CN103941241A - Radiation correction method suitable for non-linear track SAR imaging - Google Patents

Abstract

The invention belongs to the field of SAR image radiation correction method for radar signal processing, in particular to a radiation correction method suitable for non-linear track SAR imaging. The radiation correction method suitable for the non-linear track SAR imaging comprises a step of range compression and a step of direction matching, wherein the direction matching comprises a step of motion compensation and a step of illuminance compensation. The radiation correction method includes: enabling a sensor to measure and acquire a data sample of a position posture vector of a radar aircraft; initializing the sample data, and carrying out filtering processing and interpolation processing to acquire an estimated value of the position posture vector of the radar aircraft; according to the estimated value of the position posture vector of the radar aircraft, reconstructing an image for the target position through a back projection algorithm, and carrying out motion compensation and illuminance compensation on each pixel point of the imaging area to obtain a focused SAR image with equal gain. The radiation correction method suitable for the non-linear track SAR imaging is capable of simultaneously correcting the unbalance of the radiation intensity along each of the range and azimuth, eliminating the bad influence of the platform posture jitter on the imaging quality, and effectively improving the non-linear track SAR imaging quality.

Description

A kind of radiation correction method that is applicable to non-rectilinear flight path SAR imaging

Technical field

The invention belongs to SAR in Radar Signal Processing (synthetic-aperture radar, Synthetic Aperture Radar) image radiation bearing calibration field, be specifically related to a kind of radiation correction method that is applicable to non-rectilinear flight path SAR imaging.

Background technology

Conventional SAR is conventionally by obtaining the SAR image of ground scene and target apart from pulse pressure, azimuth match (containing motion compensation), three steps of radiant correction.The first step, apart from pulse pressure, is obtained High Range Resolution by launching large bandwidth signal; Second step azimuth match, be Caliberation Flight position of platform depart from ideal line flight path and the distance that causes to walking about and phase error, and the large aperture that utilizes platform rectilinear motion to form obtains high-resolution lateral separation picture, complete the focusing of SAR image in orientation; The 3rd step radiant correction, makes progress with 1/r in oblique distance according to radar equation ²relation carry out radiant correction (wherein r is the distance that ground scene arrives Texas tower line of motion), obtain the SAR image of gain balance.Typical application system has the CARABAS SAR of Sweden and X-SAR that domestic Zhong electricity group 38 develops and the P-SAR of National University of Defense technology's development.

Unmanned aerial vehicle platform is compared and is had stronger maneuverability, disguise and security with other unloaded platform with spaceborne.UAV system SAR is owing to can the situations such as road, bridge, crossing site, landform, landforms in area of operations being scouted, can be used for disclosing camouflage or evaluate camouflage effectiveness, the feature such as also have that working service is convenient, no one was injured dangerous, will become a kind of very important scouting detection means in the Situation Awareness of future battlefield.Yet, be subject to the impact of air-flow and flight stability etc., especially, when low-latitude flying, usually there is violent positional jitter in small-sized unmanned plane, and flight track often shows as non-rectilinear flight path, simultaneously with violent attitude jitter.Research shows, when flight track non-rectilinear, the skew of position and the variation of attitude will produce considerable influence to the focusing of SAR image.Domestic and international imaging situation when for flight track non-rectilinear, main is all the motion state of utilizing sensor measurement platform, the distance causing due to position deviation ideal line flight path is proofreaied and correct with phase error to walking about, with the SAR image that obtains focusing on.Yet above-mentioned algorithm does not all have to consider that the variation due to platform stance causes that beam center points to the impact of shake on SAR imaging, so these SAR formation methods can not meet the demand of the lower meticulous SAR imaging of high resolving power of platform non-rectilinear flight path motion.

Summary of the invention

The technical problem to be solved in the present invention is to change the shortcoming on image quality impact for ignoring controlling antenna wave beam to point in existing non-rectilinear flight path SAR imaging, proposes a kind of radiation correction method that is applicable to SAR imaging under non-rectilinear flight path condition.

