CN104977579B - A kind of many bright spot target space-time detecting methods based on random covariance matrix - Google Patents
A kind of many bright spot target space-time detecting methods based on random covariance matrix Download PDFInfo
- Publication number
- CN104977579B CN104977579B CN201510388924.8A CN201510388924A CN104977579B CN 104977579 B CN104977579 B CN 104977579B CN 201510388924 A CN201510388924 A CN 201510388924A CN 104977579 B CN104977579 B CN 104977579B
- Authority
- CN
- China
- Prior art keywords
- target
- data
- tested
- covariance matrix
- estimation
- 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
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
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/539—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 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
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/02—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems using reflection of acoustic waves
Abstract
The present invention relates to a kind of many bright spot target space-time detecting methods based on random covariance matrix, test problems are solved using two step likelihood ratio test criterions, the detection to many bright spot target echoes is realized, this method includes：The echo received by sonar battle array obtains one group of echo data, regard above-mentioned one group of echo data of acquisition as data to be tested；Assuming that under reverberation covariance matrix and goal orientation vector known case, the maximal possibility estimation substitution likelihood ratio function of target reflection factor, interference index, target index is obtained into the first test statistics.It is distributed according to the inverse obedience of above-mentioned reverberation covariance matrix against Fu Wei Saudi Arabia, obtains the MAP estimation of the matrix, and then obtain the second test statistics, it is compared with threshold value, completes target whether there is judgement.The present invention to priori by making full use of, it is ensured that the robustness of high-resolution sonar system in the presence of without assistance data, and the rejection ability to being disturbed from secondary lobe.
Description
Technical field
The present invention relates to signal detection field, when especially designing a kind of many bright spot target empties based on random covariance matrix
Detection method.
Background technology
Space-time adaptive detection (STAD) is the self-adaptive processing skill by space-time joint for framework, for the purpose of target detection
Art, it calculates detection statistic using the observation data received, directly judges the presence or absence of target, it is achieved thereby that Reverberation Rejection and
The integration of target detection.In essence, Reverberation Rejection is the process of Data Whitening, and for STAD, this process is lain in
In detector, so not needing extra Reverberation Rejection step.With the space-time adaptive processing method detected after first Reverberation Rejection
Compare, STAD can more effectively utilize observation data, so as to obtain preferably detection performance.In recent years, STAD is in high-resolution
The research in active sonar field is very active.Due to the raising of range resolution ratio, the target echo of high-resolution active sonar is past
It is past to occupy multiple range cells, referred to as many bright spot targets.Due to the particularity of many bright spot targets, its STAD problems are sonar work
Author proposes new problem.
The STAD detections of many bright spot targets under Reverberation are distributed for complex Gaussian, traditional method is usually assumed that can be with
Uniform assistance data is obtained, to estimate the reverberation covariance matrix of to-be-measured cell (master data), so as to construct adaptive inspection
Survey statistic.The range cell that assistance data is typically closed on from master data is obtained.But in actual applications, high-resolution sonar
Non-homogeneous environment is usually operated at, assistance data is difficult often to obtain.For this problem, foreign scholar is based on Generalized Likelihood Ratio
(GLRT) test criterion, it is proposed that independent of many bright spot object detection methods of assistance data, such method can be according to target
The prior information of bright spot number, Automatic sieve selects target data and assistance data from data to be tested, realizes to many bright spot mesh
Target self-adapting detecting.
In actual applications, high-resolution sonar system is usually operated at non-homogeneous environment, causes the original of non-homogeneous phenomenon
Because having a lot, such as changeable sea-floor surficial types, the shoal of fish, Isolated interferers.In such a case, it is possible to the uniform observation obtained
Data length is very limited, and the performance for causing existing many bright spot object detection methods is declined to a great extent.
The content of the invention
The purpose of the present invention is to realize to make full use of priori, effectively improves the robust of high-resolution sonar system
Property.
