CN106918807B - A kind of Targets Dots condensing method of radar return data - Google Patents
A kind of Targets Dots condensing method of radar return data Download PDFInfo
- Publication number
- CN106918807B CN106918807B CN201710111371.0A CN201710111371A CN106918807B CN 106918807 B CN106918807 B CN 106918807B CN 201710111371 A CN201710111371 A CN 201710111371A CN 106918807 B CN106918807 B CN 106918807B
- Authority
- CN
- China
- Prior art keywords
- data
- detection
- sliding window
- value
- connected domain
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention belongs to Radar Signal Processing Technology fields, disclose a kind of Targets Dots condensing method of radar return data, comprising: obtain the raw radar data of radar multiple scan periods；Envelope detection and multiple-pulse non-inherent accumulation are successively carried out to raw radar data, the echo data after obtaining pulse accumulation；CFAR detection is carried out to the echo data after pulse accumulation, and then successively carries out two-value quantification treatment to the echo data after CFAR detection and is detected along the two-value sliding window of azimuth dimension；Using the echo data after the detection of two-value sliding window as binary image data, the operation for first expanding post-etching is carried out to binary image data, to obtain all connected domains in binary image data；The first connected domain and the second connected domain are filtered out, the remaining connected domain comprising Targets Dots is obtained；Target information is obtained according to the remaining connected domain comprising Targets Dots, realizes the cohesion to Targets Dots；It can be improved target component estimated accuracy and target resolution capability.
Description
Technical field
The invention belongs to a kind of cohesions of the Targets Dots of Radar Signal Processing Technology field more particularly to radar return data
Method, pretreatment and target component for radar return data are extracted.
Background technique
With the development of Radar Technology, people require radar to have more functions and effect.Traditional radar
It is mainly used for measuring the essential informations such as distance, orientation, the pitching of target, modern radar is required in increasingly complex environment (interference
With clutter) under realize to the detection of target, plot extraction and target following.
Modern radar technology develops towards intelligent, miniaturization, information-based and precise treatment direction.People require radar can
More accurate target information is obtained, and realizes more stable target following.Plot coherence technology is exactly by radar
Echo data carries out the process that target component estimation and plot extraction is realized in processing, point mark of the radar data processing system to admission
Carry out the tracking display that flight path processing realizes target.The country starts late to the research of radar data processing technique compared to foreign countries,
Radar system of the research of Plot coherence processing technique mainly for MTD system.The method that Plot coherence mainly uses at this stage
There are the Plot coherence based on sliding window method, Plot coherence based on image procossing etc..
The problem of Plot coherence is primarily present at this stage be under complex environment how the accurate differentiation to Targets Dots, and
How method that different radar systems under Plot coherence handled is improved.Plot coherence method based on sliding window method is more common
A kind of Plot coherence method, the problem of being primarily present are the division that will appear Targets Dots when target end of a period threshold value is larger, mesh
The division being marked in orientation is solved often through the method for reducing threshold value, and threshold value selection cannot be too low, too low
Threshold value cause False Intersection Points mark excessive, influence a mark quality.The classical method based on image procossing is using image outline
It searches the method for seeking mass center and realizes Plot coherence, be unable to improve a mark in the fragmentation problem in upper and orientation.
Summary of the invention
The shortcomings that for the above-mentioned prior art, the purpose of the present invention is to provide a kind of Targets Dots of radar return data
Condensing method is not detected only with the two-value sliding window in sliding window method, while using the Morphological Filtering Algorithm in mathematics to connection
Domain carries out dilation erosion operation, can not only improve the Targets Dots of sliding window method appearance in side by choosing suitable structural element
Separating phenomenon on position, while may be implemented to improve Targets Dots apart from upper separating phenomenon.
In order to achieve the above objectives, the present invention is realised by adopting the following technical scheme.
A kind of Targets Dots condensing method of radar return data, described method includes following steps:
Step 1, the raw radar data of multiple scan periods of the acquisition radar Jing Guo process of pulse-compression, described original time
Wave number is according to for comprising the 2-D data apart from peacekeeping azimuth dimension, and the raw radar data of the multiple scan period is along azimuth dimension
It is arranged successively；
Step 2, envelope detection and non-inherent accumulation are successively carried out to the raw radar data, after obtaining pulse accumulation
Echo data；
Step 3, CFAR detection is carried out to the echo data after the pulse accumulation, the echo after obtaining CFAR detection
Data, and then successively carry out two-value quantification treatment to the echo data after the CFAR detection and slided along the two-value of azimuth dimension
Window detection, the echo data after obtaining the detection of two-value sliding window.
Step 4, corresponding binary image data, and binary map are converted by the echo data after two-value sliding window detection
The length of picture corresponds to the distance unit of the echo data after the two-value sliding window detection, and the width of bianry image corresponds to the two-value
The umber of pulse of echo data after sliding window detection carries out the operation for first expanding post-etching to the binary image data, thus
All connected domains into the binary image data；
Step 5, in all connected domains, the first connected domain and the second connected domain are filtered out, first connected domain is
It only include the connected domain of an isolated point mark, second connected domain is the point mark composition that threshold value is enlarged beyond in distance dimension
Connected domain, to obtain the remaining connected domain comprising Targets Dots；It is obtained according to the remaining connected domain comprising Targets Dots
To target information, the cohesion to Targets Dots is realized.
The characteristics of technical solution of the present invention and further improvement are as follows:
(1) after the step 1, and before the step 2, the method also includes: to the raw radar data
Cancellation is carried out, the clutter in raw radar data is inhibited.
(2) step 2 specifically:
Envelope detection is carried out to the raw radar data, to obtain raw radar data in respective distances-orientation list
The amplitude information of member；The distance unit is sampled point of the raw radar data in distance dimension, and the localizer unit is
Sampled point of the raw radar data in azimuth dimension.
Sliding window accumulation is carried out on adjacent pulse to the echo data after envelope detection, the number of echoes after obtaining pulse accumulation
According to.
(3) step 3 specifically:
Sliding window detection is carried out along distance unit to the echo data after the pulse accumulation, by reference units all in window
Mean value multiplied by the value that threshold factor obtains be set as sliding window detection threshold value；By the amplitude of unit to be detected and the sliding window
The threshold value of detection is compared, if the amplitude of unit to be detected is greater than the threshold value of sliding window detection, by list to be detected
The data of member retain, and otherwise set 0 for the data of the unit to be detected；
And then the data that amplitude is greater than to the unit to be detected of the threshold value of sliding window detection are set as 1, complete to perseverance
Echo data after false-alarm detection carries out the process of two-value quantification treatment；
To when carrying out the echo data after two-value quantification treatment and carrying out the detection of two-value sliding window along azimuth dimension, using M/N criterion,
Wherein, M indicates the detection threshold of two-value sliding window detection, and N indicates the length of window used when the detection of two-value sliding window.
(4) in step 4, the operation for first expanding post-etching is carried out to the binary image data, to obtain the two-value
All connected domains in image data, specifically include:
Selecting structure element carries out the operation for first expanding post-etching with the structural element of selection, so to binary image data
The lookup of connected domain is carried out to the binary image data for first expanding post-etching afterwards.
(5) in step 5, the remaining connected domain according to comprising Targets Dots obtains target information, realizes to mesh
The cohesion of punctuate mark, specifically includes:
The centroid position for obtaining the remaining connected domain comprising Targets Dots, by the centroid position in original echo number
Range information of the corresponding distance unit as target in, by the centroid position in raw radar data corresponding orientation
Azimuth information of the unit as target；
The corresponding raw radar data of remaining connected domain comprising Targets Dots is obtained, according to described comprising target point
Range-azimuth unit and corresponding envelope detection where the corresponding raw radar data of remaining connected domain of mark is as a result, ask
Obtain range information, azimuth information and the amplitude information of target.
Compared with prior art, the present invention having the advantage that
(1) it is handled since echo data is converted to bianry image by the present invention, it can be using software realization and offline
Processing；Have the advantages that based on image procossing and based on sliding window method Plot coherence method respectively, and in classical Plot coherence side
Further a mark is handled using shape filtering method on the basis of method, the deficiency of two kinds of classical ways is compensated for, into one
Step improves the separating phenomenon of Targets Dots, while with higher mark quality；
(2) Targets Dots condensing method provided by the invention uses bianry image connected domain lookup method, can accurately really
Set the goal a mark distributed areas, has higher resolution ratio to Targets Dots, after lookup using centroid method determine region mass center or
Target component is estimated using echo amplitude method of weighting, target component estimation is more more acurrate than sliding window method；
(3) more effective for filtering out for the differentiation of Targets Dots and False Intersection Points mark, it combines be based on sliding window method target first
The differentiation of point mark and Plot coherence criterion and target distribution feature, can formulate such target-recognition criterion: on adjusting the distance
Extend the point mark that biggish Targets Dots are filtered out and isolated to be filtered out, it is secondly lesser to connected domain area to filter out
(design parameter will according to circumstances be set, ununified standard), these object judgements filter out criterion with point mark can be effective
Identify Targets Dots, while can flexible control point mark quantity；
(4) influence caused by Target Splitting can be efficiently reduced by carrying out connected domain merging by image expansion etching operation,
Further improve the effect based on sliding window method Plot coherence.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention, for those of ordinary skill in the art, it can also be obtained according to these attached drawings other attached
Figure.Without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of process signal of Targets Dots condensing method of radar return data provided in an embodiment of the present invention
Figure；
Fig. 2 is two-value sliding window detection process schematic diagram provided in an embodiment of the present invention；
Fig. 3 is expansion results schematic diagram of the different structure element provided in an embodiment of the present invention to same target；
Fig. 4 is Corrosion results schematic diagram of the different structure element provided in an embodiment of the present invention to same target；
Fig. 5 is pixel syntople schematic diagram provided in an embodiment of the present invention；
Fig. 6 is the connectivity schematic diagram of P, Q, S point provided in an embodiment of the present invention；
Fig. 7 is connected domain search procedure schematic diagram provided in an embodiment of the present invention；
Fig. 8 is raw radar data schematic diagram provided in an embodiment of the present invention；
Fig. 9 is the echo data schematic diagram after non-inherent accumulation provided in an embodiment of the present invention；
Figure 10 is the echo data schematic diagram after CFAR detection result provided in an embodiment of the present invention；
Figure 11 is two-value sliding window testing result schematic diagram provided in an embodiment of the present invention；
Figure 12 is bianry image dilation erosion result schematic diagram provided in an embodiment of the present invention；
Figure 13 is that Targets Dots provided in an embodiment of the present invention agglomerate result schematic diagram.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The demand for development radar system of Radar Technology provides higher target component, after antenna beam is scanned target
Echo data separating phenomenon can be generated after signal processing and target detection, be divided into distance division and orientation division.For two
Plot coherence is exactly the extraction to target primary information (apart from orientation) for dimensional Radar.Usually the extraction of target information is adopted
With binary Slide-window detector, the degree that Slide-window detector reduces false-alarm depends on detection threshold and length of window.Plot coherence one
Aspect requires to filter out False Intersection Points mark, on the other hand requires have higher resolution capability.Target Splitting can be reduced to bring
Influence and meanwhile to improve target component quality be an important index for Plot coherence.It is provided in an embodiment of the present invention
Plot coherence method one side algorithm based on image procossing is realized in bianry image, is not influenced by other conditions, is more held
Easily realize；Another aspect image procossing is combined with based on sliding window method Plot coherence, increases the criterion to mark, favorably
In raising Targets Dots quality.Plot coherence method provided in an embodiment of the present invention based on image procossing is based on sliding window method point
The result that the thought and principle of mark cohesion are combined with image procossing related algorithm, and improved on this basis.
The embodiment of the present invention provides a kind of Targets Dots condensing method of radar return data, as shown in Figure 1, the method
Include the following steps:
Step 1, the raw radar data of multiple scan periods of the acquisition radar Jing Guo process of pulse-compression, described original time
Wave number is according to for comprising the 2-D data apart from peacekeeping azimuth dimension, and the raw radar data of the multiple scan period is along azimuth dimension
It is arranged successively.
You need to add is that after the step 1, and before the step 2, the method also includes: to the original
Beginning echo data carries out cancellation, inhibits the clutter in raw radar data.
Step 2, envelope detection and multiple-pulse non-inherent accumulation are successively carried out to the raw radar data, obtains pulse product
Echo data after tired.
Pulse accumulation is divided into correlative accumulation and non-inherent accumulation.Correlative accumulation can preferably improve target signal to noise ratio, but by
It is higher in computational complexity, and the Plot coherence evaluated error after correlative accumulation is larger, especially when target Doppler frequency
The effect agglomerated when appearing in multiple channels or when target tangentially moves is bad, for two-coordinate radar, target
Major parameter is distance, orientation and velocity, and the data after being detected according to signal target extract target component.Using non-phase
Ginseng accumulation computational complexity is low, and adaptability is very strong under many scenes, therefore technical solution of the present invention uses the non-coherent of multiple-pulse
Accumulation.
The step 2 specifically:
Envelope detection is carried out to the raw radar data, to obtain raw radar data in respective distances-orientation list
The amplitude information of member；The distance unit is sampled point of the raw radar data in distance dimension, and the localizer unit is
Sampled point of the raw radar data in azimuth dimension.
Sliding window accumulation is carried out on adjacent pulse to the echo data after envelope detection, the number of echoes after obtaining pulse accumulation
According to.
Step 3, CFAR detection is carried out to the echo data after the pulse accumulation, the echo after obtaining CFAR detection
Data, and then successively carry out two-value quantification treatment to the echo data after the CFAR detection and slided along the two-value of azimuth dimension
Window detection, the echo data after obtaining the detection of two-value sliding window, further decreases the influence of False Intersection Points mark bring.The two-value sliding window
Echo data after detection is 2-D data, wherein one-dimensional representation distance unit, the umber of pulse of another one-dimensional representation azimuth dimension；
The CFAR detection is using unit average constant false alarm detection (CA-CFAR).
In the step 3, so the echo data after the CFAR detection is successively carried out two-value quantification treatment and
Two-value sliding window along azimuth dimension detects, the echo data after obtaining the detection of two-value sliding window, specifically:
First the echo data after the pulse accumulation that step 2 processing obtains examined using unit average constant false alarm
Survey processing, unit average constant false alarm detection is to carry out sliding window detection along distance unit, by the mean value of reference units all in window
It is set as the threshold value of sliding window detection multiplied by the value that threshold factor obtains, the amplitude of unit to be detected and the sliding window are detected
Threshold value is compared, and is retained threshold data is crossed, and is not crossed threshold data and is set as 0；Then the data for crossing threshold value are arranged
It is 1, to complete the process for carrying out CFAR detection and two-value quantification treatment to the echo data after the pulse accumulation；
To when carrying out the echo data after two-value quantification treatment and carrying out the detection of two-value sliding window along azimuth dimension, using M/N criterion,
Wherein, M indicates the detection threshold of two-value sliding window detection, and N indicates the length of window used when the detection of two-value sliding window.
The target starting thresholding of two-value sliding window detection influences less the resolution of Targets Dots, so setting target is whole here
Thresholding is identical as starting thresholding, and empirically formula M=1.5*sqrt (N) is chosen the relationship of M and N, but in order to subtract
Few Targets Dots division, the detection threshold of two-value sliding window detection answer reduction appropriate.
It should be noted that target end of a period threshold value refers to when differentiating in sliding window detection process to the point mark of target,
Two-value sliding window accumulating value is a change procedure, judges Targets Dots in orientation by choosing starting thresholding and end of a period thresholding
Starting point and end of a period point, to complete resolution of the Targets Dots in orientation.
