CN112686095B - Ship target track correlation method for stationary orbit staring satellite remote sensing image - Google Patents
Ship target track correlation method for stationary orbit staring satellite remote sensing image Download PDFInfo
- Publication number
- CN112686095B CN112686095B CN202011410712.2A CN202011410712A CN112686095B CN 112686095 B CN112686095 B CN 112686095B CN 202011410712 A CN202011410712 A CN 202011410712A CN 112686095 B CN112686095 B CN 112686095B
- Authority
- CN
- China
- Prior art keywords
- track
- ship target
- remote sensing
- sensing image
- satellite
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 83
- 101150080778 INPP5D gene Proteins 0.000 claims description 292
- 238000003384 imaging method Methods 0.000 claims description 29
- 230000009466 transformation Effects 0.000 claims description 22
- 230000008859 change Effects 0.000 claims description 9
- 239000000126 substance Substances 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 8
- 238000012417 linear regression Methods 0.000 claims description 8
- 238000013213 extrapolation Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000013519 translation Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000010998 test method Methods 0.000 claims 2
- 230000000875 corresponding effect Effects 0.000 description 9
- 238000012544 monitoring process Methods 0.000 description 6
- 238000005070 sampling Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000012937 correction Methods 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Images
Landscapes
- Radar Systems Or Details Thereof (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method for correlating a ship target track of a staring satellite remote sensing image of a stationary orbit, which belongs to the technical field of satellite remote sensing, introduces ship automatic identification system information, firstly realizes the correlation of a synthetic ship target, effectively reduces the number of correlation matching combinations, then divides the correlation into two correlation stages, correlates non-cooperative targets, not only reduces the number of matching combinations in a correlation algorithm and improves the speed of ship target track correlation, but also overcomes the influence of the conditions of medium-low resolution and low frame rate on the intermittent track correlation of the ship target, reduces the correlation complexity, improves the correlation accuracy, finally connects the intermittent track on the correlation, and improves the integrity of ship target situation display.
Description
Technical Field
The invention relates to the technical field of satellite remote sensing, in particular to a method for correlating a ship target track of a remote sensing image of a staring satellite in a stationary orbit.
Background
The traditional target track association method aims at two types of data: one is radar and electronic reconnaissance data, and the data has the characteristics of higher data sampling rate and is mainly realized based on target motion characteristics; the other is multi-camera video data, which is characterized by higher image spatial resolution and is mainly realized based on target image characteristics.
The stationary orbit staring satellite can stay above a fixed area for a long time, can quickly adjust an imaging monitoring area as required, can acquire the track of a ship target in a related sea area during each observation, and can continuously observe a plurality of important ship targets or key sea areas by associating the tracks of the ship targets at different imaging moments, namely associating the tracks of the ship targets intermittently, so that the stationary orbit staring satellite has the quick response capability to key events and the near-real-time monitoring capability to key targets, is very suitable for long-time monitoring and quick imaging access, and plays an important role in reconnaissance and monitoring of moving targets on the sea.
The prior art requires that data have higher sampling rate or higher image spatial resolution, however, a remote sensing image sequence acquired by a fixed orbit staring satellite has the characteristics of less sampling points, low spatial resolution, short continuous observation time and the like, so that the traditional target intermittent track association method is difficult to adapt to ship target track association of the remote sensing image of the fixed orbit staring satellite, thereby causing low association accuracy rate in the prior art and even failing to realize effective target track association.
Disclosure of Invention
The invention aims to provide a method for correlating a ship target track of a remote sensing image of a stationary orbit staring satellite, which solves the problem of intermittent track correlation of the ship target under the conditions of medium-low resolution and low frame rate, reduces the correlation complexity and improves the correlation efficiency and accuracy by respectively correlating a cooperative target and a non-cooperative target.
In order to achieve the purpose, the invention provides the following scheme:
a method for correlating ship target tracks of stationary orbit staring satellite remote sensing images, comprising the following steps:
acquiring a ship target track set of a stationary orbit staring satellite remote sensing image sequence and an AIS ship target track set provided by a ship automatic identification system in an imaging observation time period corresponding to the stationary orbit staring satellite remote sensing image sequence; the ship target track set of the stationary orbit staring satellite remote sensing image sequence comprises the following steps: a first remote sensing image sequence ship target track set of a stationary orbit staring satellite and a second remote sensing image sequence ship target track set of the stationary orbit staring satellite;
determining that a certain track in the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a certain track in the second remote sensing image sequence ship target track set of the stationary orbit staring satellite are intermittent tracks of the same cooperative ship target according to the AIS ship target track set to obtain a stationary orbit staring satellite remote sensing image sequence cooperative ship target intermittent track association result;
screening out a non-cooperative ship target set of the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a non-cooperative ship target set of the second remote sensing image sequence ship target track set of the stationary orbit staring satellite according to the intermittent track correlation result of the cooperative ship target;
associating a first remote sensing image sequence non-cooperative ship target set of a stationary orbit staring satellite with a second remote sensing image sequence non-cooperative ship target set of the stationary orbit staring satellite to obtain a stationary orbit staring satellite remote sensing image sequence non-cooperative ship target intermittent track association result;
and processing the intermittent track correlation result of the stationary orbit staring satellite remote sensing image sequence cooperation ship target and the intermittent track correlation result of the stationary orbit staring satellite remote sensing image sequence non-cooperation ship target by utilizing polynomial fitting to obtain a stationary orbit staring satellite remote sensing image sequence interruption track connection.
Optionally, the method may be characterized in that,
the AIS ship target track set is
T i Representing the target track of the ith AIS ship, and Q representing the number of target tracks provided by AIS data in the imaging observation time period of the stationary orbit staring satellite;
the staring satellite remote sensing image sequence of the stationary orbit is integrated with the naval vessel target track
T n Representing a certain ship target track, wherein N respectively represents the number of ship target tracks observed by the imaging of the stationary orbit staring satellite;
Optionally, the method includes determining, according to the AIS naval target track set, that a certain track in the first remote sensing image sequence naval target track set of the stationary orbit staring satellite and a certain track in the second remote sensing image sequence naval target track set of the stationary orbit staring satellite are intermittent tracks of the same cooperative naval target, and obtaining an intermittent track association result of the stationary orbit staring satellite remote sensing image sequence cooperative naval target, and specifically includes:
performing space-time alignment on the ship target track set of the stationary orbit staring satellite remote sensing image sequence and the AIS ship target track set A to realize time alignment between the ship target track of the stationary orbit staring satellite image and the AIS ship target track set A;
finding AIS ship target point trace set A at k moment by using iteration closest point method k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
And staring at the satellite second remote sensing image sequence ship target point trace set relative to the stationary orbit
New point set after rigid body transformation
Obtaining a new point set after the AIS ship target rigid body transformation by using an optimal correlation method
Staring at the first remote sensing image sequence ship target point trace set of the satellite with the stationary orbit
The first point trace correlation result and the new point set after the AIS ship target rigid body transformation
A ship target point trace set of a second remote sensing image sequence staring at the satellite on the stationary orbit
Second trace point association result of (2);
respectively checking the first track association result and the second track association result by using 3/5 criteria, eliminating error association and obtaining a corresponding first track relationResult of the combination phi 1 And the second track correlation result phi 2 ；
If the same AIS ship target track T exists i Satisfies (T) i ,T n )∈Φ 1 And (T) i ,T m )∈Φ 2 Target track T of ship n And T m The method is an intermittent track of the same target, so that intermittent track association of stationary orbit staring satellite remote sensing image sequences and ship targets is realized.