Concrete technical scheme of the present invention is as follows:

A radiation correction method that is applicable to non-rectilinear flight path SAR imaging, comprises the following steps:

(S1) obtain radar return data, echo data is carried out processing apart from pulse pressure, obtain Range Profile;

(S2) sensor measurement obtains carrier of radar position and attitude vector data sample; Data sample is carried out after initialization, and by filtering, interpolation processing, obtain the estimated value of carrier of radar position and attitude vector;

(S3), according to the Range Profile of described step (S1) accumulation and the carrier of radar position and attitude vector estimated value in step (S2), rebuild SAR image f _bP(p), wherein, p represents the location point in imaging region;

(S4) definition illumination I (p), and it is as follows to calculate illumination:

Wherein, c is the light velocity, and u is the slow time, g (p _u, p) be antenna radiation pattern function, p _ufor u moment carrier aircraft position and attitude vector, R (p _u, p) be u carrier aircraft position p constantly _uto the distance of location point p in imaging region, w (p _u, p) be the window function of control punch electrical path length;

(S5) image after rebuilding in described step (S3) is carried out to illuminance compensation, the SAR image σ (p) after the focusing of acquisition gain balance, that is:

Principle of work of the present invention is as follows:

The first step, apart from pulse pressure.

The scattering function being defined as the target at regional location point p place is σ (p), and the SAR echo of p place target can be expressed as:

$s_R^0(t,p_u)=\frac{c^2}{{4R}^2(p_u,p)}\mathrm{\σ}\left(p\right)s(t-\frac{2}{c}R(p_u,p))---\left(1\right)$

Wherein, c is the light velocity, and t is the fast time, and u is the slow time, p _ufor u constantly carrier aircraft position and attitude vector (comprise distance, highly, orientation, crab angle, the angle of pitch, roll angle), s (t) is for transmitting, R (p _u, p) be u carrier aircraft position p constantly _uthe distance of ordering to imaging region p:

R(p _u,p)=||p _u-p|| ₂ (2)

The present invention considers that radar antenna has specific antenna beam pattern, g (p _u, p) be antenna radiation pattern function, the SAR echo of location point p place target can be modified to:

$s_R^1(t,p_u)=\frac{c^2}{{4R}^2(p_u,p)}g(p_u,p)\mathrm{\σ}\left(p\right)s(t-\frac{2}{c}R(p_u,p))---\left(3\right)$

Right after pulse pressure, obtain Range Profile s _r(t, p _u):

$s_R(t,p_u)=\frac{c^2}{{4R}^2(p_u,p)}g(p_u,p)\mathrm{\σ}\left(p\right)s_{\mathrm{PSF}}(t-\frac{2}{c}R(p_u,p))---\left(4\right)$

(4) S in formula _pSF() is a spreading function.

Second step, direction position coupling.

(1) sensing data pre-service.

First, the carrier aircraft position and attitude vector data sample { p that sensor measurement is obtained _u} ₀, and to { the p obtaining _u} ₀data are carried out initialization, that is: to { p _u} ₀data are carried out pre-service, reject { p _u} ₀in abnormal data (because the motion state of platform can flip-flop, the data of sudden change are abnormal data); Then, to { p _u} ₀data are carried out filtering, reduce and measure the impact of noise on estimated accuracy; Next to based on filtered { p _u} ₀data are carried out polynomial interpolation, obtain the carrier aircraft position and attitude vector data { p identical with SAR echo data turnover rate _u} ₁; Finally, conventionally in antenna phase center and Bu space, sensor measurement position in same point, according to { the p after interpolation _u} ₁antenna phase center positional value when data calculate SAR Range Profile, that is: the carrier aircraft position and attitude vector estimated value { p of corresponding SAR Range Profile data _u.

(2) carry out motion compensation and illuminance compensation.

