CN112986991B - Clutter distribution independent polarimetric synthetic aperture radar ship detection method - Google Patents

Clutter distribution independent polarimetric synthetic aperture radar ship detection method Download PDF

Info

Publication number
CN112986991B
CN112986991B CN202110163845.2A CN202110163845A CN112986991B CN 112986991 B CN112986991 B CN 112986991B CN 202110163845 A CN202110163845 A CN 202110163845A CN 112986991 B CN112986991 B CN 112986991B
Authority
CN
China
Prior art keywords
polarization
ship
target
clutter
enhancement
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110163845.2A
Other languages
Chinese (zh)
Other versions
CN112986991A (en
Inventor
徐舟
范崇祎
黄晓涛
王建
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.)
National University of Defense Technology
Original Assignee
National University of Defense Technology
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 National University of Defense Technology filed Critical National University of Defense Technology
Priority to CN202110163845.2A priority Critical patent/CN112986991B/en
Publication of CN112986991A publication Critical patent/CN112986991A/en
Application granted granted Critical
Publication of CN112986991B publication Critical patent/CN112986991B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention belongs to the technical field of remote sensing image processing, and discloses a clutter distribution independent polarized synthetic aperture radar ship detection method, which comprises the following steps: preprocessing a Sinclair matrix acquired by a polarized synthetic aperture radar; performing target enhancement by using a self-adaptive linear polarization filtering method, and modeling the selection problem of the self-adaptive filter as an optimization model by taking the maximum target-background contrast as a criterion; alternately optimizing the optimal polarization filter and a ship index set, wherein the optimization of the polarization filter is realized by calculating generalized eigenvectors, and the ship index set is optimized by using a disturbance updating method; on the basis of only calculating clutter statistical central moment, calculating a detection threshold irrelevant to statistical distribution by using a given false alarm probability. The self-adaptive polarization enhancement technology and the detection threshold calculation method irrelevant to distribution effectively solve the problems of difficult selection of polarization characteristics and complicated clutter statistical modeling in the target detection of the polarized synthetic aperture radar ship.