Optionally, an iteration closest point method is used for searching the AIS ship target point trace set A at the moment k k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
Finding AIS ship target point trace set A at the k moment by using an iteration closest point method k Staring a second remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
In the same manner as in (B) k Included
And
new point set A after rigid body transformation k* Comprises a first component and a second component
Corresponding to
And with
Corresponding to
Then the AIS ship target point trace set A at the k moment is searched by utilizing an iterative closest point method k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
And staring at the target point trace set of the ship in the second remote sensing image sequence of the satellite relative to the stationary orbit
New point set after rigid body transformation
The method specifically comprises the following steps:
(1) let the AIS ship target point trace set at the time k be expressed as
The set of target points of the ship in the satellite remote sensing image at the moment k is expressed as
Wherein:
for the course estimation of the ship target,
navigation for naval vessel targetsEstimating the speed;
(2) at point set A k Selecting an initial set of points
Wherein A is k Representing an AIS ship target point trace set at the time k;
Wherein R is earth Is the average earth radius;
wherein, Delta theta is the course, and Delta s is the speed threshold;
(3) from A k(r-1) Point of (5)
In B k Searching out its nearest point
Forming a point pair, finding out all point pairs in two point sets, forming a point pair set, and sharing N point pairs;
(4) according to the point pair set, calculating a rotation matrix under the weighted least square
And translation vector
Wherein the content of the first and second substances,
w (-) is a weighting function;
(5) according to
And
computing
Get the point set A k(r-1) New point set A after rigid body transformation k(r) ；
(6) Repeating the iteration process from the step (2) to the step (4) until the convergence condition is met or the preset iteration number is reached, and outputting a new point set after rigid body transformation
The convergence condition is
e (r) Convergence error at the r-th iteration, e (r-1) Is the convergence error at the r-1 st iteration, ε e Is a threshold value.
Optionally, obtaining a new point set after the AIS ship target rigid body transformation by using an optimal association method
A ship target point trace set of a first remote sensing image sequence staring at the satellite on the stationary orbit
The first point trace correlation result and the optimal correlation method are utilized to obtain a new point set after the AIS ship target rigid body transformation
Staring at a second remote sensing image sequence ship target point trace set of the satellite with the stationary orbit
The method of the second trace point correlation result is the same, and the corresponding process of the method specifically comprises the following steps:
obtaining the new point set after the rigid body transformation
According to
Obtaining a value of a two-dimensional distribution variable which minimizes the weighted sum of the distribution costs;
definition a in Satisfy the requirement of
To obtain a in The result of (1); a is a in ＝{0,1}，a in 1 represents two ship target point trace associations, otherwise two ship orders are representedAnd (4) the punctuation marks are not related, namely the first punctuation mark correlation result and the second punctuation mark correlation result are obtained.
Optionally, the associating the non-cooperative ship target set of the first remote sensing image sequence of the stationary orbit staring satellite with the non-cooperative ship target set of the second remote sensing image sequence of the stationary orbit staring satellite to obtain the discontinuous track association result of the non-cooperative ship target of the remote sensing image sequence of the stationary orbit staring satellite specifically includes:
setting up
A non-cooperative ship target track set in a first remote sensing image sequence of the staring satellite at the stationary orbit is used,
for the non-cooperative ship target track set in the second remote sensing image sequence of the stationary orbit staring satellite, the first imaging time is before the second imaging time, namely A 1 ' is an old track set, A 2 ' is a new track set; q s The number of successful associations in the association result of the intermittent track of the satellite target cooperated with the remote sensing image sequence of the fixed orbit staring satellite; setting all possible flight path combinations at the moment k as phi;
the state estimation of the old track at the k moment is obtained by utilizing the backward extrapolation of the old track set
Obtaining the state estimation of the new track at the k moment by utilizing the forward extrapolation of the new track set as
Further obtain
Estimating the state of old flight path at the moment
The state of the new track is estimated as
Obtaining a preliminary association result phi of the old track set and the new track set according to a preliminary association constraint condition v ；
Further associating the flight path segments by adopting a hypothesis testing method to obtain a further association set phi h ；
According to the further association set phi h Obtaining a first-stage coarse correlation result phi by using an optimal track correlation method w ；
According to the preliminary correlation result phi v And said first stage coarse correlation result phi w Carrying out second-stage fine correlation by utilizing the motion information and the amplitude information of the target to obtain phi c ；
Merging correlation results totals of phi w And phi c And obtaining a non-cooperative ship target track correlation result of the remote sensing image sequence of the stationary orbit staring satellite.
Optionally, the preliminary association constraint condition is:
wherein s is th Representing the speed, theta th Indicating heading, p th Representing a location association threshold;
wherein the content of the first and second substances,
for the position change caused by the speed change,
for position change due to direction change, s mean Average velocity for vessel targetsAnd Δ T is the interrupt time interval.
Optionally, the hypothesis testing method is adopted to further correlate the track segments to obtain a further correlation set Φ h The method specifically comprises the following steps:
setting:
H 0 :
and
the state estimation of new and old tracks of the same target at the same moment;
H 1 :
and
new and old tracks not belonging to the same target;
at H 0 In the step (1), the first step,
old track of time T i With new track T j The estimation error of (c) is:
covariance of the corresponding error is
Under the confidence of 1-q, the further associated set of the flight paths is as follows:
wherein, the first and the second end of the pipe are connected with each other,
obey n x Chi of degree of freedom 2 Distribution, n x Is the dimension of the state vector.