The process of motion compensation is: utilize and estimate { p _uand Range Profile s _r(t, p _u), according to typical time domain imaging algorithms (that is: rear orientation projection (BP-Back-Projection) algorithm), reconstruct the image f at p place, imaging region position _bP(p), proofreaied and correct flying platform position deviation ideal line flight path and the Range Profile that causes is walked about and phase error, that is:

$f_{\mathrm{BP}}\left(p\right)=\∫\∫w(p_u,p)s_R(t,p_u)d{p}_u\mathrm{dt}---\left(5\right)$

(5) w (p in formula _u, p) be the window function of control punch electrical path length, for guaranteeing that image orientation resolution is not with change of distance, its average weighted form is:

In formula, the angle between ∠ () statement two vectors, Φ ₁for selected accumulation angular breadth.

To in (4) formula substitution (5) formula, obtain:

$f_{\mathrm{BP}}\left(p\right)=\∫\∫w(p_u,p)\frac{c^2}{4R^2(p_u,p)}g(p_u,p)\mathrm{\σ}\left(p\right)S_{\mathrm{PSF}}(t-\frac{2}{c}R(p_u,p))d{p}_u\mathrm{dt}---\left(7\right)$

Conventionally in imaging model, will put spreading function S _pSF() is approximate with impulse function δ (), and (7) formula can be write as:

$\begin{array}{c}{f}_{\mathrm{BP}}\left(p\right)=\∫\∫w(p_u,p)\frac{c^2}{{4R}^2(p_u,p)}g(p_u,p)\mathrm{\σ}\left(p\right)\mathrm{\δ}(t-\frac{2}{c}(p_u,p))d{p}_u\mathrm{dt}\\ =\mathrm{\σ}\left(p\right)\∫w(p_u,p)\frac{c^2}{{4R}^2(p_u,p)}g(p_u,p)d{p}_u\end{array}---\left(8\right)$

It is as follows that the present invention defines illumination:

$I\left(p\right)=\frac{c^2}{4}\∫\frac{g(p_u,p)}{R^2(p_u,p)}w(p_u,p)d{p}_u---\left(9\right)$

From (6) formula and (9) formula, can find out, illumination and target property are irrelevant, are the functions of imaging geometry, antenna radiation pattern and sensing thereof, accumulation aperture length.What reflect is the total intensity of the ideal point target reflection echo signal of unit amplitude in radar beam range of exposures.

According to the definition of (9) formula, (8) formula can be written as again:

f _BP(p)＝I(p)σ(p) (10)

From (10) formula, the SAR image after reconstruction is the product of ground scene and its scattering strength of target and illumination.In non-rectilinear flight path SAR imaging, the variation of carrier of radar position and attitude vector will cause the variation of brightness value, causes the SAR image illumination after rebuilding unbalanced.

Illuminance compensation process is: utilize carrier of radar position and attitude vector estimated value { p _u, and the radar directional pattern g (p measuring _u, p), according to (9) formula, can calculate the illumination I (p) at p place, imaging region position.According to (10) formula, can obtain the f at p place, imaging region position _bP(p) carry out after illuminance compensation, calculate σ (p).

(3) image forms.

Each pixel p to imaging region all carries out above-mentioned motion compensation and illuminance compensation, in traversal imaging region, after each pixel, obtains the focusing SAR image σ (p) of ground scene and target gain equilibrium.

Adopt the present invention can reach following technique effect:

The one, data strong adaptability, easy to utilize.Radar directional pattern required for the present invention can be measured in advance.Only utilize the positional information in sensing data to compare with conventional SAR imaging, the carrier aircraft position and attitude data message that the present invention takes full advantage of in sensing data is realized illuminance compensation, does not need to obtain extra data in real time imagery.Thereby the present invention is all applicable to all ideal line flight path SAR imaging, has extensive data adaptability.The 2nd, effectively improved non-rectilinear flight path SAR image quality.With conventional SAR formation method only consider to proofread and correct radiancy along distance to unbalanced comparing, adopt illuminance compensation method of the present invention, can proofread and correct simultaneously radiancy along distance to orientation to unbalanced, eliminate the adverse effect that platform stance shake brings image quality.The 3rd, be applicable to the SAR imaging in large motor-driven situation.Compare with conventional SAR imaging situation of image defocus when the large motion of automobile appears in platform, the present invention is from imaging model, when there is the large motion of automobile in platform, also can make full use of the carrier aircraft position and attitude vector data information in sensing data, target emanation value in Accurate Reconstruction scene, obtains the focusing SAR image of gain balance.

Accompanying drawing explanation

Fig. 1 is non-rectilinear flight path SAR formation method process flow diagram of the present invention;

Fig. 2 is conventional SAR formation method process flow diagram;

Fig. 3 is the platform cross track figure of emulation;

Fig. 4 is the Illumination Distribution figure of emulated data;

Fig. 5 is by the formation method of SAR shown in Fig. 2 process flow diagram emulated data test result figure;

Fig. 6 is by the formation method of SAR shown in Fig. 1 process flow diagram emulated data test result figure;

Fig. 7 is position of platform and the attitude figure of actual measurement;

Fig. 8 is the Illumination Distribution figure of measured data;

Fig. 9 is by SAR formation method process flow diagram measured data test result figure described in Fig. 1;

Figure 10 is by SAR formation method process flow diagram measured data test result figure described in Fig. 2.

Embodiment

Below in conjunction with the drawings and specific embodiments, the invention will be further described.

As shown in Figure 1, non-rectilinear flight path SAR formation method process flow diagram of the present invention.Whole flow process is mainly divided apart from pulse pressure, two steps of azimuth match (wherein specifically comprising motion compensation and illuminance compensation) processing, alignment of data process in figure is: to carrying out polynomial interpolation based on filtered carrier aircraft position and attitude data, obtain the carrier aircraft position and attitude vector data process identical with SAR echo data turnover rate.Fig. 2 is conventional SAR formation method process flow diagram, and main minute apart from pulse pressure, azimuth match (only comprising motion compensation), radiant correction.

The present invention can be in non-rectilinear flight path SAR imaging correction of movement error and the impact of uneven illumination weighing apparatus on SAR imaging simultaneously, in motion compensation and illuminance compensation, not only ideal line flight path is departed from and the Range Profile that causes is walked about and phase error in Caliberation Flight position of platform, also proofread and correct simultaneously due to attitude jitter cause radar illumination energy along orientation to the target emanation that causes of uneven distribution inhomogeneous, finally obtain the focusing SAR image of ground scene and target gain equilibrium.

The present invention has taken into full account the movement characteristic of SAR platform non-rectilinear flight path, take full advantage of sensing data, to illumination is unbalanced, compensate, effectively eliminated the adverse effect of platform stance shake to target relative amplitude relation in SAR image, improved SAR image quality.

Below, adopt respectively simulative data and flight-test data, the method for the invention to be tested respectively, image quality is all significantly improved.

Fig. 3 is the platform cross track figure of emulation, and platform is the height flight with liftoff 100m along broken line, and radar is operated under positive side-looking band pattern, i.e. antenna beam center hold sensing is vertical with track keeping, and the pitching of antenna beam center is pointed to and remained oblique lower 45 °.

Fig. 4 is the Illumination Distribution figure of emulated data.Suppose that emitting antenna azimuth beamwidth and pitching beam angle are 60 °, beam area internal antenna be each to homogeneous radiation source, accumulation angle, orientation is 60 °.Affected by constant accumulation angle, the illumination apart from the illumination to far-end apparently higher than distance to near-end; Be subject to the impact of attitude jitter, weak, the strong trend in both sides in the middle of illumination azimuth distribution presents.

Fig. 5 is by the formation method of SAR shown in Fig. 2 process flow diagram emulated data test result figure.Fig. 5-a is imaging panorama sketch, Fig. 5-b be target (orientation 100m, apart from 80m, height 0m) enlarged drawing, Fig. 5-c be target (orientation 150m, apart from 80m, height 0m) enlarged drawing.In Fig. 5-c, the amplitude of target is obviously weaker than the amplitude of target in Fig. 5-b, and this shows that conventional SAR formation method is subject to the unbalanced impact of Illumination Distribution.

Fig. 6 is by the formation method of SAR shown in Fig. 1 process flow diagram emulated data test result figure.Fig. 6-a is imaging panorama sketch, Fig. 6-b be target (orientation 100m, apart from 80m, height 0m) enlarged drawing, Fig. 6-c be target (orientation 150m, apart from 80m, height 0m) enlarged drawing.After illuminance compensation, in Fig. 6-b, in target and Fig. 6-c, the amplitude of target is basic identical.

Fig. 7 is position of platform and the attitude figure of actual measurement.Fig. 7-a is position data, and Fig. 7-b is attitude data.Except significantly shake position, attitude is also shaken acutely as seen from Figure 7.Wherein, crab angle jitter range has reached very greatly 35 ° of left and right, and the angle of pitch and roll angle also have the shake of 15 ° of left and right.

Fig. 8 is the Illumination Distribution figure of measured data.As seen from Figure 8, Illumination Distribution in orientation to upwards all having a unbalanced phenomena with distance.

Fig. 9 is by SAR formation method process flow diagram measured data test result figure described in Fig. 2.Fig. 9-a is imaging panorama sketch, and Fig. 9-b is A target in Fig. 9-a (rib length is the trihedral angle of 0.3m) enlarged drawing, and Fig. 9-c is B target in Fig. 9-a (rib length is the trihedral angle of 0.3m) enlarged drawing.Fig. 9-a distance is strong to near-end amplitude, a little less than far-end amplitude, orientation is to a little less than near near strong 320m left and right, 480m.In Fig. 9-b target than target amplitude in Fig. 9-c up to 3.64dB.Although this illustrates that conventional SAR imaging carried out radiant correction, its for be that orientation is balanced to illumination, Range Profile illumination is according to 1/r ²the special circumstances of relation decay, for general situation, although can raise far-end amplitude, cannot eliminate the unbalanced impact of Illumination Distribution.

Figure 10 is by SAR formation method process flow diagram measured data test result figure described in Fig. 1.Figure 10-a is imaging panorama sketch, and Figure 10-b is A target in Figure 10-a (rib length is the trihedral angle of 0.3m) enlarged drawing, and Figure 10-c is B target in Figure 10-a (rib length is the trihedral angle of 0.3m) enlarged drawing.After illuminance compensation, Figure 10-a brightness of image becomes evenly, and image detail is more outstanding.In Figure 10-b, in target and Figure 10-c, target amplitude difference is reduced to 0.11dB.

The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (1)

1. a radiation correction method that is applicable to non-rectilinear flight path SAR imaging, is characterized in that: comprise the following steps:

(S1) obtain radar return data, echo data is carried out processing apart from pulse pressure, obtain Range Profile;

(S2) sensor measurement obtains carrier of radar position and attitude vector data sample; Data sample is carried out after initialization, and by filtering, interpolation processing, obtain the estimated value of carrier of radar position and attitude vector;

(S3), according to the Range Profile of described step (S1) accumulation and the carrier of radar position and attitude vector estimated value in step (S2), rebuild SAR image f _bP(p), wherein, p represents the location point in imaging region;

(S4) definition illumination I (p), and it is as follows to calculate illumination:

Wherein, c is the light velocity, and u is the slow time, g (p _u, p) be antenna radiation pattern function, p _ufor u moment carrier aircraft position and attitude vector, R (p _u, p) be u carrier aircraft position p constantly _uto the distance of location point p in imaging region, w (p _u, p) be the window function of control punch electrical path length;

(S5) image after rebuilding in described step (S3) is carried out to illuminance compensation, the SAR image σ (p) after the focusing of acquisition gain balance, that is: