CN104376539A - Method and device for decomposing objective scattering ingredients of polarized SAR (synthetic aperture radar) - Google Patents

Abstract

An embodiment of the invention discloses a method and a device for decomposing objective scattering ingredients of a polarized SAR (synthetic aperture radar). The method comprises the following steps of selecting a volume scattering model based on polarized SAR image data to be processed; extracting the maximum volume scattering power of the polarized SAR image data to be processed according to the polarized SAR image data to be processed and the volume scattering model; acquiring residual polarization coherence matrix of the polarized SAR image data to be processed according to the polarized SAR image data to be processed, the volume scattering model and the maximum volume scattering power value, and performing characteristic decomposition on the residual polarization coherence matrix to obtain the characteristic value and the characteristic vector of the residual polarization coherence matrix; calculating an objective scattering mechanism decision value according to elements of the characteristic vector; and determining even-order scattering and/or surface scattering ingredients of the polarized SAR image data to be processed according to the objective scattering mechanism decision value, and determining power values corresponding to the even-order scattering and/or surface scattering ingredients.

Description

A kind of decomposition method of polarization SAR target scattering composition and device

Technical field

The present invention relates to signal processing technology, particularly relate to decomposition method and the device of a kind of polarimetric synthetic aperture radar (SAR, Synthetic Aperture Radar) target scattering composition.

Background technology

Polarization SAR utilizes the combination of different polar transmitter and polarization reception antenna, obtains the Polarimetric SAR Image information of target.Polarimetric SAR Image information can provide a lot of terrestrial object informations, as surfaceness, symmetry, directionality etc.At present, conventional polarization SAR information extracting method adopts non-negative Eigenvalues Decomposition algorithm to realize usually.

But under non-negative Eigenvalues Decomposition algorithm is only applicable to reflective symmetry condition, under non-reflective symmetric condition, this algorithm body scattering composition crosses estimation and the low problem of the accuracy of goal decomposition result.

Summary of the invention

For solving the problems of the technologies described above, the embodiment of the present invention expects the decomposition method and the device that provide a kind of polarization SAR target scattering composition, can reduce volume scattering composition, improves the accuracy of goal decomposition result.

Technical scheme of the present invention is achieved in that

First aspect, embodiments provides a kind of decomposition method of polarization SAR target scattering composition, comprising:

Based on the Feature Selection volume scattering model T of pending polarimetric SAR image data <T> _v;

According to described pending polarimetric SAR image data <T> and described volume scattering model T _vextract the largest body scattering power P of described <T> _α; Wherein, described largest body scattering power P _αfor the performance number that the volume scattering composition of described pending polarimetric SAR image data <T> is corresponding;

According to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αobtain the residual polarization coherence matrix T of described <T> _remainder, and to described residual polarization coherence matrix T _remaindercarry out feature decomposition, obtain described residual polarization coherence matrix T _remaindereigenwert and proper vector;

Target scattering mechanism decision content η is calculated according to the element of described proper vector _i;

According to described target scattering mechanism decision content η _ijudge even scattering and/or the surface scattering composition of described pending polarimetric SAR image data <T>, and determine the performance number corresponding to described even scattering and/or surface scattering composition.

Further, described according to described pending polarimetric SAR image data <T> and described volume scattering model T _vextract the largest body scattering power P obtaining described <T> _α, comprising:

According to described pending polarimetric SAR image data <T> and described volume scattering model T _vthe largest body scattering power P of the <T> that acquisition makes formula (1) set up _α:

|<T>-P _αT _v|＝0 (1)

Wherein, || for determinant of a matrix calculates symbol.

Further, described according to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αobtain residual polarization coherence matrix T _remainder, comprising:

According to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αdescribed residual polarization coherence matrix T is obtained according to formula (2) _remainder:

T _remainder＝<T>-P _αT _v(2)。

Further, the described element according to described proper vector calculates target scattering mechanism decision content η _icomprise:

According to i-th proper vector element and formula (3) calculate target scattering mechanism decision content η corresponding to proper vector _i:

${\mathrm{\&eta;}}_i=\frac{\left|k_{i1}\right|}{\sqrt{\underset{j=2}{\overset{N}{\mathrm{\&Sigma;}}}{\left|k_{\mathrm{ij}}\right|}^2}}---\left(3\right).$

Wherein, j is described i-th proper vector element numbers, N is described i-th proper vector element number, k _ijrepresent i-th proper vector a jth element.