To achieve the above object, the embodiments of the invention provide a kind of many bright spot target empties based on random covariance matrix
When detection method.Step is as follows：
The echo received by sonar battle array obtains one group of echo data, using above-mentioned one group of echo data of acquisition as to be detected
Data；Wherein, in the presence of without target, interference that may be present is described as leading with target in above-mentioned data to be tested
To vector in mutually orthogonal vector after albefaction, in the case of with the presence of target, target is contemplated as falling with above-mentioned to be detected
In data, across multiple range cells, correspondence echo signal index length is L；
Assuming that complex Gaussian distribution is obeyed in reverberation, and under reverberation covariance matrix and goal orientation vector known case, lead to
Cross structure likelihood function and obtain disturbing the maximal possibility estimation of index under no target conditions, and have the target rope under target conditions
The maximal possibility estimation and the maximal possibility estimation of target reflection factor drawn, using the above-mentioned likelihood function having under target conditions,
Maximal possibility estimation, the maximum likelihood of above-mentioned target index indexed without the above-mentioned likelihood function under target conditions, above-mentioned interference
The steering vector of estimation, the maximal possibility estimation of above-mentioned target reflection factor and target, obtains the first test statistics；
According to the inverse multiple Wei Shate distributions of the inverse obedience of above-mentioned reverberation covariance matrix, obtained using above-mentioned data to be tested
The MAP estimation of reverberation covariance matrix is stated, above-mentioned MAP estimation is substituted into and obtained in above-mentioned first test statistics
Second test statistics, above-mentioned second test statistics is compared with detection threshold, completes target whether there is judgement.
It is preferred that, above-mentioned MAP estimation is the coloured loading that sampling covariance estimates S, can be expressed as：S+(v-
N)M0, wherein, M0High-resolution sonar is mainly used to detect environment, working method, reverberation space-time characteristic, shallow water topography characteristic
To obtain prior information, the Mean Matrix built according to reverberation scattering principle, loading level is determined that N is array element by the free degree by v
Number.
It is preferred that, above-mentioned detection threshold value is determined by false-alarm probability, can be obtained by Meng Te-Caro emulation.
It is preferred that, the process that above-mentioned completion target whether there is judgement is actually the process screened to above-mentioned data to be tested,
It is represented by：
Wherein, η is detection threshold value, and v is the free degree,It is ordered series of numbersMiddle L minimum
The corresponding index of value,It is ordered series of numbersThe corresponding index of middle L maximum, wherein, rtIt is number to be detected
According to η is detection threshold, []-1Represent to matrix inverse operation, []HRepresent conjugate transposition operation, H0Indicate no target conditions, H1
Indicate target conditions.
It is preferred that, it is above-mentioned as follows the step of screened to the data to be tested：
In the case where there are target conditions, using above-mentioned data to be tested, filtered out from corresponding subclass and target highlight number
Identical minimum value；
Under without target conditions, using above-mentioned data to be tested, filtered out from corresponding subclass and target highlight number
Identical maximum.
The reverberation covariance matrix of the present invention is the random matrix for obeying inverse multiple Wei Shate distributions, and its matrix parameter can be by sound
System of receiving detection environment, working method, reverberation space-time characteristic, the priori of shallow water topography feature are obtained, realized to priori
The abundant excavation of knowledge.On this basis, many bright spot detection methods of Knowledge-based proposed by the present invention, are completed to priori
Make full use of, greatly improve the detection performance of many bright spot targets under non-homogeneous environment.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, being used required in being described below to embodiment
Accompanying drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this
For the those of ordinary skill of field, on the premise of not paying creative work, it can also obtain other according to these accompanying drawings
Accompanying drawing：
Fig. 1 is Knowledge-based adaptive detector design principle figure of the present invention；
Fig. 2 is of the invention by Meng Te-Caro emulation mode, in false-alarm probability Pfa=10-3, N=8, K=16, L=3 and
In the case of v=2N, the present invention and tradition GLRT detector detection probabilities PdWith SRR relation curve；
Fig. 3 is of the invention by Meng Te-Caro emulation mode, in false-alarm probability Pfa=10-3, N=8, K=32, L=3 and
In the case of v=2N, the present invention and tradition GLRT detector detection probabilities PdWith SRR relation curve；
Fig. 4 is of the invention by Meng Te-Caro emulation mode, in false-alarm probability Pfa=10-3, N=8, K=32, L=3 and
In the case of v=6N, the present invention and tradition GLRT detector detection probabilities PdWith SRR relation curve.
Embodiment
Below by drawings and examples, technical scheme is described in further detail.
The present invention is in the case of no assistance data, it is assumed that colored Gaussian distribution, and reverberation covariance square are obeyed in reverberation
Battle array M by two step likelihood ratio test criterions, it is known that solve test problems, detection of the realization to many bright spot targets.
Concretely comprise the following steps：
Assuming that the linear array that echo is made up of N number of array element is received, the echo data of each array element includes being received from K distance
One group of echo data of unit, as data to be tested, is expressed as rt, t ∈ Ω={ 1 ..., K }.Use set omegaT∈ Ω represent mesh
The index of range cell, set omega where marking signalTLength is L (L≤K), and the number of L correspondence target echoes is detected
Many bright spot target numbers.Under the above-described conditions, target detection problems are converted into following dualism hypothesis problem：
Wherein H0Represent without goal hypothesis, reverberation vector nt, t ∈ Ω are independent, N-dimensional zero-mean complex Gaussian vectors, its
Covariance matrix isV is that known N-dimensional leads goal orientation vector；αt, t ∈ Ω, expression target reflection factor；T ∈ Ω are that the N-dimensional introduced to improve detector to the rejection ability of side-lobe signal disturbs vector, and it is with v in albefaction space
Interior perpendicular quadrature,It can be expressed asWherein xtIt is the complex vector of N-1 dimensions, W is the complex matrix of N × (N-1) dimensions, and
Meet<M-1/2W>⊥=<M-1/2v>。
Based on conditions above, data to be tested r can be obtainedt, t ∈ Ω
Likelihood function/probability density function (PDF) in the case where there is target conditions is：
It is with likelihood function/probability density function (PDF) under without target conditions：
Wherein | | | | the determinant of a square formation is represented,Represent conjugate transposition.
It is v to obey the free degree according to M, and average is M0Inverse multiple Wei Shate (Wishart) distributions, i.e. M ∽ CW-1((v-N)
M0, v), its PDF can be expressed as
f(M)∝||M||-(v+N)exp{-(v-N)tr(M-1M0)} (4)
Wherein, ∝ represents that the variable at the symbol two ends has proportional relation, M-1Similarly obey inverse multiple Wishart distributions.
In actual applications, Mean Matrix M0High-resolution sonar is mainly used to detect environment, working method, reverberation sky
When characteristic, shallow water topography feature obtains prior information, built according to reverberation scattering principle.It can be seen that M0Prior information is received
Enter unified model, realize the abundant excavation to priori.
In detector design aspect, the present invention is former using the test problems corresponding to two step GLRT criterions solution formulas (1)
Reason figure is as shown in figure 1, the first step, assuming that in the case of known to reverberation covariance matrix M, based on data to be tested rtDesign is seemingly
So compare detector；Second step, M MAP estimation value is substituted the true value of the M in tradition GLRT detectors, so as to be known
Know base adaptive detector, referred to as K-GLRT.
The first step：Assuming that under M known conditions, the first test statistics GLRT can be expressed as
Wherein η is by false-alarm probability (Pfa) detection threshold value that determines, it can be obtained by Meng Te-Caro emulation；
H is represented respectively1And H0In the case of range cell where the target echo that is filtered out index set,Represent in H1Feelings
α under conditiontThe maximal possibility estimation of index,Represent in H0In the case of xtThe maximal possibility estimation of index.It is right that K-GLRT passes throughWithDetermination, realize to target data and the automatic screening of assistance data, complete the adaptive inspection to many bright spot targets
Survey.
Bring formula (2), (3) into formula (5), through deriving, above-mentioned first test statistics GLRT can be write as
Wherein,WithRepresent, Ω subset, can be seen that from formula (6)It is ordered series of numbersIn
The corresponding index set of L minimum value,It is ordered series of numbersMiddle L maximum is corresponding to index set.
Second step：Solve H0In the case of M MAP estimationFirst according to M PDFf (M) and data to be tested rt
'sObtain M posterior probability function f (M | r1,...,rK), it is specially：
Wherein, ∝ represents that the measurer on symbol both sides has proportional relation,For data to be tested rtSampling association side
Poor matrix, the mark of tr () representing matrix.
By M posterior probability function f (M | r1,...,rK) solve M MAP estimationsThe step of it is as follows：
To f (M | r1,...,rK) derivation, and make derivation result be equal to 0, it can obtain
M-1(S+(v-N)M0)M-1-(K+N+v)M-1=0 (8)
According to formula (8), MAP estimationFor
For formula (9), traditional GLRT methods use maximal possibility estimation, the value be by sampling covariance estimation S to
Go out, and the MAP estimation of the present inventionIt is a S coloured loading, loading matrix is M0, loading level is true by free degree v
It is fixed.
M in first test statistics GLRT of formula (6) is replaced withObtain the second test statistics, i.e., it is of the invention
Knowledge-based detector K-GLRT, be specially：
WhereinWithIt need to recalculate, i.e., using S+ (v-N) M0The true value for replacing the M in former formula is obtained.By formula
(10) data to be tested can be screened, its step is as follows：
In H1In the case of, utilize data to be tested rt, fromIn filter out and L minimum value；
In H0In the case of, utilize data to be tested rt, fromIn filter out and L maximum.
It is worth noting that, as v=N, add-in hour, K-GLRT is degenerated to traditional non-Knowledge-based GLRT, it is examined
Surveying statistic is
It is thereinWithIt is Ω subset, corresponds respectively toL minimum value in t ∈ Ω
WithThe index set of L maximum in t ∈ Ω.
Comparison expression (10) and formula (11) can see, and K-GLRT and tradition GLRT have identical structure, are a difference in that
K-GLRT uses coloured loading S+ (v-N) M0It instead of common sampling covariance estimation S.Loading free degree v determines K-
GLRT utilizes the degree of priori, i.e. v is bigger, bigger to the producing level of priori.
The present invention analyzes K-GLRT performance by Meng Te-Caro emulation mode, and is compared with traditional GLRT.It is imitative
Design parameter in very is set to N=8, L=3, false-alarm probability Pfa=10-3,
M∽CW-1((v-N)M0, v), M0=0.9|i-j|, wherein (i, j) is the coordinate of matrix element.The mixed ratio of letter is defined asNominal target steering vector v=[1 ..., 1]T/N([·]TRepresent transposition operation).
Two kinds of different K value situations, i.e. K=2N and K=4N are considered first.Fig. 2 and Fig. 3 sets forth both of these case
Lower different detectors detection probability PdWith SRR relation curve.Data length is observed it can be seen from this two width figure fewer, K-
GLRT performance advantage is more obvious, for example, work as K=2N, during SRR=16dB, K-GLRT Pd=0.9, and GLRT Pd＜ 0.1.
It can be seen that, reasonably it can effectively improve the performance of high-resolution sonar system in the presence of a harsh environment using priori.
Then consider two kinds of different v value situations, Fig. 4 consider with Fig. 3 identical simulation parameters, be not both uniquely to adopt
V=2N is substituted with v=6N.Compare Fig. 3 and Fig. 4, it can be seen that with v increase, K-GLRT detection probability is higher, i.e., v is got over
Greatly, it is higher to the producing level of priori.
Above-described embodiment, has been carried out further to the purpose of the present invention, technical scheme and beneficial effect
Describe in detail, should be understood that the embodiment that the foregoing is only the present invention, be not intended to limit the present invention
Protection domain, within the spirit and principles of the invention, any modification, equivalent substitution and improvements done etc. all should be included
Within protection scope of the present invention.
Claims (5)
1. a kind of many bright spot target space-time detecting methods based on random covariance matrix, including：
The echo received by sonar battle array obtains one group of echo data, regard one group of echo data of acquisition as number to be detected
According to；Wherein, in the presence of without target, interference that may be present is described as and goal orientation in the data to be tested
Vector is in mutually orthogonal vector after albefaction, in the case of with the presence of target, and target is contemplated as falling with the number to be detected
In, and across multiple range cells, correspondence echo signal index length is L；
Assuming that complex Gaussian distribution is obeyed in reverberation, and under reverberation covariance matrix and goal orientation vector known case, pass through structure
Build likelihood function and obtain disturbing the maximal possibility estimation of index under no target conditions, and have what the target under target conditions was indexed
The maximal possibility estimation of maximal possibility estimation and target reflection factor, using the likelihood function having under target conditions, without mesh
The likelihood function in the case of mark, the maximal possibility estimation of interference index, the maximal possibility estimation of target index,
The maximal possibility estimation of the target reflection factor and the steering vector of target, obtain the first test statistics；
According to the inverse multiple Wei Shate distributions of the inverse obedience of the reverberation covariance matrix, obtain described mixed using the data to be tested
The MAP estimation of covariance matrix is rung, the MAP estimation is substituted into first test statistics and obtains second
Test statistics, second test statistics is compared with detection threshold, completes target whether there is judgement.
2. according to the method described in claim 1, it is characterised in that the MAP estimation is sampling covariance estimation S
One coloured loading, can be expressed as：S+(v-N)M0, wherein, M0High-resolution sonar is mainly used to detect environment, work side
Formula, reverberation space-time characteristic, shallow water topography characteristic obtain prior information, the Mean Matrix built according to reverberation scattering principle, plus
Load degree is determined that N is the linear array receiver number that array element is constituted by free degree v.
3. according to the method described in claim 1, it is characterised in that the detection threshold value is determined by false-alarm probability, it can lead to
Meng Te-Caro emulation is crossed to obtain.
4. according to the method described in claim 1, it is characterised in that it is described complete target and whether there is the process of judgement be actually pair
The process of the data to be tested screening, is represented by：
Wherein, η is detection threshold value, and v is the free degree,It is ordered series of numbersMiddle L minimum value pair
The index answered,It is ordered series of numbersThe corresponding index of middle L maximum, wherein, rtIt is data to be tested, η is
Detection threshold, []-1Represent to matrix inverse operation, []HRepresent conjugate transposition operation, H0Indicate no target conditions, H1Indicate
Target conditions.
5. method according to claim 4, it is characterised in that it is described the step of screened to the data to be tested such as
Under：
In the case where there are target conditions, using the data to be tested, filtered out from corresponding subclass identical with target highlight number
Minimum value；
Under without target conditions, using the data to be tested, filtered out from corresponding subclass identical with target highlight number
Maximum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510388924.8A CN104977579B (en) | 2015-07-03 | 2015-07-03 | A kind of many bright spot target space-time detecting methods based on random covariance matrix |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510388924.8A CN104977579B (en) | 2015-07-03 | 2015-07-03 | A kind of many bright spot target space-time detecting methods based on random covariance matrix |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104977579A CN104977579A (en) | 2015-10-14 |
| CN104977579B true CN104977579B (en) | 2017-07-28 |
Family
ID=54274259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510388924.8A Active CN104977579B (en) | 2015-07-03 | 2015-07-03 | A kind of many bright spot target space-time detecting methods based on random covariance matrix |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104977579B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105548994B (en) * | 2015-12-08 | 2017-12-12 | 天津工业大学 | A kind of underwater more bright spot mesh object detection methods based on compressed sensing |
| CN105676211B (en) * | 2015-12-30 | 2017-12-01 | 中国科学院声学研究所 | A kind of more bright spot target space-time detection methods based on contraction technology |
| CN105974400A (en) * | 2016-04-27 | 2016-09-28 | 中国科学院声学研究所 | Robust space-time detection method based on symmetric spectral characteristic |
| CN107479050B (en) * | 2017-08-13 | 2020-04-14 | 中国科学院声学研究所 | Target detection method and device based on symmetric spectral characteristics and sub-symmetric characteristics |
| CN107561540B (en) * | 2017-08-13 | 2020-02-14 | 中国科学院声学研究所 | Sonar leakage target detection method and device based on reverberation symmetric spectrum characteristics |
| CN107958256A (en) * | 2017-10-09 | 2018-04-24 | 中国电子科技集团公司第二十八研究所 | It is a kind of based on the assumption that examine the recognition methods of public sentiment number of topics and system |
| CN108008393A (en) * | 2017-10-19 | 2018-05-08 | 中国科学院声学研究所 | A kind of guiding vector corrects space-time adaptive detection method |
| CN108593767A (en) * | 2018-01-24 | 2018-09-28 | 天津大学 | A kind of method for building up of shoal buried pipes supersonic sounding echo model |
| CN111090089B (en) * | 2018-10-24 | 2021-12-07 | 中国科学院声学研究所 | Space-time adaptive detection method based on two types of auxiliary data |
| CN109782261A (en) * | 2018-12-27 | 2019-05-21 | 西北工业大学 | Cause the unsupervised learning detection method for receiving signal intensity to underwater intrusion target |
| CN112505665B (en) * | 2020-11-10 | 2023-06-23 | 中国科学院声学研究所 | Space-time self-adaptive detection method and system suitable for partial uniform reverberation environment |
| CN113933808A (en) * | 2021-09-29 | 2022-01-14 | 中国电子科技集团公司第二十九研究所 | Airborne radar moving target detection method, device, equipment and storage medium |
| CN115685082B (en) * | 2022-11-10 | 2023-09-05 | 山东工商学院 | Wald-based detection method for distance expansion target under interference and noise background |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100361414C (en) * | 2004-06-01 | 2008-01-09 | 华为技术有限公司 | Grouping vertical layered space-time detecting method and device |
| US8447130B2 (en) * | 2007-12-31 | 2013-05-21 | Intel Corporation | History-based spatio-temporal noise reduction |
| CN103033815B (en) * | 2012-12-19 | 2014-11-05 | 中国科学院声学研究所 | Detection Method and detection device of distance expansion target based on reverberation covariance matrix |
-
2015
- 2015-07-03 CN CN201510388924.8A patent/CN104977579B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104977579A (en) | 2015-10-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104977579B (en) | A kind of many bright spot target space-time detecting methods based on random covariance matrix | |
| Zhi et al. | Predictive mapping of seabed substrata using high-resolution multibeam sonar data: A case study from a shelf with complex geomorphology | |
| CN103033815B (en) | Detection Method and detection device of distance expansion target based on reverberation covariance matrix | |
| CN106872958B (en) | Radar target self-adapting detecting method based on linear fusion | |
| CN111965632B (en) | Radar target detection method based on Riemann manifold dimensionality reduction | |
| CN103954943B (en) | Networking radar system anti-Deceiving interference method | |
| CN108802722A (en) | It is a kind of based on tracking before the Faint target detection virtually composed | |
| CN106997042A (en) | A kind of Target Signal Detection and device | |
| JP6773606B2 (en) | Radar device and its radar signal processing method | |
| CN107561540A (en) | The detection method and device of sonar leakage target based on the symmetrical spectral property of reverberation | |
| CN104751185A (en) | SAR image change detection method based on mean shift genetic clustering | |
| Olea | A practical primer on geostatistics | |
| CN104237861A (en) | Method for obtaining CFAR detection threshold in unknown clutter background | |
| CN110475246A (en) | Malice anchor node detection method based on isolated forest and sequential probability ratio test | |
| CN110208806A (en) | A kind of marine radar image rainfall recognition methods | |
| CN107102308A (en) | A kind of distributed radar anomeric signals level fusion object detection method | |
| CN104198998B (en) | Clustering treatment based CFAR (Constant False Alarm Rate) detection method under non-uniform background | |
| CN104268574A (en) | SAR image change detecting method based on genetic kernel fuzzy clustering | |
| CN105954739A (en) | Knowledge-aided nonparametric constant false alarm detection method | |
| CN104880696B (en) | One kind is based on the preferable multi-channel target detection method of passage | |
| CN105866748A (en) | Constant false alarm detection method based on former detection information-constant window | |
| CN110751201B (en) | SAR equipment task failure cause reasoning method based on textural feature transformation | |
| CN105676211B (en) | A kind of more bright spot target space-time detection methods based on contraction technology | |
| CN106443612A (en) | Information entropy-based multi-radar hybrid detection method | |
| CN109696662A (en) | A kind of object detection method based on K statistical distribution pattern background |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |