CN109375189A - Polarimetric radar remote sensing images city goal decomposition method based on cross scatter model - Google Patents
Polarimetric radar remote sensing images city goal decomposition method based on cross scatter model Download PDFInfo
- Publication number
- CN109375189A CN109375189A CN201811594285.0A CN201811594285A CN109375189A CN 109375189 A CN109375189 A CN 109375189A CN 201811594285 A CN201811594285 A CN 201811594285A CN 109375189 A CN109375189 A CN 109375189A
- Authority
- CN
- China
- Prior art keywords
- scattering
- polarization
- model
- cross
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000010287 polarization Effects 0.000 claims abstract description 97
- 239000011159 matrix material Substances 0.000 claims abstract description 57
- 230000007246 mechanism Effects 0.000 claims description 24
- 230000001427 coherent effect Effects 0.000 claims description 19
- 238000009826 distribution Methods 0.000 claims description 7
- 239000004576 sand Substances 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 230000017105 transposition Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 6
- 238000010276 construction Methods 0.000 abstract description 2
- 101100348856 Aspergillus sp. (strain MF297-2) notF gene Proteins 0.000 description 2
- 238000005388 cross polarization Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
A kind of polarimetric radar remote sensing images city goal decomposition method based on cross scatter model, comprising: the step of PolSAR image haplopia polarization scattering matrix is converted into polarization covariance matrix or polarization coherence matrix；The step of PolSAR image pixel polarization orientation angle is calculated by polarization covariance matrix or polarization coherence matrix；By polarization orientation angle combination dihedral angle reflection come the step of constructing building cross scatter model；The step of solving equation according to surface scattering model, even scattering model, volume scattering model, spiral scattering model and cross scatter model construction Polarization target decomposition, and the step of solving each model scattering coefficient；The step of amendment obtains the scattering weighting coefficient of each model, carries out surface scattering power, even scattered power, volume scattering power, spiral scattered power and cross scatter power calculation using the scattering weighting coefficient of each model, completes goal decomposition.The present invention can effectively describe the HV scattering of building.
Description
Technical Field
The invention belongs to the field of remote sensing image interpretation, and relates to a method for decomposing urban targets of a polarized radar remote sensing image based on a cross scattering model, which is used for estimating target scattering mechanism components in the image and providing important characteristic information for target classification and identification.
Background
The radar polarization remote sensing image target decomposition is the sum of a plurality of basic scattering mechanisms with different physical meanings, which represents a received polarization scattering matrix, a covariance matrix or a coherent matrix, is an important means for describing and extracting the target characteristics of the radar remote sensing image, is also the basis of target detection and identification, image segmentation and classification and the like, and is a key technology in the field of radar polarization remote sensing image processing. In 1998, Freeman and Durden proposed a three-component polarization object decomposition method based on a scattering model. The method breaks down the target scattering mechanism into three types, surface scattering, dihedral scattering, and bulk scattering. Since the method is performed under the assumption of reflection symmetry, most urban targets are reflection asymmetric, and thus the method is not perfect in urban target decomposition. In consideration of the non-reflection symmetry of urban areas, Yamaguchi et al propose a four-component decomposition algorithm based on Freeman decomposition, i.e., the decomposition of target scattering into surface scattering, dihedral scattering, volume scattering, and helical scattering. Helical scattering has been shown to be effective in describing urban scattering and to distinguish urban targets from natural terrestrial objects to some extent. To further suppress the urban volume scattering power and improve the dihedral scattering power, Yamaguchi et al apply polarization azimuth compensation to the quartered component decomposition, enhancing the urban decomposition results. However, for buildings with larger polarization azimuth angles, the scattering type of the buildings in the decomposition result is still similar to the scattering type of natural ground objects such as forest vegetation.
At present, scholars generally think that forest vegetation and buildings which are not parallel to radar can both generate cross scattering, and energy is contributed to cross scattering items of a scattering matrix, so that how to distinguish the two different ground object scattering types is a hotspot of polarization decomposition research, and the research of urban cross scattering models is paid more and more attention by scholars. On the basis of Freeman three-component decomposition, Moriyama and the like establish a three-component decomposition method suitable for building decomposition, and the method divides urban backscattering into three scattering components of odd scattering, even scattering and cross scattering, and provides a cross scattering model aiming at urban scattering. The method can effectively obtain the cross scattering of the urban area so as to reduce the volume scattering estimation, but the volume scattering of the forest area is also underestimated at the same time, so that the scattering mechanism of the forest area is not very accurate. Until now, the construction of a cross scattering model of a building is still a difficult point, and how to effectively describe cross scattering of urban areas and simultaneously not influence the volume scattering of forest areas is a key problem.
Disclosure of Invention
The invention aims to provide a method for decomposing urban targets of polarized radar remote sensing images based on a cross scattering model, wherein the cross scattering model can effectively describe HV scattering components of buildings, so that the cross scattering model is effectively distinguished from HV scattering generated by forests, and the problem of scattering mechanism confusion caused by overestimation of bulk scattering power and underestimation of even scattering power in the prior art is solved.
In order to achieve the purpose, the technical scheme adopted by the invention comprises the following steps:
converting the single-view polarization scattering matrix of the PolSAR image into a polarization covariance matrix or a polarization coherent matrix;
calculating a polarization azimuth angle of the PolSAR image pixel by a polarization covariance matrix or a polarization coherent matrix;
a step of constructing a cross scattering model of the building by combining the polarization azimuth angle and the dihedral reflectors;
constructing a polarized target decomposition solving equation according to the surface scattering model, the even scattering model, the volume scattering model, the spiral scattering model and the cross scattering model, and solving scattering coefficients of the models;
and correcting to obtain the scattering weighting coefficients of the models, and calculating the surface scattering power, the even scattering power, the volume scattering power, the spiral scattering power and the cross scattering power by using the scattering weighting coefficients of the models to complete target decomposition.
The conversion steps of the polarization covariance matrix and the polarization coherent matrix are as follows:
the radar polarization scattering matrix is represented as:
in the formula, H and V represent horizontal polarization and vertical polarization, respectively, SPQ(P, Q ═ H, V) is the target backscatter coefficient of the scattering matrix when transmitting in Q polarization, and receiving in P polarization, fullS in the case of back scattering according to the reciprocity theoremHV＝SVH；
Pauli scattering vector is expressed as:
wherein, the upper labelTRepresenting a matrix transposition;
the polarization coherence matrix is represented as:
wherein, the upper labelRepresents the conjugate transpose of the matrix, the superscript denotes the complex conjugate;
the polarization covariance matrix is expressed as:
wherein,
the PolSAR image pixel polarization azimuth angle theta is estimated according to the following formula:
wherein, Re { T23Means T23Real part of, T22Characterizing the even-order scattering component, T, for the intermediate term of the polarization coherence matrix23The reflection asymmetry component is characterized for the polarization coherence matrix off-diagonal elements.
The distribution of the azimuthal polarization of a dihedral scattering structure is defined as:
wherein, thetadomIs the polarization azimuth of the building, the cross scattering coherent matrix model of the building<[T]>crossIs defined as:
wherein, Td(theta) is a coherent matrix scattering model of a standard dihedral reflector;
the above equation is further derived as:
the steps of constructing a polarized target decomposition solving equation are as follows:
(1) the expression of a multi-component polarization target decomposition equation based on a cross scattering model is listed as follows:
wherein f iss,fd,fv,fhAnd fcroWeighting coefficients for surface scattering, even scattering, volume scattering, helical scattering and cross scattering, respectively, [ T]s,[T]d,[T]v,[T]hAnd [ T]crossA standard coherent scattering model representing the corresponding scattering mechanism;
(2) expanding the equation in the step (1) to obtain the following equation system:
fs+fd|α|2+fv/2＝T11(a)
fs|β|2+fd+fv/4+fh/2+m22fcro＝T22(b)
fv/4+fh/2+m33fcro＝T33(c)
fsβ*+fdα＝T12(d)
fh/2＝|Im(T23)| (e)
wherein m is22And m33Second and third diagonal elements, Im (T), respectively, of the cross scatter model23) Is T23An imaginary part of (d);
first, the parameter f is obtained by equation (e)hAccording to the decision conditionFixing two unknown parameters fsAnd fd：
And further obtaining four equations containing four unknown variables, and solving to obtain all unknown parameters.
The system of equations in step (2) is simplified in the following manner, based on equations (b) and (c):
neglect (m)22-m33)fcroSolving all unknown parameters;
when in useSometimes:
when in useSometimes:
if f occurscroIf < 0, f is setcroAnd obtaining other unknown parameters according to a parameter solving method of four-component decomposition.
And calculating the scattering power of each model after decomposition according to the following formula by combining the scattering weighting coefficients of each model:
wherein, PcroIs the cross scatter power.
Compared with the prior art, the invention has the following beneficial effects: the method comprises the steps of firstly obtaining a polarization azimuth angle of PolSAR image pixels through calculation, and then providing a cross scattering model suitable for a building by combining a dihedral reflector, wherein the cross scattering model can effectively describe HV scattering of the building and is a scattering model suitable for a non-parallel radar azimuth building. Aiming at HV scattering generated by forests, a traditional volume scattering model can be used for describing, so that the cross scattering model solves the problem of confusion of scattering mechanisms of non-parallel radar buildings and forests. By applying the model of the invention to the multi-component polarized target decomposition, the polarized decomposition result can effectively describe the scattering mechanism of buildings, the urban scattering description accuracy is improved, and the target classification recognition result based on the scattering mechanism is more reasonable, effective and accurate.
Drawings
FIG. 1 is a flow chart of the operation of the object decomposition method of the present invention;
FIG. 2 shows various scattered power images corresponding to Radarsat-2C band data in the San Francisco region:
FIGS. (a) - (e) are the surface scattering, even scattering, volume scattering, helical scattering and cross scattering, respectively, obtained by the present invention;
FIGS. (f) - (h) are graphs synthesized from the decomposition results obtained by the method of the present invention, Y4R and MCSM, respectively;
FIG. 3 shows the percentage of scattering components in urban and forest regions selected from the results of polarization decomposition of Radarsat-2C band data: graphs (a) - (c) are the decomposition result and the scattering component percentage of the polarized target in the region A, graphs (e) - (g) are the decomposition result and the scattering component percentage of the polarized target in the region B, and graphs (d) and (h) are two regional optical images and the polarization azimuth angle distribution of the corresponding PolSAR images.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
Referring to fig. 1, the polarized SAR urban area target decomposition method based on the cross scattering model of the present invention comprises the following specific implementation steps:
step one, the radar polarization scattering matrix of each pixel in the PolSAR single vision complex data can be expressed asWherein H and V respectively represent a horizontal polePolarization and vertical polarization. SPQ(P, Q ═ H, V) is transmitted in Q polarization mode, the target backscattering coefficient of scattering matrix in P polarization mode receiving, S is present in backscattering case satisfying reciprocal theoremHV＝SVH。
Further, the Pauli scattering vector can be expressed as
Step two, scattering vector k by Pauli3PAnd calculating to obtain a polarization coherent matrix T of the pixel, wherein the calculation formula is as follows:
wherein, the upper labelRepresenting the conjugate transpose of the matrix, superscript denotes the complex conjugate,<·>is a set average;
step three, calculating a polarization azimuth angle theta according to the pixel polarization coherent matrix, wherein the calculation formula is as follows:
wherein, Re { T23Means T23The real part of (a);
deducing a building cross scattering coherent matrix model by using the dihedral angle reflectors and combining the polarization azimuth angles;
the method comprises the following specific steps:
4a) the distribution of the azimuthal polarization of a dihedral scattering structure is defined as:
wherein, thetadomThe building polarization azimuth angle can be calculated through the third step.
Cross scatter coherence matrix model for buildings<[T]>crossIs calculated by the formulaWherein T isd(θ) is a coherent matrix scattering model of the standard dihedral reflectors, and the above equation can be further calculated as:
4b) the cross scatter coherence matrix model is adaptive and closely related to the polarization azimuth of the building. When the polarization azimuth is zero, the model can be expressed asWhen the building polarization azimuth is 45 deg., the model becomesWhen the building polarization azimuth is 22.5 °, the model becomes
Therefore, the scattering model can obtain different cross scattering components according to different polarization azimuth angles of the building;
step five, building a multi-component polarization target decomposition equation on the basis of Yamaguchi four-component decomposition by combining a building cross scattering model; the method comprises the following specific steps:
5a) yamaguchi proposes that Four-component polarization object decomposition (Yamaguchi Four-component decomposition-Y4O) is performed on the basis of Freeman three-component object decomposition, taking human into considerationThe object of manufacture not satisfying the conditions of reflection symmetry, i.e.Description of scattering mechanism applicable to urban targets. The equation is<[C]>＝fs[C]s+fd[C]d+fv[C]v+fh[C]hWherein<[C]>In the form of a polarization covariance matrix,<·>for multi-view processing or spatial ensemble averaging. f. ofs,fd,fvAnd fhWeight coefficients for surface scattering, even scattering, volume scattering and helical scattering, [ C ], respectively]s,[C]d,[C]vAnd [ C]hThe standard covariance scattering model representing the corresponding scattering mechanism is:
and
wherein α and β represent the ratio of the scatterers in horizontal polarization to vertical polarization;
5b) on the basis of four-component polarized target decomposition, a cross scattering model-based multi-component polarized target decomposition expression is as follows
Wherein f iss,fd,fv,fhAnd fcroWeighting coefficients for surface scattering, even scattering, volume scattering, helical scattering and cross scattering, respectively, [ T]s,[T]d,[T]v,[T]hAnd [ T]crossThe standard coherent scattering model representing the corresponding scattering mechanism, the above equation uses a polarized coherent matrix.
Calculating weighting coefficients of surface scattering, even scattering, volume scattering, helical scattering and cross scattering according to a multi-component polarized target decomposition expression, and specifically comprising the following steps of:
6a) from the expansion of the polarized target decomposition expression, the following system of equations can be obtained:
fs+fd|α|2+fv/2＝T11(a)
fs|β|2+fd+fv/4+fh/2+m22fcro＝T22(b)
fv/4+fh/2+m33fcro＝T33(c)
fsβ*+fdα＝T12(d)
fh/2＝|Im(T23)| (e)
wherein m is22And m33Respectively represent the second and third diagonal elements, Im (T) in the cross scatter coherence matrix scattering model23) Is T23The imaginary part of (c). It can be seen that fhCan be directly obtained from the formula (e) of the equation set;
6b) when f is solvedhThe system then has four equations remaining, containing six unknowns. In accordance with the Yamaguchi solution of four-component decomposition method, the decision condition is also requiredTo fix two unknown parameters fsAnd fdNamely:
if it is notF is thens＝0
If it is notF is thend＝0
Four equations are obtained containing four unknown variables so that all unknown parameters can be solved.
6c) In order to make the result analytic expression more concise, the equation set needs to be simplified. From equations (b) and (c) we can obtain:
fd+(m22-m33)fcro＝T22-T33,
fs|β|2+(m22-m33)fcro＝T22-T33,
in the above two equations, because (m)22-m33)fcroIs much smaller than fdAnd fs|β|2And therefore may be omitted.
F can be obtained by combining the above formula with equation (d)dAnd fsα and β in combination with the above equations, all unknown parameters can be solved, i.e., whenSometimes:
when in useSometimes:
if f occurscroIf < 0, f can be setcroAnd obtaining other unknown parameters according to a parameter solving method of the original four-component decomposition.
Step seven, after obtaining all the weighting coefficients of the scattering mechanism, obtaining each decomposed scattering power according to the above calculation result, wherein the specific calculation formula is as follows:
Ps＝fs(1+|β|2),Pd＝fd(1+|α|2),Pv＝fv
Ph＝fh,Pcro＝fcro
wherein P iscroIs the cross scatter power.
The principle of the invention which adopts the polarized target decomposition based on the cross scattering model is as follows: aiming at the cross scattering characteristic that a building is different from forest body scattering and aiming at solving the problem of confusion of scattering mechanisms of the building and the forest in the non-parallel radar flight direction, a cross scattering coherent matrix model of the building is provided, the cross scattering model can better describe the HV scattering characteristic of the building and can be distinguished from the forest body scattering characteristic, and therefore the scattering mechanisms of different ground objects can be distinguished more accurately. After the building scattering model is improved, the obtained result is necessarily more reasonable, effective and accurate, and accords with the physical significance.
The decomposition method of the invention is verified through experiments: the PolSAR data adopted by the experimental object is San Francisco (San Francisco) regional full polarization data acquired by a satellite-borne Radarsat-2C wave band, the azimuth resolution is 4.82 meters, the distance direction is 4.73 meters, and the PolSAR data comprises various ground object types such as forests, urban buildings, oceans and the like.
The experimental procedure was as follows: the Radarsat-2PolSAR data is planned and decomposed by a cross-scattering-based multi-component polarization target decomposition method, and in addition, two typical urban building polarization target decomposition methods, namely a rotational polarization azimuth-based four-component decomposition (Y4R) and an original multi-component decomposition (MCSM), are used as comparison methods. The polarization covariance matrix was calculated using a 3 x 3 sliding window average on the polarization data. And simultaneously, building and forest typical areas are respectively selected from the three polarization decomposition results, the percentage of the power occupied by each scattering mechanism in each area is calculated, and the polarization azimuth angle distribution condition of each area is analyzed.
The experimental results and analysis are set forth below:
it can be seen from fig. 2(b) that the smaller the azimuthal angle of polarization of the building, the stronger the even scattering, and some buildings with larger azimuthal angles of polarization have almost zero even scattering. This phenomenon is exactly the opposite of the cross-scatter shown in fig. 2(e), and it can be seen that the larger the azimuthal angle of polarization of the building, the stronger the cross-scatter. Furthermore, it can also be seen from fig. 2(e) that although both the forest and the non-parallel radar flight direction buildings contribute to the cross scatter term of the coherence matrix, the cross scatter of the forest is very weak in fig. 2(e) because the proposed cross scatter model is building specific. The cross scattering similar to that of buildings in forest (e) occurs because of the penetration of radar waves, and the secondary or tertiary scattering occurs in forest, and the scattering mechanism is similar to that of building wall-ground-wall scattering. In addition, it can be seen that the buildings with larger polarization azimuth angles, such as the upper right building with the polarization azimuth angle of 45 °, have the strongest cross scattering, and are clearly distinguished from other ground objects, which verifies the effectiveness of the cross scattering model provided by the invention for the buildings, and shows that the model can effectively describe the cross scattering of the building area and can be distinguished from the body scattering of the forest. Fig. 2(d) shows the helical scatter component, which primarily describes the reflection asymmetry of the target. It can be seen that the artificial target, especially the building with a large polarization azimuth angle, has the strongest helical scattering, but the distinction between the building and the natural region is not obvious, and the helical scattering power of the natural region such as forest is also high, so that the model cannot effectively describe the cross scattering generated by the building which is not parallel to the radar.
Referring to fig. 2(f) - (h), the power synthesis of even order scattering, helical scattering, and cross scattering or line scattering (MCSM) of the present invention is characterized as urban scattering, displayed with brighter gray values; the volume scattering power is characterized as forest vegetation scattering and is displayed by adopting a medium gray value; the surface scattering power is mainly characterized by the scattering of flat natural objects such as water surface and bare land, and is characterized by dark gray. In fig. 2(f), it can be seen that due to the effect of cross scatter, buildings that are not parallel to the direction of radar flight appear noticeably bright, consistent with fig. 2(e), which is clearly distinguishable from natural areas such as forests. In fig. 2(g), the Y4R decomposition result cannot effectively extract cross-polarization scattering of buildings, and the buildings which are not parallel to the radar exhibit the same scattering mechanism as forests, which indicates that scattering mechanism confusion still exists. In fig. 2(h), although the MCSM can enhance the even-order scattering power and the line scattering power of urban areas, the volume scattering of forests is underestimated compared to the Y4R result, making the scattering mechanism description of natural areas inaccurate. In fig. 2(f), the volume scattering and surface scattering in natural areas such as forest can be well maintained. The method can better represent the scattering mechanism of the building area, and can not influence the description of the scattering mechanism of the natural area.
To quantitatively measure the scattering component percentages, an urban area (area a) and a forest area (area B) were selected from the results obtained by the three decomposition methods in fig. 2, and their corresponding scattering component percentages are shown in fig. 3. And simultaneously, optical images of the two regions and corresponding polarization azimuth angle distribution of the PolSAR images are given. As can be seen from the figure, even-order scattering power and helical scattering power obtained by the cross scattering model-based polarization decomposition method and the Y4R method are relatively similar in the building region, but the volume scattering power obtained by the former method is greatly reduced, the helical scattering power is enhanced, and the surface scattering power in the urban region is slightly increased. This demonstrates that the present invention can effectively separate cross scatter from buildings and bulk scatter from forests from total cross polarization scatter. In fig. 3(c), the MCSM method can significantly enhance even-order scattering in urban areas, but the surface scattering is also strong and the line scattering power is not significant, and compared with fig. 3(g), the distinction between buildings and forests is not significant after the MCSM decomposition. Comparing fig. 3(e) and fig. 3(f), it can be seen that the present invention has a better retention of forest region scattering mechanism, the cross scattering power is only 0.2%, and the bulk scattering still occupies the main body. Whereas in the results of the MCSM, the percentage of bulk scattering was much less and the line scattering was still 6.6%. As can be seen from fig. 3(d) and 3(h), although the polarization azimuth angle distributions of the two regions are similar, both are around 45 °, the cross scattering of the two regions is completely different, which indicates that the cross scattering model proposed by the present invention is effective for buildings.
Claims (7)
1. A method for decomposing urban targets of polarized radar remote sensing images based on a cross scattering model is characterized by comprising the following steps:
converting the single-view polarization scattering matrix of the PolSAR image into a polarization covariance matrix or a polarization coherent matrix;
calculating a polarization azimuth angle of the PolSAR image pixel by a polarization covariance matrix or a polarization coherent matrix;
a step of constructing a cross scattering model of the building by combining the polarization azimuth angle and the dihedral reflectors;
constructing a polarized target decomposition solving equation according to the surface scattering model, the even scattering model, the volume scattering model, the spiral scattering model and the cross scattering model, and solving scattering coefficients of the models;
and correcting to obtain the scattering weighting coefficients of the models, and calculating the surface scattering power, the even scattering power, the volume scattering power, the spiral scattering power and the cross scattering power by using the scattering weighting coefficients of the models to complete target decomposition.