Further, described according to described target scattering mechanism decision content η _ijudge even scattering and the surface scattering composition of described pending polarimetric SAR image data <T>, and determine the performance number corresponding with surface scattering composition to described even scattering, can comprise:

As described η _iduring >1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is surface scattering, and correspondingly, the scattering composition performance number of described surface scattering is and described proper vector corresponding eigenvalue λ _i;

As described η _iduring <1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is even scattering, and correspondingly, the scattering composition performance number of described even scattering is and described proper vector corresponding eigenvalue λ _i.

Second aspect, embodiments provides a kind of decomposer of polarimetric synthetic aperture radar SAR target scattering composition, comprising: choose module, extraction module, acquisition module, feature decomposition module, computing module, determination module, wherein:

Describedly choose module, for the Feature Selection volume scattering model T based on pending polarimetric SAR image data <T> _v;

Described extraction module, for according to described pending polarimetric SAR image data <T> and the described volume scattering model T choosing module and choose _vextract the largest body scattering power P of described <T> _α; Wherein, described largest body scattering power P _αfor the performance number that the volume scattering composition of described pending polarimetric SAR image data <T> is corresponding;

Described acquisition module, for according to described pending polarimetric SAR image data <T>, described in choose the volume scattering model T that module chooses _vand the largest body scattering power value P that described extraction module extracts _αobtain the residual polarization coherence matrix T of described <T> _remainder;

Described feature decomposition module, for the residual polarization coherence matrix T obtained described acquisition module _remaindercarry out feature decomposition, obtain described residual polarization coherence matrix T _remaindereigenwert and proper vector;

Described computing module, the element for the proper vector obtained according to described feature decomposition module calculates target scattering mechanism decision content η _i;

Described determination module, for the target scattering mechanism decision content η calculated according to described computing module _ijudge even scattering and/or the surface scattering composition of described pending polarimetric SAR image data <T>, and determine the performance number corresponding to described even scattering and/or surface scattering composition.

Further, described extraction module, also for according to described pending polarimetric SAR image data <T> and the described volume scattering model T choosing module and choose _vthe largest body scattering power P of the <T> that acquisition makes formula (4) set up _α:

|<T>-P _αT _v|＝0 (4)

Wherein, || for determinant of a matrix calculates symbol.

Further, described acquisition module, also for according to described pending polarimetric SAR image data <T>, described in choose the volume scattering model T that module chooses _vand the largest body scattering power value P that described extraction module extracts _αdescribed residual polarization coherence matrix T is obtained according to formula (5) _remainder:

T _remainder＝<T>-P _αT _v(5)。

Further, described computing module, i-th proper vector also for obtaining according to described feature decomposition module element and formula (6) calculate target scattering mechanism decision content η corresponding to proper vector _i:

${\mathrm{\&eta;}}_i=\frac{\left|k_{i1}\right|}{\sqrt{\underset{j=2}{\overset{N}{\mathrm{\&Sigma;}}}{\left|k_{\mathrm{ij}}\right|}^2}}---\left(6\right).$

Wherein, j is described i-th proper vector element numbers, N is described i-th proper vector element number, k _ijrepresent i-th proper vector a jth element.

Further, described determination module, also for:

As the η that described computing module calculates _iduring >1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is surface scattering, and correspondingly, the scattering composition performance number of described surface scattering is and described proper vector corresponding eigenvalue λ _i;

As the η that described computing module calculates _iduring <1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is even scattering, and correspondingly, the scattering composition performance number of described even scattering is and described proper vector corresponding eigenvalue λ _i.

Embodiments provide a kind of decomposition method and device of polarization SAR target scattering composition, after volume scattering composition is extracted to pending Polarimetric SAR Image, the residual polarization coherence matrix obtained is carried out treatment and analysis further, thus the surface scattering obtained in pending Polarimetric SAR Image and/or even scattering composition, can avoid occurring that the phenomenon of estimation is crossed in volume scattering under non-reflective symmetric condition, the scope of application of expansion non-negative Eigenvalues Decomposition algorithm, improves the accuracy of goal decomposition result.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the decomposition method of polarization SAR target scattering composition in the embodiment of the present invention;

Fig. 2 is Google earth image corresponding to the pending Polarimetric SAR Image that provides in the embodiment of the present invention;

Fig. 3 be in the embodiment of the present invention for the process flow diagram of the decomposition method of the volume scattering model polarization SAR target scattering composition that is third-order matrix;

Fig. 4 is the False color image figure of the decomposition result of under reflective symmetry condition, Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention;

Fig. 5 is the False color image figure of the decomposition result of under non-reflective symmetric condition, Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention;

Fig. 6 is the False color image figure of the decomposition result of under reflective symmetry condition, the topography of Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention;

Fig. 7 is the False color image figure of the decomposition result of under non-reflective symmetric condition, the topography of Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention;

Fig. 8 is the structural drawing of the decomposer of polarization SAR target scattering composition in the embodiment of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described.

It should be noted that, the scattering of occurring in nature comprises surface scattering, even scattering and volume scattering.Wherein surface scattering is mainly from the reflection of rough surface, and the atural object surface major part as occurring in nature all belongs to this situation; Even scattering is mainly from trunk and ground, scattering between buildings and ground; Volume scattering is synthesized towards dipole scattering at random by a series of, such as the vegetation area of a large amount of leaf composition.And the basic thought of the embodiment of the present invention is: after extracting volume scattering composition to pending Polarimetric SAR Image, the residual polarization coherence matrix obtained is carried out treatment and analysis further, thus the surface scattering obtained in pending Polarimetric SAR Image and/or even scattering composition, the technical scheme that the embodiment of the present invention is provided can be avoided occurring that volume scattering composition crosses the problem of estimation.

Fig. 1 is the process flow diagram of the decomposition method of polarization SAR target scattering composition in the embodiment of the present invention, and shown in figure 1, the method comprises:

S101: based on the Feature Selection volume scattering model T of pending polarimetric SAR image data <T> _v;

It should be noted that, select suitable volume scattering model can improve the accuracy of goal decomposition result according to the image characters of ground object in pending polarimetric SAR image data <T>.

Pending polarimetric SAR image data <T> is made up of pixel, and each pixel can be considered as a third-order matrix, and correspondingly, volume scattering model is third-order matrix too.Understandably, in embodiments of the present invention, step in S102-S105 is all that volume scattering model does computing based on each pixel, when processing all pixels of pending polarimetric SAR image data <T>, the correlation parameter of <T> can be obtained.Be briefly described concrete technical scheme in order to clear, the single pixel for pending polarimetric SAR image data <T> in the embodiment of the present invention is described.

Particularly, if the pending Polarimetric SAR Image that pending Polarimetric SAR Image is artificial atural object takes as the leading factor, volume scattering model T _vcan be chosen for and simplify volume scattering model, specifically such as formula shown in (1):

$T_v=\left[\begin{array}{ccc}0.5& 0& 0\\ 0& 0.25& 0\\ 0& 0& 0.25\end{array}\right]---\left(1\right)$

If the type of ground objects that pending Polarimetric SAR Image comprises is a lot, the versatility of consideration method, volume scattering model T _vgeneral volume scattering model can be chosen for, specifically such as formula shown in (2):

$T_v\left(\mathrm{\&theta;}\right)=T_{\mathrm{\&alpha;}}+\frac{2n}{n+1}{T}_{\mathrm{\&beta;}}\left(2\mathrm{\&theta;}\right)+\frac{n(n-1)}{(n+1)(n+2)}{T}_{\mathrm{\&gamma;}}\left(4\mathrm{\&theta;}\right)---\left(2\right)$

Wherein, parameter θ is random parameters, and n is mean obliquity, T _α,t _β(2 θ) and T _γ(4 θ) is such as formula shown in (3):

$\left\{\begin{array}{c}{T}_{\mathrm{\&alpha;}}=\left[\begin{array}{ccc}0.5& 0& 0\\ 0& 0.25& 0\\ 0& 0& 0.25\end{array}\right]\\ T_{\mathrm{\&beta;}}\left(2\mathrm{\&theta;}\right)=\frac{1}{4}\left[\begin{array}{ccc}0& -\mathrm{co\; s}& \sin \left(2\mathrm{\&theta;}\right)\\ -\cos \left(2\mathrm{\&theta;}\right)& 0& 0\\ \sin \left(2\mathrm{\&theta;}\right)& 0& 0\end{array}\right]\\ T_{\mathrm{\&gamma;}}\left(4\mathrm{\&theta;}\right)=\frac{1}{4}\left[\begin{array}{ccc}0& 0& 0\\ 0& \cos \left(4\mathrm{\&theta;}\right)& -\sin \left(4\mathrm{\&theta;}\right)\\ 0& -\sin \left(4\mathrm{\&theta;}\right)& -\cos \left(4\mathrm{\&theta;}\right)\end{array}\right]\end{array}\right.---\left(3\right)$

S102: according to pending polarimetric SAR image data <T> and volume scattering model T _vextract the largest body scattering power P of <T> _α;

Wherein, largest body scattering power P _αfor the performance number that the volume scattering composition of pending polarimetric SAR image data <T> is corresponding.

Exemplarily, step S102 specifically can comprise:

According to pending polarimetric SAR image data <T> and volume scattering model T _vthe largest body scattering power P of the <T> that acquisition makes formula (4) set up _α;

|<T>-P _αT _v|＝0 (4)

Wherein, symbol || the determinant computation symbol of representing matrix.

S103: according to pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αobtain the residual polarization coherence matrix T of <T> _remainder, and to residual polarization coherence matrix T _remaindercarry out feature decomposition, obtain residual polarization coherence matrix T _remaindereigenwert and corresponding proper vector;

Exemplarily, according to pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αobtain residual polarization coherence matrix T _remainder, specifically can comprise:

According to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αdescribed residual polarization coherence matrix T is obtained according to formula (5) _remainder:

T _remainder＝<T>-P _αT _v(5)

Wherein, T _remainderrepresent the residual polarization coherence matrix after pending polarimetric SAR image data removing body scattering composition.

Exemplarily, to residual polarization coherence matrix T _remaindercarry out feature decomposition, obtain eigenwert and corresponding proper vector, specifically can be expressed as formula (6):

Wherein, M represents T _remaindereigenwert quantity, H represents conjugate transpose operation; λ _irepresent T _remainderi-th eigenwert, represent i-th eigenwert characteristic of correspondence vector.

Understandably, the character according to proper value of matrix: the value of matrix determinant is the product of all eigenwerts, therefore T _remainderthe value of determinant is can obtain according to formula (4) and formula (5) again | T _remainder|=| <T>-P _at _v|=0, therefore, T _remaindereigenwert in have at least an eigenwert to be zero.

S104: calculate target scattering mechanism decision content η according to the element of proper vector _i;

Wherein, target scattering mechanism decision content η _ifor as the important evidence judging target scattering mechanistic class.

Exemplarily, step S104 specifically can comprise:

According to i-th proper vector element and formula (7) calculate target scattering mechanism decision content η corresponding to proper vector _i:

${\mathrm{\&eta;}}_i=\frac{\left|k_{i1}\right|}{\sqrt{\underset{j=2}{\overset{N}{\mathrm{\&Sigma;}}}{\left|k_{\mathrm{ij}}\right|}^2}}---\left(7\right)$

Wherein, j is described i-th proper vector element numbers, N is described i-th proper vector element number.

It should be noted that, the eigenwert of the matrix obtained in Polarization target decomposition is corresponding with each scattering composition, and the ability value of each scattering composition is all non-negative, and the eigenwert therefore adopting Polarization target decomposition method to obtain also is non-negative.

S105: according to target scattering mechanism decision content η _ijudge even scattering and/or the surface scattering composition of pending polarimetric SAR image data <T>, and determine the performance number corresponding to even scattering and/or surface scattering composition;

Exemplarily, according to target scattering mechanism decision content η _ijudge even scattering and/or the surface scattering composition of pending polarimetric SAR image data <T>, and determine the performance number corresponding to even scattering and/or surface scattering composition, can comprise:

Work as η _iduring >1, determine in pending polarimetric SAR image data <T> with proper vector correspondence really qualitative objective is surface scattering, and correspondingly, the scattering composition performance number of surface scattering is and proper vector corresponding eigenvalue λ _i;

Work as η _iduring <1, determine in pending polarimetric SAR image data <T> with proper vector correspondence really qualitative objective is even scattering, and correspondingly, the scattering composition performance number of even scattering is and proper vector corresponding eigenvalue λ _i.

In conjunction with the embodiment shown in earlier figures 1, next one that provides for the embodiment of the present invention shown in third-order matrix and Fig. 2 for volume scattering model pending Polarimetric SAR Image is described in detail, as shown in Figure 2, the atural object that this Polarimetric SAR Image mainly comprises has forest, city, farmland and airport etc.Therefore, the decomposition method detailed process of carrying out polarization SAR target scattering composition as shown in Figure 3, can comprise:

S301: choosing general volume scattering model is volume scattering model T _v;

S302: according to pending polarimetric SAR image data <T> and volume scattering model T _vacquisition makes | <T>-P _αt _v| the largest body scattering power P of=0 <T> set up _α;

S303: according to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αaccording to T _remainder=<T>-P _αt _vobtain described residual polarization coherence matrix T _remainder, and to residual polarization coherence matrix T _remaindercarry out feature decomposition, obtain eigenwert and corresponding proper vector, specifically can be expressed as

Wherein, due to residual polarization coherence matrix T _remainderfor third-order matrix, therefore, λ ₁, λ ₂, λ ₃represent respectively residual polarization coherence matrix T _remaindercarry out three eigenwerts that feature decomposition obtains; represent and three eigenvalue λ respectively ₁, λ ₂, λ ₃corresponding proper vector.

It should be noted that, due to T _remaindereigenwert in have at least an eigenwert to be zero, for the ease of subsequent step analysis, set eigenvalue λ in embodiments of the present invention ₃equal zero, two other eigenvalue λ ₁, λ ₂be more than or equal to zero; Correspondingly two nonzero eigenvalue λ ₁, λ ₂corresponding proper vector be non-vanishing vector.

S304: the non-zero characteristics vector obtained according to S303 and calculate respectively with proper vector corresponding target scattering mechanism decision content η ₁and η ₂;

Wherein, the sequence number of i representation feature vector, k _i1, k _i2, k _i3for proper vector element.

S305: according to target scattering mechanism decision content η _ijudge even scattering and/or the surface scattering composition of pending polarimetric SAR image data <T>, and determine the performance number corresponding to even scattering and/or surface scattering composition.

Particularly, in the present embodiment, according to target scattering mechanism decision content η _ijudge that even scattering and/or the surface scattering composition of pending polarimetric SAR image data <T> can comprise following four kinds of situations:

The first, if η ₁>1 and η ₂<1, then residual polarization coherence matrix T _remaindercan be decomposed into surface scattering and even is scattering into a point sum, the performance number of the scattering composition of its surface scattering is λ ₁, the performance number λ of the scattering composition of even scattering ₂;

The second, if η ₂>1 and η ₁<1, then residual polarization coherence matrix T _remaindercan be decomposed into surface scattering and even is scattering into a point sum, the performance number of the scattering composition of its surface scattering is λ ₂, the performance number λ of the scattering composition of even scattering ₁;

The third, if η ₁>1 and η ₂>1, then residual polarization coherence matrix T _remaindercan be analyzed to two independently surface scatterings, total scattering power of its surface scattering is (λ ₁+ λ ₂);

4th kind, if η ₁<1 and η ₂<1, then residual polarization coherence matrix T _remaindercan be analyzed to two independently even scatterings, total scattering power of its even scattering is (λ ₁+ λ ₂).

Understandably, after by S301 to S305 scattering ingredient breakdown being carried out to each pixel in Fig. 2, the scattering composition of all targets in Fig. 2 can just be obtained.

Next, verify the decomposition method that the embodiment of the present invention proposes, in order to the surface scattering of polarimetric SAR image data more pending more intuitively, the size of the energy value of even scattering and volume scattering, can synthesize pseudo color image by these three components and represent.Wherein, the color assignment in False color image image and natural color adapt, and namely use blue presentation surface scattering, represent volume scattering by green, represent even scattering by redness.The change of size of surface scattering, even scattering and volume scattering can be found out according to the depth of color.

The False color image figure of Fig. 4 to be the False color image figure of the decomposition result of under reflective symmetry condition, Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention, Fig. 5 be decomposition result of under non-reflective symmetric condition, Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention.

Comparison diagram 4 and Fig. 5 can find, the green components in Fig. 5 is more shallow than the green components in Fig. 4, illustrates, after the decomposition method of the polarization SAR target scattering composition adopting the embodiment of the present invention to provide, the volume scattering power of all pixels reduces all to some extent.

Particularly, by under the volume scattering power sum of all pixels in the decomposition result of under reflective symmetry condition, Fig. 2 being carried out to polarization SAR target scattering composition and non-reflective symmetric condition, the volume scattering power sum that Fig. 2 carries out all pixels in the decomposition result of polarization SAR target scattering composition is compared and can be obtained: the latter reduces 6.15% than the former volume scattering power averaging.This illustrates, the decomposition method of the polarization SAR target scattering composition adopting the embodiment of the present invention to provide, can effectively suppress volume scattering to cross estimation under non-reflective symmetric condition.

The False color image figure of Fig. 6 to be the False color image figure of the decomposition result of under reflective symmetry condition, the topography of Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention, Fig. 7 be decomposition result of under non-reflective symmetric condition, the topography of Fig. 2 being carried out to polarization SAR target scattering composition in the embodiment of the present invention.

It should be noted that, image in red elliptic is highway, and the leading scattering mechanism of highway is the even scattering that the highway side of being erected by height and ground return are formed, therefore, the color of the False color image figure of highway should be red in theory.Comparison diagram 6 and Fig. 7 can find, Fig. 6 high speed highway is green, and highway is red in the figure 7, this shows, the goal decomposition result of the Polarization target decomposition method that the employing embodiment of the present invention in Fig. 7 provides under non-reflective symmetric condition meets the actual scattering signatures of atural object more, obtains the goal decomposition result that accuracy is higher.

In sum, the invention provides a kind of decomposition method and device of polarization SAR target scattering composition, after volume scattering composition is extracted to pending Polarimetric SAR Image, the residual polarization coherence matrix obtained is carried out treatment and analysis further, thus the surface scattering obtained in pending Polarimetric SAR Image and/or even scattering composition, can avoid occurring that the phenomenon of estimation is crossed in volume scattering under non-reflective symmetric condition, the scope of application of expansion non-negative Eigenvalues Decomposition algorithm, improves the accuracy of goal decomposition result.

The embodiment of the present invention also provides a kind of decomposer of polarization SAR target scattering composition, and this device can realize the decomposition method of above-mentioned polarization SAR target scattering composition.

Fig. 8 is the structural drawing of the decomposer of polarization SAR target scattering composition in the embodiment of the present invention, shown in figure 8, this device comprises: choose module 801, extraction module 802, acquisition module 803, feature decomposition module 804, computing module 805, determination module 806, wherein:

Choose module 801, for the Feature Selection volume scattering model T based on pending polarimetric SAR image data <T> _v;

Extraction module 802, for according to pending polarimetric SAR image data <T> with choose the volume scattering model T that module 801 chooses _vextract the largest body scattering power P of <T> _α; Wherein, largest body scattering power P _αfor the performance number that the volume scattering composition of pending polarimetric SAR image data <T> is corresponding;

Further, extraction module 802, also for according to pending polarimetric SAR image data <T> with choose the volume scattering model T that module 801 chooses _vacquisition makes formula <T>-P _αt _v| the largest body scattering power P of=0 <T> set up _α.

Acquisition module 803, for according to pending polarimetric SAR image data <T>, choose the volume scattering model T that module 801 chooses _vand the largest body scattering power value P that extraction module 802 extracts _αobtain the residual polarization coherence matrix T of <T> _remainder;

Further, acquisition module 803, also for according to pending polarimetric SAR image data <T>, choose the volume scattering model T that module 801 chooses _vand the largest body scattering power value P that extraction module 802 extracts _αaccording to formula T _remainder=<T>-P _αt _vobtain residual polarization coherence matrix T _remainder.

Feature decomposition module 804, for the residual polarization coherence matrix T obtained acquisition module 803 _remaindercarry out feature decomposition, obtain residual polarization coherence matrix T _remaindereigenwert and proper vector;

Computing module 805, the element for the proper vector obtained according to feature decomposition module 804 calculates target scattering mechanism decision content η _i;

Further, computing module 805, i-th proper vector also for obtaining according to feature decomposition module 804 element and formula calculate the target scattering mechanism decision content η that proper vector is corresponding _i.

Wherein, j is i-th proper vector element numbers, N is i-th proper vector element number, k _ijrepresent i-th proper vector a jth element.

Determination module 806, for the target scattering mechanism decision content η calculated according to computing module 805 _ijudge even scattering and/or the surface scattering composition of pending polarimetric SAR image data <T>, and determine the performance number corresponding to even scattering and/or surface scattering composition.

Further, determination module 806, also for:

As the η that computing module 805 calculates _iduring >1, determine in pending polarimetric SAR image data <T> with proper vector correspondence really qualitative objective is surface scattering, and correspondingly, the scattering composition performance number of surface scattering is and proper vector corresponding eigenvalue λ _i;

As the η that computing module 805 calculates _iduring <1, determine in pending polarimetric SAR image data <T> with proper vector correspondence really qualitative objective is even scattering, and correspondingly, the scattering composition performance number of even scattering is and proper vector corresponding eigenvalue λ _i.

Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt the form of hardware embodiment, software implementation or the embodiment in conjunction with software and hardware aspect.And the present invention can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory and optical memory etc.) of computer usable program code.

The present invention describes with reference to according to the process flow diagram of the method for the embodiment of the present invention, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can being provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, making the instruction performed by the processor of computing machine or other programmable data processing device produce device for realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

These computer program instructions also can be loaded in computing machine or other programmable data processing device, make on computing machine or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computing machine or other programmable devices is provided for the step realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

The above, be only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.

Claims (10)

1. a decomposition method for polarimetric synthetic aperture radar SAR target scattering composition, is characterized in that, comprising:

Based on the Feature Selection volume scattering model T of pending polarimetric SAR image data <T> _v;

According to described pending polarimetric SAR image data <T> and described volume scattering model T _vextract the largest body scattering power P of described <T> _α; Wherein, described largest body scattering power P _αfor the performance number that the volume scattering composition of described pending polarimetric SAR image data <T> is corresponding;

According to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αobtain the residual polarization coherence matrix T of described <T> _remainder, and to described residual polarization coherence matrix T _remaindercarry out feature decomposition, obtain described residual polarization coherence matrix T _remaindereigenwert and proper vector;

Target scattering mechanism decision content η is calculated according to the element of described proper vector _i;

According to described target scattering mechanism decision content η _ijudge even scattering and/or the surface scattering composition of described pending polarimetric SAR image data <T>, and determine the performance number corresponding to described even scattering and/or surface scattering composition.

2. decomposition method according to claim 1, is characterized in that, described according to described pending polarimetric SAR image data <T> and described volume scattering model T _vextract the largest body scattering power P obtaining described <T> _α, comprising:

According to described pending polarimetric SAR image data <T> and described volume scattering model T _vthe largest body scattering power P of the <T> that acquisition makes formula (1) set up _α:

|<T>-P _αT _v|＝0 (1)

Wherein, || for determinant of a matrix calculates symbol.

3. decomposition method according to claim 1, is characterized in that, described according to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αobtain residual polarization coherence matrix T _remainder, comprising:

According to described pending polarimetric SAR image data <T>, volume scattering model T _vand largest body scattering power value P _αdescribed residual polarization coherence matrix T is obtained according to formula (2) _remainder:

T _remainder＝<T>-P _αT _v(2)。

4. decomposition method according to claim 1, is characterized in that, the described element according to described proper vector calculates target scattering mechanism decision content η _icomprise:

According to i-th proper vector element and formula (3) calculate target scattering mechanism decision content η corresponding to proper vector _i:

${\mathrm{\&eta;}}_i=\frac{\left|k_{i1}\right|}{\sqrt{\underset{j=2}{\overset{N}{\mathrm{\&Sigma;}}}{\left|k_{\mathrm{ij}}\right|}^2}}---\left(3\right).$

Wherein, j is described i-th proper vector element numbers, N is described i-th proper vector element number, k _ijrepresent i-th proper vector a jth element.

5. decomposition method according to claim 1, is characterized in that, described according to described target scattering mechanism decision content η _ijudge even scattering and the surface scattering composition of described pending polarimetric SAR image data <T>, and determine the performance number corresponding with surface scattering composition to described even scattering, can comprise:

As described η _iduring >1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is surface scattering, and correspondingly, the scattering composition performance number of described surface scattering is and described proper vector corresponding eigenvalue λ _i;

As described η _iduring <1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is even scattering, and correspondingly, the scattering composition performance number of described even scattering is and described proper vector corresponding eigenvalue λ _i.

6. a decomposer for polarimetric synthetic aperture radar SAR target scattering composition, is characterized in that, described device comprises: choose module, extraction module, acquisition module, feature decomposition module, computing module, determination module, wherein:

Describedly choose module, for the Feature Selection volume scattering model T based on pending polarimetric SAR image data <T> _v;

Described extraction module, for according to described pending polarimetric SAR image data <T> and the described volume scattering model T choosing module and choose _vextract the largest body scattering power P of described <T> _α; Wherein, described largest body scattering power P _αfor the performance number that the volume scattering composition of described pending polarimetric SAR image data <T> is corresponding;

Described acquisition module, for according to described pending polarimetric SAR image data <T>, described in choose the volume scattering model T that module chooses _vand the largest body scattering power value P that described extraction module extracts _αobtain the residual polarization coherence matrix T of described <T> _remainder;

Described feature decomposition module, for the residual polarization coherence matrix T obtained described acquisition module _remaindercarry out feature decomposition, obtain described residual polarization coherence matrix T _remaindereigenwert and proper vector;

Described computing module, the element for the proper vector obtained according to described feature decomposition module calculates target scattering mechanism decision content η _i;

Described determination module, for the target scattering mechanism decision content η calculated according to described computing module _ijudge even scattering and/or the surface scattering composition of described pending polarimetric SAR image data <T>, and determine the performance number corresponding to described even scattering and/or surface scattering composition.

7. decomposer according to claim 6, is characterized in that, described extraction module, also for according to described pending polarimetric SAR image data <T> and the described volume scattering model T choosing module and choose _vthe largest body scattering power P of the <T> that acquisition makes formula (4) set up _α:

|<T>-P _αT _v|＝0 (4)

Wherein, || for determinant of a matrix calculates symbol.

8. decomposer according to claim 6, is characterized in that, described acquisition module, also for according to described pending polarimetric SAR image data <T>, described in choose the volume scattering model T that module chooses _vand the largest body scattering power value P that described extraction module extracts _αdescribed residual polarization coherence matrix T is obtained according to formula (5) _remainder:

T _remainder＝<T>-P _αT _v(5)。

9. decomposer according to claim 6, is characterized in that, described computing module, i-th proper vector also for obtaining according to described feature decomposition module element and formula (6) calculate target scattering mechanism decision content η corresponding to proper vector _i:

${\mathrm{\&eta;}}_i=\frac{\left|k_{i1}\right|}{\sqrt{\underset{j=2}{\overset{N}{\mathrm{\&Sigma;}}}{\left|k_{\mathrm{ij}}\right|}^2}}---\left(6\right).$

Wherein, j is described i-th proper vector element numbers, N is described i-th proper vector element number, k _ijrepresent i-th proper vector a jth element.

10. decomposer according to claim 6, is characterized in that, described determination module, also for:

As the η that described computing module calculates _iduring >1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is surface scattering, and correspondingly, the scattering composition performance number of described surface scattering is and described proper vector corresponding eigenvalue λ _i;

As the η that described computing module calculates _iduring <1, determine in described pending polarimetric SAR image data <T> with described proper vector correspondence really qualitative objective is even scattering, and correspondingly, the scattering composition performance number of described even scattering is and described proper vector corresponding eigenvalue λ _i.