Optionally, the set Φ is further related according to h Obtaining a first-stage coarse correlation result phi by using an optimal track correlation method w The method specifically comprises the following steps:
according to the
Old track of time T i With new track T j Determines the associated cost function as:
determining a two-dimensional distribution variable which minimizes the weighted sum of the distribution costs by using an optimal track association method:
wherein, b ij Satisfy the requirement of
b ij ＝{0,1}，b ij 1 means that two track segments are associated, otherwise they are not relevant;
using phi w ＝{(T i ,T j )∈Φ h ,(T i ,T j ):b ij 1, obtaining the first-stage coarse correlation result phi w 。
Optionally, the correlation result Φ is obtained according to the preliminary correlation v And said first stage coarse correlation result phi w The second stage is carried out by using the motion information and amplitude information of the objectFine correlation to obtain phi c The method specifically comprises the following steps:
setting the first stage coarse correlation result phi w In the step (1), the first step,
for the magnitude of the new track in the ith associated track pair,
the amplitude of the old track in the ith associated track pair is the average value of the target amplitude at each moment under current observation;
obtaining the relation between the amplitude of the new track in the ith associated track pair and the amplitude of the old track in the ith associated track pair by utilizing a linear regression relation
Wherein epsilon is a target amplitude error, and is subjected to Gaussian distribution with a mean value of zero and a standard deviation of sigma;
obtaining linear regression model and error standard deviation estimated value by least square estimation
Wherein K is phi a The number of medium ship association combinations;
according to the preliminary correlation result phi v And said first stage coarse correlation result phi w Using a correlation model
Eliminating error correlation pairs through amplitude relation detection;
detecting the effectiveness of the correlation by adopting a standard deviation criterion of 3 times, wherein the correlation is effective when the amplitude prediction error is less than 3 times of the standard deviation, otherwise, the correlation is regarded as abnormal correlation deviating from linearity and is eliminated, so that the correlation result of the ship target is obtained
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: the invention introduces the information of the ship automatic identification system, firstly realizes the association of the cooperative ship target, effectively reduces the number of associated matching combinations, then divides the association into two association stages, and associates the non-cooperative targets, thereby not only reducing the number of matching combinations in the association algorithm and improving the speed of ship target track association, but also overcoming the influence of low resolution on association, reducing the association complexity, improving the association accuracy, finally connecting the intermittent tracks on the association and improving the integrity of ship target situation display.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a method for correlating a ship target track in a remote sensing image of a stationary orbit staring satellite provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," "third," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the objects so described are interchangeable under appropriate circumstances. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions.
In this patent document, the drawings discussed below and the embodiments used to describe the principles of the present disclosure are by way of illustration only and should not be construed in any way to limit the scope of the present disclosure. Those skilled in the art will understand that the principles of the invention may be implemented in any suitably arranged system. Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. Further, a terminal according to an exemplary embodiment will be described in detail with reference to the accompanying drawings. Like reference symbols in the various drawings indicate like elements.
The terms used in the present specification are only used to describe specific embodiments, and are not intended to show the concept of the present invention. Unless the context clearly dictates otherwise, expressions used in the singular form encompass expressions in the plural form. In the present specification, it is to be understood that terms such as "comprising," "having," and "containing" are intended to specify the presence of stated features, integers, steps, acts, or combinations thereof, as taught in the present specification, and are not intended to preclude the presence or addition of one or more other features, integers, steps, acts, or combinations thereof. Like reference symbols in the various drawings indicate like elements.
The invention aims to provide a method for correlating a ship target track of a remote sensing image of a stationary orbit staring satellite, which solves the problem of intermittent track correlation of the ship target under the conditions of medium-low resolution and low frame rate, reduces the correlation complexity and improves the correlation efficiency and accuracy by respectively correlating a cooperative target and a non-cooperative target.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a schematic flow diagram of a method for correlating a ship target track in a remote sensing image of a stationary orbit staring satellite, which is provided by the invention, and the method comprises the following steps:
step 101: acquiring a ship target track set of a stationary orbit staring satellite remote sensing image sequence and an AIS ship target track set provided by a ship automatic identification system in an imaging observation time period corresponding to the stationary orbit staring satellite remote sensing image sequence; the satellite remote sensing image sequence ship target track set of the stationary orbit staring comprises the following steps: the method comprises the steps of staring at a satellite at a stationary orbit for a first remote sensing image sequence naval target track set and staring at a satellite at a stationary orbit for a second remote sensing image sequence naval target track set.
Step 102: and determining that a certain track in the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a certain track in the second remote sensing image sequence ship target track set of the stationary orbit staring satellite are intermittent tracks of the same cooperative ship target according to the AIS ship target track set, and obtaining a stationary orbit staring satellite remote sensing image sequence cooperative ship target intermittent track association result.
Step 103: and screening out a non-cooperative ship target set of the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a non-cooperative ship target set of the second remote sensing image sequence ship target track set of the stationary orbit staring satellite according to the intermittent track correlation result of the cooperative ship target.
Step 104: and associating the first remote sensing image sequence non-cooperative ship target set of the stationary orbit staring satellite with the second remote sensing image sequence non-cooperative ship target set of the stationary orbit staring satellite to obtain a stationary orbit staring satellite remote sensing image sequence non-cooperative ship target intermittent track association result.
Step 105: and processing the intermittent track correlation result of the stationary orbit staring satellite remote sensing image sequence cooperation ship target and the intermittent track correlation result of the stationary orbit staring satellite remote sensing image sequence non-cooperation ship target by utilizing polynomial fitting to obtain a stationary orbit staring satellite remote sensing image sequence interruption track connection.
The invention introduces the Automatic Identification System (AIS) information of the ship, firstly realizes the association of the cooperative ship target, effectively reduces the number of associated matching combinations, then associates the non-cooperative target in two association stages, not only reduces the number of matching combinations in an association algorithm and improves the speed of ship target track association, but also overcomes the influence of the conditions of medium and low resolution and low frame rate on the intermittent track association of the ship target, reduces the association complexity, improves the association accuracy, finally connects the intermittent tracks on the association and improves the integrity of ship target situation display.
The remote sensing image sequence obtained by the single staring imaging observation of the stationary orbit staring satellite is processed by a ship target detection and tracking algorithm, and the tracks of the cooperative ship target and the non-cooperative ship target can be obtained.
The ship automatic identification system can provide static data (such as ship name, type, size and the like) and dynamic data (such as position, course to ground, speed to ground and the like) of the cooperative ship targets, and a certain number of cooperative ship targets exist in adjacent two imaging areas under a maneuvering inspection mode due to the wide coverage range of a single-shot remote sensing image of a stationary orbit staring satellite. According to an RPC model (remote sensing image geometric correction model) of a stationary orbit staring satellite remote sensing image, calculating the latitude and longitude coordinate range of a remote sensing image sequence imaging area observed by two times of imaging, selecting AIS data covering the range of the two times of imaging observation area and the imaging time period, and screening a common cooperation ship target of the first time of imaging observation and the second time of imaging observation according to ship identity information such as ship name, MMSI and the like provided by the AIS data.