2. The urban target decomposition method for the polarized radar remote sensing image based on the cross scattering model as claimed in claim 1, wherein the conversion steps of the polarized covariance matrix and the polarized coherence matrix are as follows:
the radar polarization scattering matrix is represented as:
in the formula, H and V represent horizontal polarization and vertical polarization, respectively, SPQ(P, Q ═ H, V) is transmitted in Q polarization mode, the target backward complex scattering coefficient of scattering matrix in P polarization mode receiving, S is existed in backward scattering condition satisfying reciprocal theoremHV＝SVH；
Pauli scattering vector is expressed as:
wherein, the upper labelTRepresenting a matrix transposition;
the polarization coherence matrix is represented as:
wherein, the upper labelCommon to the representation matricesYoke transpose, superscript denotes complex conjugate;
the polarization covariance matrix is expressed as:
wherein,
3. the method for decomposing urban targets of polarized radar remote sensing images based on cross scattering models according to claim 1, wherein the polarization azimuth angle θ of the PolSAR image pixels is estimated according to the following formula:
wherein, Re { T23Means T23Real part of, T22Characterizing the even-order scattering component, T, for the intermediate term of the polarization coherence matrix23The reflection asymmetry component is characterized for the polarization coherence matrix off-diagonal elements.
4. The method for decomposing urban targets of polarized radar remote sensing images based on the cross scattering model according to claim 1, wherein the polarization azimuth angle distribution of the dihedral angle scattering structure body is defined as:
wherein, thetadomIs the polarization azimuth of the building, the cross scattering coherent matrix model of the building<[T]>crossIs defined as:
wherein, Td(theta) is a coherent matrix scattering model of a standard dihedral reflector;
the above equation is further derived as:
5. the urban target decomposition method for the polarized radar remote sensing image based on the cross scattering model according to claim 1, wherein the step of constructing a polarized target decomposition solving equation is as follows:
(1) the expression of a multi-component polarization target decomposition equation based on a cross scattering model is listed as follows:
wherein f iss,fd,fv,fhAnd fcroWeighting coefficients for surface scattering, even scattering, volume scattering, helical scattering and cross scattering, respectively, [ T]s,[T]d,[T]v,[T]hAnd [ T]crossA standard coherent scattering model representing the corresponding scattering mechanism;
(2) expanding the equation in the step (1) to obtain the following equation system:
fs+fd|α|2+fv/2＝T11(a)
fs|β|2+fd+fv/4+fh/2+m22fcro＝T22(b)
fv/4+fh/2+m33fcro＝T33(c)
fsβ*+fdα＝T12(d)
fh/2＝|Im(T23)| (e)
wherein m is22And m33Second and third diagonal elements, Im (T), respectively, of the cross scatter model23) Is T23An imaginary part of (d);
first, the parameter f is obtained by equation (e)hAccording to the decision conditionFixing two unknown parameters fsAnd fd：
And further obtaining four equations containing four unknown variables, and solving to obtain all unknown parameters.
6. The method for decomposing urban targets in polarized radar remote sensing images based on the cross-scattering model according to claim 5, wherein the system of equations in the step (2) is simplified according to the following equations (b) and (c):
neglect (m)22-m33)fcroSolving all unknown parameters;
when in useSometimes:
when in useSometimes:
if f occurscroIf < 0, f is setcroAnd obtaining other unknown parameters according to a parameter solving method of four-component decomposition.