Description

Clutter distribution independent polarimetric synthetic aperture radar ship detection method
Technical Field
The invention belongs to the technical field of remote sensing image processing, and particularly relates to a method for detecting marine ships by utilizing a polarization synthetic aperture radar image.
Background art:
in recent years, ocean surveillance has become a focus of research. Synthetic Aperture Radars (SAR) play an important role in this field due to their unique advantages, such as all weather, all time, etc. In contrast, a polar synthetic aperture radar (polar synthetic aperture radar, polar sar) can obtain the electromagnetic scattering property of a target from a polarization dimension, and can better complete target detection and classification tasks.
In marine surveillance, an important research direction is ship detection and identification, and in the past thirty years, the PolSAR ship detection method has received extensive attention and achieved relatively good research results. From the current technical means, the PolSAR ship detection method mainly transforms a multidimensional polarization domain to a one-dimensional characteristic domain, and then performs Constant False Alarm Rate (CFAR) detection in the characteristic domain, and the method relates to two key technologies, wherein the first technology is characteristic transformation or characteristic enhancement, namely, suitable characteristics and transformation are searched to achieve the purposes of enhancing the ship target and inhibiting background clutter; and secondly, the clutter statistical modeling is carried out, namely the proper probability distribution is searched for describing the statistical characteristics of the clutter. And finally, confirming the points different from the clutter as ship targets. However, in most cases, finding a suitable feature transformation and a clutter statistical model is very complicated, because the ship scattering feature and the clutter statistical characteristic of PolSAR are affected by various factors such as geographic environment, radar parameters, meteorological conditions, and the like, and it is difficult to find a universal method. When the scatter signature and the clutter statistic signature are mismatched from the actual situation, the performance of the detector is drastically degraded.
Therefore, the difficulty brought by the feature transformation and clutter statistical modeling can be effectively solved by researching the adaptive polarization feature enhancement technology of the PolSAR data and the detection technology irrelevant to clutter distribution.
The invention content is as follows:
aiming at the difficulty in the process of detecting the PolSAR ship at present, the invention provides a clutter distribution-independent polarized synthetic aperture radar ship detection method, which can adaptively enhance the polarization characteristics according to PolSAR data, and the subsequent detection threshold is not set depending on clutter distribution.
The technical scheme adopted by the invention is as follows:
a clutter distribution independent polarimetric synthetic aperture radar ship detection method comprises the following implementation steps:
s1 polarization data preprocessing
Directly obtaining a Sinclair scattering matrix S of a scene based on PolSAR, analyzing a distributed target by using second-order expected data, preprocessing the S and further determining a polarization coherent matrix pair T and a polarization covariance matrix C; performing further matrix decomposition on the polarization coherent matrix T and the polarization covariance matrix C, and extracting polarization scattering entropy, polarization average Alpha angle and polarization scattering anisotropy characteristics;
the next adaptive polarization target enhancement modeling step takes the vector form of the preprocessed polarization data as input, and the vector form is the vector form of a T or C matrix or the vector form formed by polarization scattering entropy, polarization average Alpha angle and polarization scattering anisotropy characteristics;
s2 adaptive polarization target enhancement modeling
After the preprocessed polarization characteristic data are obtained, polarization enhancement is carried out on the target in a self-adaptive mode; determining a set containing ship indexes as theta, performing polarization enhancement on a ship target by adopting a linear polarization filter so as to enable ships in the theta to be maximally distinguished from background clutter after filtering, taking a filtered target clutter ratio as an optimization index, and establishing an optimization model of adaptive polarization enhancement by taking the theta and a polarization filter coefficient as optimization variables at the same time;
s3 adaptive polarization target enhancement implementation
The adaptive polarization target enhancement realizes that ship target marking and contrast enhancement processing are finished by an alternate iteration method, and specifically comprises 2 implementation steps: optimizing a polarization filter, and optimizing and calibrating a ship index set; alternately performing the steps until convergence based on the 2 steps, and further completing adaptive polarization enhancement of the ship target, wherein the polarization filter optimization is obtained by solving a generalized Rayleigh quotient problem under a ship index determining condition, and the ship index set optimization and calibration are updated by using a disturbance method under the polarization filter determining condition;
s4 detection threshold setting
After the polarization enhancement filter is obtained through optimization, a square detector is adopted to detect a ship target; under the condition of giving the false alarm probability, determining a detection threshold through the statistical central moment of the clutter and the false alarm probability on the basis of a Markov inequality without modeling clutter distribution; the elements that eventually exceed the threshold are considered ship targets and the actual false alarm rate of the detector is bounded above by a given false alarm probability.
Description of the drawings:
in order to more clearly illustrate the technical solution of the present invention, the drawings required in the solution are briefly described below.
FIG. 1 is a schematic block diagram of a clutter distribution independent polarimetric synthetic aperture radar vessel detection process proposed by the present invention;
fig. 2 is a schematic diagram of a total power image of PolSAR according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a ship target polarization enhanced image provided by an embodiment of the invention;
fig. 4 is a schematic diagram of a ship target detection result binary image according to an embodiment of the present invention.
The specific implementation mode is as follows:
for the purpose of better illustrating the invention and for the purpose of facilitating an understanding thereof, the invention will now be described in detail by way of specific embodiments with reference to the accompanying drawings.
A clutter distribution independent polarimetric synthetic aperture radar ship detection method comprises the following implementation steps:
s1 polarization data preprocessing
Directly acquiring a Sinclair scattering matrix S of a target by a polarized synthetic aperture radar:
Figure GDA0003633009460000021
wherein S isHHIndicating that the electromagnetic wave is transmitted with horizontal polarization, received with horizontal polarization, SHVIndicating that the electromagnetic wave is transmitted in a vertically polarized manner, received in a horizontally polarized manner, SVHElectromagnetic waves are transmitted in a horizontally polarized manner, received in a vertically polarized manner, SVVThe electromagnetic wave is transmitted in a vertical polarization mode and received in a vertical polarization mode;
for distributed targets, S is converted into Pauli vectors
Figure GDA0003633009460000022
And Lexicogrphic vector
Figure GDA0003633009460000031
Construction poleQuantizing the coherence matrix T and the polarization covariance matrix C to obtain a second order property of the polarization scattering matrix:
Figure GDA0003633009460000032
wherein, E [. cndot. ] represents taking mathematical expectation, and the spatial averaging replaces the taking of mathematical expectation;
further processing the polarization coherent matrix T and the polarization covariance matrix C to extract more fine polarization scattering entropy, polarization average Alpha angle and polarization characteristics of polarization scattering anisotropy;
the data used in this embodiment is in the form of vectors of the preprocessed polarization data, which is in the form of vectors of T or C matrices, or in the form of vectors composed of polarization scattering entropy, polarization mean Alpha angle, and polarization scattering anisotropy characteristics, and fig. 2 is a schematic diagram of a total power image of PolSAR provided in this embodiment of the present invention. S2 adaptive polarization enhanced modeling
The preprocessed polarization characteristic data is expressed as V, and a data matrix V is formed by using complete PolSAR data1,v2,...,vN]V contains N elements in total of ship data and clutter data; PolSAR ship detection is that ship data are separated from V;
firstly, a ship target is polarized and enhanced by using a linear polarization filter w, so that the ship and the background have larger discrimination, and the specific expression is g-wHV, and realizing self-adaptive processing by making the linear polarization filter w self-adaptively change along with the data matrix V, and the implementation is as follows:
the set containing the index of the vessel index is denoted as set Θ, and the average energy of the polarization enhanced vessel is denoted as
Figure GDA0003633009460000033
Wherein v isiThe ith element in the V is represented, and the | theta | represents the number of the elements in the theta; representing the average energy of clutter as
Figure GDA0003633009460000034
The design of the optimal polarization filter is realized by taking the contrast of the maximized target and the clutter as a criterion, so that the subsequent target detection is facilitated, and the optimal polarization filter is realized by optimizing the following problems
Figure GDA0003633009460000035
Wherein, w*For the designed optimal filter weight, σ2Represents the system noise power to prevent the denominator from being 0;