AIS ship target track setIs composed of
T i The target track of the ith ship is shown, and Q represents the number of target tracks provided by AIS data in the observation period of imaging of the stationary orbit staring satellite.
T i The target track of the ith AIS ship is shown, and Q represents the number of target tracks provided by AIS data in the observation period of imaging of the stationary orbit staring satellite.
For a stationary orbit staring satellite remote sensing image sequence
T n And N respectively represents the number of the ship target tracks observed by the stationary orbit staring satellite in the imaging.
And (3) inputting the detection result of the remote sensing image sequence of the stationary orbit satellite into the multi-hypothesis tracker, and outputting the state estimation result of the ship target.
For the tracking of a staring satellite in a stationary orbit, the obtained ship target track n can be described as
Wherein the content of the first and second substances,
and
for the first and last state update times of the ship target track n,
as a result of estimating the motion state of the ship target at the time k,
and
respectively representing positions in the latitude and longitude directions,
which represents the course of the heading,
the basic units representing speed, latitude, longitude and heading are degrees (°), and the unit of speed is knots (kn).
To improve the accuracy of the target track estimation, the state estimation result of the target is used
Observation result for replacing ship target position
Namely that
The estimated covariance matrix is denoted P n (k | k). While a heading estimate of the target may be obtained
And speed estimation
The method for obtaining the intermittent track correlation result of the remote sensing image sequence of the stationary orbit staring satellite cooperating with the ship target by using the remote sensing image sequence measured by the stationary orbit staring satellite twice specifically comprises the following steps:
(1) and performing space-time alignment on the ship target track set of the stationary orbit staring satellite remote sensing image sequence and the AIS ship target track set A, so as to realize time alignment between the ship target track of the stationary orbit staring satellite image and the AIS ship target track set A.
(2) Finding AIS ship target point trace set A at k moment by using iteration closest point method k Staring a first remote sensing image sequence ship target point trace set of satellite relative to the stationary orbit
New point set after rigid body transformation
And staring at the target point trace set of the ship in the second remote sensing image sequence of the satellite relative to the stationary orbit
New point set after rigid body transformation
(3) Obtaining the new point set after the AIS ship target rigid body transformation by using an optimal correlation method
Staring at the first remote sensing image sequence ship target point trace set of the satellite with the stationary orbit
The first point trace correlation result and the new point set after the AIS ship target rigid body transformation
A ship target point trace set of a second remote sensing image sequence staring at the satellite on the stationary orbit
And (3) associating the results with the second trace.
(4) Respectively checking the first point track correlation result and the second point track correlation result by using 3/5 criteria, eliminating error correlation and obtaining a corresponding first track correlation result phi 1 And said second track correlation result Φ 2 。
(5) If the same AIS ship target track T exists i Satisfies (T) i ,T n )∈Φ 1 And (T) i ,T m )∈Φ 2 Target track T of ship n And T m The method is an intermittent track of the same target, so that intermittent track association of stationary orbit staring satellite remote sensing image sequences and ship targets is realized.
For a given stationary orbital staring satellite image, spatio-temporal filtering is required to select those AIS messages that are useful for data fusion. The AIS data may be organized based on the ship's Marine Mobile Service Identity (MMSI) number. While AIS data is usually continuously generated, while stationary orbit staring satellite images are acquired at fixed intervals, so the AIS track needs to be projected to the satellite image acquisition time, thereby achieving time alignment between the ship target track of the satellite image and the AIS ship target track. Therefore, space-time alignment is needed before the association of the cooperative ship target, and the time alignment between the ship target track of the stationary orbit staring satellite image and the ship target track set A is realized. The method is to linearly interpolate or extrapolate the AIS data to the acquisition time of the satellite images. In particular, the use of Near Infrared (NIR) bands for ship target detection requires a time delay of about 40s between the NIR band and the acquisition time.
For AIS, the errors are mainly GPS positioning errors (less than 100m) and interpolation errors. The AIS location may be considered a true target location because the error is small. However, since the GEO optical satellite is far from the earth and there is systematic bias of the RPC, the positioning error of the satellite image will be an order of magnitude greater than the AIS. In order to solve the problems, the invention converts the track association into the point pattern matching problem for processing, and uses an iterative Closest point ICP (iterative Closest point) method to search the rigid transformation relation between the AIS ship target track and the ship target track of the stationary orbit staring satellite remote sensing image sequence.
The invention utilizes ICP to search AIS ship target point trace set A at the k moment k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
And staring at the target point trace set of the ship in the second remote sensing image sequence of the satellite relative to the stationary orbit
New point set after rigid body transformation
In the method, the AIS ship target point trace set A at the k moment is searched by utilizing an iteration closest point method k Relative to the stationary railTrack staring satellite first remote sensing image sequence ship target point trace set
New point set after rigid body transformation
And searching the AIS ship target point trace set A at the k moment by using an iterative closest point method k Staring at a ship target point trace set of a second remote sensing image sequence of the satellite relative to the stationary orbit
New point set after rigid body transformation
B is set in the same way k Included
And
new point set A after rigid body transformation k* Comprises and
corresponding to
And with
Corresponding to
The method specifically comprises the following steps:
(1) let the AIS ship target point trace set at the k moment be expressed as
The set of target points of the ship in the satellite remote sensing image at the moment k is expressed as
Wherein:
is an estimate of the heading of the vessel target,
a speed estimate for the ship target;
(2) at point set A k Selecting an initial set of points
Wherein A is k Representing an AIS ship target point trace set at the k moment;
for the AIS ship target point trace data at the time k,
satisfy the requirement of
Δ p is a position threshold value for determining correlation, and Q' is the number of the initial point sets;
is specifically calculated by
Wherein R is earth Is the average earth radius.
wherein, Δ θ is the heading and Δ s is the speed threshold.
(3) From A k(r-1) Point of (1)
In B k Searching out its nearest point
Forming a point pair, finding out all point pairs in two point sets, forming a point pair set, and having N point pairs in total.
(4) According to the point pair set, calculating a rotation matrix under the weighted least square
And translation vector
Wherein the content of the first and second substances,
w (-) is a weighting function.
(5) According to
And
computing
Get the point set A k(r-1) New point set A after rigid body transformation k(r) 。
(6) Repeating the iteration process from the step (2) to the step (4) until the convergence condition is met or the preset iteration times are reached, and outputting a new point set after rigid body transformation
The convergence condition is
e (r) Convergence error at the r-th iteration, e (r-1) Is the convergence error at the r-1 st iteration, ε e Is a threshold value.
