CN106772371A - Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application - Google Patents

Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application Download PDF

Info

Publication number
CN106772371A
CN106772371A CN201611039349.1A CN201611039349A CN106772371A CN 106772371 A CN106772371 A CN 106772371A CN 201611039349 A CN201611039349 A CN 201611039349A CN 106772371 A CN106772371 A CN 106772371A
Authority
CN
China
Prior art keywords
polarimetric
polarization
error
classification
influence
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201611039349.1A
Other languages
Chinese (zh)
Inventor
许丽颖
崔雷
涂尚坦
陈筠力
於伟民
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Institute of Satellite Engineering
Original Assignee
Shanghai Institute of Satellite Engineering
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Institute of Satellite Engineering filed Critical Shanghai Institute of Satellite Engineering
Priority to CN201611039349.1A priority Critical patent/CN106772371A/en
Publication of CN106772371A publication Critical patent/CN106772371A/en
Pending legal-status Critical Current

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
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/9021—SAR image post-processing techniques
    • G01S13/9023—SAR image post-processing techniques combined with interferometric techniques
    • 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
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/904—SAR modes
    • G01S13/9058—Bistatic or multistatic SAR
    • 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
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
    • G01S13/904—SAR modes
    • G01S13/9076—Polarimetric features in SAR
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/40—Means for monitoring or calibrating

Abstract

The invention discloses a kind of polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application, its step mainly includes：Step one, sets up the distortion model of polarimetric SAR interferometry ground object target measurement in the presence of polarimetric calibration error；Step 2, analyzes influence of the polarimetric calibration error to polarimetric SAR interferometry key system perameter under the model；Step 3, for the Error Propagation Model of terrain classification polarimetric calibration error；Step 4, polarimetric SAR interferometry is provided to two kinds of influence cases of typical classification algorithm based on Error Propagation Model, finally gives the polarimetric calibration parameter requirements based on classification application.The present invention can obtain the model of data, obtain distortion covariance matrix；Influence of the analysis polarimetric calibration error to polarimetric SAR interferometry key system perameter；The basic procedure of terrain classification is analyzed, the Error Propagation Model of polarimetric calibration error is obtained；According to systematic function and application performance correlation, directly System Parameter Design is claimed by application performance.

Description

Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application
Technical field
It is more particularly to a kind of to be based on polarimetric SAR interferometry the present invention relates to a kind of polarimetric calibration parameter requirements analysis method The polarimetric calibration parameter requirements analysis method of classification application.
Background technology
Polarimetric SAR interferometry is that for being grown up based on interference SAR measurement and polarization SAR e measurement technology in the recent period is new Cutting edge technology, mapping, mima type microrelief change detection, vegetation biomass estimate etc. various fields have important application.Enter One important prerequisite of the theoretical research of row polarimetric SAR interferometry and operational use is to obtain high-quality polarimetric SAR interferometry data, The acquisition of high-quality polarimetric SAR interferometry data is closely related with System Parameter Design：On the one hand, existing System Parameter Design Method generally needs, based on conventional interference SAR or polarization SAR system performance analysis method, to lack at present specifically designed for polarization interference The Parameters design of SAR system；On the other hand, systematic function is closely related with application performance, is but difficult to straight by application performance Connect and System Parameter Design is claimed, lack the polarimetric SAR interferometry system performance analysis method for application.This be mainly by Contacted with complicated Coherent Scattering Model between polarimetric SAR interferometry application and radar system parameters, do not had between the two There is the relational expression of direct quantitative, therefore analytically analysis polarimetric SAR interferometry system application performance is difficult to.
Although Krieger et al. has the ability for distinguishing different height scattering center according to polarimetric SAR interferometry, it is proposed that phase The system performance analysis method of position pipe, but this method is not the appraisal procedure directly against application, and it is similar to interference SAR The appraisal procedure of " decorrelation analysis " in System Performance Analysis.Lopez-Sanchez et al. qualitative analyses crops applications To polarimetric SAR interferometry system baseline, frequency bandwidth, incidence angle, obtaining mode, revisiting period requirement.Lack direct pin at present The Parameters design of the polarimetric SAR interferometry system to applying.Polarization interference packet amount containing more information, makes some poles of separation The similar atural object of some scattering properties is possibly realized in change SAR, and carrying out terrain classification using polarimetric SAR interferometry image also turns into one Individual study hotspot.Research in the urgent need to carrying out the parameter designing of the polarimetric SAR interferometry system directly against terrain classification application.
The content of the invention
The technical problems to be solved by the invention are to provide a kind of polarimetric calibration ginseng based on polarimetric SAR interferometry classification application Number requirement analysis method, its model that can obtain data obtains distortion covariance matrix；Analysis polarimetric calibration error is dry to polarization Relate to the influence of SAR key system perameters；The basic procedure of terrain classification is analyzed, the error propagation mould of polarimetric calibration error is obtained Type；According to systematic function and application performance correlation, directly System Parameter Design is claimed by application performance.
The present invention is to solve above-mentioned technical problem by following technical proposals：It is a kind of to be classified based on polarimetric SAR interferometry The polarimetric calibration parameter requirements analysis method of application, its step mainly includes：
Step one, sets up the distortion model of polarimetric SAR interferometry ground object target measurement in the presence of polarimetric calibration error；
Step 2, analyzes influence of the polarimetric calibration error to polarimetric SAR interferometry key system perameter under the model；
Step 3, for the Error Propagation Model of terrain classification polarimetric calibration error；
Step 4, provides polarimetric SAR interferometry to two kinds of influence cases of typical classification algorithm, most based on Error Propagation Model The polarimetric calibration parameter requirements based on classification application are obtained eventually.
Preferably, the step one includes：
By polarimetric SAR interferometry system acquisition data procedures and polarization Coherence optimization essence, polarimetric calibration error bag is given Including interchannel crosstalk, disequilibrium is the main error source of polarimetric SAR interferometry system application, and then obtains the measurement of ground object target Distortion model；
Using the model, the polarization scattering matrix vector quantization of target is obtained after Scattering of Vector by being calculated pole Change the target polarization covariance matrix and polarization coherence matrix in the presence of calibration error.
Preferably, the target polarization covariance matrix in the presence of the polarimetric calibration error and polarization coherence matrix mainly divide The influence optimal to polarization interference double scattering mechanism amplitude coherent of analysis polarisation error.
Preferably, the method for the step 3 including the terrain classification, influence to polarization interference Wishart graders, right The influence that polarization interference class merges.
Preferably, the method for the terrain classification comprises the following steps：
Step 11, obtains polarization interference image, carries out phase separation immunoassay to polarimetric SAR interferometry data and coherence matrix is estimated The early stage treatment of meter；
Filtered data are entered line parameter calculating by step 12, and preliminary classification is carried out using the parameter for calculating；
Step 13, the average polarization interference matrix of each class is calculated according to preliminary classification result.
Preferably, the step 4 includes the method, right that polarization interference data are classifiedGrader Influence, the influence to the grader based on Shannon entropy to resolve.
Preferably, the method that the polarization interference data are classified, one kind is using polarization decomposing method and coherence The terrain classification method being combined, typical algorithm is sorting technique；Another kind is by polarization interference Wishart graders and pole Change the sorting technique that interference decomposition method is combined, typical algorithm is divided with polarization interference Wishart using Shannon entropy to resolve The method that class device carries out terrain classification.
Positive effect of the invention is：The present invention can be by providing the polarization interference in the presence of polarimetric calibration error SAR obtains the model of data, obtains the distortion covariance matrix of ground object target in the case of polarisation error is present；Utilizing covariance Matrix analysis influence of the polarimetric calibration error to polarimetric SAR interferometry key system perameter under the model；By analyzing atural object point The basic procedure of class, obtains the Error Propagation Model of polarimetric calibration error；Polarimetric SAR interferometry pair is given further according to the model Two kinds of influence cases of typical classification algorithm, finally obtained the polarimetric calibration parameter of the polarimetric SAR interferometry based on classification application Demand.In this way, can be according to systematic function and application performance correlation, directly System Parameter Design is proposed by application performance will Ask.
Brief description of the drawings
Fig. 1 is that the flow of polarimetric calibration parameter requirements analysis method of the present invention based on polarimetric SAR interferometry classification application is shown It is intended to.
Fig. 2 is the data collecting model figure of polarimetric SAR interferometry in the present invention.
Fig. 3 is the terrain classification flow chart of polarimetric SAR interferometry in the present invention.
Fig. 4 a- Fig. 4 b be in the present invention calibration error to the optimal influence figure for thinking responsibility number.
Specific embodiment
Present pre-ferred embodiments are given below in conjunction with the accompanying drawings, to describe technical scheme in detail.
As shown in figure 1, the invention discloses polarization of the one kind based on polarimetric SAR interferometry (synthetic aperture radar) classification application Scaling parameter requirement analysis method, its step mainly includes：
Step one, sets up the distortion model of polarimetric SAR interferometry ground object target measurement in the presence of polarimetric calibration error；
Step 2, analyzes influence of the polarimetric calibration error to polarimetric SAR interferometry key system perameter under the model；
Step 3, for the Error Propagation Model of terrain classification polarimetric calibration error；
Step 4, provides polarimetric SAR interferometry to two kinds of influence cases of typical classification algorithm, most based on Error Propagation Model The polarimetric calibration parameter requirements based on classification application are obtained eventually.
The step one includes：
By polarimetric SAR interferometry system acquisition data procedures and polarization Coherence optimization essence, polarimetric calibration error bag is given Including interchannel crosstalk, disequilibrium is the main error source of polarimetric SAR interferometry system application, and then obtains the measurement of ground object target Distortion model；
Using the model, the polarization scattering matrix vector quantization of target is obtained after Scattering of Vector by being calculated pole Change the target polarization covariance matrix and polarization coherence matrix in the presence of calibration error.
The acquisition methods of two groups of view data of the same target are：Interference flight, polarimetric SAR interferometry primary antenna twice Or two groups of view data that slave antenna is obtained；Or, polarimetric SAR interferometry primary antenna and pair are in-flight obtained respectively once interfering Two groups of view data of antenna, polarimetric SAR interferometry system acquisition data procedures are as shown in Figure 2.Polarimetric SAR interferometry system mainly has Several error sources below：1) transmitting and receiving channel are uneven, signal is produced amplitude and phase distortion；2) poliarizing antenna passage Between exist between H, V passage of signal cross-talk distortion, i.e. system and there is no completely isolated, there is coupling phenomenon；3) echo-signal Decorrelation, mainly there is the decorrelation of baseline space, time decorrelation and Signal-to-Noise decorrelation；4) polarimetric SAR interferometry is concerned with The presence of spot；5) aircraft flight attitude error.Because in the optimal algorithm of polarimetric SAR interferometry, polarization base conversion requirement is carried out Accurate same polarization and cross-polarized multiple calibration, the evaluated error of interchannel crosstalk, disequilibrium and same polarization phase is The principal element of influence polarimetric SAR interferometry measurement.
Each alphabetical implication in the separate equations, such as the ps in formula (3) etc. please be describe.
Polarimetric SAR interferometry system obtains two groups of full polarimetric SAR datas, is formulated such as following formula (1)：
Polarisation error main influence matrix R and matrix T, interferometry geometric error can mainly influence system phase φ1With φ2, the influence of polarisation error and interferometric parameter deviation is mainly considered due to this section, therefore do not consider system gain K and thermal noise N。
Influence for convenience of analysis polarization interference systematic error to applying, in the case of reciprocity medium, utilizes Lexicographic bases obtain the target scattering matrix M and real target scattering square of systematic survey to collision matrix vector quantization Battle array S vector quantization formula m and s, respectively such as following formula (1) and (2)：
In the case where system gain K and thermal noise N is not considered, measurement vector is as follows with the relation of the true vector of target Formula (3)：
M=ejφPs(3)
WhereinS is the vector such as formula obtained after real target scattering matrix S vector quantizations (2)。
If δ1=δ2=δ3=δ4=δ, f1=f2=f, then received using Pauli basic matrixs to collision matrix vector quantization The coherent scattering vector of error influence is such as following formula (4) with former coherent scattering vector：
kd=Am
Wherein,Obtain closing for the coherent scattering vector that is influenceed by error and former coherent scattering vector It is k to bed=Am=ejφAPs=ejφ(APA-1) k=ejφZk, Z is distortion transfer matrix Z=APA in formula-1。
The coherent scattering vector that two secondary complete polarization antennas are obtained under the influence of polarisation error is respectively：Z1And Z2Respectively the distortion transfer matrix of the antenna of major-minor two, obtains polarization interference error Under the influence of coherence matrix be such as following formula (5)：
Wherein,It is system intervention phase deviation, it is relevant using the polarization interference influenceed by polarization interference error Matrix, theoretically influence of the derivation system parameter to polarimetric SAR interferometry application, for the quantization requirement of polarimetric calibration provides theory Foundation.
During polarimetric SAR interferometry system development, coherence is the measurement index of polarimetric SAR interferometry systematic function, is sought The problem for seeking the polarization mode that can make interference Coherence optimization is the underlying issue of polarization interference applied analysis.Main Analysis polarization is missed The difference influence optimal to polarization interference double scattering mechanism amplitude coherent is analyzed, and it is in pole that double scattering mechanism amplitude coherent is optimal Searching combines the maximum polarized state of coherence factor γ amplitudes in changing space, and mathematic(al) representation is such as following formula (6)：
Wherein, ω1And ω2It is the polarized state vector of the antenna of major-minor two, will be optimal in formula using method of Lagrange multipliers Value problem is converted into Eigenvalues Decomposition problem and obtains such as following formula (7)：
The amplitude and phase value of three optimized coherence coefficients being calculated are respectively such as following formula (8)：
arg(μ1jμ2j *)=arg ((ω1j Hk1)(ω2j Hk2)*)=arg (ω1jHΩ12ω2j) (8)
Wherein, ω1jAnd ω2jThree kinds of optimal scattering mechanisms for being calculated major-minor image are represented respectively.
Coherence optimization is carried out in the case of being influenceed by error to collision matrix to be decomposed into such as following formula (9)：
(Z2 H)-1(T22 -1Ω12 HT11 -1Ω12)Z2 Hω2j d=λj dω2j d
(Z1 H)-1(T11 -1Ω12T22 -1Ω12 H)Z1 Hω1j d=λj dω1j d (9)
Contrast learns, carried out in the case of polarisation error the matrix of Eigenvalues Decomposition to it is error free when be similar square Battle array, the characteristic value size of similar matrix is constant, but characteristic vector can change.Physical interpretation is to ensure optimized coherence system Scattering mechanism is reselected in the case that number size is constant.Scattering mechanism changes and means that the phase of coherence factor Error can be produced.The amplitude and phase value of optimized coherence coefficient under the influence of error is obtained to be respectively such as following formula (10)：
arg(μ1 dμ2 d*)=arg ((ω1 d)Hk1 d)((ω2 d)Hk2 d)*=arg (b1b2 *ω1HΩ12ω2) (10)
In the presence of error, the relation of scattering mechanism and former scattering mechanism According to constraints arg ((ω1j d)Hω2j d)=0 obtains following formula (11)：
arg((ω1j d)Hω2j d)=arg ((b1(Z1 H)-1ω1j)H·b2(Z2 H)-1ω2j)=arg (b1 *b2ω1j H(Z2 HZ1 )-1ω2j)=0 (11)
It is equivalent to arg (b1 *b2)=- arg (ω1j H(Z2 HZ1)-1ω2j)。
Contrast sees that polarization interference error will not produce influence, but the phase error that can cause on the amplitude of coherence factor, The phase error for causing is such as following formula (12)：
Take E-SAR to be tested in the L-band polarization interference data that German Oberpfafenhoffen areas obtain, lead to Cross and set different polarisation errors and calculate the data existed in the case of calibration error, the phase obtained under different error condition Responsibility number finds out that polarisation error can influence double scattering optimized coherence machine with comparison diagram ideally as shown in Figs. 4 a-b The phase of the optimized coherence coefficient being obtained, but do not interfere with the amplitude of optimized coherence coefficient.Interferometric phase error will not be right The amplitude and phase of optimized coherence coefficient produce influence.Because polarization interference optimized coherence process is in global optimization's process, The amplitude of optimized coherence coefficient will not produce change.But polarisation error presence can make the characteristic vector can change, this explanation It is, when global optimum is ensured, scattering mechanism to be reselected in complete polarization space.Interferometric phase deviation does not interfere with full pole Change scattering mechanism is reselected in space, so interferometric phase deviation will not produce influence to optimized coherence coefficient, but A skew can be produced to the overall distribution of coherence factor.
The coefficient correlation classified exactly to two images using polarization interference data is analyzed, by coherence factor with Polarization decomposing method is combined carries out terrain classification, as shown in figure 3, the method for terrain classification comprises the following steps in step 3：
Step 11, obtains polarization interference image, carries out phase separation immunoassay to polarimetric SAR interferometry data and coherence matrix is estimated The early stage treatment of meter；
Filtered data are entered line parameter calculating by step 12, and preliminary classification is carried out using the parameter for calculating；
Step 13, the average polarization interference matrix (Ji Lei centers) of each class is calculated according to preliminary classification result.
All classes indexing between any two is calculated when initial classes are excessive, merging is iterated according to that can index Treatment, the class number until reaching requirement；
Wishart clusters finally are carried out to desired class number, last classification results are obtained.
By shadow of the important system parameter such as crosstalk, disequilibrium error to each step of atural object sorting algorithm between analysis channel Sound sets up Error Propagation Model.
Influence to polarization interference Wishart graders：
It is polarization interference coherence matrix T6It is d (T, V to define Bayes's maximum likelihood classifierm)=ln | Vm|+tr (Vm -1T target to the distance at all class centers, V) are calculatedmIt is the centers scatter matrix of m classes.If the distance with m classes center Minimum, then the target be classified as m classes.If the polarization distortion matrix of major-minor antenna is identical, i.e. Z1=Z2=Z, now Td=ZTTZT H, its InZ0Represent 3 × 3 null matrix.Obtained by the property of similar matrix, obtain being influenceed by polarization interference error Under Wishart polarization interference graders be such as following formula (13)：
d(Td,Vm d)=ln | ZT|+ln|ZT H|+ln|Vm|+tr(Vm -1T) (13)
The ln in same system | ZT|+ln|ZT H| value keep it is constant, polarization interference error only influence target to each The absolute distance at class center, does not interfere with relative distance, so not interfering with Wishart classification results.In sum, polarize Interference error will not produce influence to polarization interference Wishart graders.
The influence merged to polarization interference class：
If initial class divides too many in assorting process, it is necessary to class consolidation problem is considered, by calculating all classes two Indexing between two, two classes that maximum will can be indexed in whole class set merge into a class.Liang Gelei centers are respectively ViAnd Vj's The class spacing of the i-th class and jth class is such as following formula (14)：
Within-cluster variance is defined as the pixel in class to the average distance between the characteristic dispersion coherence matrix at class center, warp The equivalent within-cluster variance for crossing the i-th class obtained after abbreviation is Dii=ln | Vi|.Ri can be indexedjIt is defined as the class between two classes The ratio between spacing and within-cluster variance, beTo be there are maximum two classes that can be indexed to be polymerized to a class, Zhi Daohe And to during required class number complete class merging process.
Polarization interference error can influence indexing between class, and then influence final classification results.Missed in polarization interference Under the influence of difference, class center is obtained for Vi dAnd Vj dThe i-th class and the spacing of jth class be such as following formula (15)：
The equivalent within-cluster variance of the i-th class is Dii d=ln | ZT|+ln|ZT H|+ln|Vi|.Same system error is identical, So, make 2 (ln | ZT|+ln|ZT H)=A, A are constant.Obtain under the influence of polarization interference distortion, Vi dAnd Vj dClass can Indexing Rij dIt is such as following formula (16)：
Find out that polarization interference error can increase the indexed generation of class influence, this influence with the increase of A, and Influence can be produced on final class amalgamation result.
The method that polarization interference data are classified, a kind of is the atural object being combined with coherence using polarization decomposing method Sorting technique, typical algorithm is sorting technique；Another kind is by polarization interference Wishart graders and polarization interference decomposition side The sorting technique that method is combined, typical algorithm is to carry out atural object using Shannon entropy to resolve and polarization interference Wishart graders The method of classification.
It is rightThe content of the influence of grader is as follows：
Cloude-Pottier decomposition methods calculate two using the characteristic value and characteristic vector of the polarization coherence matrix of target Individual parameter scattering entropy H and average angle of scatteringLee et al. basesThe scattering properties of plane by Terrain Scattering type with Wishart graders are combined and for atural object to be divided into 8 classes.L.Ferro-Famil et al. proposes optimized coherence spectrum parameter, and utilizes The parameter is finely divided to improve classification performance to 8 classes divided.Grader utilization can be indexed will be above-mentioned Classification results are merged into required class number, and using Wishart iterators cluster and obtain final classification results.
Influence of the lower surface analysis polarization interference error to sorting parameter.The coherence matrix T influenceed by errorii d, i=1,2 with Coherence matrix T ideallyii, the relation of i=1,2 is Tii d=ZTiiZH, i=1,2, because polarization interference error can be to dividing Class parameter produces influence, and Z is non-unity battle array, thus polarization interference error to H andInfluence cannot quantify derive.
The amplitude of the polarization interference optimized coherence coefficient obtained according to polarization interference amplitude optimized coherence theory | γi|, it is fixed The adopted optimal entropy H of polarization interferenceintWith polarization interference optimized coherence degree of anisotropy AintFor
PiIt is calculated by the amplitude of optimized coherence coefficient,
Find out because polarization interference error does not influence on the amplitude of optimized coherence coefficient, polarization Coherence optimization spectrum parameter is not Influenceed by polarization interference error.
To sum up analysis learns that polarization interference error is to polarization parameterHave an impact, to polarization interference spectrum parameter Hint/ AintWithout influence, have an impact to separating degree, polarization interference Wishart graders are not influenceed.Missed to quantitatively analyze calibration Influence of the difference to above-mentioned each factor and rightThe influence of classification.Using the classification results of preferable echo data as ginseng Standard is examined, is calculated under the influence of different polarisation errors,Sorting technique,The relative classification of sorting technique The degree of accuracy, is listed in table 1
Table 1WithThe relative classification degree of accuracy of sorting technique
The influence of surface channel unbalance in phase and unbalance in phase to polarization interference terrain classification accuracy rate is most aobvious in table 1 Write, influence of the polarization isolation to the degree of accuracy of classifying is relatively small.Generally polarization isolation is higher, and terrain classification is more accurate Really.The polarization interference classification when polarization isolation is less than 0.5dB less than -30dB, unbalance in phase less than 5 °, amplitude imbalance The degree of accuracy is in and receives scope.
Influence to the grader based on Shannon entropy to resolve：
Shannon entropys are the variational a kind of means for representing random scatter region, and expression random scatter region higher is more It is unstable.Polarization interference data are decomposed into three parts for representing different physical messages using shannon entropys, this three part difference It is energy information entropy SI[T], polarization information entropy Sp[T] and interference information entropy Sμ[T], pole is carried out using shannon entropy to resolve result Changing interference terrain classification can make full use of polarization interference information to obtain preferable classifying quality.
Such as following formula (17), (18)：
S [T]=SI[T]+Sp[T]+Sμ[T] (17)
T in formulaiiRepresent the covariance matrix of major-minor image, Ii, i=1,2 represents the major-minor image of polarization interference image Polarization scattering general power.
Such as following formula (19)：
P in formulai, i=1,2 represent the Barakat polarizabilities of collision matrix ,-∞ ＜ Sp[T]≤0 represents that polarization is right The contribution of Shannon entropys, low Sp[T] value correspond to certainty scattering mechanism, S highp[T] value correspondence random scatter region.
Such as following formula (20)：
μ in formulai, i=1,2,3 is polarization interference amplitude optimized coherence coefficient [15] ,-∞ ＜ Sμ[T]≤0 represents interference point Measure the contribution to entropy, low Sμ[T] value correspondence coherent scattering region, S highμ[T] value correspondence incoherent scattering region.
The presence of polarization interference systematic error can influence the classify degree of accuracy, the scattering strength portion influenceed by polarization interference error Point entropy isI in formula1 dAnd I2 dRepresent respectively and receive pole Change the scattering general power of the major-minor image under the influence of interference error, Tii d, i=1,2 represent by polarization interference error influenceed it is major-minor The covariance matrix of image, then Ii dIt is written as Ii d=tr (Tii d)=tr (ZTiiZH), i=1,2, contrast, missed by polarization interference Difference influence, the error to the entropy production of scattering strength part is such as following formula (21)：
The polarized portions entropy influenceed by polarization interference error is , polarization interference error is such as following formula (22) to the error of polarized portions entropy production：
The interference portion entropy influenceed by polarization interference error is Sμ[Td]=log { (1- (μ1 d)2)(1-(μ2 d)2)(1-(μ3 d )2), μ in formulai d, i=1,2,3 represent the three kinds of optimal scattering coefficients of amplitude received under the influence of polarization interference error, then polarization interference Error does not interfere with three kinds of amplitudes of the optimal scattering mechanism of amplitude coherent, so such as following formula (23)：
Sμ[Td]=Sμ[T] (23)
In sum, polarization interference error can influence energy information entropy and polarization information entropy, but on not influenceing interference information Entropy.It is same to be existed using E-SAR in order to quantitatively analyze influence of the polarization interference error to the sorting technique based on Shannon entropys The L-band polarization interference data that German Oberpfafenhoffen areas obtain are tested, the classification degree of accuracy for obtaining such as table 2 It is shown.
The relative classification degree of accuracy of sorting technique of the Oberpfafenhoffen of table 2 areas based on Shannon entropys
Learnt by the result of table 2, experimental result meets theory analysis, and polarization interference error is to interference portion entropy substantially without shadow Ring.Learnt by the result of table 2, the uneven influence to polarization interference terrain classification accuracy rate of surface channel is most notable, and phase is uneven The influence weighed to the degree of accuracy of classifying is relatively small.Generally, polarisation error is smaller, and terrain classification is more accurate.When polarization every The degree of accuracy of polarization interference classification is higher than when being less than 5 ° less than 0.5dB, unbalance in phase less than -30dB, channel imbalance from degree 90%.
Particular embodiments described above, technical problem, technical scheme and beneficial effect to solution of the invention are carried out Further describe, should be understood that and the foregoing is only specific embodiment of the invention, be not limited to The present invention, all any modification, equivalent substitution and improvements within the spirit and principles in the present invention, done etc., should be included in this Within the protection domain of invention.

Claims (7)

1. a kind of polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application, it is characterised in that its step Mainly include：
Step one, sets up the distortion model of polarimetric SAR interferometry ground object target measurement in the presence of polarimetric calibration error；
Step 2, analyzes influence of the polarimetric calibration error to polarimetric SAR interferometry key system perameter under the model；
Step 3, for the Error Propagation Model of terrain classification polarimetric calibration error；
Step 4, polarimetric SAR interferometry is provided to two kinds of influence cases of typical classification algorithm based on Error Propagation Model, final to obtain To the polarimetric calibration parameter requirements based on classification application.
2. the polarimetric calibration parameter requirements analysis method of polarimetric SAR interferometry classification application is based on as claimed in claim 1, and it is special Levy and be, the step one includes：
By polarimetric SAR interferometry system acquisition data procedures and polarization Coherence optimization essence, providing polarimetric calibration error includes leading to Intertrack crosstalk, disequilibrium are the main error source of polarimetric SAR interferometry system application, and then obtain the measurement distortion of ground object target Model；
Using the model, the polarization scattering matrix vector quantization of target is determined after obtaining Scattering of Vector by being calculated polarization Target polarization covariance matrix and polarization coherence matrix in the presence of mark error.
3. the polarimetric calibration parameter requirements analysis method of polarimetric SAR interferometry classification application is based on as claimed in claim 2, and it is special Levy and be, the target polarization covariance matrix in the presence of the polarimetric calibration error and the Main Analysis polarization of polarization coherence matrix are missed The difference influence optimal to polarization interference double scattering mechanism amplitude coherent.
4. the polarimetric calibration parameter requirements analysis method of polarimetric SAR interferometry classification application is based on as claimed in claim 1, and it is special Levy and be, the method for the step 3 including terrain classification, the influence to polarization interference Wishart graders, to polarization interference The influence that class merges.
5. the polarimetric calibration parameter requirements analysis method of polarimetric SAR interferometry classification application is based on as claimed in claim 4, and it is special Levy and be, the method for the terrain classification comprises the following steps：
Step 11, obtains polarization interference image, and polarimetric SAR interferometry data are carried out with what phase separation immunoassay and coherence matrix were estimated Early stage treatment；
Filtered data are entered line parameter calculating by step 12, and preliminary classification is carried out using the parameter for calculating；
Step 13, the average polarization interference matrix of each class is calculated according to preliminary classification result.
6. the polarimetric calibration parameter requirements analysis method of polarimetric SAR interferometry classification application is based on as claimed in claim 1, and it is special Levy and be, the step 4 includes the method, right that polarization interference data are classifiedThe influence of grader, Influence to the grader based on Shannon entropy to resolve.
7. the polarimetric calibration parameter requirements analysis method of polarimetric SAR interferometry classification application is based on as claimed in claim 6, and it is special Levy and be, the method that the polarization interference data are classified is a kind of to be combined with coherence using polarization decomposing method Terrain classification method, typical algorithm is sorting technique；Another kind is by polarization interference Wishart graders and polarization interference point The sorting technique that solution method is combined, typical algorithm is carried out with polarization interference Wishart graders using Shannon entropy to resolve The method of terrain classification.
CN201611039349.1A 2016-11-21 2016-11-21 Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application Pending CN106772371A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611039349.1A CN106772371A (en) 2016-11-21 2016-11-21 Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611039349.1A CN106772371A (en) 2016-11-21 2016-11-21 Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application

Publications (1)

Publication Number Publication Date
CN106772371A true CN106772371A (en) 2017-05-31

Family

ID=58974953

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611039349.1A Pending CN106772371A (en) 2016-11-21 2016-11-21 Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application

Country Status (1)

Country Link
CN (1) CN106772371A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108398666A (en) * 2018-02-09 2018-08-14 复旦大学 The polar system Parameters design of satellite-borne synthetic aperture radar
CN109633576A (en) * 2018-10-30 2019-04-16 上海无线电设备研究所 A kind of polarization SAR high accuracy Scaling algorithm under channel noise
CN110988879A (en) * 2019-12-24 2020-04-10 中南大学 Vegetation parameter inversion method, terminal equipment and storage medium
CN111458689A (en) * 2020-04-14 2020-07-28 上海无线电设备研究所 Multipath scattering characteristic classification method based on polarization scattering center
CN117214843A (en) * 2023-11-07 2023-12-12 中国科学院空天信息创新研究院 Full-polarization SAR system performance evaluation method based on scaler group

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110175771A1 (en) * 2007-05-08 2011-07-21 Raney Russell K Synthetic Aperture Radar Hybrid-Quadrature-Polarity Method and Architecture for Obtaining the Stokes Parameters of Radar Backscatter
CN102253377A (en) * 2011-04-22 2011-11-23 哈尔滨工业大学 Target detection method for polarimetric interferometry synthetic aperture radar on basis of eigenvalue analysis
CN104181530A (en) * 2014-07-21 2014-12-03 中国科学院电子学研究所 Determination method and device for polarization isolation index
CN104463207A (en) * 2014-12-05 2015-03-25 西安电子科技大学 Knowledge self-encoding network and polarization SAR image terrain classification method thereof
US9287891B1 (en) * 2015-01-20 2016-03-15 Mediatek Inc. Successive approximation register analog to digital converters
CN105608465A (en) * 2015-12-15 2016-05-25 西安电子科技大学 Fully polarimetric SAR (Synthetic Aperture Radar) coastline terrain classification method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110175771A1 (en) * 2007-05-08 2011-07-21 Raney Russell K Synthetic Aperture Radar Hybrid-Quadrature-Polarity Method and Architecture for Obtaining the Stokes Parameters of Radar Backscatter
CN102253377A (en) * 2011-04-22 2011-11-23 哈尔滨工业大学 Target detection method for polarimetric interferometry synthetic aperture radar on basis of eigenvalue analysis
CN102253377B (en) * 2011-04-22 2012-11-21 哈尔滨工业大学 Target detection method for polarimetric interferometry synthetic aperture radar on basis of eigenvalue analysis
CN104181530A (en) * 2014-07-21 2014-12-03 中国科学院电子学研究所 Determination method and device for polarization isolation index
CN104463207A (en) * 2014-12-05 2015-03-25 西安电子科技大学 Knowledge self-encoding network and polarization SAR image terrain classification method thereof
US9287891B1 (en) * 2015-01-20 2016-03-15 Mediatek Inc. Successive approximation register analog to digital converters
CN105608465A (en) * 2015-12-15 2016-05-25 西安电子科技大学 Fully polarimetric SAR (Synthetic Aperture Radar) coastline terrain classification method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
张林涛: "PollnSAR极化误差对最优相干相位的影响研究", 《电子与信息学报》 *
李昊: "基于图像分类的极化SAR系统的极化误差分析", 《舰船电子对抗》 *
马晓鹏: "极化SAR定标算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108398666A (en) * 2018-02-09 2018-08-14 复旦大学 The polar system Parameters design of satellite-borne synthetic aperture radar
CN109633576A (en) * 2018-10-30 2019-04-16 上海无线电设备研究所 A kind of polarization SAR high accuracy Scaling algorithm under channel noise
CN109633576B (en) * 2018-10-30 2020-09-08 上海无线电设备研究所 Polarized SAR high-precision calibration algorithm under channel noise
CN110988879A (en) * 2019-12-24 2020-04-10 中南大学 Vegetation parameter inversion method, terminal equipment and storage medium
CN111458689A (en) * 2020-04-14 2020-07-28 上海无线电设备研究所 Multipath scattering characteristic classification method based on polarization scattering center
CN111458689B (en) * 2020-04-14 2022-07-05 上海无线电设备研究所 Multipath scattering characteristic classification method based on polarization scattering center
CN117214843A (en) * 2023-11-07 2023-12-12 中国科学院空天信息创新研究院 Full-polarization SAR system performance evaluation method based on scaler group

Similar Documents

Publication Publication Date Title
CN106772371A (en) Polarimetric calibration parameter requirements analysis method based on polarimetric SAR interferometry classification application
Zhang et al. Fully polarimetric SAR image classification via sparse representation and polarimetric features
Gemba et al. Multi-frequency sparse Bayesian learning for robust matched field processing
CN106526530B (en) 2-L type array arrival direction estimation algorithm based on propagation operator
Jacobs et al. Automatic target recognition using sequences of high resolution radar range-profiles
Ammari et al. Multistatic imaging of extended targets
US10107904B2 (en) Method and apparatus for mapping and characterizing sea ice from airborne simultaneous dual frequency interferometric synthetic aperture radar (IFSAR) measurements
Maghsoudi et al. Polarimetric classification of Boreal forest using nonparametric feature selection and multiple classifiers
CN108133232A (en) A kind of Radar High Range Resolution target identification method based on statistics dictionary learning
CN104036289A (en) Hyperspectral image classification method based on spatial and spectral features and sparse representation
CN101464956A (en) Method for classifying polarization synthetic aperture radar image based on sub-aperture analysis
CN110389316B (en) Near-field and far-field mixed information source positioning method based on two-step matrix difference
CN103824093A (en) SAR (Synthetic Aperture Radar) image target characteristic extraction and identification method based on KFDA (Kernel Fisher Discriminant Analysis) and SVM (Support Vector Machine)
CN106501765A (en) A kind of Maximum Likelihood DOA Estimation based on quadratic sum and Semidefinite Programming
CN110412499A (en) Broadband DOA Estimation algorithm based on the RSS algorithm under compressive sensing theory
Anderson et al. Analysis of ASCAT ocean backscatter measurement noise
Tao et al. Joint time–frequency mask and convolutional neural network for real-time separation of multipath in GNSS deformation monitoring
Liang et al. A data fusion orientation algorithm based on the weighted histogram statistics for vector hydrophone vertical array
Hou et al. A unified framework for comparing the classification performance between quad-, compact-, and dual-polarimetric SARs
Shengqi et al. Full-polarization HRRP recognition based on joint sparse representation
CN102183755A (en) Novel high-resolution orientation-estimating method based on Cauchy Gaussian model
CN114926696A (en) Remote sensing image classification method based on complex matrix and multi-feature collaborative learning
Li et al. Multiple target three-dimensional coordinate estimation for bistatic MIMO radar with uniform linear receive array
Ferro-Famil et al. SAR imaging using coherent modes of diversity: SAR polarimetry, interferometry and tomography
Guo et al. Channel imbalance calibration based on the zero helix of Bragg-like targets

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20170531