in the polarization enhancement model given by the formula (III), the target data contained in theta is subjected to polarization enhancement, the clutter data not contained in theta is suppressed, the estimation of theta is simultaneously completed, and the polarization enhancement objective function is recorded
Figure GDA0003633009460000036
The polarization enhancement model was obtained as follows
Figure GDA0003633009460000037
Accordingly, the optimum polarization filter w*Optimal estimate of the ship index set Θ*(IV) is the optimal solution for (IV);
s3 adaptive polarization target enhancement implementation
The adaptive polarization target enhancement realizes that ship target marking and contrast enhancement processing are completed through an alternate iteration method so as to solve the optimal estimation of an optimal polarization filter and a ship index set, and specifically comprises 2 implementation steps: optimizing a polarization filter, and optimizing and calibrating a ship index set; the steps of polarization filter optimization, ship index set optimization and calibration are alternately carried out until convergence, and finally the adaptive polarization enhancement of the ship target is completed; the method comprises the following specific steps:
s3.1 polarization filter optimization
The ship index set theta at the k-th iteration is determinedkUnder the condition of (2), polarizing filter wkThe optimization problem is converted into a generalized Rayleigh quotient problem, and the optimal solution of the generalized Rayleigh quotient problem is given by the following generalized eigenvector
Figure GDA0003633009460000041
Wherein, INAn identity matrix of dimension N is represented,
Figure GDA0003633009460000042
representing the principal eigenvector of matrix A;
s3.2 Ship index set optimization and correction
The ship index set theta at the k-th iteration is determinedkAnd a polarization filter wkUnder the condition of (1), a disturbance updating method is used for obtaining a ship index set theta at the k +1 th timek+1；
Order to
Figure GDA0003633009460000043
Representing an index set ΘkIf i ∈ ΘkThen determine
Figure GDA0003633009460000044
Otherwise
Figure GDA0003633009460000045
To represent
Figure GDA0003633009460000046
In a diagonal matrix form, to obtain
Figure GDA0003633009460000047
To thetakThe ith element in the list is disturbed, the category of the ith element is changed, and the disturbed ith element is changed into the category of the ith elementIs marked as
Figure GDA0003633009460000048
Will be provided with
Figure GDA0003633009460000049
And
Figure GDA00036330094600000410
respectively combine wkSubstituting into J (w, theta), and recording the corresponding function value as JkAnd
Figure GDA00036330094600000411
comparing the sizes of the two to confirm the set thetak+1The specific expression of the elements in (1) is as follows
Figure GDA00036330094600000412
Traversing N elements in PolSAR data to obtain a target index set theta of the (k + 1) th iterationk+1；
Note that the detection of a ship target is essentially a binary problem, and in actual processing, there is a possibility of a reverse indication situation, namely Θk(ii) is wrongly labeled as a set of clutters, while targets are all excluded from the set;
considering the practical detection problem, the number of clutter is much larger than the number of targets if
Figure GDA00036330094600000413
Then the vector is indicated
Figure GDA00036330094600000414
Correction was made as follows
Figure GDA00036330094600000415
Wherein 1 isNRepresenting a full 1 vector of length N; FIG. 3 shows a ship provided for an example of the inventionA schematic of the target-only polarization-enhanced image;
s4 detection threshold setting
After obtaining the optimal polarization enhancement filter w*Then, detecting the ship target by using a square detector; for a given detection threshold T, satisfy
Figure GDA00036330094600000416
Is considered as a ship target, otherwise is considered as background noise;
without assuming clutter elements
Figure GDA0003633009460000051
Under the condition of the distribution, determining a detection threshold by a false alarm probability delimitation method; for clutter element vi，
Figure GDA0003633009460000052
Always true and will false alarm probability PfaIs shown as
Figure GDA0003633009460000053
Wherein Pr {. cndot } represents a probability;
for any r > 0, the value is obtained according to the Markov inequality
Figure GDA0003633009460000054
For a square detector with a detection threshold of T, the false alarm probability of the square detector cannot exceed that of the square detector under the condition of only utilizing r-order statistical central moment of clutter
Figure GDA0003633009460000055
For a given false alarm probability PfaDetermining the detection threshold T by
Figure GDA0003633009460000056
Wherein m isrThe r-order statistical central moment representing clutter is estimated by
Figure GDA0003633009460000057
The PolSAR ship detection method provided by the invention can adaptively complete ship target polarization enhancement and clutter suppression, the ship sea contrast is enhanced to the maximum extent, and the clutter distribution does not need to be modeled on the basis of ensuring the false alarm probability in subsequent detection processing, so that the method is high in applicability. Fig. 4 is a schematic diagram of a ship target detection result binary image provided by an example of the invention.

