CN109903307A - A kind of moving target detecting method based on SAIR - Google Patents
A kind of moving target detecting method based on SAIR Download PDFInfo
- Publication number
- CN109903307A CN109903307A CN201910096256.XA CN201910096256A CN109903307A CN 109903307 A CN109903307 A CN 109903307A CN 201910096256 A CN201910096256 A CN 201910096256A CN 109903307 A CN109903307 A CN 109903307A
- Authority
- CN
- China
- Prior art keywords
- moving target
- coordinate
- target
- sair
- image
- 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.)
- Withdrawn
Links
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a kind of moving target detecting methods based on SAIR, the method comprising the steps of: being scanned using rotary scanning pattern synthesis aperture interferometric radiometer to field range, according to the One Dimensional Projection image of time of measuring arrangement measurement, two-dimensional time projected image is obtained；Projected image is denoised, and extracts the projection coordinate of target object；The equation of locus of moving target is obtained according to motion model；Projection coordinate is substituted into equation of locus, obtains the kinematic parameter of object.The present invention can effectively track Secondary movement time-varying target, and detection accuracy is high.
Description
Technical field
The present invention relates to a kind of moving target detecting methods, belong to millimeter-wave near-field imaging field more particularly to a kind of base
In the moving target detecting method of SAIR.
Background technique
Mm-wave imaging radiometer is the powerful passive sensor of high-definition remote sensing.It is different from microwave imaging, wavelength compared with
Higher spatial resolution may be implemented in short mm-wave imaging.Compared with optics and infra-red radiation, millimeter wave can penetrate clothing
The dielectric materials such as object, plastics, timber are lost relatively small.In addition, the target of different materials has different millimeter wave properties,
And it is easy to be detected by millimeter wave imaging sensor.Based on these abilities, millimeter wave imaging sensor has been applied to a variety of
Application field, including the monitoring of geochemistry remote sensing target and precision target imaging, aircraft safety land, highway communication prison in mist
Survey, civilian remote sensing and hidden threat target detection due to safety concerns etc..
Since Moving target detection is beneficial to many applications, such as target following or classification, track of vehicle determination etc..People
It has been developed that several for detecting mobile mesh calibration method.Currently, mobile mesh object detection method can be divided into following two categories:
One is moving target is extracted from image sequence, the kinematic parameter of moving target is then estimated.Obviously it must survey first in this way
The image sequence of entire target scene is measured, the information and algorithm complexity of this method are usually larger, and its real-time performance is poor；Separately
One kind is the kinematic parameter for receiving extraction moving target in signal directly from detector.For example, SAR (synthetic aperture radar) can
The position and speed of mobile target is estimated from radar echo signal by relevant Detection of Moving Objects.Pass through this letter
Breath amount is small, and the low method of algorithm complexity can simply implement the high real-time detection to moving target.However, traditional fortune
Moving-target is detected mainly for speed stable simple motion target.These detection methods, which are not suitable for detection, has acceleration
With the order motion target of steering.
Summary of the invention
Goal of the invention: the present invention propose it is a kind of can be to the side that Secondary movement target is detected using rotary scanning type SAIR
Method.
Technical solution: a kind of moving target detecting method based on SAIR of the present invention, comprising steps of
(1) field range is scanned using rotary scanning pattern synthesis aperture interferometric radiometer, is arranged according to time of measuring
The One Dimensional Projection image for arranging measurement, obtains two-dimensional time projected image；
(2) projected image is denoised, and extracts the projection coordinate of moving target；
(3) equation of locus of moving target is obtained according to motion model；
(4) projection coordinate for obtaining step (2) substitutes into equation of locus, obtains the kinematic parameter of moving target.
Further, the target object is single order or Secondary movement target.
The utility model has the advantages that detection method of the present invention has the advantage that
(1) it is suitable for detection Secondary movement target, i.e., the moving target of the multinomial track with acceleration and steering；
(2) be suitable for track the kinematic parameter of time-varying target, according to time of measuring arrangement nearest (about 1-2s) it is one-dimensional when
Between projected image TθaTwo-dimensional time projected image T is formed according to the detection principle of RS-SAIRt-p, when making m- projected footprint with
Suitable motion model fitting finds out parameters.
Detailed description of the invention
Fig. 1 is the geometric graph of RS-SAIR of the present invention；
Fig. 2 is the one-dimensional thinned array layout of the present invention；
Fig. 3 is the mobile target scene schematic diagram of motion profile of the present invention；
Fig. 4 is image T of the present inventiont-pIn mobile target projected footprint；
Fig. 5 is the result that the present invention is fitted mobile target projection coordinate；
Fig. 6 is the result of present invention fitting moving target position.
Specific embodiment
Further description of the technical solution of the present invention with reference to the accompanying drawings and examples.
As shown in Figure 1, Sparse array is placed in XOY plane, and rotated around point O.θaIt is array rotation angle.
For giving θaThinned array, antenna is visibility function to the correlation between the reception signal of (c, l)
Sample.Thus we are available, it is seen that property sample Vc,lIt can simplify and be expressed as
Wherein, (x, y) is the coordinate of moving target, and T is the brightness temperature image of target scene, and k=2 π/λ is round wave
Number.
With exp [(- jk (Rc-Rl)] antenna is indicated to the phase difference between (c, l), this is the key factor of SAIR.?
XrOYrProjected coordinate system in, antenna is (X to the coordinate of (c, l)C, 0) and (Xl, 0), moving target coordinate is (xr,yr).It will
Target is to antenna to the distance R of (c, l)cAnd RlIt is indicated by Taylor expansion are as follows:
Formula (2) and (3) are substituted into formula (1), and definition space coordinate u=k (Xl-Xc)/R, our available angles
θaUnder one-dimensional visibility function Vθa:
Wherein,It is phase modification item, v=0, can be improved the imaging of near field SAIR at this time
Precision.TθaIt is One Dimensional Projection image, can be rebuild by FFT method
In general, having different θaOne Dimensional Projection image can be used for rebuilding brightness temperature image T.However, the letter is dedicated
In detecting mobile target, One Dimensional Projection image is directly used in the position for extracting target.For given moving target MT, work as array
Projection coordinate x when with constant angular velocity ω rotation, in One Dimensional Projection imager-MTIt may be expressed as:
xr-MT(t)=RMT(t)cos(ωt+θa0+θMT(t)) (6)
Wherein, θa0It is the initial angle of Sparse array, (xMT(t),xMT(t)) be target MT in xoy coordinate seat
Mark.For the sake of simplicity, we mainly analyze mobile target using following motion model herein:
xMT(t)=axt2+vxt+x0, yMT(t)=ayt2+vyt+y0 (8)
Wherein, [ax,vx,x0,ay,vy,y0] it is the kinematic parameter for needing to estimate, each element, which respectively indicates, indicates the direction x
Acceleration, the speed in the direction x, the initial position in the direction x, the acceleration in the direction y, the speed in the direction y, the direction y initial position；
xMT(t) motion profile of the moving target in x-axis, y are indicatedMT(t) motion profile of moving target on the y axis is indicated.It is worth note
Meaning, the track of any moving target is all continuous function, can be indicated with polynomial function.Accordingly it is also possible to pass through this
The detection method is invented to estimate Secondary movement parameter.
According to formula (6), the projection coordinate x of MTr-MT(t) will change with array rotation as θa=ω t+ θa0, and
Its track meets the cosine distribution modulated by the kinematic parameter of MT.Based on the fact that the detailed mistake of detection method of the present invention
Journey is as follows:
(1) the One Dimensional Projection image T arranged according to time of measuringθaTo constitute two-dimensional time projected image Tt-p。
(2) after subtracting surface noise, through by when m- perspective view Tt-pIn high brightness point take out pixel it is bright
The gray value of the image pixel value of the average brightness of degree > image, as the projection coordinate x of targetr-MT。
(3) motion model of the moving target according to shown in formula (8) improves formula (6), to construct the rail of MT
Mark expression formula.
(4) by the projection coordinate x of extractionr-MTIt is fitted to the track expression formula of MT, to obtain the kinematic parameter of MT.
In analysis above, it is assumed that the background of MT is clean background.However, in practice, usually existing in background
Some interference, this will reduce the extraction accuracy of the track MT.Fortunately, the image T of static backgroundt-pThere is array rotation
Periodical image.Therefore, it is necessary first to from image Tt-pIn subtract the projected image of background to be applied to complicated real scene.
Numerical-Mode fitted results: main analog parameter is listed in table 1.Minimum baseline separation delta u is 1.1cm.As shown in Figure 2,
The synthetic aperture of thinned array with 16 antennas is about DSA=90 Δ u=1m.Initial angle θa0It is set as 0.In order to accurate
The imaging process of simulation RS-SAIR, the reception that the integration operation of the radiated wave generated by all radiation sources obtains antenna are believed
Number.Then, visibility sample is calculated by receiving the cross-correlation calculation between signal.
Finally, rebuilding One Dimensional Projection image according to the FFT method in formula (5).In order to verify proposed HMTD method,
The simulated experiment of two aerial target and ground target has been carried out respectively.Kinematic parameter is set as [40m/s2,-60m/s,60m,
20m/s2,40m/s,30m]。
The simulation parameter of 1 thinned array RS-SAIR of table
| Antenna amount | 16 | Wavelength X | 3mm |
| Baseline separation delta u | 1.1cm | Detecting distance R | 5000m |
| Aperture synthesis DSA | 1m | Angular velocity omega | 0.5rad/s |
As shown in Fig. 3,4,5,6, either aerial target or ground target can accurately rebuild mobile target
Track and position.Due to the interference of complex background, the playback accuracy of ground target is slightly worse than aerial target.For objective comparison,
The kinematic parameter of measurement is listed in Table 2 below, and the RMSE of position of the target in Fig. 6 calculates as follows:
The estimated result of table 2 kinematic parameter and location error
| Parameter | [ax,vx,x0,ay,vy,y0] | RMSE |
| Aerial target | 39.0,-61.2,63.9,21.7,33.5,35.4 | 0.05% |
| Ground target | 38.5,-60.3,65.3,20.7,34.6,38.3 | 0.10% |
As shown in table 2, action reference variable is fine, and error is small, and location error (RMSE) is very small (less than 1%), even if
For ground target and so.This shows that detection method proposed by the invention can effectively track order motion target.
Claims (5)
1. a kind of moving target detecting method based on SAIR, which is characterized in that comprising steps of
(1) field range is scanned using rotary scanning pattern synthesis aperture interferometric radiometer, is arranged and is surveyed according to time of measuring
The One Dimensional Projection image of amount obtains two-dimensional time projected image；
(2) projected image is denoised, and extracts the projection coordinate of target object；
(3) equation of locus of moving target is obtained according to motion model；
(4) projection coordinate for obtaining step (2) substitutes into equation of locus, obtains the kinematic parameter of object.
2. according to claim 1, it is characterised in that: the moving target is single order or second order.
3. according to claim 1, which is characterized in that the step (2) includes:
(2.1) Sparse array is placed in XOY plane, and is rotated around point O, and array rotation angle is θa, moving target coordinate
For (x, y)；
(2.2) in XrOYrProjected coordinate system in, antenna is (X to the coordinate of (c, l)C, 0) and (Xl, 0), moving target coordinate is
(xr,yr)；Wherein (c, l) is the serial number of antenna pair, c=1 ..., N, l=1 ..., N, and N is the number of antenna pair；
(2.3) according to moving target and antenna to the distance R of (c, l)cAnd Rl, construct angle, θaUnder one-dimensional visibility function Vθa,
And One Dimensional Projection image T is rebuild by FFT methodθa, by when m- perspective view in high brightness point take out, obtain moving target MT
Projection coordinate x in One Dimensional Projection imager-MT, wherein the high brightness point is the picture for the average brightness that brightness is greater than image
Vegetarian refreshments.
4. the moving target detecting method according to claim 1 based on SAIR, which is characterized in that described in step (3)
Motion model are as follows:
xMT(t)=axt2+vxt+x0, yMT(t)=ayt2+vyt+y0
Wherein, ax,vx,x0,ay,vy,y0It is the kinematic parameter for needing to estimate, respectively indicates the acceleration in the direction x, the speed in the direction x
Degree, the initial position in the direction x, the acceleration in the direction y, the speed in the direction y, the direction y initial position.
5. the moving target detecting method according to claim 1 based on SAIR, which is characterized in that the equation of locus
Are as follows:
xr-MT(t)=RMT(t)cos(ωt+θa0+θMT(t))
θMT(t)=arctan [xMT(t)/yMT(t)]
Wherein, θa0It is the initial angle of Sparse array, (xMT(t),xMT(t)) be moving target in XOY coordinate coordinate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910096256.XA CN109903307A (en) | 2019-01-31 | 2019-01-31 | A kind of moving target detecting method based on SAIR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910096256.XA CN109903307A (en) | 2019-01-31 | 2019-01-31 | A kind of moving target detecting method based on SAIR |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109903307A true CN109903307A (en) | 2019-06-18 |
Family
ID=66944488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910096256.XA Withdrawn CN109903307A (en) | 2019-01-31 | 2019-01-31 | A kind of moving target detecting method based on SAIR |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109903307A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538372A (en) * | 2021-07-14 | 2021-10-22 | 重庆大学 | Three-dimensional target detection method and device, computer equipment and storage medium |
-
2019
- 2019-01-31 CN CN201910096256.XA patent/CN109903307A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113538372A (en) * | 2021-07-14 | 2021-10-22 | 重庆大学 | Three-dimensional target detection method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Frolind et al. | Circular-aperture VHF-band synthetic aperture radar for detection of vehicles in forest concealment | |
| Palm et al. | Radargrammetric DEM extraction over urban area using circular SAR imagery | |
| US6150972A (en) | Process for combining multiple passes of interferometric SAR data | |
| CN104849712B (en) | A kind of three-dimensional deformation monitoring system based on many base MIMO SAR | |
| JP2003500658A (en) | Procedures for radar measurements of urban area and landslide motion. | |
| CN109358325B (en) | Terrain inversion method of radar altimeter under rugged terrain background | |
| CN109061638B (en) | Phased array close-range digital imaging method | |
| CN112051568B (en) | Pitching angle measurement method of two-coordinate radar | |
| CN113093184B (en) | Interferometric measurement method based on video synthetic aperture radar | |
| EP3775989B1 (en) | Device, system and method for localization of a target in a scene | |
| Nenashev et al. | The problem of determination of coordinates of unmanned aerial vehicles using a two-position system ground radar | |
| CN107797110B (en) | System imaging method of synthetic aperture projection radiation and synthetic aperture projection radiometer | |
| Yang et al. | A 3-D electromagnetic-model-based algorithm for absolute attitude measurement using wideband radar | |
| CN109903307A (en) | A kind of moving target detecting method based on SAIR | |
| Góes et al. | Refraction Effect in SAR Processing for Focused Subsurface Tomography | |
| Nitti et al. | Automatic GCP extraction with high resolution COSMO-SkyMed products | |
| Klare et al. | DisasterManagement with the MIMO Radar MIRA-CLE Ka: Measurements of a Slowly MovingWall | |
| Wang et al. | Forward looking InSAR based field terrain mapping for unmanned ground vehicle | |
| RU2416806C2 (en) | Method of processing radar signals | |
| CN111679287B (en) | Active video three-dimensional hyperspectral imaging method | |
| Mikhalev et al. | Fusion of sensor data for source localization using the Hough transform | |
| Omori et al. | K-space decomposition based super-resolution three-dimensional imaging method for millimeter wave radar | |
| Aksu et al. | 3D scene reconstruction from multi-sensor EO-SAR data | |
| Moaca et al. | Investigation of displacement measurements performed with a ground-based fixed receiver bistatic SAR simulator | |
| Lv et al. | Vector motion parameter estimation for an approaching missile based on an extended-target model |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190618 |
|
| WW01 | Invention patent application withdrawn after publication |