CN119335489A - Target RCS time sequence feature extraction method and device - Google Patents
Target RCS time sequence feature extraction method and device Download PDFInfo
- Publication number
- CN119335489A CN119335489A CN202411369971.3A CN202411369971A CN119335489A CN 119335489 A CN119335489 A CN 119335489A CN 202411369971 A CN202411369971 A CN 202411369971A CN 119335489 A CN119335489 A CN 119335489A
- Authority
- CN
- China
- Prior art keywords
- cepstrum
- target
- frequency
- time sequence
- target rcs
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 65
- 239000011159 matrix material Substances 0.000 claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 238000004590 computer program Methods 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 13
- 238000001228 spectrum Methods 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 10
- 238000009432 framing Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 6
- 230000004044 response Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 4
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000009825 accumulation Methods 0.000 abstract description 8
- 230000000875 corresponding effect Effects 0.000 description 37
- 230000006870 function Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000704 physical effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000010287 polarization Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a device for extracting a target RCS time sequence feature, wherein the method comprises the steps of obtaining a target RCS time sequence of a target in different motion states; the method comprises the steps of superposing the target RCS time sequences, calculating to obtain a frequency value corresponding to a preset frequency accumulation contribution, determining a cepstrum frequency adopted by cepstrum analysis according to the frequency value, and obtaining a cepstrum feature extraction parameter matrix of the target in each motion state according to the cepstrum frequency. The scheme realizes the extraction of the target RCS time sequence characteristics under different states, and further identifies the motion state of the target based on the target RCS time sequence.
Description
Technical Field
The invention relates to the technical field of radar target identification, in particular to a target RCS time sequence feature extraction method and device.
Background
Because radar has all-day, all-weather and penetrability detection capability, unmanned aerial vehicle target recognition technology based on radar detection has become a research hot spot in recent years. In radar target recognition, the information amount is related to the bandwidth of a signal, and the larger the bandwidth is, the more the information amount is. Although the information quantity of the narrow-band bandwidth is limited, the structure is simple, the manufacturing cost is low, and the development is easy, so that the narrow-band radar information is still an important research object for space target identification. The radar cross-sectional area (Radar Cross Section, RCS) of a target is the primary information that a narrowband radar can acquire from the target. RCS has rich information, but the extraction difficulty is greater.
At present, related researches are carried out on the RCS feature extraction and identification problems of unmanned aerial vehicles in a plurality of documents. In the actual process, the flying state of the aerospace target can be changed along with the execution task, so that the modulation phenomenon is generated on the RCS. And Matthew, et al 2017 in IET Radar, sonar &
The publication by Navigation mentions that when the unmanned aerial vehicle with load is classified and identified, the unmanned aerial vehicle is found to have obvious difference in spectral characteristics under the flight and hover, but quantitative research is not performed. Therefore, the electromagnetic scattering characteristics are determined by the physical properties and the gesture movement characteristics of the target, so that the change rule of the dynamic RCS characteristics of the target in different movement states is different, and the method has important significance for target identification application in actual complex environments. Therefore, a method for extracting the time series characteristics of the target RCS based on different motion states is needed.
Disclosure of Invention
The invention provides a target RCS time sequence feature extraction method and device, which realize the extraction of target RCS time sequence features under different states, and further identify the motion state of a target based on the target RCS time sequence.
In a first aspect, the present invention provides a method for extracting a time series feature of a target RCS, including:
Acquiring a target RCS time sequence of a target in different motion states;
Superposing the target RCS time sequences, and calculating to obtain a frequency value corresponding to the accumulated contribution of the preset frequency;
determining a cepstrum frequency adopted by cepstrum analysis according to the frequency value;
And obtaining a cepstrum feature extraction parameter matrix of the target under each motion state according to the cepstrum frequency.
Optionally, the time sequences corresponding to the RCS time sequences of the same target in different motion states are the same, and the sampling frequencies are the same.
Optionally, the preset frequency cumulative contribution is 50%.
Optionally, the frequency value is determined by the following formula:
fb＝(K'/K)·fs
The method comprises the steps of determining a target RCS time sequence, wherein K is the number of Fourier transforms of the target RCS time sequence, f s is a frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, f b is the frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, s (N) is the target RCS time sequence in the nth motion state, N is the category number of the motion state, FFT () represents the Fourier transform, and alpha is the accumulated contribution of the preset frequency.
Optionally, the cepstrum frequency is calculated by the following formula:
Wherein f re is the cepstrum frequency, f b is the frequency value, and f is the frequency value corresponding to the target RCS time sequence.
Optionally, the obtaining a cepstrum feature extraction parameter matrix of the target in each motion state according to the cepstrum frequency includes:
For each of said target RCS time sequences in motion state, performing:
Preprocessing, framing and smoothing are sequentially carried out on the target RCS time sequence to obtain a plurality of frame sub-time sequences;
Performing discrete Fourier transform on the sub-time sequence to obtain spectrum energy distribution;
setting a triangular filter bank comprising a plurality of triangular filters according to the cepstrum frequency;
Filtering the spectrum energy distribution by the triangular filter bank, calculating the logarithm of the energy output by each triangular filter, and performing discrete cosine transform to obtain a cepstrum coefficient of the sub-time sequence;
generating a cepstrum coefficient matrix under the motion state based on each sub-time sequence;
and extracting first-order differential parameters of the cepstrum coefficient matrix, and combining to obtain a cepstrum feature extraction parameter matrix under the motion state.
Optionally, the number of triangular filters in the triangular filter bank is 22-26;
The frequency response of each triangular filter in the triangular filter bank is determined by the following formula:
wherein H m(fre) is the frequency response of the M-th triangular filter, f (M) is the cepstrum center frequency value corresponding to the M-th triangular filter, M is [0, M-1], and M is the number of the triangular filters.
In a second aspect, the present invention provides a target RCS time-series feature extraction apparatus, including:
the acquisition module is used for acquiring the target RCS time sequence of the target in different motion states;
The cepstrum processing module is used for superposing the target RCS time sequence, calculating to obtain a frequency value corresponding to the preset frequency accumulated contribution, and determining the cepstrum frequency adopted by the cepstrum analysis according to the frequency value;
and the characteristic extraction module is used for obtaining a cepstrum characteristic extraction parameter matrix of the target in each motion state according to the cepstrum frequency.
In a third aspect, an embodiment of the present invention further provides a computing device, including a memory and a target processor, where the memory stores a computer program, and the target processor implements the method described in any of the first aspects of the present specification when executing the computer program.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method according to any of the first aspects of the present description.
In a fifth aspect, embodiments of the present invention also provide a computer program product comprising computer instructions which, when executed by a processor, implement the steps of the method of any of the first aspects of the present description.
The invention provides a target RCS time sequence feature extraction method and device, which are used for acquiring target RCS time sequences under different motion states, determining a frequency value corresponding to a preset frequency accumulation contribution after all the target RCS time sequences are overlapped, and determining a cepstrum frequency during cepstrum analysis based on the frequency value so as to extract a parameter matrix based on the cepstrum frequency. Thus, the characteristics under different motion states are obtained through cepstrum analysis, and the motion state of the current radar target can be determined through identifying the characteristics, so that the task intention of the radar target is further deduced, and a theoretical basis is provided for subsequent situation awareness.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for extracting a time series feature of a target RCS according to an embodiment of the invention;
FIG. 2 is a target RCS time sequence and cepstrum feature extraction parameter matrix of an unmanned aerial vehicle in a hover state according to an embodiment of the present invention;
FIG. 3 is a target RCS time sequence and cepstrum feature extraction parameter matrix of an unmanned aerial vehicle in a rotation state according to an embodiment of the present invention;
FIG. 4 is a graph showing a target RCS time sequence and a cepstrum feature extraction parameter matrix of an unmanned aerial vehicle in a hover state according to an embodiment of the present invention;
FIG. 5 is a target RCS time sequence and cepstrum feature extraction parameter matrix of an unmanned aerial vehicle in a transverse direct flight state according to an embodiment of the present invention;
FIG. 6 is a matrix of parameters extracted from the target RCS time sequence and cepstrum features of the unmanned aerial vehicle in a vertical direct flight state according to an embodiment of the present invention;
FIG. 7 is a hardware architecture diagram of a computing device according to one embodiment of the invention;
Fig. 8 is a schematic structural diagram of a device for extracting time series characteristics of a target RCS according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments, and all other embodiments obtained by those skilled in the art without making any inventive effort based on the embodiments of the present invention are within the scope of protection of the present invention.