7. The method for decomposing urban targets of polarized radar remote sensing images based on cross scattering models according to claim 6, wherein the scattering weighting coefficients of the models are combined, and the scattering power of each model after decomposition is calculated according to the following formula:
wherein, PcroIs the cross scatter power.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811594285.0A CN109375189A (en) | 2018-12-25 | 2018-12-25 | Polarimetric radar remote sensing images city goal decomposition method based on cross scatter model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811594285.0A CN109375189A (en) | 2018-12-25 | 2018-12-25 | Polarimetric radar remote sensing images city goal decomposition method based on cross scatter model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109375189A true CN109375189A (en) | 2019-02-22 |
Family
ID=65371619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811594285.0A Pending CN109375189A (en) | 2018-12-25 | 2018-12-25 | Polarimetric radar remote sensing images city goal decomposition method based on cross scatter model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109375189A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032974A (en) * | 2019-04-12 | 2019-07-19 | 杭州世平信息科技有限公司 | A kind of description of SAR image scene characteristic and classification method |
| CN110223311A (en) * | 2019-05-24 | 2019-09-10 | 杭州世平信息科技有限公司 | Polarimetric radar edge Detection Method of Remote Sensing Images based on power drive adaptive windows |
| CN111123268A (en) * | 2020-01-02 | 2020-05-08 | 中国人民解放军国防科技大学 | Polarized target decomposition method based on fine scattering model |
| CN112597434A (en) * | 2020-11-05 | 2021-04-02 | 中国科学院国家空间科学中心 | Rapid implementation method and system for polarized KHT decomposition |
| CN112836638A (en) * | 2021-02-03 | 2021-05-25 | 自然资源部第一海洋研究所 | High sea state SAR ship detection method and application |
| CN113933839A (en) * | 2021-10-14 | 2022-01-14 | 中国人民解放军国防科技大学 | Method, apparatus, system and medium for polarization dependent pattern interpretation of radar |
| CN114429453A (en) * | 2021-12-31 | 2022-05-03 | 中国人民解放军国防科技大学 | Method for extracting artificial target of polarization synthetic aperture radar image driven by scattering mechanism |
| CN114998606A (en) * | 2022-05-10 | 2022-09-02 | 北京科技大学 | Weak scattering target detection method based on polarization feature fusion |
| CN115166741A (en) * | 2022-09-08 | 2022-10-11 | 中国科学院空天信息创新研究院 | Simplified model-based dual-phase central polarization chromatography decomposition method |
| CN116824221A (en) * | 2023-05-25 | 2023-09-29 | 西南交通大学 | PolSAR urban classification method and system considering radar target azimuth symmetry |
| CN117908023A (en) * | 2024-03-19 | 2024-04-19 | 中国科学院空天信息创新研究院 | Self-adaptive four-component polarization interference synthetic aperture radar decomposition method |
| CN118363024A (en) * | 2024-06-19 | 2024-07-19 | 中国人民解放军海军工程大学 | Polarolysis method for marine ship target detection of PolSAR image |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103197304A (en) * | 2013-04-19 | 2013-07-10 | 哈尔滨工业大学 | PolSAR image double-layer target decomposition method based on nonreflecting symmetric scattering component extraction |
-
2018
- 2018-12-25 CN CN201811594285.0A patent/CN109375189A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103197304A (en) * | 2013-04-19 | 2013-07-10 | 哈尔滨工业大学 | PolSAR image double-layer target decomposition method based on nonreflecting symmetric scattering component extraction |
Non-Patent Citations (3)
| Title |
|---|
| DELIANG XIANG ET AL.: "Built-up Area Extraction from PolSAR Imagery with Model-Based Decomposition and Polarimetric Coherence", 《REMOTE SENSING》 * |
| DELIANG XIANG ET AL.: "Model-Based Decomposition With Cross Scattering for Polarimetric SAR Urban Areas", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 * |
| YOSHIO YAMAGUCHI ET AL.: "Four-Component Scattering Model for Polarimetric SAR Image Decomposition", 《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032974A (en) * | 2019-04-12 | 2019-07-19 | 杭州世平信息科技有限公司 | A kind of description of SAR image scene characteristic and classification method |
| CN110223311A (en) * | 2019-05-24 | 2019-09-10 | 杭州世平信息科技有限公司 | Polarimetric radar edge Detection Method of Remote Sensing Images based on power drive adaptive windows |
| CN111123268A (en) * | 2020-01-02 | 2020-05-08 | 中国人民解放军国防科技大学 | Polarized target decomposition method based on fine scattering model |
| CN112597434A (en) * | 2020-11-05 | 2021-04-02 | 中国科学院国家空间科学中心 | Rapid implementation method and system for polarized KHT decomposition |
| CN112597434B (en) * | 2020-11-05 | 2024-03-22 | 中国科学院国家空间科学中心 | Method and system for rapidly realizing polarization KHT decomposition |
| CN112836638A (en) * | 2021-02-03 | 2021-05-25 | 自然资源部第一海洋研究所 | High sea state SAR ship detection method and application |
| CN112836638B (en) * | 2021-02-03 | 2023-08-01 | 自然资源部第一海洋研究所 | High sea condition SAR ship detection method and application |
| CN113933839B (en) * | 2021-10-14 | 2024-05-28 | 中国人民解放军国防科技大学 | Method, device, system and medium for interpreting polarization-dependent directional diagram of radar |
| CN113933839A (en) * | 2021-10-14 | 2022-01-14 | 中国人民解放军国防科技大学 | Method, apparatus, system and medium for polarization dependent pattern interpretation of radar |
| CN114429453B (en) * | 2021-12-31 | 2023-11-14 | 中国人民解放军国防科技大学 | Method for extracting artificial targets from polarized synthetic aperture radar images driven by scattering mechanism |
| CN114429453A (en) * | 2021-12-31 | 2022-05-03 | 中国人民解放军国防科技大学 | Method for extracting artificial target of polarization synthetic aperture radar image driven by scattering mechanism |
| CN114998606A (en) * | 2022-05-10 | 2022-09-02 | 北京科技大学 | Weak scattering target detection method based on polarization feature fusion |
| CN114998606B (en) * | 2022-05-10 | 2023-04-18 | 北京科技大学 | Weak scattering target detection method based on polarization feature fusion |
| CN115166741A (en) * | 2022-09-08 | 2022-10-11 | 中国科学院空天信息创新研究院 | Simplified model-based dual-phase central polarization chromatography decomposition method |
| CN115166741B (en) * | 2022-09-08 | 2022-11-29 | 中国科学院空天信息创新研究院 | Simplified model-based dual-phase central polarization chromatography decomposition method |
| CN116824221A (en) * | 2023-05-25 | 2023-09-29 | 西南交通大学 | PolSAR urban classification method and system considering radar target azimuth symmetry |
| CN117908023A (en) * | 2024-03-19 | 2024-04-19 | 中国科学院空天信息创新研究院 | Self-adaptive four-component polarization interference synthetic aperture radar decomposition method |
| CN117908023B (en) * | 2024-03-19 | 2024-05-14 | 中国科学院空天信息创新研究院 | Self-adaptive four-component polarization interference synthetic aperture radar decomposition method |
| CN118363024A (en) * | 2024-06-19 | 2024-07-19 | 中国人民解放军海军工程大学 | Polarolysis method for marine ship target detection of PolSAR image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109375189A (en) | Polarimetric radar remote sensing images city goal decomposition method based on cross scatter model | |
| Xiang et al. | Unsupervised polarimetric SAR urban area classification based on model-based decomposition with cross scattering | |
| CN103593669B (en) | A kind of method that polarimetric synthetic aperture radar image four component decomposes | |
| CN107229933B (en) | Freeman/eigenvalue decomposition method of adaptive volume scattering model | |
| Wernecke et al. | Lead detection in Arctic sea ice from CryoSat-2: quality assessment, lead area fraction and width distribution | |
| Jin et al. | Polarimetric scattering and SAR information retrieval | |
| CN102901965B (en) | Radar imaging method of electrically large size target in ocean clutter environment | |
| Schvartzman et al. | Signal processing and radar characteristics (SPARC) simulator: A flexible dual-polarization weather-radar signal simulation framework based on preexisting radar-variable data | |
| CN102540157A (en) | Ground feature classifying method based on simplified polarization synthetic aperture radar data | |
| CN110412573A (en) | Polarimetric SAR image decomposition method and storage medium | |
| CN110910494B (en) | Method for analyzing backward scattering mechanism of seawater floating raft culture under random sea conditions | |
| CN115825955A (en) | Polarization time sequence InSAR method based on coherent matrix self-adaptive decomposition | |
| Long et al. | Geometrical Structure Classification of Target HRRP Scattering Centers Based on Dual Polarimetric $ H/\alpha $ Features | |
| Antropov et al. | PolSAR mosaic normalization for improved land-cover mapping | |
| Zhang et al. | Polarimetric Autocorrelation Matrix: A New Tool for Joint Characterizing of Target Polarization and Doppler Scattering Mechanism | |
| CN101976357A (en) | Method and device for classifying fully polarimetric synthetic aperture radar image | |
| CN113608218A (en) | Frequency domain interference phase sparse reconstruction method based on back projection principle | |
| Hu et al. | Model-based polarimetric target decomposition with power redistribution for urban areas | |
| CN117111063A (en) | Polarized SAR image feature extraction method | |
| CN107144841B (en) | Polarized SAR image target decomposition method based on minimum residual power | |
| CN108710816B (en) | Polarized SAR image building detection method based on probability fusion | |
| Chen et al. | Roll-invariant target parameter extraction from POLSAR data | |
| Zhang et al. | A bistatic synthetic aperture radar imagery simulation of maritime scene using the extended nonlinear chirp scaling algorithm | |
| Carratelli et al. | Reconstruction of SAR wave image effects through pseudo random simulation | |
| Chitroub et al. | A new PCA-based method for data compression and enhancement of multi-frequency polarimetric SAR imagery |
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: 20190222 |