Claims (4)

1. A clutter distribution independent method of detecting a polarized synthetic aperture radar vessel, comprising the steps of:
s1 polarization data preprocessing
Directly obtaining a Sinclair scattering matrix S of a scene based on PolSAR, analyzing a distributed target by using second-order expected data, and preprocessing the S to further determine a polarization coherent matrix T and a polarization covariance matrix C; performing further matrix decomposition on the polarization coherent matrix T and the polarization covariance matrix C, and extracting polarization scattering entropy, polarization average Alpha angle and polarization scattering anisotropy characteristics;
the next adaptive polarization target enhancement modeling step takes the vector form of the preprocessed polarization data as input, and the vector form is the vector form of a T or C matrix or the vector form formed by polarization scattering entropy, polarization average Alpha angle and polarization scattering anisotropy characteristics;
s2 adaptive polarization target enhancement modeling
After the preprocessed polarization characteristic data are obtained, polarization enhancement is carried out on the target in a self-adaptive mode; determining a set containing ship indexes as theta, performing polarization enhancement on a ship target by adopting a linear polarization filter so as to enable ships in the theta to be maximally distinguished from background clutter after filtering, taking a filtered target clutter ratio as an optimization index, and establishing an optimization model of adaptive polarization enhancement by taking the theta and a polarization filter coefficient as optimization variables at the same time;
s3 adaptive polarization target enhancement implementation
The adaptive polarization target enhancement realizes that ship target marking and contrast enhancement processing are finished by an alternate iteration method, and specifically comprises 2 implementation steps: optimizing a polarization filter, and optimizing and calibrating a ship index set; alternately performing the steps until convergence based on the 2 steps, and further completing adaptive polarization enhancement of the ship target, wherein the polarization filter optimization is obtained by solving a generalized Rayleigh quotient problem under a ship index determining condition, and the ship index set optimization and calibration are updated by using a disturbance method under the polarization filter determining condition;
s4 detection threshold setting
After the polarization enhancement filter is obtained through optimization, detecting a ship target by adopting a square detector; under the condition of giving the false alarm probability, determining a detection threshold through the statistical central moment of the clutter and the false alarm probability on the basis of a Markov inequality without modeling clutter distribution; the elements that eventually exceed the threshold are considered to be ship targets and the actual false alarm rate of the detector is bounded above by a given false alarm probability.
2. A method of clutter distribution independent polar synthetic aperture radar vessel detection according to claim 1, wherein: in the step S2, a linear polarization filter is adopted to enhance the ship target in the adaptive polarization target enhancement modeling, and an optimization model is established with the filtered target clutter ratio as a target, and the specific implementation steps are as follows:
the preprocessed polarization characteristic data is expressed as V, and a data matrix V is formed by using complete PolSAR data1,v2,...,vN]V contains N elements in total of ship data and clutter data; PolSAR ship detection is that ship data are separated from V;
firstly, a linear polarization filter w is used for polarization enhancement of a ship target, so that the ship and the background have larger discrimination, and a detailed tableDa is g ═ wHV, and realizing self-adaptive processing by making the linear polarization filter w self-adaptively change along with the data matrix V, and the implementation is as follows:
the set containing the index of the vessel index is denoted as set Θ, and the average energy of the polarization enhanced vessel is denoted as
Figure FDA0003633009450000011
Wherein v isiThe ith element in the V is represented, and the | theta | represents the number of the elements in the theta; representing the average energy of clutter as
Figure FDA0003633009450000021
The design of the optimal polarization filter is realized by taking the contrast of the maximized target and the clutter as a criterion, so that the subsequent target detection is facilitated, and the optimal polarization filter is realized by optimizing the following problems
Figure FDA0003633009450000022
Wherein, w*For the designed optimal filter weight, σ2Represents the system noise power to prevent the denominator from being 0;
in the polarization enhancement model given by the formula (I), the target data contained in theta is subjected to polarization enhancement, the clutter data not contained in theta are suppressed, the estimation of theta is simultaneously completed, and a polarization enhancement target function is recorded
Figure FDA0003633009450000023
The polarization enhancement model was obtained as follows
Figure FDA0003633009450000024
Accordingly, the optimum polarization filter w*Optimal estimate of the ship index set Θ*Is the optimal solution of (II).
3. A clutter distribution independent polarimetric synthetic aperture radar vessel detection method according to claim 2, wherein: the implementation of the S3 adaptive polarization target enhancement includes 2 implementation steps: optimizing a polarization filter, and optimizing and calibrating a ship index set; the steps of polarization filter optimization, ship index set optimization and calibration are alternately carried out until convergence, and finally the adaptive polarization enhancement of the ship target is completed; the method comprises the following concrete steps:
s3.1 polarization filter optimization
The ship index set theta at the k-th iteration is determinedkUnder the condition of (1), polarizing filter wkThe optimization problem is converted into a generalized Rayleigh quotient problem, and the optimal solution of the generalized Rayleigh quotient problem is given by the following generalized eigenvector
Figure FDA0003633009450000025
Wherein, INAn identity matrix of dimension N is represented,
Figure FDA0003633009450000026
representing the principal eigenvector of matrix A;
s3.2 Ship index set optimization and correction
The ship index set theta at the k-th iteration is determinedkAnd a polarization filter wkUnder the condition of (1), a disturbance updating method is used for obtaining a ship index set theta at the k +1 th timek+1；
Order to
Figure FDA0003633009450000027
Representing an index set ΘkIf i ∈ ΘkThen determine
Figure FDA0003633009450000028
Otherwise
Figure FDA0003633009450000029
Figure FDA00036330094500000210
To represent
Figure FDA00036330094500000211
In a diagonal matrix form, to obtain
Figure FDA00036330094500000212
To thetakThe ith element in the system is disturbed, the category of the ith element is changed, and the disturbed indication vector is recorded as
Figure FDA00036330094500000213
Will be provided with
Figure FDA00036330094500000214
And
Figure FDA00036330094500000215
respectively combine wkSubstituting into J (w, theta), and recording the corresponding function value as JkAnd
Figure FDA0003633009450000031
comparing the sizes of the two to confirm the set thetak+1The specific expression of the elements in (1) is as follows
Figure FDA0003633009450000032
Traversing N elements in the PolSAR data to obtain a target index set theta of the (k + 1) th iterationk+1；
Considering the practical detection problem, the number of clutter is much larger than the number of targets if
Figure FDA0003633009450000033
Then the pair indicates the vector
Figure FDA0003633009450000034
Correction was made as follows
Figure FDA0003633009450000035
Wherein 1 isNRepresenting a full 1 vector of length N.
4. A clutter distribution independent polarimetric synthetic aperture radar vessel detection method according to claim 3, wherein: the detection threshold is designed according to the false alarm probability on the basis of a Markov inequality in the step S4 under the condition of only using the clutter statistical central moment, a clutter statistical distribution model is not needed, and the method specifically comprises the following implementation steps:
after obtaining the optimal polarization enhancement filter w*Then, detecting the ship target by using a square detector; for a given detection threshold T, satisfy
Figure FDA0003633009450000036
Is considered as a ship target, otherwise is considered as background noise;
without assuming clutter elements
Figure FDA0003633009450000037
Under the condition of the distribution, determining a detection threshold by a false alarm probability delimitation method; for clutter element vi，
Figure FDA0003633009450000038
Always true and will false alarm probability PfaIs shown as
Figure FDA0003633009450000039
Wherein Pr {. cndot } represents a probability;
for any r > 0, the value is obtained according to the Markov inequality
Figure FDA00036330094500000310
For a square detector with a detection threshold of T, the false alarm probability of the square detector cannot exceed that of the square detector under the condition of only utilizing r-order statistical central moment of clutter
Figure FDA00036330094500000311
For a given false alarm probability PfaDetermining the detection threshold T by
Figure FDA00036330094500000312
Wherein m isrThe statistical central moment of the order r representing clutter is estimated by
Figure FDA00036330094500000313
CN202110163845.2A 2021-02-05 2021-02-05 Clutter distribution independent polarimetric synthetic aperture radar ship detection method Active CN112986991B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110163845.2A CN112986991B (en) 2021-02-05 2021-02-05 Clutter distribution independent polarimetric synthetic aperture radar ship detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110163845.2A CN112986991B (en) 2021-02-05 2021-02-05 Clutter distribution independent polarimetric synthetic aperture radar ship detection method

Publications (2)

Publication Number Publication Date
CN112986991A CN112986991A (en) 2021-06-18
CN112986991B true CN112986991B (en) 2022-06-21

Family

ID=76348384

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110163845.2A Active CN112986991B (en) 2021-02-05 2021-02-05 Clutter distribution independent polarimetric synthetic aperture radar ship detection method

Country Status (1)

Country Link
CN (1) CN112986991B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104091335A (en) * 2014-07-04 2014-10-08 西安电子科技大学 Polarization SAR image ship target detection method
WO2016097890A1 (en) * 2014-12-15 2016-06-23 Airbus Group Singapore Pte. Ltd. Automated method for selecting training areas of sea clutter and detecting ship targets in polarimetric synthetic aperture radar imagery
CN107025654A (en) * 2016-02-01 2017-08-08 南京理工大学 The adaptive ship detection method of SAR image checked based on global iterative
CN107145874A (en) * 2017-05-13 2017-09-08 复旦大学 Ship Target Detection and discrimination method in complex background SAR image
CN110297244A (en) * 2019-07-26 2019-10-01 中国人民解放军国防科技大学 Method for detecting wake flow of polarized synthetic aperture radar sea surface ship
CN112147591A (en) * 2020-08-27 2020-12-29 清华大学 Polarized radar sea surface ship detection method and device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104091335A (en) * 2014-07-04 2014-10-08 西安电子科技大学 Polarization SAR image ship target detection method
WO2016097890A1 (en) * 2014-12-15 2016-06-23 Airbus Group Singapore Pte. Ltd. Automated method for selecting training areas of sea clutter and detecting ship targets in polarimetric synthetic aperture radar imagery
CN107025654A (en) * 2016-02-01 2017-08-08 南京理工大学 The adaptive ship detection method of SAR image checked based on global iterative
CN107145874A (en) * 2017-05-13 2017-09-08 复旦大学 Ship Target Detection and discrimination method in complex background SAR image
CN110297244A (en) * 2019-07-26 2019-10-01 中国人民解放军国防科技大学 Method for detecting wake flow of polarized synthetic aperture radar sea surface ship
CN112147591A (en) * 2020-08-27 2020-12-29 清华大学 Polarized radar sea surface ship detection method and device

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Faint Ship Wake Detection in PolSAR Images;Zhou Xu et al.;《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》;20180731;第15卷(第7期);全文 *
基于PCDM香农熵的全极化SAR图像船舶目标检测方法;张程等;《遥感技术与应用》;20180620(第03期);全文 *
基于极化合成孔径雷达的舰船检测方法;张鑫等;《***工程与电子技术》;20101015(第10期);全文 *

Also Published As

Publication number Publication date
CN112986991A (en) 2021-06-18

Similar Documents

Publication Publication Date Title
Kang et al. A modified faster R-CNN based on CFAR algorithm for SAR ship detection
CN103839073B (en) Polarization SAR image classification method based on polarization features and affinity propagation clustering
CN108133468B (en) Constant false alarm rate ship detection method based on adaptive parameter enhancement and trail auxiliary detection
Andriyanov et al. Anomalies detection on spatially inhomogeneous polyzonal images
CN113674308B (en) SAR image ship target rapid detection method based on image enhancement and multiple detection
Tang et al. Sonar image mosaic based on a new feature matching method
CN112147591A (en) Polarized radar sea surface ship detection method and device
CN112215832B (en) SAR trail image quality evaluation and self-adaptive detection parameter adjustment method
CN117333468B (en) Flood disaster monitoring method for multi-mode time sequence PolSAR image
CN112669332B (en) Method for judging sea-sky conditions and detecting infrared targets based on bidirectional local maxima and peak value local singularities
CN112986991B (en) Clutter distribution independent polarimetric synthetic aperture radar ship detection method
Gao et al. Fast two‐dimensional subset censored CFAR method for multiple objects detection from acoustic image
CN113222898A (en) Double-voyage SAR image trace detection method based on multivariate statistics and deep learning
Zou et al. Sonar Image Target Detection for Underwater Communication System Based on Deep Neural Network.
CN117115436A (en) Ship attitude detection method and device, electronic equipment and storage medium
CN115700739A (en) Full-polarization SAR image ship detection method
Ju et al. A SAR image segmentation method based on MLRT
CN114898224B (en) Change detection method based on physical scattering mechanism
Cui et al. Information theoretical similarity measure for change detection
Li et al. Ship target detection method based on local saliency enhancement
Wang et al. Adaptive spatial constraint sparse representation for target detection in PolSAR image
Jung et al. Local cell-averaging fast CFAR for multi-target detection in high-resolution SAR images
Madjidi et al. Percentile-Based Estimator for Generalized Rayleigh CFAR Ship Detection in SAR Imagery
CN114690182A (en) GP-PNF-based polarized SAR target orientation fuzzy elimination method
Chang et al. High resolution polarimetric SAR target classification with neural network

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
GR01 Patent grant
GR01 Patent grant