Because the existing narrow-band statistical features cannot meet the requirements of actual dynamic scenes, feature dimension expansion is urgently needed. The electromagnetic scattering characteristics are determined by the physical properties and the gesture movement characteristics of the targets, so that the change rule of the dynamic RCS characteristics of the targets generated by different target movement states is analyzed, and the method has important significance for target identification application in actual complex environments.
Based on this, as shown in fig. 1, an embodiment of the present invention provides a method for extracting a time series feature of a target RCS, which includes:
Step 100, acquiring a target RCS time sequence of a target in different motion states;
step 102, superposing the target RCS time sequences, and calculating to obtain a frequency value corresponding to a preset frequency accumulation contribution;
104, determining a cepstrum frequency adopted by cepstrum analysis according to the frequency value;
and 106, obtaining a cepstrum feature extraction parameter matrix of the target in each motion state according to the cepstrum frequency.
According to the method, target RCS time sequences under different motion states are obtained, then frequency values corresponding to the preset frequency accumulation contribution are determined after all the target RCS time sequences are overlapped, and the cepstrum frequency during cepstrum analysis is determined based on the frequency values, so that a cepstrum feature extraction parameter matrix under different motion states is lifted based on the cepstrum frequency. Thus, the characteristics under different motion states are obtained through cepstrum analysis, and the motion state of the current radar target can be determined through identifying the characteristics, so that the task intention of the radar target is further deduced, and a theoretical basis is provided for subsequent situation awareness.
It should be noted that the target referred to by the present invention may be an aerospace target such as an unmanned aerial vehicle.
For step 100, the time sequences corresponding to the target RCS time sequences of the same target in different motion states are the same, and the sampling frequencies are the same, so that the target RCS time sequences can be overlapped according to the time sequences.
Aiming at step 102, overlapping the target RCS time sequence, calculating to obtain a frequency value corresponding to the preset frequency accumulation contribution, and determining the frequency value through the following formula:
fb＝(K'/K)·fs
The method comprises the steps of determining a target RCS time sequence, wherein K is the number of Fourier transforms of the target RCS time sequence, f s is a frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, f b is the frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, s (N) is the target RCS time sequence in the nth motion state, N is the category number of the motion state, FFT () represents the Fourier transform, and alpha is the accumulated contribution of the preset frequency.
Note that 0< α <1. Although a plurality of K 'can be determined based on the above formula, K' in the present invention refers to the minimum value that makes the above inequality hold.
In the invention, the frequency value corresponding to the preset frequency accumulation contribution is used as a criterion for constructing the cepstrum frequency scale, so that the improved cepstrum frequency is obtained, and the accuracy of feature extraction and the effectiveness of the extracted feature information are further improved.
In a preferred embodiment, the preset frequency cumulative contribution is 50%, i.e. α=50%.
According to the method, the cepstrum feature extraction parameter matrix obtained by the preset frequency accumulation contribution amount of 50% is utilized, and the movement information of the target contained in the cepstrum feature extraction parameter matrix is more comprehensive, so that the recognition accuracy is higher when the movement state is recognized based on the cepstrum feature extraction parameter matrix.
In step 104, a cepstrum frequency adopted in the cepstrum analysis is determined according to the frequency value, and the cepstrum frequency is calculated by the following formula:
Wherein f re is the cepstrum frequency, f b is the frequency value, and f is the frequency value corresponding to the target RCS time sequence.
Specifically, the frequency value corresponding to the target RCS time sequence is a linear frequency, and step 104 is to convert the linear frequency into a cepstrum frequency adopted in cepstrum analysis through the above formula. For example, f is a first frequency value corresponding to the 1 st fourier transform of the target RCS time series, f re is a cepstrum frequency corresponding to the first frequency value after the conversion, and f re is a cepstrum frequency corresponding to the K frequency value after the conversion if the corresponding f is a K frequency value corresponding to the K fourier transform of the target RCS time series.
Aiming at step 106, obtaining a cepstrum feature extraction parameter matrix of the target in each motion state according to the cepstrum frequency, including:
For each of said target RCS time sequences in motion state, performing:
Preprocessing, framing and smoothing are sequentially carried out on the target RCS time sequence to obtain a plurality of frame sub-time sequences;
Performing discrete Fourier transform on the sub-time sequence to obtain spectrum energy distribution;
setting a triangular filter bank comprising a plurality of triangular filters according to the cepstrum frequency;
Filtering the spectrum energy distribution by the triangular filter bank, calculating the logarithm of the energy output by each triangular filter, and performing discrete cosine transform to obtain a cepstrum coefficient of the sub-time sequence;
generating a cepstrum coefficient matrix under the motion state based on each sub-time sequence;
and extracting first-order differential parameters of the cepstrum coefficient matrix, and combining to obtain a cepstrum feature extraction parameter matrix under the motion state.
The preprocessing includes normalizing the target RCS time sequence, framing the normalized target RCS time sequence according to a preset frequency division step, and smoothing to obtain a sub-time sequence. Wherein the frequency band length of each sub-time sequence is the same.
Specifically, for each target RCS time sequence in n motion states, carrying out normalization processing and framing processing to obtain I sub-time sequences, and carrying out smoothing processing on each individual sub-time sequence by using a Hamming window W (l) so as to minimize the discontinuity of each frame signal and improve the frequency leakage condition, wherein the expression of the Hamming window W (l) is as follows:
wherein L is the length of the sub-time sequence;
Performing discrete Fourier transform on the ith sub-time sequence (I is not less than 1) to obtain spectrum energy distribution, setting a group of triangular filter groups according to improved cepstrum frequency, calculating the logarithm of energy output by each triangular filter, performing discrete cosine transform to obtain a group of coefficients M (I), namely the cepstrum coefficients of the ith frame signal, repeating the steps to obtain a cepstrum coefficient matrix M n corresponding to all the I frame sub-time sequences in the nth motion state, extracting a first-order differential parameter delta M n of the cepstrum coefficient matrix M n, and combining to obtain a cepstrum feature extraction parameter matrix M' n,M'n＝[MnΔMn of the target RCS time sequence in the nth motion state.
In the invention, a cepstrum analysis is adopted to carry out Fourier transform on a target RCS time sequence to obtain spectrum energy distribution, then the energy obtained after filtering is logarithmic, and then discrete cosine transform is carried out to obtain cepstrum coefficients, thus having certain anti-interference capability on noise.
In a preferred embodiment, the number of triangular filters in the triangular filter bank is 22-26;
the frequency response of each triangular filter in the triangular filter bank is determined by the following equation:
Wherein H m(fre) is the frequency response of the M-th triangular filter, f (M) is the cepstrum center frequency value corresponding to the M-th triangular filter, M is [0, M-1], and M is the number of the triangular filters.
In the invention, a triangular filter bank is adopted to smooth the frequency spectrum and eliminate the function of harmonic waves, so as to highlight the formants of the original voice. Since the tone or pitch of the audio is not presented within the cepstral coefficient matrix parameters, the recognition system featuring the cepstral coefficient matrix is not affected by the tone of the input audio. The number of the triangular filters is preferably limited to 22-26, so that the triangular filter bank can fully play the functions, and the calculation amount can be further reduced.
In a specific embodiment, fig. 2 to 6 show a target RCS time sequence and a cepstral feature extraction parameter matrix of the unmanned aerial vehicle under different motion states, where the length of the target RCS time sequence is 2.5s, the number of frames of the cepstral feature extraction parameter matrix is 21 frames, and the dimension is 12, so that the size of the parameter matrix M' is 24×21. As can be seen from fig. 2 to fig. 6, there are differences in the cepstrum feature extraction parameter matrices corresponding to different motion states. For quantitative analysis of feature differences, the pearson correlation coefficient (Pearson Correlation Coefficient, PCC) is selected as a measure whose expression is
Wherein Cov (·) represents covariance operation, σ Mm、σMn represents standard deviation of M m、Mn, M is M-th motion state, and n is n-th motion state. The similarity calculation result obtained by the expression is that the similarity value of the two same motion states is 1.0, the similarity value of the hovering state and the autorotation state is 0.4029, the similarity value of the hovering state and the hovering state is 0.4369, the similarity value of the hovering state and the transverse direct flight state is 0.3427, the similarity value of the hovering state and the longitudinal direct flight state is 0.4601, the similarity value of the autorotation state and the hovering state is 0.0645, the similarity value of the autorotation state and the transverse direct flight state is 0.2431, the similarity value of the autorotation state and the longitudinal direct flight state is 0.2659, the similarity value of the hovering state and the transverse direct flight state is 0.2249, the similarity value of the hovering state and the longitudinal direct flight state is 0.2466, and the similarity value of the transverse direct flight state and the longitudinal direct flight state is 0.2649. According to the calculation result, the maximum value of the feature similarity of different motion states is 0.4601, most features are weakly correlated (rho < 0.4), and the improved cepstrum feature extraction parameter matrix can effectively represent the motion rule of the unmanned aerial vehicle.
It should be noted that, the upper graphs in fig. 2 to 6 are all schematic views of the target RCS time sequence, and the ordinate is the RCS value (dB), and the lower graphs are all the cepstrum feature extraction parameter matrices (X-band HH polarization).
In a preferred embodiment, after obtaining the cepstral feature extraction parameter matrix, the method further includes:
Acquiring an RCS time sequence of a target to be detected of the target in a current period;
Superposing the target RCS time sequences, and calculating to obtain a frequency value corresponding to the accumulated contribution of the preset frequency;
determining a cepstrum frequency adopted by cepstrum analysis according to the frequency value;
obtaining a to-be-detected cepstrum feature extraction parameter matrix of the target according to the cepstrum frequency;
Respectively carrying out similarity calculation on the to-be-detected cepstrum feature extraction parameter matrix and the cepstrum feature extraction parameter matrix of the current known motion state, and outputting a similarity value, wherein the current known motion state comprises a hovering state, a autorotation state, a spiral state, a transverse direct flight state and a longitudinal direct flight state;
judging whether a similarity value larger than a preset threshold exists or not;
And if so, determining the current known motion state corresponding to the similarity value as the current motion state of the target in the current period.
The range of values of the preset threshold is (0.46,1 ], preferably 0.7, 0.8, etc., specifically, the above-mentioned cepstrum feature extraction parameter matrix to be detected is obtained by using any one of the above-mentioned target RCS time series feature extraction methods.
In the invention, as the cepstrum feature extraction parameter matrixes under different motion states have differences, the current motion state corresponding to the cepstrum feature extraction parameter matrix to be detected can be obtained by comparing the similarity of the cepstrum feature extraction parameter matrix to be detected and the cepstrum feature extraction parameter matrix under the known motion state, the recognition of the target motion state based on the target RCS time sequence is realized, and the invention has important significance for the target recognition application under the actual complex environment.
As shown in fig. 7 and 8, the embodiment of the invention provides a device for extracting a time series characteristic of a target RCS. The apparatus embodiments may be implemented by software, or may be implemented by hardware or a combination of hardware and software. In terms of hardware, as shown in fig. 7, a hardware architecture diagram of a computing device where a target RCS time-series feature extraction apparatus provided by an embodiment of the present invention is located, where in addition to a processor, a memory, a network interface, and a nonvolatile memory shown in fig. 7, the computing device where the embodiment is located may generally include other hardware, such as a forwarding chip responsible for processing a packet, and so on. Taking a software implementation as an example, as shown in fig. 8, as a device in a logic sense, the device is formed by reading a corresponding computer program in a nonvolatile memory into a memory by a CPU of a computing device where the device is located. The embodiment provides a target RCS time-series feature extraction device, which includes:
An acquisition module 800, configured to acquire a target RCS time sequence of a target in different motion states;
The cepstrum processing module 802 is configured to superimpose the target RCS time sequences, calculate a frequency value corresponding to a preset frequency accumulation contribution, and determine a cepstrum frequency adopted by cepstrum analysis according to the frequency value;
The feature extraction module 804 is configured to obtain a cepstrum feature extraction parameter matrix of the target in each motion state according to the cepstrum frequency.
In some specific embodiments, the obtaining module 800 may be configured to perform the step 100, the cepstrum processing module 802 may be configured to perform the step 102 and the step 104, and the feature extraction module 804 may be configured to perform the step 106.
In one embodiment of the present invention, the time sequences corresponding to the RCS time sequences of the same target in different motion states are the same, and the sampling frequencies are the same.
In one embodiment of the present invention, the cepstrum processing module 302 is further configured to perform the following operations:
The frequency value is determined by the following formula:
fb＝(K'/K)·fs
The method comprises the steps of obtaining a target RCS time sequence, wherein K is the number of Fourier transforms of the target RCS time sequence, f s is a frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, f b is the frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, s (N) is the target RCS time sequence in the nth motion state, N is the category number of the motion state, FFT () represents the Fourier transform, alpha is the preset frequency accumulated contribution quantity, and alpha=50%.
In one embodiment of the present invention, the cepstrum processing module 302 is further configured to perform the following operations:
the cepstrum frequency is determined by the following formula:
Wherein f re is the cepstrum frequency, f b is the frequency value, and f is the frequency value corresponding to the target RCS time sequence.
In one embodiment of the present invention, the feature extraction module 304 is further configured to perform the following operations:
For each of said target RCS time sequences in motion state, performing:
Preprocessing, framing and smoothing are sequentially carried out on the target RCS time sequence to obtain a plurality of frame sub-time sequences;
Performing discrete Fourier transform on the sub-time sequence to obtain spectrum energy distribution;
setting a triangular filter bank comprising a plurality of triangular filters according to the cepstrum frequency, wherein the number of the triangular filters in the triangular filter bank is 22-26;
Filtering the spectrum energy distribution by the triangular filter bank, calculating the logarithm of the energy output by each triangular filter, and performing discrete cosine transform to obtain a cepstrum coefficient of the sub-time sequence;
generating a cepstrum coefficient matrix under the motion state based on each sub-time sequence;
and extracting first-order differential parameters of the cepstrum coefficient matrix, and combining to obtain a cepstrum feature extraction parameter matrix under the motion state.
In one embodiment of the present invention, the method further comprises an identification module for performing the following operations:
Acquiring an RCS time sequence of a target to be detected of the target in a current period;
Superposing the target RCS time sequences, and calculating to obtain a frequency value corresponding to the accumulated contribution of the preset frequency;
determining a cepstrum frequency adopted by cepstrum analysis according to the frequency value;
obtaining a to-be-detected cepstrum feature extraction parameter matrix of the target according to the cepstrum frequency;
Respectively carrying out similarity calculation on the to-be-detected cepstrum feature extraction parameter matrix and the cepstrum feature extraction parameter matrix of the current known motion state, and outputting a similarity value, wherein the current known motion state comprises a hovering state, a autorotation state, a spiral state, a transverse direct flight state and a longitudinal direct flight state;
judging whether a similarity value larger than a preset threshold exists or not;
And if so, determining the current known motion state corresponding to the similarity value as the current motion state of the target in the current period.
It will be appreciated that the structure illustrated in the embodiments of the present invention does not constitute a specific limitation on one type of target RCS time series feature extraction device. In other embodiments of the invention, a target RCS time-series feature extraction device may include more or fewer components than shown, or may combine certain components, or may split certain components, or may have a different arrangement of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
The content of information interaction and execution process between the modules in the device is based on the same conception as the embodiment of the method of the present invention, and specific content can be referred to the description in the embodiment of the method of the present invention, which is not repeated here.
The embodiment of the invention also provides a computing device which comprises a memory and a target processor, wherein the memory stores a computer program, and when the target processor executes the computer program, the target RCS time sequence feature extraction method in any embodiment of the invention is realized.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program which, when being executed by a processor, causes the processor to execute the target RCS time sequence feature extraction method in any embodiment of the invention.