Since the ICP method cannot guarantee a one-to-one correspondence between the tracks, correlation errors may be caused. Therefore, the new point set after the AIS ship target rigid body transformation is obtained by using an optimal correlation method
Staring at the first remote sensing image sequence ship target point trace set of the satellite with the stationary orbit
The first point trace correlation result and the new point set after the AIS ship target rigid body transformation
Staring at a second remote sensing image sequence ship target point trace set of the satellite with the stationary orbit
The second trace point association result specifically includes:
obtaining the new point set after the rigid body transformation
According to
And obtaining a value of the two-dimensional distribution variable which minimizes the weighted sum of the distribution costs.
Limit a in Satisfy the requirement of
To obtain a in The result of (1); a is in ＝{0,1}，a in And (2) representing that the target point traces of the two ships are associated with each other, otherwise, representing that the target point traces of the two ships are not associated with each other, and obtaining the first point trace association result and the second point trace association result.
To further derive the track correlation results, criteria 3/5 is employed to check whether a correlation is maintained on the continuous track, thereby eliminating false correlations and deriving a track correlation result between the AIS ship target and the stationary orbit staring satellite ship target.
The stationary orbit staring satellite has few sampling points, low spatial resolution and short continuous observation time, and is easily shielded by cloud layers, so that the condition that the ship target track observed by the stationary orbit staring satellite is interrupted is easily caused. However, AIS data has a large number of sampling points and is generally continuous in time domain distribution. Therefore, the target track sets in the two imaging observation time periods of the geostationary orbit staring satellite are respectively
And
if the same AIS ship target track T exists i The target track T of the stationary orbit staring satellite ship at different imaging time periods n ∈B 1 And T m ∈B 2 If the correlation relationship is satisfied, the ship target track T can be judged n And T m Is a break of the same objectAnd (4) continuing the track, thereby realizing the intermittent track association of the stationary orbit staring satellite remote sensing image sequence cooperation ship target.
The unassociated target in the target track set of the staring satellite in the stationary orbit is regarded as the non-cooperative ship target track set, and the non-cooperative ship target track set is set
For a set of non-cooperative vessel target tracks in the sequence of first remote sensing images of the stationary orbiting gaze satellite,
for the non-cooperative ship target track set in the second remote sensing image sequence of the stationary orbit staring satellite, the first imaging time is before the second imaging time, namely A 1 ' is an old track set, A 2 ' is a new track set; q s The number of successful associations in the association result of the intermittent track of the ship target is cooperated for the remote sensing image sequence of the stationary orbit staring satellite; setting all possible flight path combinations at the moment k as phi;
obtaining the state estimation of old flight path at the k moment by using the backward extrapolation of the old flight path set as
Obtaining the state estimation of the new track at the k moment by utilizing the forward extrapolation of the new track set as
Further obtain
The state of the old track at the moment is estimated as
The state of the new track is estimated as
Suppose an old flight path
And new flight path
Two tracks of the same ship target need to meet certain constraint conditions in speed, direction and distance. By setting rough constraints, a preliminary association combination can be obtained:
wherein s is th Representing the speed, theta th Indicating heading, p th Representing a location association threshold;
wherein the content of the first and second substances,
for the position change caused by the speed change,
for positional changes due to changes in direction, s mean Δ T is the interruption time interval for the average velocity of the ship target.
In practical application, corresponding threshold values need to be set according to the maneuvering condition of the target in the monitoring area. In the present invention, s is set th 、s mean 20kn and 10kn, theta, respectively th Is 60 degrees.
Further associating the flight path segments by adopting a hypothesis testing method to obtain a further association set phi h . As the estimation error statistics of the track state of the same target at the same moment are independent and obey Gaussian distribution, the method can adopt Chi 2 The distributed hypothesis testing method is further associated with the track segment. Let H 0 ，H 1 Two respectively indicating whether track segments are related or notSuppose, H 0 And H 1 Is defined as:
H 0 :
and
is the state estimation of the new and old tracks of the same target at the same moment.
H 1 :
And
new and old flight paths which do not belong to the same target.
At H 0 In (1),
old track of time T i With new track T j The estimation error of (c) is:
covariance of the corresponding error is
Under the confidence of 1-q, the further associated set of the flight paths is as follows:
wherein, the first and the second end of the pipe are connected with each other,
obey n x Chi of degree of freedom 2 Distribution, n x Is the dimension of the state vector.
According to said stepAssociation set phi h Obtaining a first-stage coarse correlation result phi by using an optimal track correlation method w . According to the above
Old track of time T i With new track T j Determines the associated cost function as:
determining a two-dimensional distribution variable which minimizes the weighted sum of distribution costs by using an optimal track association method:
wherein, b ij Satisfy the requirement of
b ij ＝{0,1}，b ij 1 means that two track segments are associated, otherwise not relevant;
by using phi w ＝{(T i ,T j )∈Φ h ,(T i ,T j ):b ij 1, obtaining the first-stage coarse correlation result phi w 。
And the hypothesis test is carried out on the motion state of the target, and the method belongs to a conservative association strategy, so that the reliability of track segment association can be ensured. And under a long time interval, part of ship targets can show some maneuverability, such as turning, accelerating and the like, if the rest tracks are associated only by using the motion state, some tracks can be wrongly associated, so that the invention utilizes the motion information and the amplitude information of the targets to carry out second-stage precise association. According to the preliminary correlation result phi v And said first stage coarse correlation result phi w Performing second-stage fine correlation by using the motion information and the amplitude information of the target to obtain phi c 。
(1) Amplitude relation solving
Under the influence of observation time, the conditions such as illumination and the like in two times of observation are different, so that the target amplitude of the same target in the new and old tracks is different, but has certain correlation. Under ideal conditions, the amplitude of the image can be adjusted by adopting an image histogram registration mode, namely relative radiation correction, so that the invariance of the amplitude of the same ship target in two times of observation is realized. However, the histogram matching is an integral matching, and the proportion of the sea surface background in the image is large, so that the ship target occupies fewer pixels, which causes non-uniformity during histogram matching and inaccurate ship target amplitude correction. Therefore, the invention directly utilizes the ship target pair which is related in the front and adopts a linear regression model to analyze the corresponding relation of the ship target amplitude under two observations. Setting the ship target amplitude set meeting the correlation condition as
Wherein the content of the first and second substances,
and
respectively representing the amplitudes of the new track and the old track in the ith associated track pair, wherein the amplitudes are the average value of the target amplitudes at all times under current observation.
Assuming that the relationship between the target amplitudes under two observations satisfies a linear regression, i.e.
Where ε is the target amplitude error, which is assumed to follow a Gaussian distribution with a mean of zero and a standard deviation of σ.
Obtaining linear regression models and error standard deviation estimated values by least square estimation through the amplitude of the correlated track segment pairs
Wherein K is phi a And the number of the medium ship association combinations.
(2) Amplitude relationship anomaly detection
Preliminary association combination Φ for coarse association in the first stage v The correlation result phi in, but not in, the first stage w The flight path segment in (1) is mainly a target with stronger maneuverability, and in order to further improve the confidence degree of flight path association, the association result
And eliminating error correlation pairs through amplitude relation detection. To measure whether the associated track segment pairs are anomalous, i.e. obey a linear regression equation.
In the invention, a standard deviation criterion of 3 times is adopted, when the amplitude prediction error is less than 3 times of standard deviation, the correlation is considered to be effective, namely the correct correlation track segment pair is obtained, otherwise, the correlation is considered to be abnormal correlation deviating from linearity, and the abnormal correlation is removed, thereby obtaining the ship target correlation result
Combining the correlation results to total phi w And phi c And obtaining a non-cooperative ship target track correlation result of the remote sensing image sequence of the stationary orbit staring satellite.
And processing the intermittent track correlation result of the stationary orbit staring satellite remote sensing image sequence cooperation ship target and the intermittent track correlation result of the stationary orbit staring satellite remote sensing image sequence non-cooperation ship target by utilizing polynomial fitting to obtain a stationary orbit staring satellite remote sensing image sequence interruption track connection.
In order to improve the integrity and continuity of tracks in a monitoring area, the tracks are required to be connected on the basis of track segment association, polynomial fitting is adopted to connect track segments meeting the association relation, and AIS ship position information is used for position information in new and old track segments for a cooperative target during polynomial fitting; for non-cooperative targets, estimated ship location information is used.
Between two observations of a stationary orbit satellite, the motion state of a ship target may change, causing the ship position to deviate from equiangular course travel. In order to smoothly connect the old track and the new track, the time is used as an independent variable, and N-order polynomials are adopted to respectively fit the position state estimated values of longitude and latitude direction axes, so that the track of the target in the whole time period is obtained.
When the fitting order is larger, the fitting curve has larger volatility and is not consistent with the target motion state; when the fitting order is small, such as 1 st order straight line fitting, it is difficult to accurately describe a complex motion state. According to the general motion situation of the ship target in practice, 3-order fitting is selected in the invention.
The principle and the implementation mode of the invention are explained by applying a specific example, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (9)
1. A method for correlating ship target tracks of a remote sensing image of a staring satellite at a stationary orbit is characterized by comprising the following steps:
acquiring a ship target track set of a fixed-orbit staring satellite remote sensing image sequence and an AIS ship target track set provided by a ship automatic identification system in an imaging observation time period corresponding to the fixed-orbit staring satellite remote sensing image sequence; the ship target track set of the stationary orbit staring satellite remote sensing image sequence comprises the following steps: a first remote sensing image sequence ship target track set of a staring satellite at a stationary orbit and a second remote sensing image sequence ship target track set of the staring satellite at the stationary orbit;
determining that a certain track in the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a certain track in the second remote sensing image sequence ship target track set of the stationary orbit staring satellite are intermittent tracks of the same cooperative ship target according to the AIS ship target track set to obtain a stationary orbit staring satellite remote sensing image sequence cooperative ship target intermittent track association result;
screening out a non-cooperative ship target set of the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a non-cooperative ship target set of the second remote sensing image sequence ship target track set of the stationary orbit staring satellite according to the intermittent track correlation result of the cooperative ship target;
the method specifically comprises the following steps of associating a non-cooperative ship target set of a first remote sensing image sequence of a stationary orbit staring satellite with a non-cooperative ship target set of a second remote sensing image sequence of a stationary orbit staring satellite to obtain an intermittent track association result of the non-cooperative ship target of the remote sensing image sequence of the stationary orbit staring satellite, wherein the method specifically comprises the following steps:
setting up
For a set of non-cooperative vessel target tracks in the sequence of first remote sensing images of the stationary orbiting gaze satellite,
for the non-cooperative ship target track set in the second remote sensing image sequence of the stationary orbit staring satellite, the first imaging time is before the second imaging time, namely A 1 ' is an old track set, A 2 ' is a new track set; q s The number of successful associations in the association result of the intermittent track of the satellite target cooperated with the remote sensing image sequence of the fixed orbit staring satellite; setting all possible flight path combinations at the moment k as phi;
obtaining the state estimation of old flight path at the k moment by using the backward extrapolation of the old flight path set as
Obtaining the k time of the new track by using the forward extrapolation of the new track setThe state is estimated as
Further obtain
The state of the old track at the moment is estimated as
The state of the new track is estimated as
Obtaining a preliminary association result phi of the old track set and the new track set according to a preliminary association constraint condition v ；
Adopting a hypothesis test method to further correlate the track segments to obtain a further correlation set phi h ；
According to the further association set phi h Obtaining a first-stage coarse correlation result phi by using an optimal track correlation method w ；
According to the preliminary correlation result phi v And said first stage coarse correlation result Φ w Carrying out second-stage fine correlation by utilizing the motion information and the amplitude information of the target to obtain phi c ；
Merging correlation results totals of phi w And phi c Obtaining a non-cooperative ship target track correlation result of the stationary orbit staring satellite remote sensing image sequence;
and processing the discontinuous track correlation result of the stationary orbit staring satellite remote sensing image sequence cooperation naval vessel target and the discontinuous track correlation result of the stationary orbit staring satellite remote sensing image sequence non-cooperation naval vessel target by utilizing polynomial fitting to obtain the stationary orbit staring satellite remote sensing image sequence interruption track connection.
2. The method for correlating ship target tracks on stationary orbit staring satellite remote sensing images according to claim 1,
the AIS ship target track set is
T i Representing the ith AIS ship target track, and Q representing the number of target tracks provided by AIS data in a stationary orbit staring satellite imaging observation time period;
the staring satellite remote sensing image sequence of the stationary orbit is integrated as follows
T n Representing a certain ship target track, wherein N respectively represents the number of ship target tracks observed by the imaging of the stationary orbit staring satellite;
3. The method for correlating ship target tracks based on remote sensing image of stationary orbit staring satellite according to claim 2, wherein the step of determining that a certain track in the first remote sensing image sequence ship target track set of the stationary orbit staring satellite and a certain track in the second remote sensing image sequence ship target track set of the stationary orbit staring satellite are intermittent tracks of the same cooperative ship target according to the AIS ship target track set to obtain intermittent track correlation results of the remote sensing image sequence cooperative ship target of the stationary orbit staring satellite specifically comprises the steps of:
performing space-time alignment on the ship target track set of the stationary orbit staring satellite remote sensing image sequence and the AIS ship target track set A to realize time alignment between the ship target track of the stationary orbit staring satellite image and the AIS ship target track set A;
finding AIS ship target point trace set A at k moment by using iteration closest point method k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
And staring at the satellite second remote sensing image sequence ship target point trace set relative to the stationary orbit
New point set after rigid body transformation
Obtaining a new point set after the AIS ship target rigid body transformation by using an optimal correlation method
A ship target point trace set of a first remote sensing image sequence staring at the satellite on the stationary orbit
The first point trace correlation result and the new point set after the AIS ship target rigid body transformation
A ship target point trace set of a second remote sensing image sequence staring at the satellite on the stationary orbit
Second trace point association result of (2);
respectively checking the first point track correlation result and the second point track correlation result by using 3/5 criteria, eliminating error correlation and obtaining a corresponding first track correlation result phi 1 And second track correlation result phi 2 ；
If the same AIS ship target track T exists i Satisfies (T) i ,T n )∈Φ 1 And (T) i ,T m )∈Φ 2 Target track T of ship n And T m The method is an intermittent track of the same target, so that intermittent track association of stationary orbit staring satellite remote sensing image sequences and ship targets is realized.
4. The method for correlating ship target tracks on stationary orbit staring satellite remote sensing images according to claim 3,
finding AIS ship target point trace set A at k moment by using iteration closest point method k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
And searching the AIS ship target point trace set A at the k moment by using an iterative closest point method k Staring a second remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
New point set after rigid body transformation
In the same manner as in (B) k Included
And
new point set A after rigid body transformation k* Comprises a first component and a second component
Corresponding to
And with
Corresponding to
Then the AIS ship target point trace set A at the k moment is searched by utilizing an iterative closest point method k Staring a first remote sensing image sequence ship target point trace set of the satellite relative to the stationary orbit
After rigid body transformationPoint set
And staring at the target point trace set of the ship in the second remote sensing image sequence of the satellite relative to the stationary orbit
New point set after rigid body transformation
The method specifically comprises the following steps:
(1) let the AIS ship target point trace set at the time k be expressed as
The set of target points of the ship in the satellite remote sensing image at the moment k is expressed as
Wherein:
for the AIS ship target point trace data at the time k,
(2) at point set A k Selecting an initial set of points
Wherein A is k Representing an AIS ship target point trace set at the time k;
wherein, Delta theta is the course, and Delta s is the speed threshold;
(3) from A k(r-1) Point of (1)
In B k Searching out its nearest point
Forming a point pair, finding out all point pairs in two point sets, forming a point pair set, and sharing N point pairs;
(4) according to the point pair set, calculating a rotation matrix under the weighted least square
And translation vector
Wherein, the first and the second end of the pipe are connected with each other,
w (-) is a weighting function;
(5) according to
And
computing
Obtain point set A k(r-1) New point set A after rigid body transformation k (r) ；
(6) Repeating the iteration process from the step (2) to the step (4) until the convergence condition is met or the preset iteration times are reached, and outputting a new point set after rigid body transformation
The convergence condition is
e (r) Convergence error at the r-th iteration, e (r-1) Is the convergence error, ε, at the r-1 st iteration e Is a threshold value.
5. The method for correlating ship target tracks on stationary orbit staring satellite remote sensing images according to claim 4,
obtaining the new point set after the AIS ship target rigid body transformation by using an optimal correlation method
A ship target point trace set of a first remote sensing image sequence staring at the satellite on the stationary orbit
The first point trace correlation result and the optimal correlation method are utilized to obtain a new point set after the AIS ship target rigid body transformation
A ship target point trace set of a second remote sensing image sequence staring at the satellite on the stationary orbit
The method of the second trace point correlation result is the same, and the corresponding process of the method specifically comprises the following steps:
obtaining the new point set after the rigid body transformation
According to
Obtaining a value of a two-dimensional distribution variable which minimizes the weighted sum of the distribution costs;
limit a in Satisfy the requirement of
To obtain a in The result of (1); a is a in ＝{0,1}，a in And (2) representing that the target point traces of the two ships are associated with each other, otherwise, representing that the target point traces of the two ships are not associated with each other, and obtaining the first point trace association result and the second point trace association result.
6. The method for correlating ship target tracks on stationary orbit staring satellite remote sensing images according to claim 1, wherein the preliminary correlation constraint conditions are as follows:
wherein s is th Representing the speed, theta th Indicating heading, p th Representing a location association threshold;
wherein the content of the first and second substances,
for the position change caused by the speed change,
for positional changes due to changes in direction, s mean Δ T is the interruption time interval for the average speed of the ship target.
7. The stationary orbit staring satellite remote sensing image ship target track gate as claimed in claim 1The association method is characterized in that the hypothesis test method is adopted to further associate the track segments to obtain a further association set phi h The method specifically comprises the following steps:
setting:
at H 0 In the step (1), the first step,
old track of time T i With new track T j The estimation error of (2) is:
covariance of corresponding error of
At a confidence of 1-q, the set of further correlations of the flight path is:
wherein, the first and the second end of the pipe are connected with each other,
obey n x Chi of degree of freedom 2 Distribution, n x Is the dimension of the state vector.
8. The method for correlating ship target tracks on stationary orbit staring satellite remote sensing images according to claim 1, wherein the method is characterized in that according to the further correlation set Φ h Obtaining a first-stage coarse correlation result phi by using an optimal track correlation method w The method specifically comprises the following steps:
according to the above
Old track of time T i With new track T j Determines the associated cost function as:
determining a two-dimensional distribution variable which minimizes the weighted sum of the distribution costs by using an optimal track association method:
wherein, b ij Satisfy the requirements of
b ij ＝{0,1}，b ij 1 means that two track segments are associated, otherwise they are not relevant;
using phi w ＝{(T i ,T j )∈Φ h ,(T i ,T j ):b ij 1, obtaining the first-stage coarse correlation result phi w 。
9. Root of herbaceous plantsThe method for correlating ship target tracks based on geostationary-orbit staring satellite remote sensing images according to claim 1, wherein the correlation is performed according to the preliminary correlation result Φ v And said first stage coarse correlation result Φ w Carrying out second-stage fine correlation by utilizing the motion information and the amplitude information of the target to obtain phi c The method specifically comprises the following steps:
setting the first stage coarse correlation result phi w In (1),
for the magnitude of the new track in the ith associated track pair,
the amplitude of the old track in the ith associated track pair is the average value of the target amplitude at each moment under current observation;
obtaining the relation between the amplitude of the new track in the ith associated track pair and the amplitude of the old track in the ith associated track pair by utilizing a linear regression relation
Wherein epsilon is a target amplitude error, and is subjected to Gaussian distribution with a mean value of zero and a standard deviation of sigma;
obtaining linear regression model and error standard deviation estimated value by least square estimation
Wherein K is phi a The number of the medium ship association combinations;
according to the preliminary correlation result phi v And said first stage coarse correlation result phi w Using a correlation model
Rejecting wrong associated pairs through amplitude relation detection;
detecting the effectiveness of the correlation by adopting a standard deviation criterion of 3 times, when the amplitude prediction error is less than 3 times of the standard deviation, the correlation is effective, otherwise, the correlation is regarded as abnormal correlation deviating from linearity and is eliminated, and thus a ship target correlation result is obtained
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011410712.2A CN112686095B (en) | 2020-12-04 | 2020-12-04 | Ship target track correlation method for stationary orbit staring satellite remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011410712.2A CN112686095B (en) | 2020-12-04 | 2020-12-04 | Ship target track correlation method for stationary orbit staring satellite remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112686095A CN112686095A (en) | 2021-04-20 |
| CN112686095B true CN112686095B (en) | 2022-08-19 |
Family
ID=75446151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011410712.2A Active CN112686095B (en) | 2020-12-04 | 2020-12-04 | Ship target track correlation method for stationary orbit staring satellite remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112686095B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113393497B (en) * | 2021-07-07 | 2022-08-09 | 中国人民解放军海军航空大学 | Ship target tracking method, device and equipment of sequence remote sensing image under condition of broken clouds |
| CN113467250B (en) * | 2021-07-27 | 2022-01-07 | 浙江中裕通信技术有限公司 | Ship course and region estimation method based on Beidou navigation and data driving |
| CN116188519B (en) * | 2023-02-07 | 2023-10-03 | 中国人民解放军海军航空大学 | Ship target motion state estimation method and system based on video satellite |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104391281A (en) * | 2014-11-21 | 2015-03-04 | 武汉大学 | Method for improving sky-wave radar sea surface ship target tracking and positioning precision |
| CN109034075A (en) * | 2018-07-31 | 2018-12-18 | 中国人民解放军61646部队 | The method of face battle array gazing type remote sensing satellite tracking Ship Target |
| CN111626129A (en) * | 2020-04-27 | 2020-09-04 | 中国人民解放军军事科学院国防科技创新研究院 | Ship target joint detection method based on satellite AIS and infrared camera |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10330794B2 (en) * | 2016-04-04 | 2019-06-25 | Spire Global, Inc. | AIS spoofing and dark-target detection methodology |
| CN110458089B (en) * | 2019-08-08 | 2020-11-06 | 中国人民解放军军事科学院国防科技创新研究院 | Marine target association system and method based on high-low orbit optical satellite observation |
| CN110686679B (en) * | 2019-10-29 | 2021-07-09 | 中国人民解放军军事科学院国防科技创新研究院 | High-orbit optical satellite offshore target interruption track correlation method |
| CN110888126B (en) * | 2019-12-06 | 2023-01-17 | 西北工业大学 | Unmanned ship information perception system data comprehensive processing method based on multi-source sensor |
| CN111402299B (en) * | 2020-04-08 | 2023-05-05 | 中国人民解放军海军航空大学 | Remote sensing image target tracking method and device based on static orbit staring satellite |
-
2020
- 2020-12-04 CN CN202011410712.2A patent/CN112686095B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104391281A (en) * | 2014-11-21 | 2015-03-04 | 武汉大学 | Method for improving sky-wave radar sea surface ship target tracking and positioning precision |
| CN109034075A (en) * | 2018-07-31 | 2018-12-18 | 中国人民解放军61646部队 | The method of face battle array gazing type remote sensing satellite tracking Ship Target |
| CN111626129A (en) * | 2020-04-27 | 2020-09-04 | 中国人民解放军军事科学院国防科技创新研究院 | Ship target joint detection method based on satellite AIS and infrared camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112686095A (en) | 2021-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112686095B (en) | Ship target track correlation method for stationary orbit staring satellite remote sensing image | |
| US9989625B2 (en) | Sky polarization and sun sensor system and method | |
| JP5724907B2 (en) | Integrated method of satellite information and ship information for integrated ship monitoring | |
| US9734399B2 (en) | Context-aware object detection in aerial photographs/videos using travel path metadata | |
| Liu et al. | GF-4 satellite and automatic identification system data fusion for ship tracking | |
| CN112346096B (en) | High-low orbit remote sensing satellite ship target track and point track correlation method and system | |
| CN110389366B (en) | Maritime target motion estimation method based on multi-source SAR satellite | |
| US9453731B2 (en) | System and method for determining orientation relative to earth | |
| CN105352509A (en) | Unmanned aerial vehicle motion target tracking and positioning method under geographic information space-time constraint | |
| CN110780327B (en) | Marine target cooperative positioning method based on satellite-borne AIS and infrared camera | |
| CN108334885A (en) | A kind of video satellite image space object detection method | |
| Bhamidipati et al. | SLAM-based integrity monitoring using GPS and fish-eye camera | |
| CN110686679B (en) | High-orbit optical satellite offshore target interruption track correlation method | |
| Helgesen et al. | Sensor combinations in heterogeneous multi-sensor fusion for maritime target tracking | |
| CN111912430A (en) | On-orbit geometric calibration method, device, equipment and medium for high-orbit optical satellite | |
| KR20240007457A (en) | Method and apparatus for correcting geometry of synthetic aperture radar image observed at sea | |
| CN110458089B (en) | Marine target association system and method based on high-low orbit optical satellite observation | |
| CN117538873A (en) | SAR (synthetic aperture radar) offshore target positioning method and system based on Doppler displacement estimation | |
| Negahdaripour | Calibration of DIDSON forward-scan acoustic video camera | |
| CN117173215A (en) | Inland navigation ship whole-course track identification method and system crossing cameras | |
| Gwon et al. | Nontemporal relative pose estimation for opti-acoustic bundle adjustment | |
| Ruscio et al. | Information communication technology (ict) tools for preservation of underwater environment: A vision-based posidonia oceanica monitoring | |
| Chen et al. | A preliminary study on positioning of an underwater vehicle based on feature matching of Seafloor Images | |
| Baruwa et al. | Video-based Real-time Automated Distance Estimation at Sea (RADES) for marine mammal mitigation | |
| CN116188519B (en) | Ship target motion state estimation method and system based on video satellite |
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 |