CN113985384B - Spatial target translation compensation method and device based on template matching - Google Patents
Spatial target translation compensation method and device based on template matching Download PDFInfo
- Publication number
- CN113985384B CN113985384B CN202111615766.7A CN202111615766A CN113985384B CN 113985384 B CN113985384 B CN 113985384B CN 202111615766 A CN202111615766 A CN 202111615766A CN 113985384 B CN113985384 B CN 113985384B
- Authority
- CN
- China
- Prior art keywords
- matching
- template
- contour points
- time
- frequency 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.)
- Active
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
-
- 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
- G01S7/415—Identification of targets based on measurements of movement associated with the target
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 discloses a spatial target translation compensation method and a spatial target translation compensation device based on template matching, wherein the method comprises the following steps: s01, receiving radar echoes and performing time-frequency analysis to obtain time-frequency data; s02, filtering the time frequency data and then carrying out binarization processing to obtain binary time frequency data; s03, performing primary template matching by using a pre-constructed matching template and the binary time-frequency data to obtain all class contour points; s04, calculating the structural similarity between a neighborhood taking the similar contour points as the center and the matching template, and screening out all true contour points after eliminating all false contour points according to the structural similarity; and S05, fitting a trend curve according to the true contour points and carrying out translation parameter estimation to realize space target translation compensation. The invention has the advantages of simple realization method, high efficiency and precision, strong application flexibility and the like.
Description
Technical Field
The invention relates to the technical field of spatial target motion compensation, in particular to a spatial target translation compensation method and device based on template matching.
Background
Space target flight is often accompanied by bait, and great difficulty is brought to radar target identification. The conventional recognition method based on the target characteristics has difficulty in recognizing the true target from the lure, and thus it is required to improve the recognition accuracy of the target in terms of the micro-motion information of the target. The target micro-motion form can be rotation, conical rotation, swing and the like, in the prior art, the micro-motion characteristic extraction and characteristic research is mostly carried out based on micro-motion, however, when the target moves in an actual space, two motions of translation and micro-motion can occur, the translation can damage the structure of micro Doppler, and adverse effects such as inclination, folding, translation and the like are generated. Therefore, translation compensation is required before the spatial target micro-doppler characteristic is performed.
For the translation compensation of a space target, the following methods are generally adopted at present:
1. translational compensation method based on (extended) Radon/Hough transformation
The method comprises the steps of converting the translation parameters into a parameter domain in a line integration mode, estimating a focus point in the parameter domain, and reversely deducing the target translation parameters to realize target translation compensation. However, the time complexity of the method is high, long time is needed for realizing translation compensation, and great limitation is caused on realizing target translation real-time compensation.
2. Translational motion compensation method based on empirical mode decomposition
The method separates out micro Doppler curves of different scattering points through a traditional micro Doppler separation method (such as a Viterbi algorithm, a trend estimation-based method, a nearest neighbor method and the like), performs empirical mode decomposition on a single micro Doppler curve to obtain a residual term (namely a target translation trend term), and then estimates a translation parameter by using the residual term, thereby realizing translation compensation.
However, in the method, the micro-doppler curves of all scattering points need to be separated in advance, empirical mode decomposition is performed according to a single micro-doppler curve, and the residual terms are taken as the translation trend for compensation. When the micro-doppler curve coupling is severe, it is difficult to realize the precise separation of the micro-doppler lines, resulting in a large error in the estimation of the subsequent translation trend.
3. Conjugate multiplication-based translation compensation method
The method generally multiplies an original conjugate signal by a delay signal or multiplies two symmetrical delay conjugate signals to realize the reduction of a translation equation and the inhibition of a cross term, and then estimates a translation parameter by utilizing Radon/Hough transform or fractional Fourier transform to realize target translation compensation. However, the method has high requirement on the delay amount, and when the delay amount cannot meet the requirement, the estimation accuracy of the target translation parameter is sharply reduced.
4. Translational compensation method based on fractional Fourier transform
The method realizes the estimation of the second-order motion target translation parameter according to the correlation between the linear frequency modulation signal and the fractional Fourier transform, thereby realizing the target translation compensation and generally realizing the compensation of only a second-order motion track. However, when higher-order translational compensation is performed, the method needs to be combined with other algorithms, and the problem of accuracy reduction is also caused.
In summary, in the prior art, each spatial target translation compensation method has respective defects, such as high complexity and low efficiency, or low precision and large delay, and cannot simultaneously consider both the compensation efficiency and precision. Therefore, it is desirable to provide a spatial target translation compensation method, so as to simultaneously achieve the complexity, efficiency and precision of compensation.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides a spatial target translation compensation method and device based on template matching, which have the advantages of simple implementation method, high efficiency and precision and strong applicability and flexibility.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a spatial target translation compensation method based on template matching comprises the following steps:
s01, receiving radar echoes and performing time-frequency analysis to obtain time-frequency data;
s02, filtering the time frequency data and then carrying out binarization processing to obtain binary time frequency data;
s03, performing preliminary template matching by using a pre-constructed matching template and the binary time-frequency data to obtain all contour points, wherein the matching template is used for simulating a peak area of a micro Doppler curve;
s04, calculating the structural similarity between the neighborhood taking the similar contour points as the center and the matching template, and screening out all true contour points after eliminating all false contour points according to the structural similarity;
and S05, fitting a trend curve according to the true contour points and estimating translation parameters to realize space target translation compensation.
Further, before step S01, the method includes generating a corresponding radar echo according to a preset radar parameter, where the radar parameter includes any one or more of a radar electromagnetic parameter, a target translation parameter, a target micro-motion parameter, and a target structure parameter, the radar electromagnetic parameter includes any one or more of a radar carrier frequency, a sampling frequency, and a bandwidth, the target translation parameter includes any one or more of a speed, an acceleration, and a second-order acceleration, the target micro-motion parameter includes any one or more of a spin frequency, a conic spin frequency, and a precession angle, and the target structure parameter includes the number of scattering points and/or the positions of the scattering points.
Further, the step S03 includes: and performing convolution operation on the matching template and the binary time-frequency data, and eliminating the area with the matching degree smaller than a preset threshold value in the binary time-frequency data to obtain all the similar contour points.
Further, before the step S03, a matching template construction step is further included, including: and constructing the matching template by using an upper and lower symmetrical parabola to simulate a peak area of a micro Doppler curve, wherein the size of the matching template is configured according to the sharpness degree of an upper peak and a lower peak of the micro Doppler curve.
Further, the step S04 includes: calculating the structural similarity between a neighborhood taking each type of contour point as a center and with a specified size and the matching template, and screening true contour points and false contour points according to the structural similarity, wherein when the structural similarity is greater than a preset similarity threshold value, the true contour points are judged and reserved; and when the structural similarity is smaller than a preset similarity threshold, judging to be a pseudo contour point and removing, wherein the upper contour point uses the lower half part of the matching template to screen, and the lower contour point uses the upper half part of the matching template to screen.
Further, after the step S03 and before the step S04, the method further includes segmenting the time-frequency data after the preliminary matching in the step S03 according to a time axis, where the number of segments isi_nAnd determining according to the number of the upper contour points and the lower contour points, and detecting the maximum value and the minimum value of the contour-like points contained in each section of time frequency data.
Further, in step S04, the screening of the authenticity contour points is performed by using a serial calculation method or a parallel calculation method for each segment of data formed by segmenting the preliminarily matched time-frequency data according to a time axis.
Further, the step S05 includes: and fitting the upper and lower true contour points obtained in the step S04 to obtain upper and lower trend curves, averaging the upper and lower trend curves to obtain a translation equation of the target, performing reverse estimation by using the translation equation to obtain the translation parameter, and performing conjugate multiplication to realize translation full compensation.
A spatial target translation compensation device based on template matching comprises:
the receiving module is used for receiving the radar echo and performing time-frequency analysis to obtain time-frequency data;
the binarization module is used for filtering the time frequency data and then performing binarization processing to obtain binary time frequency data;
the template matching module is used for carrying out preliminary template matching on the binary time-frequency data by using a pre-constructed matching template to obtain all kinds of contour points;
the true and false contour point screening module is used for calculating the structural similarity between a neighborhood taking the similar contour point as a center and the matching template, eliminating all false contour points according to the structural similarity and screening all true contour points;
and the parameter estimation module is used for fitting a trend curve according to the true contour points and estimating translation parameters to realize space target translation compensation.
A computer apparatus comprising a processor and a memory, the memory being arranged to store a computer program, the processor being arranged to execute the computer program, and the processor being arranged to execute the computer program to perform the method as described above.
Compared with the prior art, the invention has the advantages that:
1. according to the method, binarization processing is carried out on time-frequency data of radar echoes, and then primary matching is carried out on the binary time-frequency data by using a matching template to obtain all kinds of contour points; then screening true and false contour points through structural similarity, and eliminating all false contour points; and finally, fitting a trend curve according to the true contour points, further estimating translation parameters, realizing space target translation compensation, considering complexity, efficiency and precision of compensation, and realizing high-efficiency and high-precision space target translation compensation.
2. According to the invention, the matching template is specifically designed according to the structure of the micro Doppler line, the preliminary matching of binary time frequency data is realized based on the matching template, the problems of target micro Doppler structure damage and the like caused by translation motion can be solved, and the size of the matching template is configured, so that the method can be suitable for the detection of the upper and lower contour points of the micro Doppler line under different parameters, and has high detection precision and strong applicability.
3. The invention further segments the matched binary time-frequency graph, thereby effectively reducing the waste of unnecessary detection time of a non-detection area, and the upper and lower contour points of each segment of interval only appear at the top or bottom after segmentation, reducing the detection times of each segment of interval to two, greatly reducing the operation time and greatly reducing the time complexity without influencing the screening result.
Drawings
Fig. 1 is a schematic flow chart illustrating an implementation principle of the spatial target translation compensation method based on template matching according to this embodiment.
Fig. 2 is a schematic flow chart of a specific implementation of the spatial target translation compensation method based on template matching according to the present embodiment.
Fig. 3 is a schematic structural diagram of the matching template in the present embodiment.
Fig. 4 is a schematic flow chart illustrating a detailed implementation of spatial target translation compensation in the present embodiment using the serial mode.
Fig. 5 is a schematic flow chart of a detailed implementation of spatial target translation compensation in the embodiment using the parallel mode.
Detailed Description
The invention is further described below with reference to the drawings and specific preferred embodiments of the description, without thereby limiting the scope of protection of the invention.
As shown in fig. 1 and 2, the spatial target translation compensation method based on template matching in this embodiment includes the steps of:
s01, receiving radar echoes and performing time-frequency analysis to obtain time-frequency data;
s02, filtering the time frequency data and then carrying out binarization processing to obtain binary time frequency data;
s03, performing preliminary template matching by using a pre-constructed matching template and binary time-frequency data to obtain all kinds of contour points, wherein the matching template is used for simulating a peak area of a micro Doppler curve;
s04, calculating the structural similarity between the neighborhood taking the similar contour point as the center and the matching template, eliminating all false contour points according to the structural similarity, and screening out all true contour points;
and S05, fitting a trend curve according to the true contour points and estimating translation parameters to realize space target translation compensation.
According to the method, binarization processing is carried out on the time-frequency data of the radar echo, and then preliminary matching is carried out on the binary time-frequency data by using a matching template simulating a peak area of a micro Doppler curve, so that all kinds of contour points are obtained; then screening true and false contour points through structural similarity, and eliminating all false contour points; and finally, fitting a trend curve according to the true contour points, further estimating translation parameters, realizing space target translation compensation, considering complexity, efficiency and precision of compensation, and realizing high-efficiency and high-precision space target translation compensation.
In this embodiment, before step S01, a corresponding radar echo is generated according to a preset radar parameter, where the radar parameter includes a radar electromagnetic parameter, a target translation parameter, a target micro-motion parameter, a target structure parameter, and the like, where the radar electromagnetic parameter includes a radar carrier frequency, a sampling frequency, a bandwidth, and the like, the target translation parameter includes a speed, an acceleration, a second-order acceleration, and the like, the target micro-motion parameter includes a spin frequency, a cone spin frequency, a precession angle, and the target structure parameter includes the number of scattering points, the position of scattering points, and the like.
In a specific application embodiment, after the receiver receives the generated radar echo, radar micro-motion signal processing is performed, and the radar echo received by the receiver is converted into a time-frequency image so as to represent a time-varying frequency spectrum of the radar echo. The radar transmitting signal can be a broadband signal or a narrowband signal, the time-frequency data acquisition mode is specifically determined according to a signal system, wherein for the broadband signal, a pulse compression method is adopted to obtain an echo signal time-frequency image, and if dechirp (frequency modulation release) processing is firstly carried out, then pulse compression processing is carried out, and time-frequency data are obtained by slow time accumulation; for the narrow-band signal, the time-frequency transformation is carried out by adopting short-time Fourier transformation and the like to obtain an echo signal time-frequency image. In data acquisition, the micro-motion echo model specifically utilizes a sliding scattering point micro-Doppler change rule to reversely deduce a micro-motion distance change model, the micro-motion echo model can specifically adopt a precession echo model, and the translation model adopts a third-order polynomial model.
In step S02 of this embodiment, down-sampling is specifically adopted to reduce the image (for example, to 512 × 512), a gaussian filtering method is used to filter out part of the noise, and then binarization processing is performed to obtain binary time-frequency data.
In step S01, a micro doppler line can be obtained by time-frequency analysis. The peak of the outer envelope of the micro-doppler line can be approximated to be parabolic, so the sharpness of the peak can be characterized by the opening width and the distance from the vertex to the cross section of the parabola. In this embodiment, before the step S03, a matching template construction step is further included, including: a matching template is constructed by using an up-and-down symmetrical parabola, as shown in fig. 3 (a) corresponds to the up-and-down symmetrical parabola, and fig. 3 (b) corresponds to the constructed matching template), so as to simulate a peak region of a micro-doppler curve, wherein the size of the matching template is the same as that of the peak region of the micro-doppler curveH×WAccording to micro Doppler lineThe sharpness of the upper and lower peaks is determined, whereinHIn order to match the height of the template,Wto match the width of the template. The construction of the matching template is that the size of the template is determined according to the sharp degree of the peak of the micro Doppler line, the template structure is two parabolas which are symmetrical up and down, the height and the width of the matching template correspond to the opening width of the parabola and the distance from the peak to the cross section, and the sharp degree of the micro Doppler line can be represented by the height and the width of the matching template. The matching template is designed specifically according to the structure of the micro Doppler line, preliminary matching of binary time-frequency data is achieved based on the matching template, the problems that the target micro Doppler structure is damaged due to translation motion and the like can be solved, the size of the matching template is configured, the method can be suitable for detection of upper and lower contour points of the micro Doppler line under different parameters, detection precision is high, and applicability is high.
And if the matching degree between the matching template and the binary time-frequency data is higher, the matching template is generally the similar contour point, and if the matching degree is lower, the contour point which does not belong to the class can be eliminated. Step S03 in this embodiment specifically includes: and performing convolution operation on the matching template and the binary time-frequency data, and eliminating the regions with the matching degree smaller than a preset threshold value in the binary time-frequency data, namely eliminating the regions with low matching degree and reserving the regions with high matching degree by using a threshold value method to obtain all contour points.
The step S04 of the present embodiment includes: the calculation takes various contour points as the center,H×WScreening true and false contour points according to the structural similarity of the neighborhood of the size (namely the size of the matched template) and the matched template, and judging as the true contour points and reserving the true contour points when the structural similarity is greater than a preset similarity threshold; and when the structural similarity is smaller than the preset similarity threshold, judging to be false contour points and removing, wherein the upper contour points use the lower half part of the matching template to screen true contour points, and the lower contour points use the upper half part of the matching template to screen true contour points.
In this embodiment, after the step S03 and before the step S04, the method further includes segmenting the preliminarily matched time-frequency data according to the time axis, where the number of segments isi_nAccording to the upper and lower wheelsProfile Point quantity determination (details)i_nAnd subtracting one from the sum of the upper and lower contour points), and detecting the maximum value and the minimum value of the contour points contained in each section of time frequency data, namely, only taking the maximum contour point and the minimum contour point in each section of time frequency data for subsequent true and false contour point screening in each section. By segmenting the matched binary time-frequency graph, the waste of unnecessary detection time of a non-detection area is effectively reduced, the upper and lower contour points of each segment of interval only appear at the top or the bottom after segmentation, the detection times of each segment of interval are reduced to two times, the operation time is greatly shortened, and the time complexity can be greatly reduced under the condition of not influencing the screening result.
In step 04, after the time-frequency data obtained after matching is divided into multiple segments according to the time axis, the true and false contour point screening may be performed on each segment of data in a serial computing manner (serial mode), that is, the true and false contour point screening is sequentially performed on each segment. As shown in fig. 4, in the serial mode, after segmenting the matched time-frequency data according to the time axis, from the first interval, determining a region of structural similarity with the matching template according to the contour-like point obtained from the matching result, where the region is a neighborhood determined by taking the contour-like point as the center and the size of the matching template; and the results obtained by each screening are sequentially stored in the same vector, and the next screening is performed after the previous screening is finished.
The segments may also perform the screening of the authenticity contour points in a parallel computing manner (parallel mode), i.e. the segments perform the screening of the authenticity contour points in parallel. As shown in fig. 5, in the parallel mode, after segmenting the matched time-frequency diagram according to the time axis, the storage vector of the screening result of each segment is allocated in advance, and the process of determining the region for comparing the structural similarity is the same as that in the serial mode, but the difference is that each segment is screened at the same time, so that the execution efficiency can be further improved.
The following specific steps for screening the true and false contour points are implemented by taking a serial mode as an example:
step S401: and determining the number of the segments.
Determining the number of segments according to the number of upper and lower contour pointsi_nI.e. byIs the sum of the upper and lower contour points minus one.
Step S402: and determining a structural similarity comparison area.
From the first segmentation interval, the contour point of the top (maximum value) and the bottom (minimum value) is found out, the contour point is taken as the center, and the neighborhood determined by the size of the matching template is the contrast area.
Step S403: and calculating the structural similarity.
Calculating the structural similarity between the contrast area and the corresponding part of the matching template and comparing the structural similarity with a similarity threshold
Making comparison when the threshold value is exceeded
When the real contour points are detected, the real contour points are considered as the real contour points and are sequentially recorded in a storage vector; when less than the threshold value
And (4) when the image is a false contour point, removing the contour point.
The embodiment can meet different requirements by performing segmentation processing on the matched time-frequency data and performing true and false contour point screening according to a serial mode or a parallel mode. Compared with the two modes, the serial mode has lower space complexity and higher time complexity; the parallel mode has a high spatial complexity but a low temporal complexity. The method of the invention can reduce the time complexity as a whole no matter the serial mode or the parallel mode is adopted.
In this embodiment, translation parameter estimation is implemented through matching template construction, template matching and true-false contour point screening. The template matching method comprises the steps of constructing a matching template, wherein the size of the template is determined according to the sharp degree of the peak of a micro Doppler line, the template structure is two parabolas which are symmetrical up and down, the template matching is to convolute the matching template and a binarized time-frequency image, and points which are larger than a preset matching threshold value are reserved; and the true and false contour point screening is to judge by using the structural similarity, screen out the true contour points which are greater than the similarity threshold value, and estimate the translation parameters according to the screened true contour points.
The step S05 of the present embodiment includes: and fitting the upper and lower true contour points obtained in the step S04 to obtain upper and lower trend curves, averaging the upper and lower trend curves to obtain a translation equation of the target, performing reverse estimation by using the translation equation to obtain translation parameters, and performing conjugate multiplication to realize translation full compensation. The above-mentioned concrete method can adopt polynomial fitting method to make curve fitting. Namely, after the translation parameter is estimated, a translation equation is determined, conjugate multiplication is utilized, namely, a compensation signal is constructed by utilizing the estimated parameter, and the original signal is multiplied by the conjugate signal of the compensation signal, so that the translation full compensation is realized.
In order to better adapt to the compensation effect under different translation orders, the self-adaptive determination of the fitting order can be further realized by setting a highest-order coefficient threshold, namely when the current highest-order coefficient is greater than the threshold, the actual translation order is higher, the current fitting order is increased for continuous judgment, otherwise, the translation order is reduced by one for the current fitting order, so that the self-adaptive compensation is realized under different environments.
The invention relates to a spatial target translation compensation method based on template matching, which comprises the steps of obtaining original radar echoes and converting the original radar echoes into time-frequency data; the translation parameter estimation is realized after the construction of the matched template, the template matching and the screening of the true and false contour points in sequence, the translation equation is determined according to the estimated translation parameter, the translation compensation is realized by conjugate multiplication, and the high-efficiency and high-precision spatial target translation compensation can be realized.
The spatial target translation compensation device based on template matching in the embodiment comprises:
the receiving module is used for receiving the radar echo and performing time-frequency analysis to obtain time-frequency data;
the binarization module is used for filtering the time frequency data and then performing binarization processing to obtain binary time frequency data;
the template matching module is used for carrying out primary template matching on the binary time-frequency data by using a pre-constructed matching template to obtain all kinds of contour points;
the true and false contour point screening module is used for calculating the structural similarity between a neighborhood taking the similar contour point as the center and the matching template, eliminating all false contour points according to the structural similarity and screening all true contour points;
and the parameter estimation module is used for fitting a trend curve according to the true contour points and estimating the translation parameters to realize space target translation compensation.
In the template matching module of this embodiment, a convolution operation is performed on the matching template and the binary time-frequency data, and a region with a matching degree smaller than a preset threshold in the binary time-frequency data is removed, so as to obtain all the contour points.
A matching template construction module is further connected in front of the template matching module in this embodiment, and is configured to construct the matching template by using an up-down symmetrical parabola so as to simulate a peak region of a micro doppler curve, wherein the size of the matching template is configured according to the sharpness degree of an upper peak and a lower peak of the micro doppler curve.
In the screening module for the authenticity contour points, the structural similarity between the neighborhood taking each contour point as the center and with the specified size and the matching template is calculated, and the authenticity contour points are screened according to the structural similarity, wherein when the structural similarity is larger than a preset similarity threshold value
Judging the contour points to be true contour points and reserving the true contour points; and when the structural similarity is smaller than the threshold value, judging as a false contour point and removing, wherein the upper contour point uses the lower half part of the matching template to screen, and the lower contour point uses the upper half part of the matching template to screen.
In this embodiment, a segmentation module is further connected between the template matching module and the true/false contour point screening module, and is configured to segment the primarily matched time-frequency data according to a time axis, where the number of segments isi_nAnd determining according to the number of the upper contour points and the lower contour points, and detecting the maximum value and the minimum value of the contour-like points contained in each section of time frequency data.
In this embodiment, the screening module performs screening of the authenticity contour points on each segment of data formed by segmenting the preliminarily matched time-frequency data according to a time axis by using a serial computing manner or a parallel computing manner.
In the parameter estimation module of this embodiment, an upper trend curve and a lower trend curve are obtained by fitting according to upper and lower true contour points obtained in the true and false contour point screening module, the upper trend curve and the lower trend curve are averaged to obtain a translation equation of a target, the translation equation is used for reverse estimation to obtain the translation parameter, and conjugate multiplication is used for realizing translation full compensation.
The computer device of the embodiment includes a processor and a memory, the memory is used for storing computer programs, the processor is used for executing the computer programs, and the processor is used for executing the computer programs to execute the method.
The foregoing is considered as illustrative of the preferred embodiments of the invention and is not to be construed as limiting the invention in any way. Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention should fall within the protection scope of the technical scheme of the present invention, unless the technical spirit of the present invention departs from the content of the technical scheme of the present invention.
Claims (9)
1. A spatial target translation compensation method based on template matching is characterized by comprising the following steps:
s01, receiving radar echoes and carrying out time-frequency analysis to obtain time-frequency data;
s02, filtering the time frequency data and then carrying out binarization processing to obtain binary time frequency data;
s03, performing preliminary template matching by using a pre-constructed matching template and the binary time-frequency data to obtain all contour points, wherein the matching template is used for simulating a peak area of a micro Doppler curve;
s04, calculating the structural similarity between the neighborhood taking the similar contour points as the center and the matching template, and screening out all true contour points after eliminating all false contour points according to the structural similarity;
s05, fitting a trend curve according to the true contour points and estimating translation parameters to realize space target translation compensation;
before the step S03, the method further includes a matching template construction step, including: and constructing the matching template by using an upper and lower symmetrical parabola to simulate a peak area of a micro Doppler curve, wherein the size of the matching template is configured according to the sharpness degree of an upper peak and a lower peak of the micro Doppler curve.
2. The spatial target translation compensation method based on template matching according to claim 1, wherein before step S01, the method includes generating corresponding radar echoes according to preset radar parameters, where the radar parameters include any one or more of radar electromagnetic parameters, target translation parameters, target micro-motion parameters, and target structure parameters, the radar electromagnetic parameters include any one or more of radar carrier frequency, sampling frequency, and bandwidth, the target translation parameters include any one or more of velocity, acceleration, and second-order acceleration, the target micro-motion parameters include any one or more of spin frequency, conic rotation frequency, and precession angle, and the target structure parameters include number of scattering points and/or positions of scattering points.
3. The spatial target translation compensation method based on template matching according to claim 1, wherein the step S03 includes: and performing convolution operation on the matching template and the binary time-frequency data, and eliminating the area with the matching degree smaller than a preset threshold value in the binary time-frequency data to obtain all the similar contour points.
4. The spatial target translation compensation method based on template matching according to claim 1, wherein the step S04 includes: calculating the structural similarity between the neighborhood with the specified size and the matching template by taking each type of contour point as a center, and screening true contour points according to the structural similarity, wherein when the structural similarity is greater than a preset similarity threshold value, the true contour points are judged and reserved; and when the structural similarity is smaller than the preset similarity threshold, judging to be a pseudo contour point and removing, wherein the upper contour point uses the lower half part of the matching template to screen, and the lower contour point uses the upper half part of the matching template to screen.
5. The spatial target translation compensation method based on template matching according to any one of claims 1 to 4, wherein after the step S03 and before the step S04, the method further comprises segmenting the time-frequency data after the preliminary matching in the step S03 according to a time axis, wherein the number of segments isi_nAnd determining according to the number of the upper contour points and the lower contour points, and detecting the maximum value and the minimum value of the contour-like points contained in each section of time frequency data.
6. The spatial target translation compensation method according to claim 5, wherein in step S04, the screening of the false contour points is performed in a serial calculation manner or a parallel calculation manner on each segment of data formed by segmenting the preliminarily matched time-frequency data according to a time axis.
7. The spatial target translation compensation method based on template matching according to any one of claims 1 to 4, wherein the step S05 includes: and fitting the upper and lower true contour points obtained in the step S04 to obtain upper and lower trend curves, averaging the upper and lower trend curves to obtain a translation equation of the target, performing reverse estimation by using the translation equation to obtain the translation parameter, and performing conjugate multiplication to realize translation full compensation.
8. A spatial target translation compensation device based on template matching is characterized by comprising:
the receiving module is used for receiving the radar echo and performing time-frequency analysis to obtain time-frequency data;
the binarization module is used for filtering the time frequency data and then performing binarization processing to obtain binary time frequency data;
the template matching module is used for performing preliminary template matching on the binary time frequency data by using a pre-constructed matching template to obtain all profile points, constructing the matching template by using top and bottom symmetrical parabolas to simulate a peak area of a micro Doppler curve, wherein the size of the matching template is configured according to the sharpness degree of top and bottom peaks of the micro Doppler curve;
the true and false contour point screening module is used for calculating the structural similarity between a neighborhood taking the class contour point as a center and the matching template, eliminating all false contour points according to the structural similarity and screening out all true contour points;
and the parameter estimation module is used for fitting a trend curve according to the true contour points and estimating translation parameters to realize space target translation compensation.
9. A computer arrangement comprising a processor and a memory, the memory being adapted to store a computer program, the processor being adapted to execute the computer program, wherein the processor is adapted to execute the computer program to perform the method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111615766.7A CN113985384B (en) | 2021-12-28 | 2021-12-28 | Spatial target translation compensation method and device based on template matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111615766.7A CN113985384B (en) | 2021-12-28 | 2021-12-28 | Spatial target translation compensation method and device based on template matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113985384A CN113985384A (en) | 2022-01-28 |
| CN113985384B true CN113985384B (en) | 2022-05-24 |
Family
ID=79734545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111615766.7A Active CN113985384B (en) | 2021-12-28 | 2021-12-28 | Spatial target translation compensation method and device based on template matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113985384B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102788969A (en) * | 2012-07-04 | 2012-11-21 | 中国人民解放军海军航空工程学院 | Sea surface micromotion target detection and feature extraction method based on short-time fractional Fourier transform |
| CN104463866A (en) * | 2014-12-04 | 2015-03-25 | 无锡日联科技有限公司 | Local shape matching method based on outline random sampling |
| CN106548173A (en) * | 2016-11-24 | 2017-03-29 | 国网山东省电力公司电力科学研究院 | A kind of improvement no-manned plane three-dimensional information getting method based on classification matching strategy |
| KR20180135546A (en) * | 2017-06-13 | 2018-12-21 | 한국과학기술원 | Drone detection system of various angles using microdoppler and method thereof |
| CN111175724A (en) * | 2020-01-08 | 2020-05-19 | 中国科学院上海光学精密机械研究所 | Target micro Doppler component extraction method based on centroid calculation |
| CN112184849A (en) * | 2020-09-25 | 2021-01-05 | 中南大学 | Intelligent processing method and system for complex dynamic multi-target micro-motion signals |
| CN112835005A (en) * | 2020-12-31 | 2021-05-25 | 江苏云禾峰智能科技有限公司 | Micro Doppler feature extraction method based on super-resolution target tracking |
-
2021
- 2021-12-28 CN CN202111615766.7A patent/CN113985384B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102788969A (en) * | 2012-07-04 | 2012-11-21 | 中国人民解放军海军航空工程学院 | Sea surface micromotion target detection and feature extraction method based on short-time fractional Fourier transform |
| CN104463866A (en) * | 2014-12-04 | 2015-03-25 | 无锡日联科技有限公司 | Local shape matching method based on outline random sampling |
| CN106548173A (en) * | 2016-11-24 | 2017-03-29 | 国网山东省电力公司电力科学研究院 | A kind of improvement no-manned plane three-dimensional information getting method based on classification matching strategy |
| KR20180135546A (en) * | 2017-06-13 | 2018-12-21 | 한국과학기술원 | Drone detection system of various angles using microdoppler and method thereof |
| CN111175724A (en) * | 2020-01-08 | 2020-05-19 | 中国科学院上海光学精密机械研究所 | Target micro Doppler component extraction method based on centroid calculation |
| CN112184849A (en) * | 2020-09-25 | 2021-01-05 | 中南大学 | Intelligent processing method and system for complex dynamic multi-target micro-motion signals |
| CN112835005A (en) * | 2020-12-31 | 2021-05-25 | 江苏云禾峰智能科技有限公司 | Micro Doppler feature extraction method based on super-resolution target tracking |
Non-Patent Citations (4)
| Title |
|---|
| "Micro-Doppler feature extraction using single-frequency radar for high-speed targets";Cai Wang等;《IEEE Transactions on Geoscience and Remote Sensing》;20190514;全文 * |
| "Micro-Doppler Signature Extraction from High Speed Targets with Micro-Motions";Rujiang Guo等;《2014 IEEE Radar Conference-20140814-2016 CIE International Conference on Radar (RADAR)》;20171005;全文 * |
| "一种改进的复合运动空间目标微动周期估计方法";赵燕等;《电子测量与仪器学报》;20200228;全文 * |
| "基于趋势估计的微多普勒分离与特征提取算法 ";彭正红等;《系统工程与电子技术》;20210712;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113985384A (en) | 2022-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108154519B (en) | Method and device for segmenting blood vessels in fundus image and storage medium | |
| CN110415208B (en) | Self-adaptive target detection method and device, equipment and storage medium thereof | |
| US10497128B2 (en) | Method and system for sea background modeling and suppression on high-resolution remote sensing sea images | |
| US9076075B2 (en) | Method and device for following an object in a sequence of at least two images | |
| CN107167810B (en) | Method for rapidly extracting underwater target by side-scan sonar imaging | |
| CN103761731A (en) | Small infrared aerial target detection method based on non-downsampling contourlet transformation | |
| CN112130142B (en) | Method and system for extracting micro Doppler features of complex moving target | |
| CN101482969B (en) | SAR image speckle filtering method based on identical particle computation | |
| CN113837198B (en) | Improved self-adaptive threshold Canny edge detection method based on three-dimensional block matching | |
| CN114002658B (en) | Radar target micro-motion feature extraction method based on point trace curve association curve separation | |
| CN109858438A (en) | A kind of method for detecting lane lines based on models fitting | |
| CN108647693B (en) | Sea surface infrared target detection method based on binary significance characteristics | |
| CN108305268B (en) | Image segmentation method and device | |
| CN109829902B (en) | Lung CT image nodule screening method based on generalized S transformation and Teager attribute | |
| CN109584256B (en) | Pulsar dispersion value estimation method based on Hough line detection | |
| Zhou et al. | An efficient and robust corner detection algorithm | |
| CN113985384B (en) | Spatial target translation compensation method and device based on template matching | |
| Kondaveeti et al. | Abridged shape matrix representation for the recognition of aircraft targets from 2D ISAR imagery | |
| CN113689449B (en) | Inversion method and system for characteristic parameters of mesoscale vortices | |
| CN106952241B (en) | Electromagnetic image blocking method based on morphological method and Meanshift algorithm | |
| Abu et al. | Robust image denoising for sonar imagery | |
| CN113608190B (en) | Sea surface target detection method and system based on three characteristics of singular space | |
| CN113128474B (en) | Structural modal identification method based on computer vision and variational modal decomposition | |
| CN115147613A (en) | Infrared small target detection method based on multidirectional fusion | |
| Koo et al. | An area-based stereo matching using adaptive search range and window size |
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 |