Specifically, two-value sliding window detection process is as shown in Figure 2, wherein Fig. 2 (a) is the input letter after two-value quantification treatment
Number, Fig. 2 (b) is that two-value sliding window detects cumulative process, and Fig. 2 (c) is two-value sliding window testing result schematic diagram；M/N criterion refers to window
Mouth length N=5, detection threshold M=3, are considered as detecting target when two-value accumulation result is more than or equal to threshold value in window
It is denoted as 1, is otherwise denoted as 0.
You need to add is that the raw radar data refers to the number after antenna scanning after extra pulse compression processing
According to the processing to echo data includes clutter recognition, signal accumulation and CFAR detection.
For two-coordinate radar, target component refers to the distance and bearing information of target, here to echo data
Non-inherent accumulation is carried out, clutter recognition is realized using canceller.
It, can be to echo data only
It carries out simple signal accumulation and omits cancellation process, the present invention is to echo data processing using sliding window accumulation, constant false alarm
The method of detection and the detection of two-value sliding window.
Step 4, using the echo data after two-value sliding window detection as binary image data, to the bianry image number
According to the operation for carrying out first expanding post-etching, to obtain all connected domains in the binary image data.
Specifically, carrying out dilation erosion operation using Morphological Filtering Algorithm to the binary image data, it is therefore an objective to realize
The merging of connected domain.
The expansion of image and etching operation, from the point of view of image processing point, binary Images Processing is exactly by a small knot
Constitutive element is moved and is compared point by point on big bianry image, makes corresponding processing according to the result of the comparison.Structural elements
Element refers to the background image with certain size, the shape that structural element is not fixed, while design form converts algorithm
Need to carry out the design of structural element according to input picture and required information.Usual structural element shape have square, rectangle,
It is round and linear etc..Briefly, expansion is exactly that the background dot around object is merged into object, and object is outside by expansion
Two adjacent objects are possible to be coupled together, will appear as the conjunction of connected domain in bianry image by extension in this way
And.Opposite, the etching operation of image is exactly the boundary point for eliminating object, in addition eliminates lesser object.
(1) expansion process of image can be simply described as structural element and be traversed on bianry image, with two-value
For image black color dot, when structural element origin is identical as certain point pixel in bianry image, then structural element and bianry image
Corresponding all the points all become black color dots.Different structure element is illustrated in figure 3 to the expansion results schematic diagram of same target, figure
3 (a) be structural element 1, and Fig. 3 (b) is structural element 2, and Fig. 3 (c) is former bianry image, and Fig. 3 (d) is right using structural element 1
It is that former bianry image is expanded as a result, Fig. 3 (e) be expanded using the former bianry images of 2 pairs of structural element as a result, its
In, symbol "+" indicates coordinate origin position in structural element.
(2) etching operation of image can be briefly described for structural element when being traversed on bianry image, with two-value
For image black color dot, when structural element origin is identical as certain point pixel in bianry image, if rest of pixels in structural element
Have that one or more pixel is different from bianry image corresponding pixel points, then the point of bianry image corresponding to origin becomes white
Color dot.Different structure element is illustrated in figure 4 to the Corrosion results schematic diagram of same target, Fig. 4 (a) is structural element 1, Fig. 4
It (b) is structural element 2, Fig. 4 (c) is former bianry image, and Fig. 4 (d) is corroded using 1 pair of structural element former bianry image
As a result, Fig. 4 (e) is the result corroded using 2 pairs of structural element former bianry images.
Structural element used in the embodiment of the present invention is line segment shape structural element, and central point is in line segment center, length
Using the 1/4 of coherent processing umber of pulse, post-etching is first expanded to bianry image, dilation erosion respectively carries out once.Bianry image is swollen
The swollen merging for realizing connected domain, etching operation restores the bianry image after expansion, but image after restoring and original
Bianry image is different.
Specifically, obtaining needing when all connected domains in the binary image data to after dilation erosion in step 4
Connected domain is searched.
It is adjacent with to be connected to be fundamental relation between pixel, in bianry image in addition to edge pixel, around each pixel
There are 8 abutment points, distinguished according to the position of abutment points and there are 4 abutment points.As shown in figure 5, being 8 around Fig. 5 (a) black color dots
Schematic diagram is abutted, is 4 adjacent schematic diagrames around Fig. 5 (b) black color dots.
If two points are adjacent according to 4 syntoples or 8 by a series of identical point sequence of gray scales in bianry image
Connecing relationship and connecting together so just claims the two points to be connected domain, and the set of the points being connected to the two points all in this way is just
Constitute connected domain, as shown in fig. 6, point P, Q, S that pixel value is 1, wherein P and Q are 8 to be connected to, and S and Q are 4 to be connected to, S and
P is 8 connections.
By taking 8 connections as an example, the pixel for being 1 to pixel value carries out connection domain lookup, it is assumed that A is a company in bianry image
Lead to domain, a point is known as P in A, then can pass through following iterative progress for the lookup of connected domain A:
X0=P
Work as Xk=Xk-1When, algorithmic statement, and A=Xk.Wherein, B indicates structural element,Indicate image expansion operation, Y
For original bianry image.
It is illustrated in figure 7 connection domain lookup schematic diagram, using 8 connected relations, structural element such as Fig. 7 (a), Fig. 7 (b) are to connect
The starting point P of logical domain lookup, black elements are extracted element, and grey is still undrawn element, and Fig. 7 (c) is the company of lookup
Logical domain first time iteration as a result, Fig. 7 (d) be search second of iteration of connected domain as a result, Fig. 7 (e) is to be connected to domain lookup most
Whole result schematic diagram.According to the method for connection domain lookup, the connected domain needs after lookup are marked, until all pixels are
Until 1 point has all marked, this completes the lookups of connected domain.
Step 5, in all connected domains, the first connected domain and the second connected domain are filtered out, first connected domain is
It only include the connected domain of an isolated point mark, second connected domain is the point mark composition that threshold value is enlarged beyond in distance dimension
Connected domain, to obtain the remaining connected domain comprising Targets Dots；It is obtained according to the remaining connected domain comprising Targets Dots
To target information, the cohesion to Targets Dots is realized.
In step 5, the remaining connected domain according to comprising Targets Dots obtains target information, realizes to target point
The cohesion of mark, specifically includes:
The centroid position for obtaining the remaining connected domain comprising Targets Dots, by the centroid position in original echo number
Range information of the corresponding distance unit as target in, by the centroid position in raw radar data corresponding orientation
Azimuth information of the unit as target；
The corresponding raw radar data of remaining connected domain comprising Targets Dots is obtained, according to described comprising target point
Distance unit, localizer unit and corresponding envelope detection knot where the corresponding raw radar data of remaining connected domain of mark
Fruit acquires range information, azimuth information and the amplitude information of target.
Specifically, in the embodiment of the present invention, when measured data target echo is weaker, in order to retain Targets Dots as far as possible,
To which the point mark used filters out criterion are as follows:
(1) independent mark is filtered out, according to target echo characteristic distributions, usual target is in orientation or distance
On can extend, therefore can according to Targets Dots distance or orientation extension size extension filter out some False Intersection Points marks；
(2) corresponding to echo data to be filtered out apart from the more point mark of upper extension, the maximum value of distance unit according to
The estimation parameter of target and system determines；
(3) in the case where system allows, more point mark should be retained as far as possible, so point mark is differentiated and filters out criterion and want
Increase according to the actual situation or reduces.
Experiment content and result
Experiment 1, echo data are the echo data of certain air search radar admission, are illustrated in figure 8 certain interception to empty data
Part, data are spliced by the data in the same orientation and distance segment that intercept in 9 scan periods of radar.
Target echo data are shown as shown in figure 8, the echo data after sliding window accumulation is as shown in Figure 9.It is as shown in Figure 10
Treated for constant false alarm as a result, it is after two-value sliding window detects to be as shown in figure 11 as a result, Figure 11 (a) is two-value sliding window detection M/N
In criterion, testing result when M=2, N=5, Figure 11 (b) is that two-value sliding window detects in M/N criterion, detection when M=3, N=5
As a result.It is as shown in figure 12 after bianry image dilation erosion as a result, as shown in figure 13 is target provided in an embodiment of the present invention
The Plot coherence result of Plot coherence method.
As seen from Figure 8, there are the distribution of stronger clutter around target, the data of single interception are difficult to judge target point
The position of mark, therefore can substantially see by spliced data the position of target, the estimation to target position can be passed through
Plot coherence effect is verified.
As seen from Figure 9, target is not obvious the improvement situation of weak signal target, still after non-inherent accumulation
Still there is certain benefit, mainly have certain smoothing effect to radially distributed meteorological clutter, to followed by permanent empty
Alert detection is advantageous.
As seen from Figure 10, Targets Dots and False Intersection Points mark are influenced by signal processing and CFAR detection in left point mark
It is bigger, it compares after CFAR detection with raw radar data, can see together by the substantially judgement to target position
The point mark separating phenomenon of one target is more obvious, and place marked in the figure is the point mark of Target Splitting.
As seen from Figure 11, the detection of two-value sliding window can effectively improve the influence of Target Splitting bring, while reduce false-alarm,
But when using M/N criterion, threshold value crosses point mark of the conference loss compared with weak signal target, and sliding window will appear Targets Dots division when detecting
Phenomenon is usually solved by the method for reducing threshold value, and the too low False Intersection Points mark of threshold value can increase, on mark quality influence compared with
Greatly.Consider so threshold value needs to compromise.As seen from Figure 12, expansion corruption is carried out by the bianry image after detecting to sliding window
Erosion operation latter aspect ensure that on the other hand target resolution capability is also fairly effective to the improvement of Target Splitting.
As seen from Figure 13, improved method is successfully realized Plot coherence, and Targets Dots cohesion achieves preferably
Effect.
To sum up, according to the experimental results, the method for the present invention can effectively realize Plot coherence, while reduce Target Splitting
Influence to Plot coherence.
Technical solution provided in an embodiment of the present invention compares the Plot coherence of sliding window method, the Plot coherence based on image procossing
Technology substantially increases target component precision and target resolution capability, and the related algorithm such as present invention uses in image two
Value image expansion caustic solution can further improve the problem of there may be Target Splittings in sliding window method Plot coherence.Sliding window method
Benefit be that can effectively overcome Target Splitting and false-alarm bring to influence, but the selection of sliding window method threshold value is to Plot coherence shadow
Sound is larger, and this point can be improved by shape filtering in image procossing.So in conjunction with the excellent of sliding window method and image procossing
A mark is transformed to image area and handled by point, using connected domain lookup algorithm and image expansion corrosion treatment, is on the one hand used
The detection of two-value sliding window is handled target echo data and is put mark judgement, on the other hand be can use image processing method and is overcome
Point mark division, to obtain the target state estimator parameter of better quality.The method that this patent is mentioned combines at sliding window method and image
Reason method, improved method are mainly reflected in the processing first is that using Slide-window detector and CFAR detection to echo data, and second
It is the influence for improving Targets Dots division using shape filtering method in image procossing.Improved Plot coherence method is solidifying to mark
The research of poly- technology has important reference value and research significance, and measured data demonstrates the validity of proposed method.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (6)
1. a kind of Targets Dots condensing method of radar return data, which is characterized in that described method includes following steps:
Step 1, the raw radar data of multiple scan periods of the radar Jing Guo process of pulse-compression, the original echo number are obtained
According to for comprising the 2-D data apart from peacekeeping azimuth dimension, and the raw radar data of the multiple scan period along azimuth dimension successively
Arrangement；
Step 2, envelope detection and non-inherent accumulation are successively carried out to the raw radar data, the echo after obtaining pulse accumulation
Data；
Step 3, CFAR detection is carried out to the echo data after the pulse accumulation, the number of echoes after obtaining CFAR detection
According to, and then two-value quantification treatment and the two-value sliding window along azimuth dimension are successively carried out to the echo data after the CFAR detection
Detection, the echo data after obtaining the detection of two-value sliding window, the echo data after the two-value sliding window detection is 2-D data, wherein
One-dimensional representation distance unit, the umber of pulse of another one-dimensional representation azimuth dimension；
Step 4, corresponding binary image data, and bianry image number are converted by the echo data after two-value sliding window detection
According to length correspond to the distance unit of the echo data after two-value sliding window detection, described in the width of binary image data is corresponding
The umber of pulse of echo data after the detection of two-value sliding window carries out the operation for first expanding post-etching to the binary image data, from
And obtain all connected domains in the binary image data；
Step 5, in all connected domains, the first connected domain and the second connected domain are filtered out, first connected domain is only to wrap
Connected domain containing an isolated point mark, second connected domain are that the company of the point mark composition of threshold value is enlarged beyond in distance dimension
Logical domain, to obtain the remaining connected domain comprising Targets Dots；Mesh is obtained according to the remaining connected domain comprising Targets Dots
Information is marked, realizes the cohesion to Targets Dots.
2. a kind of Targets Dots condensing method of radar return data according to claim 1, which is characterized in that described
After step 1, and before the step 2, the method also includes: cancellation is carried out to the raw radar data, is inhibited
Clutter in raw radar data.
3. a kind of Targets Dots condensing method of radar return data according to claim 1, which is characterized in that the step
Rapid 2 specifically:
Envelope detection is carried out to the raw radar data, to obtain raw radar data in respective distances-localizer unit
Amplitude information；Distance unit is sampled point of the raw radar data in distance dimension, and localizer unit is the original echo
Sampled point of the data in azimuth dimension；
Sliding window accumulation is carried out on adjacent pulse to the echo data after envelope detection, the echo data after obtaining pulse accumulation.
4. a kind of Targets Dots condensing method of radar return data according to claim 1, which is characterized in that the step
Rapid 3 specifically:
Sliding window detection is carried out along distance unit to the echo data after the pulse accumulation, respectively by reference units all in window
The mean value of amplitude is set as the threshold value of sliding window detection multiplied by the value that threshold factor obtains；By the amplitude of unit to be detected with it is described
The threshold value of sliding window detection is compared, will be to be checked if the amplitude of unit to be detected is greater than the threshold value of sliding window detection
The data for surveying unit retain, and otherwise set 0 for the data of the unit to be detected；The threshold factor is preset
Value；
And then the data that will be greater than the unit to be detected of the threshold value of the sliding window detection are set as 1, complete to CFAR detection
Echo data afterwards carries out the process of two-value quantification treatment；
When carrying out the detection of two-value sliding window using M/N criterion along azimuth dimension to the echo data after progress two-value quantification treatment, M is indicated
The detection threshold of two-value sliding window detection, N indicate the length of window used when the detection of two-value sliding window.
5. a kind of Targets Dots condensing method of radar return data according to claim 1, which is characterized in that step 4
In, the operation for first expanding post-etching is carried out to the binary image data, to obtain all in the binary image data
Connected domain specifically includes:
Selecting structure element carries out the operation for first expanding post-etching to binary image data with the structural element of selection, obtains elder generation
The binary image data of post-etching is expanded, the lookup of connected domain is then carried out to the binary image data for first expanding post-etching, from
And obtain all connected domains in the binary image data.
6. a kind of Targets Dots condensing method of radar return data according to claim 1, which is characterized in that step 5
In, target information is obtained according to the remaining connected domain comprising Targets Dots, realizes the cohesion to Targets Dots, it is specific to wrap
It includes:
The centroid position for obtaining the remaining connected domain comprising Targets Dots, by the centroid position in raw radar data
Range information of the corresponding distance unit as target, by the centroid position in raw radar data corresponding localizer unit
Azimuth information as target；
Obtain it is described include Targets Dots the corresponding raw radar data of remaining connected domain, according to described comprising Targets Dots
Range-azimuth unit and corresponding envelope detection where the corresponding raw radar data of remaining connected domain is as a result, acquire mesh
Target range information, azimuth information and amplitude information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710111371.0A CN106918807B (en) | 2017-02-28 | 2017-02-28 | A kind of Targets Dots condensing method of radar return data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710111371.0A CN106918807B (en) | 2017-02-28 | 2017-02-28 | A kind of Targets Dots condensing method of radar return data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106918807A CN106918807A (en) | 2017-07-04 |
| CN106918807B true CN106918807B (en) | 2019-07-09 |
Family
ID=59454618
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710111371.0A Active CN106918807B (en) | 2017-02-28 | 2017-02-28 | A kind of Targets Dots condensing method of radar return data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106918807B (en) |
Families Citing this family (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107561518B (en) * | 2017-07-27 | 2020-10-09 | 中国船舶重工集团公司第七二四研究所 | Three-coordinate radar point trace condensation method based on two-dimensional sliding window local extreme value |
| CN107861107B (en) * | 2017-10-23 | 2021-02-05 | 电子科技大学 | Double-threshold CFAR (computational fluid dynamics) and trace point agglomeration method suitable for continuous wave radar |
| CN108254727A (en) * | 2017-12-08 | 2018-07-06 | 西安电子科技大学 | A kind of radar plot condensing method based on Contour extraction |
| CN108490410B (en) * | 2018-03-06 | 2021-08-17 | 中国船舶重工集团公司第七二四研究所 | Two-coordinate radar sea target joint detection and tracking method |
| CN108120961B (en) * | 2018-03-08 | 2023-07-14 | 合肥乾升电子科技有限公司 | Method and device for realizing PPI display enhancement of detection video based on target azimuth expansion |
| CN108680904A (en) * | 2018-05-02 | 2018-10-19 | 深圳市华讯方舟雷达技术装备有限公司 | A kind of suppressing method of marine navigation radar co-channel interference |
| CN108614264B (en) * | 2018-05-04 | 2023-08-04 | 武汉华讯国蓉科技有限公司 | Radar target point trace condensation method based on connected marking rule |
| CN108663667A (en) * | 2018-05-11 | 2018-10-16 | 清华大学 | Range extension target detection method and system under the incomplete observation of radar |
| CN109061593B (en) * | 2018-07-31 | 2020-12-01 | 安徽四创电子股份有限公司 | Target extraction method based on ground monitoring radar spectrum data |
| CN109100697B (en) * | 2018-07-31 | 2020-12-22 | 安徽四创电子股份有限公司 | Target condensation method based on ground monitoring radar system |
| CN109375177B (en) * | 2018-08-30 | 2020-12-08 | 安徽四创电子股份有限公司 | Moving target detection method for airport scene surveillance radar system |
| CN109254271B (en) * | 2018-08-30 | 2021-05-04 | 安徽四创电子股份有限公司 | Static target suppression method for ground monitoring radar system |
| CN109375185A (en) * | 2018-10-12 | 2019-02-22 | 中国电子科技集团公司第十四研究所 | A kind of point mark filtering technique based on bianry image connected domain area measurement |
| CN111175709B (en) * | 2019-12-27 | 2023-02-24 | 中国电子科技集团公司第十四研究所 | Error suppression-based large-range meteorological radar-oriented jigsaw method |
| CN111366924B (en) * | 2020-05-09 | 2022-11-25 | 苏州雷森电子科技有限公司 | Radar target point trace aggregation method based on multi-dimensional connected domain processing |
| CN112051568B (en) * | 2020-09-01 | 2023-06-30 | 西安电子工程研究所 | Pitching angle measurement method of two-coordinate radar |
| CN112541872B (en) * | 2020-12-08 | 2024-02-27 | 陕西长岭电子科技有限责任公司 | Meteorological radar display effect optimization method |
| CN112305508B (en) * | 2020-12-23 | 2021-04-30 | 北京海兰信数据科技股份有限公司 | Echo splitting processing method and system based on marine radar |
| CN112835001B (en) * | 2020-12-29 | 2023-08-08 | 西安电子科技大学 | Sea surface target radar trace condensation method, electronic equipment and storage medium |
| CN113868496A (en) * | 2021-09-14 | 2021-12-31 | 中国人民解放军海军航空大学航空作战勤务学院 | Data association method and device, electronic equipment and computer readable storage medium |
| CN113820685B (en) * | 2021-10-19 | 2024-07-19 | 中安锐达(北京)电子科技有限公司 | Primary trace point condensation method based on fan block structure |
| CN114415123B (en) * | 2022-04-01 | 2022-07-19 | 北京海兰信数据科技股份有限公司 | Non-coherent neighborhood based weighting pulse accumulation processing method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102288945A (en) * | 2011-05-11 | 2011-12-21 | 成都成电电子信息技术工程有限公司 | Image enhancing method for ship radar |
| CN104036146A (en) * | 2014-06-26 | 2014-09-10 | 中国电子科技集团公司第二十八研究所 | Trace point clustering method for clustering trace points of radar targets |
| CN104391294A (en) * | 2014-11-27 | 2015-03-04 | 中国船舶重工集团公司第七二四研究所 | Connection domain characteristic and template matching based radar plot correlation method |
| CN105044699A (en) * | 2015-07-08 | 2015-11-11 | 中国电子科技集团公司第二十八研究所 | Radar plot centroid method based on Radon-Fourier transformation |
| CN105259553A (en) * | 2015-11-11 | 2016-01-20 | 西安电子科技大学 | Micro-motion target scattering point track association method based on distance-instant Doppler image |
| RU2596610C1 (en) * | 2015-06-16 | 2016-09-10 | Федеральное государственное унитарное предприятие "Государственный научно-исследовательский институт авиационных систем" | Method of search and detection of object |
| CN106093946A (en) * | 2016-05-27 | 2016-11-09 | 四川九洲空管科技有限责任公司 | A kind of target condensing method being applicable to scene surveillance radar and device |
| CN106093890A (en) * | 2016-08-26 | 2016-11-09 | 零八电子集团有限公司 | The method setting up residual clutter suppression radar residual spur |
-
2017
- 2017-02-28 CN CN201710111371.0A patent/CN106918807B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102288945A (en) * | 2011-05-11 | 2011-12-21 | 成都成电电子信息技术工程有限公司 | Image enhancing method for ship radar |
| CN104036146A (en) * | 2014-06-26 | 2014-09-10 | 中国电子科技集团公司第二十八研究所 | Trace point clustering method for clustering trace points of radar targets |
| CN104391294A (en) * | 2014-11-27 | 2015-03-04 | 中国船舶重工集团公司第七二四研究所 | Connection domain characteristic and template matching based radar plot correlation method |
| RU2596610C1 (en) * | 2015-06-16 | 2016-09-10 | Федеральное государственное унитарное предприятие "Государственный научно-исследовательский институт авиационных систем" | Method of search and detection of object |
| CN105044699A (en) * | 2015-07-08 | 2015-11-11 | 中国电子科技集团公司第二十八研究所 | Radar plot centroid method based on Radon-Fourier transformation |
| CN105259553A (en) * | 2015-11-11 | 2016-01-20 | 西安电子科技大学 | Micro-motion target scattering point track association method based on distance-instant Doppler image |
| CN106093946A (en) * | 2016-05-27 | 2016-11-09 | 四川九洲空管科技有限责任公司 | A kind of target condensing method being applicable to scene surveillance radar and device |
| CN106093890A (en) * | 2016-08-26 | 2016-11-09 | 零八电子集团有限公司 | The method setting up residual clutter suppression radar residual spur |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106918807A (en) | 2017-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106918807B (en) | A kind of Targets Dots condensing method of radar return data | |
| US7741990B2 (en) | Method and apparatus for detection of moving objects by SAR images | |
| Wen et al. | Video SAR moving target detection using dual faster R-CNN | |
| US6897802B1 (en) | Fusion of shape and multiscale features for unknown target rejection | |
| US7116265B2 (en) | Recognition algorithm for the unknown target rejection based on shape statistics obtained from orthogonal distance function | |
| Ran et al. | An adaptive fast factorized back-projection algorithm with integrated target detection technique for high-resolution and high-squint spotlight SAR imagery | |
| CN112184749B (en) | Moving target tracking method based on video SAR cross-domain combination | |
| CN112731307B (en) | RATM-CFAR detector based on distance-angle joint estimation and detection method | |
| CN107942329A (en) | Motor platform single-channel SAR is to surface vessel object detection method | |
| Inggs et al. | Ship target recognition using low resolution radar and neural networks | |
| Kang et al. | ISAR cross-range scaling via joint estimation of rotation center and velocity | |
| CN109100696A (en) | Microinching target removing method based on point clutter figure | |
| CN116027318A (en) | Method, device, electronic equipment and storage medium for multi-sensor signal fusion | |
| CN108020834B (en) | Improved EDPCA (enhanced data analysis component analysis) -based moving target detection method and device and electronic equipment | |
| CN112435249A (en) | Dynamic small target detection method based on periodic scanning infrared search system | |
| EP1515160B1 (en) | A target shadow detector for synthetic aperture radar | |
| Kim et al. | Deep-learning based multi-object detection and tracking using range-angle map in automotive radar systems | |
| Li et al. | Multi-scale ships detection in high-resolution remote sensing image via saliency-based region convolutional neural network | |
| Liu et al. | High-precision imaging and dense vehicle target detection method for airborne single-pass circular synthetic aperture radar | |
| KR101813357B1 (en) | Clutter elimination Method and Apparatus for applying variable data map and variable statistic characteristics information for identifying underwater fixed target and clutter | |
| CN115508833A (en) | GNSS BI-SAR river boundary detection system | |
| Tan et al. | A ship detection method based on YOLOv7 in range-compressed SAR data | |
| CN116027288A (en) | Method and device for generating data, electronic equipment and storage medium | |
| Wang et al. | An ISAR Autofocus Imaging Algorithm Based on FCN and Transfer Learning | |
| Li et al. | Ship Focusing and Positioning Based on 2-D Ambiguity Resolving for Single-Channel SAR Mounted on High-Speed Maneuvering Platforms With Small Aperture |
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 |