Embodiments of the present application also provide a computer program product comprising a computer program, which is read from a computer readable storage medium by a processor of a computer device, the computer program being executed by the processor to cause the computer device to perform a target RCS time series feature extraction method according to any of the above embodiments.
Specifically, a system or apparatus provided with a storage medium on which a software program code realizing the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or apparatus may be caused to read out and execute the program code stored in the storage medium.
In this case, the program code itself read from the storage medium may realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code form part of the present invention.
Examples of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-R, CD-RWs, DVD-ROMs, DVD-RAMs, DVD-RWs, DVD+RWs), magnetic tapes, nonvolatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer by a communication network.
Further, it should be apparent that the functions of any of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like on the computer to perform a part or all of the actual operations based on the instructions of the program code.
Further, it is understood that the program code read out by the storage medium is written into a memory provided in an expansion board inserted into a computer or into a memory provided in an expansion module connected to the computer, and then a CPU or the like mounted on the expansion board or the expansion module is caused to perform part and all of actual operations based on instructions of the program code, thereby realizing the functions of any of the above embodiments.
It is noted that relational terms such as first and second, and the like, are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one.," does not exclude that an additional identical element is present in a process, method, article, or apparatus that comprises the element.
It will be appreciated by those of ordinary skill in the art that implementing all or part of the steps of the above method embodiments may be accomplished by hardware associated with program instructions, and that the above program may be stored in a computer readable storage medium which, when executed, performs the steps comprising the above method embodiments, where the above storage medium includes various media that may store program code, such as ROM, RAM, magnetic or optical disks.
It should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention, and not for limiting the same, and although the present invention has been described in detail with reference to the above-mentioned embodiments, it should be understood by those skilled in the art that the technical solution described in the above-mentioned embodiments may be modified or some technical features may be equivalently replaced, and these modifications or substitutions do not make the essence of the corresponding technical solution deviate from the spirit and scope of the technical solution of the embodiments of the present invention.
Claims (10)
1. A method for extracting a time series feature of a target RCS, comprising:
Acquiring a target RCS time sequence of a target in different motion states;
Superposing the target RCS time sequences, and calculating to obtain a frequency value corresponding to the accumulated contribution of the preset frequency;
determining a cepstrum frequency adopted by cepstrum analysis according to the frequency value;
And obtaining a cepstrum feature extraction parameter matrix of the target under each motion state according to the cepstrum frequency.
2. The method of claim 1, wherein the time sequences corresponding to the target RCS time sequences of the same target in different motion states are the same, and the sampling frequencies are the same;
and/or the number of the groups of groups,
The preset frequency cumulative contribution is 50%.
3. The method of claim 1, wherein the frequency value is determined by the formula:
fb＝(K'/K)·fs
The method comprises the steps of determining a target RCS time sequence, wherein K is the number of Fourier transforms of the target RCS time sequence, f s is a frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, f b is the frequency value corresponding to the Kth Fourier transform in the target RCS time sequence, s (N) is the target RCS time sequence in the nth motion state, N is the category number of the motion state, FFT () represents the Fourier transform, and alpha is the accumulated contribution of the preset frequency.
4. The method of claim 1, wherein the cepstral frequency is calculated by the formula:
Wherein f re is the cepstrum frequency, f b is the frequency value, and f is the frequency value corresponding to the target RCS time sequence.
5. The method according to any one of claims 1 to 4, wherein said deriving a matrix of cepstral feature extraction parameters of said target in each of said motion states from said cepstral frequencies comprises:
For each of said target RCS time sequences in motion state, performing:
Preprocessing, framing and smoothing are sequentially carried out on the target RCS time sequence to obtain a plurality of frame sub-time sequences;
Performing discrete Fourier transform on the sub-time sequence to obtain spectrum energy distribution;
setting a triangular filter bank comprising a plurality of triangular filters according to the cepstrum frequency;
Filtering the spectrum energy distribution by the triangular filter bank, calculating the logarithm of the energy output by each triangular filter, and performing discrete cosine transform to obtain a cepstrum coefficient of the sub-time sequence;
generating a cepstrum coefficient matrix under the motion state based on each sub-time sequence;
and extracting first-order differential parameters of the cepstrum coefficient matrix, and combining to obtain a cepstrum feature extraction parameter matrix under the motion state.
6. The method of claim 5, wherein the number of triangular filters in the triangular filter bank is 22-26;
The frequency response of each triangular filter in the triangular filter bank is determined by the following formula:
wherein H m(fre) is the frequency response of the M-th triangular filter, f (M) is the cepstrum center frequency value corresponding to the M-th triangular filter, M is [0, M-1], and M is the number of the triangular filters.
7. A target RCS time-series feature extraction device, comprising:
the acquisition module is used for acquiring the target RCS time sequence of the target in different motion states;
The cepstrum processing module is used for superposing the target RCS time sequence, calculating to obtain a frequency value corresponding to the preset frequency accumulated contribution, and determining the cepstrum frequency adopted by the cepstrum analysis according to the frequency value;
and the characteristic extraction module is used for obtaining a cepstrum characteristic extraction parameter matrix of the target in each motion state according to the cepstrum frequency.
8. A computing device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the method of any of claims 1-6 when the computer program is executed.
9. A computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of any of claims 1-6.
10. A computer program product comprising computer instructions which, when executed by a processor, implement the steps of the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411369971.3A CN119335489A (en) | 2024-09-29 | 2024-09-29 | Target RCS time sequence feature extraction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411369971.3A CN119335489A (en) | 2024-09-29 | 2024-09-29 | Target RCS time sequence feature extraction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN119335489A true CN119335489A (en) | 2025-01-21 |
Family
ID=94267035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411369971.3A Pending CN119335489A (en) | 2024-09-29 | 2024-09-29 | Target RCS time sequence feature extraction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119335489A (en) |
-
2024
- 2024-09-29 CN CN202411369971.3A patent/CN119335489A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112257521B (en) | CNN underwater acoustic signal target identification method based on data enhancement and time-frequency separation | |
| US9754603B2 (en) | Speech feature extraction apparatus and speech feature extraction method | |
| CN111724770B (en) | Audio keyword identification method for generating confrontation network based on deep convolution | |
| CN113646833B (en) | Speech adversarial sample detection method, device, equipment and computer-readable storage medium | |
| CN112329914B (en) | Fault diagnosis method and device for buried transformer substation and electronic equipment | |
| CN108022587A (en) | Audio recognition method, device, computer equipment and storage medium | |
| Zhao et al. | Using supervised machine learning to distinguish microseismic from noise events | |
| Shadlou Jahromi et al. | Feature extraction in fractional Fourier domain for classification of passive sonar signals | |
| CN113177514B (en) | Unmanned aerial vehicle signal detection method and device and computer readable storage medium | |
| Wang et al. | Radar HRRP target recognition in frequency domain based on autoregressive model | |
| Li et al. | Animal sound recognition based on double feature of spectrogram in real environment | |
| CN115758082A (en) | Fault diagnosis method for rail transit transformer | |
| Gunawan et al. | Repurposing transfer learning strategy of computer vision for owl sound classification | |
| CN117909659B (en) | Method and device for processing aerodynamic force signals of balance in wind tunnel dynamic test of variant aircraft | |
| EP3254282A1 (en) | Determining features of harmonic signals | |
| CN117690421B (en) | Speech recognition method, device, equipment and medium of noise reduction recognition combined network | |
| CN119335489A (en) | Target RCS time sequence feature extraction method and device | |
| Messaoud et al. | Using multi-scale product spectrum for single and multi-pitch estimation | |
| Prasanna Kumar et al. | Single-channel speech separation using combined EMD and speech-specific information | |
| CN114168788A (en) | Audio audit processing method, device, equipment and storage medium | |
| Li et al. | Animal sound recognition based on double feature of spectrogram | |
| Malfante et al. | Machine learning for automatic classification of volcano-seismic signatures | |
| CN114692693A (en) | Distributed optical fiber signal identification method, device and storage medium based on fractal theory | |
| Jiang et al. | VMD and HMM based rolling bearing fault diagnosis | |
| CN118447842B (en) | AI speech recognition, reasoning, generation system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |