0002822355010000031
Extracting to obtain time-frequency diagram, wherein STFTs(t, ω) time-frequency diagram, SM×NAnd (u) is distance-pulse data corresponding to the target distance-pulse diagram, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a window function, and j is an imaginary number unit.
A data set augmentation system for radar micro-motion target identification, comprising:
the data acquisition module is used for acquiring echo data of the target radar;
a range-pulse map determination module for determining an echo range-pulse map from the echo data, the echo range-pulse map representing a relationship between a range gate count and a pulse count;
the time-frequency diagram determining module is used for extracting the micro Doppler characteristics of the distance-pulse diagram by adopting a time-frequency analysis method to obtain a time-frequency diagram;
the color display range adjusting module is used for respectively adjusting the color display ranges of the distance-pulse diagram and the time-frequency diagram to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram;
the cutting module is used for respectively cutting the target area of the adjusted distance-pulse diagram and the target area of the adjusted time-frequency diagram according to a preset cutting size to obtain the cut distance-pulse diagram and the cut time-frequency diagram, wherein the target area is an area containing micro-motion characteristics;
and the data set determining module is used for performing feature fusion on the cut distance-pulse graph and the cut time-frequency graph to obtain an expanded data set.
Optionally, the data set expansion system for radar micro-motion target recognition further includes:
and the cancellation module is used for performing inter-pulse cancellation on the echo distance-pulse diagram by adopting a moving target display technology to obtain a processed echo distance-pulse diagram.
Optionally, the time-frequency diagram determining module includes:
the target distance-pulse graph determining unit is used for selecting a distance-pulse graph segment corresponding to the target distance unit from the distance-pulse graph as a target distance-pulse graph;
and the time-frequency diagram determining unit is used for extracting the target distance-pulse diagram by adopting short-time Fourier transform to obtain a time-frequency diagram.
Optionally, the color display range adjusting module includes:
a first color display range adjusting module for utilizing imagesc (C)1,clims1) Adjusting the color display range of the distance-pulse diagram by the function to obtain an adjusted distance-pulse diagram, wherein clims1＝[cmin1 cmax1]，C1Is the distance-pulse diagram, clims1Color display range, cmin, to be adjusted for the distance-pulse diagram1For the minimum value of the color display range to be adjusted by the distance-pulse diagram, cmax1 is the maximum value of the color display range to be adjusted by the distance-pulse diagram
A second color display range adjusting module for utilizing imagesc (C)2,clims2) Adjusting the color display range of the time-frequency diagram by a function to obtain an adjusted time-frequency diagram, wherein clims2＝[cmin2 cmax2]， C2Is said time-frequency diagram, clims2Color display Range, cmin, to be adjusted for said time-frequency diagram2For the minimum value, cmax, of the color display range to be adjusted for the time-frequency diagram2And the maximum value of the color display range to be adjusted of the time-frequency graph is obtained.
Optionally, the time-frequency diagram determining unit includes:
a time-frequency diagram determining subunit for applying a formula to the target distance-pulse diagram
Figure DEST_PATH_FDA0002822354000000021
Extracting to obtain time-frequency diagram, wherein STFTs(t, ω) is a time-frequency diagram, SM×NAnd (u) is distance-pulse data corresponding to the target distance-pulse diagram, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a window function, and j is an imaginary number unit.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: according to the invention, the distance-pulse diagram and the time-frequency diagram are adjusted in color display range, so that the distance-pulse diagram and the time-frequency diagram have wider color display range, the sample size of the image is effectively expanded, the problem of less image sample data size in the conventional radar target identification is solved, the useful characteristic information of the target is merged and fused by using a characteristic fusion mode, the target identification classification precision and reliability are effectively improved, and the high-precision identification of the radar target is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flowchart of a data set expansion method for radar micro-motion target identification according to an embodiment of the present invention;
FIG. 2 is a block diagram of a data set expansion system for radar micro-motion target identification according to an embodiment of the present invention;
FIG. 3 is a flow chart of a more specific data set expansion method for radar micro-motion target identification according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating the result of adjusting the color display range of the distance-pulse diagram according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating a result of adjusting a color display range of a time-frequency diagram according to an embodiment of the present invention;
fig. 6 is a flowchart for performing feature fusion on a distance-pulse diagram and a time-frequency diagram according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the present embodiment provides a data set expansion method for radar micro-motion target recognition, the method includes:
101: and acquiring echo data of the target radar.
102: an echo range-pulse map is determined from the echo data. The echo distance-pulse diagram represents the relationship between the number of distance gates and the number of pulses.
103: and (4) performing micro Doppler feature extraction on the distance-pulse diagram by adopting a time-frequency analysis method to obtain a time-frequency diagram.
104: and respectively adjusting the color display ranges of the distance-pulse diagram and the time-frequency diagram to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram.
105: and respectively cutting the target area of the adjusted distance-pulse diagram and the target area of the adjusted time-frequency diagram according to a preset cutting size to obtain the cut distance-pulse diagram and the cut time-frequency diagram. The target region is a region containing a micro-motion feature.
106: and performing feature fusion on the cut distance-pulse graph and the cut time-frequency graph to obtain an expanded data set.
In practical application, before 103, the method further comprises:
and performing inter-pulse cancellation on the echo distance-pulse diagram by adopting a moving target display technology to obtain a processed echo distance-pulse diagram.
In practical application, 103 is specifically:
and selecting a distance-pulse segment corresponding to the target distance unit from the distance-pulse map as a target distance-pulse map.
And extracting the target distance-pulse diagram by adopting short-time Fourier transform to obtain a time-frequency diagram.
In practical application, the target distance-pulse diagram is extracted by short-time fourier transform to obtain a time-frequency diagram, which specifically comprises the following steps:
applying a formula to the target distance-pulse diagram
Figure BDA0002822355010000061
Extracting to obtain time-frequency diagram, wherein STFTs(t, ω) is a time-frequency diagram representing the original signal at the time when u equals t, SM×NAnd (u) is distance-pulse data corresponding to the target distance-pulse diagram, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a window function representing that different t is brought in, the window function continuously slides so as to know frequency components in different time periods, and j is an imaginary number unit.
In practical application, 104 is specifically:
using imagesc (C)1,clims1) Adjusting the color display range of the distance-pulse diagram by the function to obtain an adjusted distance-pulse diagram, wherein clims1＝[cmin1 cmax1]，C1Is the distance-pulse diagram, clims1Color display range, cmin, to be adjusted for the distance-pulse diagram1For the minimum value, cmax, of the color display range to be adjusted for the distance-pulse diagram1Maximum value of color display range to be adjusted for the distance-pulse diagram
Using imagesc (C)2,clims2) Adjusting the color display range of the time-frequency diagram by a function to obtain an adjusted time-frequency diagram, wherein clims2＝[cmin2 cmax2]，C2Is said time-frequency diagram, clims2Color display Range, cmin, to be adjusted for said time-frequency diagram2For the minimum value, cmax, of the color display range to be adjusted for the time-frequency diagram2And the maximum value of the color display range to be adjusted of the time-frequency graph is obtained.
As shown in fig. 2, the present embodiment further provides a data set expansion system for radar micro-motion target recognition, the system includes:
and the data acquisition module A1 is used for acquiring echo data of the target radar.
A range-pulse map determining module A2 for determining an echo range-pulse map from the echo data, the echo range-pulse map representing a range gate count versus a pulse count.
And the time-frequency diagram determining module A3 is used for extracting the micro Doppler characteristics of the distance-pulse diagram by adopting a time-frequency analysis method to obtain the time-frequency diagram.
And the color display range adjusting module A4 is configured to adjust the color display ranges of the distance-pulse diagram and the time-frequency diagram, respectively, to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram.
And the cutting module A5 is configured to cut the target area of the adjusted distance-pulse graph and the target area of the adjusted time-frequency graph according to a preset cutting size, so as to obtain a cut distance-pulse graph and a cut time-frequency graph, where the target area is an area containing a fine motion feature.
And the data set determining module A6 is used for performing feature fusion on the clipped distance-pulse graph and the clipped time-frequency graph to obtain an expanded data set.
As an optional implementation, the data set expansion system for radar micro-motion target recognition further includes:
and the cancellation module is used for performing inter-pulse cancellation on the echo distance-pulse diagram by adopting a moving target display technology to obtain a processed echo distance-pulse diagram.
As an optional implementation manner, the time-frequency diagram determining module includes:
and the target distance-pulse map determining unit is used for selecting the distance-pulse map segment corresponding to the target distance unit from the distance-pulse map as the target distance-pulse map.
And the time-frequency diagram determining unit is used for extracting the target distance-pulse diagram by adopting short-time Fourier transform to obtain a time-frequency diagram.
As an optional implementation manner, the color display range adjustment module includes:
a first color display range adjusting module for utilizing imagesc (C)1,clims1) The function adjusts the color display range of the distance-pulse diagram to obtain the adjusted color display rangeDistance-pulse diagram of (1), wherein clims1＝[cmin1 cmax1]，C1Is the distance-pulse diagram, clims1Color display range, cmin, to be adjusted for the distance-pulse diagram1For the minimum value, cmax, of the color display range to be adjusted for the distance-pulse diagram1Maximum value of color display range to be adjusted for the distance-pulse diagram
A second color display range adjusting module for utilizing imagesc (C)2,clims2) Adjusting the color display range of the time-frequency diagram by a function to obtain an adjusted time-frequency diagram, wherein clims2＝[cmin2 cmax2]，C2Is said time-frequency diagram, clims2Color display Range, cmin, to be adjusted for said time-frequency diagram2For the minimum value, cmax, of the color display range to be adjusted for the time-frequency diagram2And the maximum value of the color display range to be adjusted of the time-frequency graph is obtained.
As an optional implementation manner, the time-frequency diagram determining unit includes:
a time-frequency diagram determining subunit for applying a formula to the target distance-pulse diagram
Figure BDA0002822355010000081
Extracting to obtain time-frequency diagram, wherein STFTs(t, ω) is a time-frequency diagram, SM×NAnd (u) is distance-pulse data corresponding to the target distance-pulse diagram, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a window function representing that different t is brought in, the window function continuously slides so as to know frequency components in different time periods, and j is an imaginary number unit.
This example provides a more specific implementation, as shown in fig. 3:
step 1: the method comprises the steps of obtaining radar echo data, preprocessing the radar echo data by utilizing a moving target display technology, performing pulse-to-pulse cancellation on echo distance-pulse data, realizing high-pass filtering processing of echoes, filtering fixed clutter, and obtaining a two-dimensional distance-pulse diagram.
Storing a radar echo range-pulse two-dimensional data (radar echo data) matrix of
S M×N1, ·, M, { s (i, j) | i ═ 1, 2; j is 1,2, the.. N }, M is the distance unit number, N is the number of all echo pulses, echo distance-pulse data are subjected to inter-pulse cancellation by adopting a moving target display (MTI) technology, and the inter-pulse cancellation is performed on SM×NThe MTI can adopt methods of secondary pulse cancellation, tertiary pulse cancellation and the like to realize high-pass filtering of distance-pulse data and filter out fixed clutter.
Step 2: selecting a target distance unit from the distance-pulse graph, carrying out time-frequency transformation on pulse dimensional data of the target distance unit, extracting the micro Doppler characteristics of the target distance unit by using a time-frequency analysis method, and obtaining a target micro Doppler two-dimensional time-frequency graph after characteristic extraction, and a target micro Doppler two-dimensional time-frequency graph STFTs(t,ω)：
Figure BDA0002822355010000082
Wherein S isM×NAnd (u) is a target echo signal (an echo signal corresponding to a target distance unit) after moving target display processing, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a movable window function, and variable parameters are window length of time-frequency transformation, so that a target micro Doppler two-dimensional time-frequency diagram is obtained. Taking the walking motion state of the human body as an example, selecting a distance unit 5-25 where the target is located, and extracting the micro-motion characteristics of the human body motion by using STFT or S transformation to obtain a time-frequency diagram.
And step 3: expanding the data set of the obtained two-dimensional distance-pulse diagram and the obtained time-frequency diagram by changing the spectrum color display amplitude range; the characteristic that the target distance changes along with time is enhanced or weakened by controlling the color range display of the two-dimensional distance-pulse, and a data set is expanded; and the color range display of the micro Doppler features in the time-frequency diagram is controlled, so that the micro Doppler features of the target are enhanced or weakened, and the data set is expanded. Different color ranges can be properly set for display according to the feature display of the spectrogram, and data sets with different scales are obtained.
And (3) performing data set expansion on the distance-pulse graph and the time-frequency graph which obtain the micro-motion of the target, displaying data in a target array C (the obtained distance-pulse graph and the time-frequency graph) as an image by using imagesc (C, clims) and setting a color range, wherein the color range is specified as a binary element vector in a clims ═ cmin cmax form, and the cmax is greater than cmin. Values in C less than or equal to cmin map to the first color in the color column (cmin), values greater than or equal to cmax map to the last color in the color column (cmax), and values between cmin and cmax map to the closest color in the color column.
Different color ranges can be properly set for display according to the feature display of the spectrogram, and data sets with different scales are obtained. Taking the walking motion state of the human body as an example, the distance-pulse diagram is set with a color display range of [ A,0], the values of A are respectively designated as-30, -40, -50 and-60, and the data set of the distance-pulse diagram is expanded as shown in FIG. 4. The color display range of the time-frequency diagram is set as [ A,0], the numerical values of A are respectively-20, -30, -40 and-50, and the data set expansion of the time-frequency diagram is shown in figure 5.
And 4, step 4: the resulting augmented data set is cropped and feature fused as shown in FIG. 6.
And 4-1, selecting an area containing the target micro-motion characteristics from the expanded distance-pulse diagram and the expanded time-frequency diagram obtained in the step 3 for cutting, wherein the size of the cutting area has a uniform size, and naming the image of the cutting area containing the target micro-motion characteristics according to the motion state of the target micro-motion to obtain the uniform size of the cut image containing the target micro-motion characteristics.
Step 4-2, the feature fusion is to merge and fuse the distance-pulse graph and the time-frequency graph containing the target micro-motion features after being cut, the merging and fusion is to merge and fuse two images which have the same name for the distance-pulse graph and the time-frequency graph and have the difference value of the color display range as a set value, for example: correspondingly fusing the clipped distance-pulse diagram containing the color display range of the walking state of the human body, which is between-30, 0, 40,0, 50,0 and 60,0, with the clipped time-frequency diagram containing the color display range of the walking state of the human body, which is between-20, 0, 30,0, 40,0 and 50,0, and naming the merged and fused data by the same naming in the distance-pulse diagram or the time-frequency diagram, namely, carrying out label calibration on the data set. The merging and fusing requires that the image widths of the distance-pulse graph and the time-frequency graph are kept consistent. The merged fusion of the images can be batch processed using drawing software or based on Python programming algorithms.
The data set expansion method for radar micro-motion target identification classification has the advantages that:
(1) the method can effectively solve the problem of small image sample data amount in the existing radar target identification, and meanwhile, can enhance the micro-motion target characteristic information and realize the high-precision identification of the radar target.
(2) According to the method, data expansion of the micro Doppler characteristic time-frequency graph can be realized only by setting the color display amplitude of the spectrogram, the requirement on the number in the data acquisition process is effectively reduced, and the labor cost is greatly reduced.
(3) The method does not need complex operation and extra calculation, greatly reduces the operation amount, and is suitable for engineering application.
(4) The extracted distance pulse profile and the extracted time-frequency graph can effectively expand the image sample size by setting the color display amplitude, and useful characteristic information of the target is merged and fused by using a characteristic fusion mode, so that the target identification classification precision and reliability are effectively improved.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims (10)

1. A data set expansion method for radar micro-motion target recognition is characterized by comprising the following steps:
acquiring echo data of a target radar;
determining an echo range-pulse map from the echo data, the echo range-pulse map representing a relationship between a range gate count and a pulse count;
performing micro Doppler feature extraction on the distance-pulse diagram by adopting a time-frequency analysis method to obtain a time-frequency diagram;
respectively adjusting the color display ranges of the distance-pulse diagram and the time-frequency diagram to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram;
respectively cutting the target area of the adjusted distance-pulse diagram and the target area of the adjusted time-frequency diagram according to a preset cutting size to obtain a cut distance-pulse diagram and a cut time-frequency diagram, wherein the target area is an area containing micro-motion characteristics;
and performing feature fusion on the cut distance-pulse graph and the cut time-frequency graph to obtain an expanded data set.
2. The method of claim 1, wherein the step of performing micro-doppler feature extraction on the range-pulse diagram by using a time-frequency analysis method to obtain a time-frequency diagram further comprises:
and performing inter-pulse cancellation on the echo distance-pulse diagram by adopting a moving target display technology to obtain a processed echo distance-pulse diagram.
3. The method for expanding a data set for radar micro-motion target identification according to claim 1, wherein the distance-pulse diagram is subjected to micro-doppler feature extraction by using a time-frequency analysis method to obtain a time-frequency diagram, and specifically comprises:
selecting a distance-pulse graph segment corresponding to a target distance unit from the distance-pulse graph as a target distance-pulse graph;
and extracting the target distance-pulse diagram by adopting short-time Fourier transform to obtain a time-frequency diagram.
4. The method for expanding a data set for radar micro-motion target identification according to claim 1, wherein the adjusting the color display ranges of the range-pulse diagram and the time-frequency diagram respectively to obtain an adjusted range-pulse diagram and an adjusted time-frequency diagram specifically comprises:
using imagesc (C)1,clims1) Adjusting the color display range of the distance-pulse diagram by the function to obtain an adjusted distance-pulse diagram, wherein clims1＝[cmin1 cmax1]，C1Is the distance-pulse diagram, clims1Color display range, cmin, to be adjusted for the distance-pulse diagram1For the minimum value, cmax, of the color display range to be adjusted for the distance-pulse diagram1A maximum value of a color display range to be adjusted for the distance-pulse map;
using imagesc (C)2,clims2) Adjusting the color display range of the time-frequency diagram by a function to obtain an adjusted time-frequency diagram, wherein clims2＝[cmin2 cmax2]，C2Is said time-frequency diagram, clims2Color display Range, cmin, to be adjusted for said time-frequency diagram2For the minimum value, cmax, of the color display range to be adjusted for the time-frequency diagram2And the maximum value of the color display range to be adjusted of the time-frequency graph is obtained.
5. The method for expanding the data set for radar micro-motion target identification according to claim 3, wherein the target distance-pulse diagram is extracted by short-time Fourier transform to obtain a time-frequency diagram, specifically:
applying a formula to the target distance-pulse diagram
Figure FDA0002822354000000021
Extracting to obtain time-frequency diagram, wherein STFTs(t, ω) is a time-frequency diagram, SM×NAnd (u) is distance-pulse data corresponding to the target distance-pulse diagram, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a window function, and j is an imaginary number unit.
6. A data set augmentation system for radar micro-motion target identification, comprising:
the data acquisition module is used for acquiring echo data of the target radar;
a range-pulse map determination module for determining an echo range-pulse map from the echo data, the echo range-pulse map representing a relationship between a range gate count and a pulse count;
the time-frequency diagram determining module is used for extracting the micro Doppler characteristics of the distance-pulse diagram by adopting a time-frequency analysis method to obtain a time-frequency diagram;
the color display range adjusting module is used for respectively adjusting the color display ranges of the distance-pulse diagram and the time-frequency diagram to obtain an adjusted distance-pulse diagram and an adjusted time-frequency diagram;
the cutting module is used for respectively cutting the target area of the adjusted distance-pulse diagram and the target area of the adjusted time-frequency diagram according to a preset cutting size to obtain the cut distance-pulse diagram and the cut time-frequency diagram, wherein the target area is an area containing micro-motion characteristics;
and the data set determining module is used for performing feature fusion on the cut distance-pulse graph and the cut time-frequency graph to obtain an expanded data set.
7. The system of claim 6, further comprising:
and the cancellation module is used for performing inter-pulse cancellation on the echo distance-pulse diagram by adopting a moving target display technology to obtain a processed echo distance-pulse diagram.
8. The system of claim 6, wherein the time-frequency diagram determining module comprises:
the target distance-pulse graph determining unit is used for selecting a distance-pulse graph segment corresponding to the target distance unit from the distance-pulse graph as a target distance-pulse graph;
and the time-frequency diagram determining unit is used for extracting the target distance-pulse diagram by adopting short-time Fourier transform to obtain a time-frequency diagram.
9. The system of claim 6, wherein the color display range adjustment module comprises:
a first color display range adjusting module for utilizing imagesc (C)1,clims1) Adjusting the color display range of the distance-pulse diagram by the function to obtain an adjusted distance-pulse diagram, wherein clims1＝[cmin1 cmax1]，C1Is the distance-pulse diagram, clims1Color display range, cmin, to be adjusted for the distance-pulse diagram1For the minimum value, cmax, of the color display range to be adjusted for the distance-pulse diagram1Maximum value of color display range to be adjusted for the distance-pulse diagram
A second color display range adjusting module for utilizing imagesc (C)2,clims2) Adjusting the color display range of the time-frequency diagram by a function to obtain an adjusted time-frequency diagram, wherein clims2＝[cmin2 cmax2]，C2Is said time-frequency diagram, clims2Color display Range, cmin, to be adjusted for said time-frequency diagram2For the minimum value, cmax, of the color display range to be adjusted for the time-frequency diagram2And the maximum value of the color display range to be adjusted of the time-frequency graph is obtained.
10. The system of claim 8, wherein the time-frequency diagram determining unit comprises:
a time-frequency diagram determining subunit for applying a formula to the target distance-pulse diagram
Figure FDA0002822354000000021
Extracting to obtain time-frequency diagram, wherein STFTs(t, ω) is a time-frequency diagram, SM×NAnd (u) is distance-pulse data corresponding to the target distance-pulse diagram, u is a discrete number of time sampling, t is time, omega is angular frequency, g (u-t) is a window function, and j is an imaginary number unit.
CN202011441343.3A 2020-12-08 2020-12-08 Radar micro-motion target identification data set expansion method and system Active CN112710996B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011441343.3A CN112710996B (en) 2020-12-08 2020-12-08 Radar micro-motion target identification data set expansion method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011441343.3A CN112710996B (en) 2020-12-08 2020-12-08 Radar micro-motion target identification data set expansion method and system

Publications (2)

Publication Number Publication Date
CN112710996A true CN112710996A (en) 2021-04-27
CN112710996B CN112710996B (en) 2022-08-23

Family

ID=75542932

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011441343.3A Active CN112710996B (en) 2020-12-08 2020-12-08 Radar micro-motion target identification data set expansion method and system

Country Status (1)

Country Link
CN (1) CN112710996B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102426360A (en) * 2011-09-01 2012-04-25 西安电子科技大学 Two-dimensional ISRA imaging method of object with micro rotation in air
CN108693509A (en) * 2018-04-08 2018-10-23 中国人民解放军海军航空大学 Frequency control battle array radar Ullage frequency focuses moving-target integration detection method
CN109407067A (en) * 2018-10-13 2019-03-01 中国人民解放军海军航空大学 Radar moving targets detection and classification integral method based on time-frequency figure convolutional neural networks
CN109740626A (en) * 2018-11-23 2019-05-10 杭州电子科技大学 The detection method of cancerous area in breast cancer pathological section based on deep learning
CN110428377A (en) * 2019-07-26 2019-11-08 北京百度网讯科技有限公司 Data extending method, apparatus, equipment and medium
CN110764082A (en) * 2019-08-12 2020-02-07 武汉理工大学 Two-dimensional ground penetrating radar three-dimensional imaging method based on MATLAB
CN111856422A (en) * 2020-07-03 2020-10-30 西安电子科技大学 Lip language identification method based on broadband multichannel millimeter wave radar

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102426360A (en) * 2011-09-01 2012-04-25 西安电子科技大学 Two-dimensional ISRA imaging method of object with micro rotation in air
CN108693509A (en) * 2018-04-08 2018-10-23 中国人民解放军海军航空大学 Frequency control battle array radar Ullage frequency focuses moving-target integration detection method
CN109407067A (en) * 2018-10-13 2019-03-01 中国人民解放军海军航空大学 Radar moving targets detection and classification integral method based on time-frequency figure convolutional neural networks
CN109740626A (en) * 2018-11-23 2019-05-10 杭州电子科技大学 The detection method of cancerous area in breast cancer pathological section based on deep learning
CN110428377A (en) * 2019-07-26 2019-11-08 北京百度网讯科技有限公司 Data extending method, apparatus, equipment and medium
CN110764082A (en) * 2019-08-12 2020-02-07 武汉理工大学 Two-dimensional ground penetrating radar three-dimensional imaging method based on MATLAB
CN111856422A (en) * 2020-07-03 2020-10-30 西安电子科技大学 Lip language identification method based on broadband multichannel millimeter wave radar

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
CHEN XIAOLONG ET AL.: ""Detection of a Low Observable Sea-Surface Target With Micromotion via the Radon-Linear Canonical Transform"", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 *
LELE YUAN: ""A time-frequency feature fusion algorithm based on neural network for HRRP"", 《PROGRESS IN ELECTROMAGNETICS RESEARCH M》 *
ZHIYU QU ET AL.: ""Radar Signal Intra-Pulse Modulation Recognition Based on Convolutional Neural Network and Deep Q-Learning Network"", 《IEEE ACCESS》 *
张文涛 等: ""基于深度自编码多维特征融合的慢动目标检测"", 《计算机工程》 *
朱宇涛等: "一种基于时频分析的自适应积累成像算法", 《空军雷达学院学报》 *
胡玉兰 等: ""高分辨雷达一维距离像的融合特征识别"", 《微型机与应用》 *
苏宁远 等: ""基于卷积神经网络的海上微动目标检测与分类方法"", 《雷达学报》 *

Also Published As

Publication number Publication date
CN112710996B (en) 2022-08-23

Similar Documents

Publication Publication Date Title
DE3417660C2 (en) Device for ultrasound examination of a circulatory system on living biological structures, and method for ultrasound imaging of parameters of a circulatory system
EP1092148B1 (en) System for rapidly calculating expansion images from high-frequency ultrasonic echo signals
DE69725670T2 (en) Highly accurate time-frequency signal analysis with low distortion using rotated window spectrograms
CN106918807B (en) A kind of Targets Dots condensing method of radar return data
DE69127764T2 (en) Phase difference autofocus setting for imaging a radar with a synthetic aperture
WO2017206999A1 (en) Method for evaluating image data of a vehicle camera
DE3882707T2 (en) RADAR WITH SYNTHETIC APERTURE.
JPH06511339A (en) Method and image processing system for determining sensor motion and scene structure
KR101524529B1 (en) Method for sar processing
US6873285B2 (en) Method and system for providing along-track alignment and formatting of synthetic aperture radar (SAR) data, and SAR image formation algorithms using such method and system
DE68920769T2 (en) Device for measuring the speed of organs and blood flow using correlation methods.
CN108318879B (en) ISAR image transverse calibration method based on IAA spectrum estimation technology
CN111830502A (en) Data set establishing method, vehicle and storage medium
DE4122592C1 (en)
CA2262848A1 (en) Pulse compression radar
CN112710996B (en) Radar micro-motion target identification data set expansion method and system
CN111856422A (en) Lip language identification method based on broadband multichannel millimeter wave radar
US3660647A (en) Automatic signal delay tracking system
KR970010077B1 (en) Doppler tracking method for object imaging from radar returns
US20240188825A1 (en) Laser scanning imaging method and system, storage medium, and computer program
EP3071989A1 (en) Method and a device for measuring a relative velocity by means of an acoustic sensor
CN112686297B (en) Radar target motion state classification method and system
EP3575829A1 (en) A method of determining a transformation matrix
Abdulla et al. DUET using automatic peak detection and histogram thresholding
Son et al. Efficient target motion compensation method for stepped-frequency ISAR signatures

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