CN113985406B - Target track splicing method for marine radar - Google Patents

Target track splicing method for marine radar Download PDF

Info

Publication number
CN113985406B
CN113985406B CN202111593274.2A CN202111593274A CN113985406B CN 113985406 B CN113985406 B CN 113985406B CN 202111593274 A CN202111593274 A CN 202111593274A CN 113985406 B CN113985406 B CN 113985406B
Authority
CN
China
Prior art keywords
radar
data
track
target
marine
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
CN202111593274.2A
Other languages
Chinese (zh)
Other versions
CN113985406A (en
Inventor
王文亮
王金魁
祁凌云
朱浩纲
俞鸿源
匡望来
曾鹏
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.)
Cssc Zhejiang Ocean Technology Co ltd
Original Assignee
Cssc Zhejiang Ocean Technology Co ltd
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 Cssc Zhejiang Ocean Technology Co ltd filed Critical Cssc Zhejiang Ocean Technology Co ltd
Priority to CN202111593274.2A priority Critical patent/CN113985406B/en
Publication of CN113985406A publication Critical patent/CN113985406A/en
Application granted granted Critical
Publication of CN113985406B publication Critical patent/CN113985406B/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
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G01S7/414—Discriminating targets with respect to background clutter
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/277—Analysis of motion involving stochastic approaches, e.g. using Kalman filters
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Electromagnetism (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a marine radar target track splicing method. The method comprises the following steps: s1: acquiring a radar original data image, and performing feature extraction, sea clutter rejection and data correction on a sea target in the image; s2: performing track association on track loss occurring in the same marine target navigation process, and establishing an association relation after calculating the distances of multi-factor Euclidean and Mahalanobis of tracks before and after the track data loss section; s3: and aiming at the flight path with the established association relation, point position data at the forefront and the last moment of the data missing section are respectively obtained, intermediate data interpolation splicing is completed, and the standard and complete flight path after splicing is output and displayed in real time. In the scheme, point position association relation is adopted from multiple factors to confirm and screen; the radar original data are analyzed and preprocessed through feature extraction, sea clutter removal, data correction and the like, the processing process of the radar original data is fully considered, and the accuracy of the radar original data is improved.

Description

Target track splicing method for marine radar
Technical Field
The invention relates to the field of track splicing, in particular to a method for splicing a target track of a marine radar.
Background
The target tracking and track history backtracking of the marine radar provide vital reference data for relevant marine supervision departments and ship navigation. In the process of tracking the marine radar target, due to the influence of external conditions such as transmission channel congestion, obstacle shielding and sea clutter interference, the problems of radar target loss, track temporary loss and the like are inevitable.
The common marine target track splicing algorithm has two problems to be optimized: on one hand, after the radar data has the inevitable data missing condition, the data before and after the missing data segment is not subjected to more thorough data preprocessing; on the other hand, when the radar missing data section track splicing is carried out, multi-factor point position incidence relation confirmation and screening are not adopted.
For example, a chinese patent document discloses a "ship trajectory planning method based on AIS trajectory manipulation unit splicing", which is disclosed in CN110108280A, based on AIS data of a given ship, an improved sliding window compression algorithm is used to divide the AIS trajectory manipulation unit to obtain an available AIS trajectory manipulation unit set of the current ship, and then an optimization algorithm is used to obtain an AIS trajectory manipulation unit that can be smoothly spliced with the current state according to the requirements of the current voyage, and finally the AIS trajectory manipulation unit is converted into a corresponding operation flow and reported to an operator. The scheme lacks multi-factor point position association relationship confirmation and screening during track splicing.
Disclosure of Invention
The method mainly solves the problems that data before and after the missing data section is not subjected to relatively thorough data preprocessing and multi-factor point location incidence relation confirmation and screening are not adopted when the track splicing of the radar missing data section is carried out in the prior art; a marine radar target track splicing method is provided.
The technical problem of the invention is mainly solved by the following technical scheme:
a marine radar target track splicing method comprises the following steps:
s1: acquiring a radar original data image, and performing feature extraction, sea clutter rejection and data correction on a sea target in the image;
s2: performing track association on track loss occurring in the same marine target navigation process, and establishing an association relation after calculating the distances of multi-factor Euclidean and Mahalanobis of tracks before and after the track data loss section;
s3: and aiming at the flight path with the established association relation, point position data at the forefront and the last moment of the data missing section are respectively obtained, intermediate data interpolation splicing is completed, and the standard and complete flight path after splicing is output and displayed in real time.
According to the scheme, the processing process of the radar original data is fully considered, a radar original data analyzing and preprocessing method is provided, and the accuracy of the radar original data is improved; the scheme provides a track association algorithm and a weight distribution method based on multi-factor Euclidean and Mahalanobis distance maneuvering distribution association, and point location association relation is adopted from multiple factors to confirm and screen; the method realizes the track splicing fitting method based on the linear interpolation method, and provides complete and reliable radar target track data for related personnel.
Preferably, the step S1 includes the following steps:
s101: acquiring a radar original data image through a radar data interface and communication equipment and inputting the radar original data image into a system in real time;
s102: converting the radar original data image from a polar coordinate to a radar data image in a Cartesian coordinate through a coordinate conversion algorithm;
s103: identifying and acquiring a marine target and characteristics thereof in a radar data image, and converting coordinates of the marine target on the image into geographic coordinates through conversion and traversal matching of the marine target characteristics;
s104: in the sea clutter removing stage, sea clutter is removed by adopting a Kalman filtering algorithm and a point tracking method;
s105: and performing data correction on the data subjected to the sea clutter rejection to complete data preprocessing and reconstruction.
The analysis and reconstruction of the navigation data of the marine radar target are realized through data preprocessing technologies such as marine target feature extraction, sea clutter rejection, data correction and the like based on an improved image matching algorithm. And in the sea clutter removing stage, sea clutter is removed by adopting a Kalman filtering algorithm and a point tracking method. In the point tracking stage, the characteristics of disorder, short-time property, speed shock property and the like of the sea clutter are considered, the tracked sea clutter is subjected to methods such as short-time speed, course, no data updating for a long time and the like, and a threshold value is set for removing.
Preferably, the specific process of step S103 is as follows:
identifying a marine target in the radar data image through a correlation matching algorithm, and evaluating the characteristics of the marine target by an expert evaluation method;
establishing a template binary gray map, and converting the binary gray map into a matrix through MATLAB to obtain a binary matrix of the offshore target;
traversing and matching the binary matrix of the marine target in the binary matrix of the radar data image to obtain a pixel coordinate of the marine target on the radar data image;
matching in proportion by taking the pixel coordinates and longitude and latitude coordinates of the radar base station, the radar data image pixel length and the radar detection actual distance as references, and converting the pixel coordinates of the offshore target on the radar data image into geographic coordinates;
and storing the acquired visual data of the latitude, the longitude and the distance of the marine target into a database.
And converting the coordinates of the target on the image into geographic coordinates, and storing the acquired visual data of the radar target such as longitude and latitude, distance and the like into a database.
Preferably, in the process of traversing matching calculation, the marine target traverses the whole radar data image through the marine target image, and the absolute difference between the marine target image and a subgraph in the whole radar data image is calculated in a comparison manner, wherein the minimum value of the absolute difference is the optimal matching position; the minimum error method calculation model is expressed as:
Figure DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 132061DEST_PATH_IMAGE002
taking the absolute difference as the optimal matching position when the minimum value is taken;
the size of the marine target image T is M multiplied by N; s is the overall radar data image,
Figure DEST_PATH_IMAGE003
is a subgraph;
s (m, n) is a subgraph in the range of (1,1) - (m, n) in the pixel coordinates in the subgraph;
t (m, n) is a subgraph in the range of (1,1) - (m, n) in pixel coordinates in the marine target image;
m is (1, M) and N is (1, N).
And obtaining the position of the offshore target through traversal matching.
Preferably, the step S2 includes the following steps:
s201: calculating the deviation of the front point position and the rear point position of the track data missing section in the aspects of course, speed and time by taking the front point position and the rear point position of the track data missing section as a calculation basis, and establishing coarse association in a parameter threshold range;
s202: calculating multi-factor Euclidean distance and Mahalanobis distance of key parameters before and after the missing section of the track data;
s203: and distributing, weighting, judging and combining the calculated weights of the multi-factor Euclidean distance and the Mahalanobis distance by an analytic hierarchy process to obtain the overall track distance, wherein the obtained shortest distance is the optimal track matching.
The track rough association method is used for saving computing resources and improving computing speed, and further provides a track association algorithm and a weight allocation method based on multi-factor Euclidean and Mahalanobis distance maneuvering allocation association, and point location association relation is adopted from multiple factors to confirm and screen.
Preferably, the specific process of step S202 is:
firstly, calculating the Euclidean distance of the speed between two targets, calculating the Euclidean distance of the position between the two targets, and then weighting and combining the two Euclidean distances;
the calculation formula of the Euclidean distance is
Figure 314781DEST_PATH_IMAGE004
In the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE005
euclidean distance;
Figure 281469DEST_PATH_IMAGE006
represents
Figure DEST_PATH_IMAGE007
Coordinates;
Figure 824749DEST_PATH_IMAGE008
represents
Figure DEST_PATH_IMAGE009
Coordinates;
the formula for the weighted combination of two euclidean distances is:
Figure DEST_PATH_IMAGE011
in the formula (I), the compound is shown in the specification,
Figure 967018DEST_PATH_IMAGE012
is the position euclidean distance;
Figure DEST_PATH_IMAGE013
is the velocity euclidean distance;
Figure 179693DEST_PATH_IMAGE014
the combined Euclidean distance;
Figure DEST_PATH_IMAGE015
representing a velocity euclidean distance coefficient;
Figure 305781DEST_PATH_IMAGE016
representing a position Euclidean distance coefficient;
the mahalanobis distance is calculated multidimensional as follows:
Figure 973523DEST_PATH_IMAGE018
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE019
expressed as a sample vector X to
Figure 563773DEST_PATH_IMAGE020
N sample vectors X1-Xn, S is the covariance matrix,
Figure 557137DEST_PATH_IMAGE020
is the mean vector;
(Vector)
Figure DEST_PATH_IMAGE021
and
Figure 232837DEST_PATH_IMAGE022
the mahalanobis distance between them is expressed as:
Figure 173112DEST_PATH_IMAGE024
two conditions need to be met during the mahalanobis distance calculation: the overall number of samples is greater than the sample dimension; the inverse of the covariance matrix exists.
And confirming and screening point position association relation from multiple factors based on a multi-factor Euclidean and Mahalanobis distance maneuvering allocation associated track association algorithm and a weight allocation method.
Preferably, the step S3 specifically includes the following steps:
s301: finishing data interpolation of the correlated track data missing segment by a linear interpolation method, performing gradient recursion perfection on all navigation parameters of the track data missing segment, and fitting a complete radar track;
s302: and (4) verifying the fitted radar target track through the continuously updated radar data, and outputting and displaying the standard and complete track after splicing in real time.
Preferably, the verification process is as follows:
Figure 696497DEST_PATH_IMAGE026
ab represents a line segment connecting a starting point and an end point of the front-section radar target track;
cd represents a line segment connecting the starting point and the end point of the target track of the rear-stage radar;
p represents a judgment threshold.
When two sections of radar tracks describe the same marine target, the two sections of radar tracks have similar motion trends, and the tracks of the two associated radar targets are approximately parallel or on the same straight line. The trigonometric function can calculate the included angle between two line segments, and the cosine cos function is used for verifying the association relation. When the angle of the target trajectory segment is close to 0, the cos value of the two segments is close to 1.
The invention has the beneficial effects that:
1. the radar original data are analyzed and preprocessed by feature extraction, sea clutter removal, data correction and the like, the processing process of the radar original data is fully considered, and the accuracy of the radar original data is improved;
2. confirming and screening point position association relation from multiple factors based on a track association algorithm and a weight distribution method for multi-factor Euclidean and Mahalanobis distance maneuvering distribution association;
3. the method realizes the track splicing fitting method based on the linear interpolation method, and provides complete and reliable radar target track data for related personnel.
4. The method is used for saving computing resources and improving computing speed through a track coarse correlation method.
Drawings
FIG. 1 is a flow chart of the marine radar target track splicing method.
FIG. 2 is an experimental verification diagram of the data preprocessing and splicing method of the present invention.
Detailed Description
The technical scheme of the invention is further specifically described by the following embodiments and the accompanying drawings.
Example (b):
the marine radar target track splicing method of the embodiment, as shown in fig. 1, includes the following steps:
step 1, obtaining a radar original data image through a radar data interface, and realizing analysis and reconstruction of marine radar target navigation data through data preprocessing technologies such as marine target feature extraction, sea clutter rejection, data correction and the like based on an improved image matching algorithm.
And 1.1, acquiring a radar original data image through a radar data interface and communication equipment and inputting the radar original data image into a system in real time.
And 1.2, converting the radar original data image from a polar coordinate to a radar data image in a Cartesian coordinate through a coordinate conversion algorithm.
And step 1.3, identifying and acquiring a marine target and characteristics thereof in the radar data image, and converting the coordinates of the marine target on the image into geographic coordinates through conversion and traversal matching of the marine target characteristics.
Identifying a marine target in the radar data image through a correlation matching algorithm, and evaluating the characteristics of the marine target by an expert evaluation method. The characteristics such as the length, the width, the plane figure, the direction and the like of the ship are considered in the characteristic selection.
And establishing a template binary gray map, and converting the binary gray map into a matrix through MATLAB to obtain a binary matrix of the offshore target.
Evaluating the characteristics of the marine target according to an expert evaluation method and constructing a binarization matrix of the marine target as follows:
Figure DEST_PATH_IMAGE027
and traversing and matching the binary matrix of the marine target in the binary matrix of the radar data image to obtain the pixel coordinate of the marine target on the radar data image.
And matching in proportion by taking the pixel coordinates and longitude and latitude coordinates of the radar base station, the radar data image pixel length and the radar detection actual distance as references, and converting the pixel coordinates of the marine target on the radar data image into geographic coordinates.
And storing the acquired visual data of the latitude, the longitude and the distance of the marine target into a database.
In the traversing matching calculation process, the sea target traverses on the whole radar data image through the sea target image, and the absolute difference between the sea target image and the subgraph in the whole radar data image is compared and calculated, wherein the minimum value of the absolute difference is the optimal matching position.
The minimum error method calculation model is expressed as:
Figure 782133DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 23759DEST_PATH_IMAGE028
taking the absolute difference as the optimal matching position when the minimum value is taken;
the size of the marine target image T is M multiplied by N; s is the overall radar data image,
Figure DEST_PATH_IMAGE029
is a subgraph;
s (M, N) is a subgraph in the range of (1,1) - (M, N) in the pixel coordinates in the subgraph;
t (M, N) is a subgraph in the range of (1,1) - (M, N) in the pixel coordinates of the marine target image;
m is (1, M) and N is (1, N).
And 2, performing track association on the track data missing problem in the same radar target navigation process, and establishing an association relation after the tracks before and after the missing section are calculated through the characteristic distance.
And 2.1, in a track coarse correlation stage, calculating the deviation of the front point position and the rear point position of the track data missing section in the aspects of course, speed and time by taking the front point position and the rear point position of the track data missing section as a calculation basis, and establishing coarse correlation within a parameter threshold range.
For the example of 1 minute tracking, the speed deviation threshold is set to 1kn, the heading deviation threshold is set to 15 °, and the time to miss is set to 300 s.
And 2.2, calculating key parameters Euclidean and Mahalanobis distances before and after the radar track missing segment by using a multi-factor Euclidean distance and Mahalanobis distance maneuvering distribution association algorithm through an offshore target track association method based on an optimized nearest neighbor algorithm.
The calculation formula of the Euclidean distance is as follows:
Figure 423516DEST_PATH_IMAGE004
in the formula (I), the compound is shown in the specification,
Figure 535829DEST_PATH_IMAGE030
euclidean distance;
Figure 339837DEST_PATH_IMAGE006
represents
Figure 990130DEST_PATH_IMAGE007
Coordinates;
Figure DEST_PATH_IMAGE031
represents
Figure 272207DEST_PATH_IMAGE009
And (4) coordinates.
In this equation, i is used to represent the value of the current calculation step, and n is the upper limit of the value.
The Euclidean distance of the speed between the two targets is calculated, the Euclidean distance of the position between the two targets is calculated, and then the two Euclidean distances are weighted and combined.
The formula for weighted combination of two euclidean distances is:
Figure 691555DEST_PATH_IMAGE011
in the formula (I), the compound is shown in the specification,
Figure 932044DEST_PATH_IMAGE032
is the position euclidean distance;
Figure DEST_PATH_IMAGE033
is the velocity euclidean distance;
Figure 69633DEST_PATH_IMAGE034
the combined Euclidean distance;
Figure 217718DEST_PATH_IMAGE015
representing a velocity euclidean distance coefficient;
Figure 570202DEST_PATH_IMAGE016
representing the position euclidean distance coefficient.
The Mahalanobis distance considers the characteristic relation among elements forming the vector, so that the dimensions among the elements are irrelevant, the influence of original data measurement units is avoided, and the possibility of dimension influence is eliminated. The Mahalanobis distance is calculated multidimensional as follows:
Figure 450433DEST_PATH_IMAGE018
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE035
expressed as a sample vector X to
Figure 75318DEST_PATH_IMAGE020
HorseThe 'k' distance, where k sample vectors X1-Xk, S is the covariance matrix,
Figure 636881DEST_PATH_IMAGE020
is the mean vector;
(Vector)
Figure 827559DEST_PATH_IMAGE036
and
Figure DEST_PATH_IMAGE037
the mahalanobis distance between them is expressed as:
Figure DEST_PATH_IMAGE039
two conditions need to be met during the mahalanobis distance calculation: the overall number of samples is greater than the sample dimension; the inverse of the covariance matrix exists.
In this embodiment, the last time point location data of the existing reliable track is taken as a reference point, the point location which has undergone the coarse data association is taken as a point location to be associated, and the fine association instance verification is performed.
As shown in table 1, the last time point of the reliable track is recorded as c, the speed is 12.6kn, the heading is 357.0 °, and the longitude and latitude positions are (30.975 ° N, 122.735 ° E); and 3 associated candidate points accord with the Mahalanobis distance calculation condition, and the result of the Mahalanobis distance calculation is preferentially adopted.
Associated candidate point 1, speed 13.5kn, heading 350.8 °, and reference point c, distance 206.04m, calculated from the euclidean distance:
Figure DEST_PATH_IMAGE041
the euclidean distance between the reference point c and the associated candidate point is obtained as 168.9063.
Calculating formula by Mahalanobis distance:
Figure DEST_PATH_IMAGE043
in the above formula
Figure 268905DEST_PATH_IMAGE044
、
Figure DEST_PATH_IMAGE045
、
Figure 584349DEST_PATH_IMAGE046
For the variance of each parameter, finding the mahalanobis distance between the reference point c and the associated candidate point as 41.5322;
similarly, the associated candidate point 2 has a navigational speed of 12.5kn, a heading of 354.6 degrees, and a distance of 45.74m from the reference point c, and has an Euclidean distance of 45.8034 and a Mahalanobis distance of 39.6832; the associated candidate point 3 was calculated to have a navigational speed of 12.4kn, a heading of 11.5 °, and a reference point c distance of 83.92m, and was substituted into the formula to calculate a euclidean distance of 85.1663 and a mahalanobis distance of 42.2186.
Therefore, the euclidean and mahalanobis distances of the associated point 2 are both minimal, and are the best associated data point locations.
The data shows that the Euclidean distance and the Mahalanobis distance can establish a better association relationship, but the Euclidean distance is greatly influenced by the component dimension, and the Mahalanobis distance can also establish the association relationship better under the condition of eliminating the dimension influence.
TABLE 1 Multi-factor Euclidean distance and Mahalanobis distance data sheet
Serial number Speed kn Course (°) Distance m Euclidean distance Mahalanobis distance
Reference point c 12.6 357.0 0 0 0
1 13.5 350.8 126.04 168.9063 41.5322
2 12.4 354.6 45.74 45.8034 39.6832
3 11.9 11.5 83.92 85.1663 42.2186
And 2.3, distributing, weighting, judging and combining the calculated weights of the multi-factor Euclidean and Mahalanobis distances by an analytic hierarchy process to obtain the overall track distance, wherein the obtained shortest distance is the optimal track matching.
After data values of speed V, longitude and latitude L and heading C describing the same target track are obtained, the 3 key factor weights are solved by an analytic hierarchy process.
And dividing Euclidean and Mahalanobis distance reliability weights according to the empirical value, wherein the Euclidean distance reliability weight is 0.4, and the Mahalanobis distance reliability weight is 0.6.
Determining membership functions U (V), U (L) and U (C) of navigation speed V, longitude and latitude L and heading C by an Analytic Hierarchy Process (AHP) according to the following weight steps:
(1) and constructing a judgment matrix. The decision matrix criteria are U (V), U (L), U (C), and different criteria have different weight ratios in the subjective impressions of different decision makers. Definition of
Figure 74236DEST_PATH_IMAGE047
The factor i is the important degree factor of the factor j, and
Figure 135733DEST_PATH_IMAGE047
and forming an A-U judgment matrix P.
Figure DEST_PATH_IMAGE048
(2) An importance ranking is calculated. According to the judgment matrix, the maximum characteristic root is obtained
Figure 357767DEST_PATH_IMAGE049
The corresponding feature vector w. The equation is as follows:
Figure 957244DEST_PATH_IMAGE051
the obtained feature vector w is normalized to obtain
Figure 985243DEST_PATH_IMAGE052
W is the importance ranking of each evaluation factor, corresponding to each
Figure 822618DEST_PATH_IMAGE053
Namely each commentPrice factor importance ranking, i.e. weight assignment.
(3) In order to verify whether the obtained weight distribution is reasonable, consistency check needs to be performed on the judgment matrix. The test uses the formula:
Figure 684395DEST_PATH_IMAGE055
in the formula, CR is the random consistency ratio of the judgment matrix;
CI is a general consistency index of the decision matrix, which is given by:
Figure 53059DEST_PATH_IMAGE056
RI is the average random consistency index of the judgment matrix, and e is the order of the judgment matrix P.
When judging the matrix P
Figure 337279DEST_PATH_IMAGE057
When or at the time
Figure DEST_PATH_IMAGE058
When P is considered to have satisfactory consistency, otherwise, the elements in P need to be adjusted to have satisfactory consistency. After a matrix form P with a satisfactory consistency is obtained,
Figure 842210DEST_PATH_IMAGE052
can be used as
Figure 671625DEST_PATH_IMAGE059
The weight coefficient of (2) is distributed to each component weight to form a data fusion strategy.
And 3, acquiring point location data of the forefront and the last moment of the data missing section respectively aiming at the flight path with the established association relation, finishing interpolation splicing of intermediate data, and outputting and displaying the standard and complete flight path after splicing in real time.
And 3.1, completing data interpolation of the correlated flight path data missing segment by a linear interpolation method, and performing gradient recursion and perfection on all navigation parameters of the flight path data missing segment to fit a complete radar flight path.
And 3.2, checking the fitted radar target track through the continuously updated radar data, and outputting and displaying the standard and complete track after splicing in real time.
The checking method comprises the following steps: when two sections of radar tracks describe the same target, the two sections of radar tracks have similar motion trends, and the tracks of the two associated radar targets are approximately parallel or on the same straight line. The trigonometric function can calculate the included angle between two line segments, and the cosine cos function is used for verifying the association relation. When the angle of the target trajectory segment is close to 0, the cos value of the two segments is close to 1.
The judgment formula is shown as follows:
Figure DEST_PATH_IMAGE060
ab represents a line segment connecting a starting point and an end point of the front-section radar target track;
cd represents a line segment connecting the starting point and the end point of the target track of the rear-stage radar;
p represents the judgment threshold and is 0.9.
As shown in fig. 2, it is an experimental verification diagram of the data preprocessing and splicing method. The lake is used as an experimental water area, two shelters are arranged in the experimental process, the west bank center is used as a test starting point, and the remote control boat sails linearly 100m in the northeast direction at a constant speed of 2m/s is used as experimental data. The level of the wind waves of the experimental water area is 1 level soft wind, the wind speed is 0-0.2m/s, and the maximum wave height is 0.1 m. The experimental data are shown in table 2.
TABLE 2 Experimental data sheet
Figure DEST_PATH_IMAGE062
According to the scheme, the radar original data are analyzed and subjected to preprocessing methods such as feature extraction, sea clutter rejection and data correction, the processing process of the radar original data is fully considered, and the accuracy of the radar original data is improved; confirming and screening point position association relation from multiple factors based on a multi-factor Euclidean and Mahalanobis distance maneuvering allocation associated track association algorithm and a weight allocation method; the method realizes the track splicing fitting method based on the linear interpolation method, and provides complete and reliable radar target track data for related personnel.
It should be understood that the examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.

Claims (5)

1. A marine radar target track splicing method is characterized by comprising the following steps:
s1: acquiring a radar original data image, and performing feature extraction, sea clutter rejection and data correction on a sea target in the image; the step S1 includes the following steps:
s101: acquiring a radar original data image through a radar data interface and communication equipment and inputting the radar original data image into a system in real time;
s102: converting the radar original data image from a polar coordinate to a radar data image in a Cartesian coordinate through a coordinate conversion algorithm;
s103: identifying and acquiring a marine target and characteristics thereof in the radar data image, and converting coordinates of the marine target on the image into geographic coordinates through conversion and traversal matching of the marine target characteristics;
s104: in the sea clutter removing stage, sea clutter is removed by adopting a Kalman filtering algorithm and a point tracking method;
s105: performing data correction on the data subjected to the sea clutter rejection to complete data preprocessing and reconstruction;
s2: performing track association on track loss occurring in the same marine target navigation process, and establishing an association relation after calculating the distances of multi-factor Euclidean and Mahalanobis of tracks before and after the track data loss section;
the step S2 includes the following steps:
s201: calculating the deviation of the front point position and the rear point position of the track data missing section in the aspects of course, speed and time by taking the front point position and the rear point position of the track data missing section as a calculation basis, and establishing coarse association in a parameter threshold range;
s202: calculating multi-factor Euclidean distance and Mahalanobis distance of key parameters before and after the missing section of the track data;
s203: distributing, weighting, judging and combining the calculated weights of the multi-factor Euclidean distance and the Mahalanobis distance by an analytic hierarchy process to obtain the whole track distance, wherein the obtained shortest distance is the optimal track matching;
s3: aiming at the flight path with the established association relationship, point position data at the forefront and the last moment of a data missing section are respectively obtained, intermediate data interpolation splicing is completed, and standard and complete flight path after splicing is output and displayed in real time;
the step S3 specifically includes the following steps:
s301: finishing data interpolation of the correlated track data missing segment by a linear interpolation method, performing gradient recursion perfection on all navigation parameters of the track data missing segment, and fitting a complete radar track;
s302: and (4) verifying the fitted radar target track through the continuously updated radar data, and outputting and displaying the standard and complete track after splicing in real time.
2. The marine radar target track splicing method according to claim 1, wherein the specific process of the step S103 is as follows:
identifying a marine target in the radar data image through a correlation matching algorithm, and evaluating the characteristics of the marine target by an expert evaluation method;
establishing a template binary gray map, and converting the binary gray map into a matrix through MATLAB to obtain a binary matrix of the offshore target;
traversing and matching the binary matrix of the marine target in the binary matrix of the radar data image to obtain a pixel coordinate of the marine target on the radar data image;
matching in proportion by taking the pixel coordinates and longitude and latitude coordinates of a radar base station, the radar data image pixel length and the radar detection actual distance as references, and converting the pixel coordinates of the marine target on the radar data image into geographic coordinates;
and storing the acquired visual data of the latitude, the longitude and the distance of the marine target into a database.
3. The marine radar target track splicing method according to claim 1 or 2, wherein in the process of traversing matching calculation, the marine target traverses on the whole radar data image through the marine target image, and the absolute difference of sub-images in the marine target image and the whole radar data image is calculated in a comparison manner, wherein the minimum value of the absolute difference is an optimal matching position; the minimum error method calculation model is expressed as:
Figure FDA0003510081480000021
wherein E (i, j) is an absolute difference, and the minimum value is taken as an optimal matching position;
the size of the marine target image T is M multiplied by N; s is an overall radar data image, Si，jIs a subgraph;
s (m, n) is a subgraph in the range of (1,1) - (m, n) in the pixel coordinates in the subgraph;
t (m, n) is a subgraph in the range of (1,1) - (m, n) in pixel coordinates in the marine target image;
m is (1, M) and N is (1, N).
4. The marine radar target track splicing method according to claim 1, wherein the specific process of the step S202 is as follows:
firstly, calculating the Euclidean distance of the speed between two targets, calculating the Euclidean distance of the position between the two targets, and then weighting and combining the two Euclidean distances;
the calculation formula of the Euclidean distance is
Figure FDA0003510081480000022
In the formula, dist Euclidean distance;
xirepresents the x coordinate;
yirepresents the y coordinate;
the formula for weighted combination of two euclidean distances is:
distsp＝a×dists+b×distp
in the formula, distpIs the position euclidean distance;
distsis the velocity euclidean distance;
distspthe combined Euclidean distance;
a represents a velocity Euclidean distance coefficient;
b represents a position Euclidean distance coefficient;
the mahalanobis distance is calculated multidimensional as follows:
Figure FDA0003510081480000031
wherein d (X) is the Mahalanobis distance from the sample vector X to mu, wherein n sample vectors X1-Xn are provided, S is the covariance matrix of the sample vectors, and mu is the mean vector of the sample vectors;
vector XiAnd XjThe mahalanobis distance between them is expressed as:
Figure FDA0003510081480000032
two conditions need to be met during the mahalanobis distance calculation: the overall number of samples is greater than the sample dimension; the inverse of the covariance matrix exists.
5. The marine radar target track splicing method according to claim 1, wherein the verification process comprises the following steps:
Figure FDA0003510081480000041
ab represents a line segment connecting a starting point and an end point of the front-section radar target track;
cd represents a line segment connecting the starting point and the end point of the target track of the rear-stage radar;
p represents a judgment threshold.
CN202111593274.2A 2021-12-24 2021-12-24 Target track splicing method for marine radar Active CN113985406B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111593274.2A CN113985406B (en) 2021-12-24 2021-12-24 Target track splicing method for marine radar

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111593274.2A CN113985406B (en) 2021-12-24 2021-12-24 Target track splicing method for marine radar

Publications (2)

Publication Number Publication Date
CN113985406A CN113985406A (en) 2022-01-28
CN113985406B true CN113985406B (en) 2022-05-10

Family

ID=79734172

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111593274.2A Active CN113985406B (en) 2021-12-24 2021-12-24 Target track splicing method for marine radar

Country Status (1)

Country Link
CN (1) CN113985406B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114814777B (en) * 2022-06-27 2022-09-27 中国人民解放军32035部队 Pattern matching correlation method and system for multi-radar dense target
CN117491987B (en) * 2023-12-29 2024-04-09 海华电子企业(中国)有限公司 Ship track splicing method based on LSTM neural network and space-time motion distance algorithm

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB201307381D0 (en) * 2013-04-24 2013-06-05 Bae Systems Plc Wind turbine mitigation in radar systems
WO2018174822A1 (en) * 2017-03-22 2018-09-27 Office National D'etudes Et De Recherches Aerospatiales Global integrity check system and associated method
CN111157982A (en) * 2019-11-20 2020-05-15 智慧航海（青岛）科技有限公司 Intelligent ship and shore cooperative target tracking system and method based on shore-based radar
CN112098993A (en) * 2020-09-16 2020-12-18 中国北方工业有限公司 Multi-target tracking data association method and system
CN113516037A (en) * 2021-05-11 2021-10-19 中国石油大学(华东) Marine vessel track segment association method, system, storage medium and equipment
CN113567951A (en) * 2021-09-26 2021-10-29 之江实验室 Cross-millimeter wave radar target tracking method based on space-time information
CN113640760A (en) * 2021-10-14 2021-11-12 中国人民解放军空军预警学院 Radar discovery probability evaluation method and equipment based on air situation data

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111650581B (en) * 2020-06-15 2023-02-28 南京莱斯电子设备有限公司 Radar global target track automatic starting method based on environment perception

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB201307381D0 (en) * 2013-04-24 2013-06-05 Bae Systems Plc Wind turbine mitigation in radar systems
WO2018174822A1 (en) * 2017-03-22 2018-09-27 Office National D'etudes Et De Recherches Aerospatiales Global integrity check system and associated method
CN111157982A (en) * 2019-11-20 2020-05-15 智慧航海（青岛）科技有限公司 Intelligent ship and shore cooperative target tracking system and method based on shore-based radar
CN112098993A (en) * 2020-09-16 2020-12-18 中国北方工业有限公司 Multi-target tracking data association method and system
CN113516037A (en) * 2021-05-11 2021-10-19 中国石油大学(华东) Marine vessel track segment association method, system, storage medium and equipment
CN113567951A (en) * 2021-09-26 2021-10-29 之江实验室 Cross-millimeter wave radar target tracking method based on space-time information
CN113640760A (en) * 2021-10-14 2021-11-12 中国人民解放军空军预警学院 Radar discovery probability evaluation method and equipment based on air situation data

Also Published As

Publication number Publication date
CN113985406A (en) 2022-01-28

Similar Documents

Publication Publication Date Title
CN113032502B (en) Ship anomaly detection method based on improved track segment DBSCAN clustering
CN113985406B (en) Target track splicing method for marine radar
US20230118438A1 (en) Spatio-temporal dp method based on ship trajectory characteristic point extraction
CN112084869B (en) Compact quadrilateral representation-based building target detection method
CN103197299B (en) Extraction and quantitative analysis system of weather radar radial wind information
CN106897681B (en) Remote sensing image contrast analysis method and system
CN106970387A (en) A kind of cruiseway Traffic flow detection method based on AIS and Radar Data Fusion
CN110619258B (en) Road track checking method based on high-resolution remote sensing image
CN111241994B (en) Deep learning remote sensing image rural highway sanded road section extraction method
CN106295503A (en) The high-resolution remote sensing image Ship Target extracting method of region convolutional neural networks
CN108562885B (en) High-voltage transmission line airborne LiDAR point cloud extraction method
CN114564545A (en) System and method for extracting ship experience course based on AIS historical data
CN114418953B (en) Sonar target detection method and system based on time history accumulated image
CN113311433B (en) InSAR interferometric phase two-step unwrapping method combining quality map and minimum cost flow
CN113920436A (en) Remote sensing image marine vessel recognition system and method based on improved YOLOv4 algorithm
CN114998744B (en) Agricultural machinery track field dividing method and device based on motion and vision dual-feature fusion
CN116958907B (en) Method and system for inspecting surrounding hidden danger targets of gas pipeline
CN115342814B (en) Unmanned ship positioning method based on multi-sensor data fusion
CN114705193A (en) Marine big data-based ship navigation path planning method and system
CN106709501B (en) Scene matching area selection and reference image optimization method of image matching system
CN115019163A (en) City factor identification method based on multi-source big data
CN117274627A (en) Multi-temporal snow remote sensing image matching method and system based on image conversion
CN107817491A (en) A kind of generation method of image, device and image forming apparatus
CN110081890B (en) Dynamic k nearest neighbor map matching method combined with deep network
CN110045363A (en) More radar track correlating methods based on relative entropy

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