CN102129665A - Image surface driving point-based imaging method of optical remote sensor - Google Patents

Abstract

The invention relates to an image surface driving point-based imaging method of an optical remote sensor. The method comprises the steps of: establishing a platform optics payload geometric distortion model by utilizing a method of coordinate transformation according to a physical model transformed from a ground coordinate system to an image surface coordinate system; , fitting a small quantity of image surface driving points into an image by utilizing polynomial interpolation; and by utilizing the model, completing the imaging of the model in different postures and analyzing the image geometric distortion caused by the postures. Therefore, the effect transformation between an object surface and an image surface is realized; the calculation amount is reduced and a higher precision is maintained; meanwhile, the problem of overlarge calculation complexity in a traditional precise geometrical model is solved and the shortcoming of lacking physical meaning in a rational function model is made up.

Description

A kind of formation method of the optical sensor based on the image planes drive point

Technical field:

The invention belongs to the optical remote sensing technology field, particularly relate to a kind of formation method of the optical sensor based on the image planes drive point.

Background technology

The high-resolution optical remote sensing device, very harsh to the requirement of platform.Existing mode transmission spacer remote sensing device adopts the line array CCD push-scanning image usually.The push-broom type imaging is very sensitive to the piecture geometry fault that the attitude instability causes, the attitude variation that the pattern distortion that the attitude variation causes is divided on rolling, pitching, three directions of driftage has caused line array CCD projection on the ground to produce distortion, and is more responsive to the influence of high-resolution optical remote sensing device especially.Analyzing satellite platform kinematic error and imaging mapping relations need set up from the geometric model between the earth axes one image coordinates system.Traditional method all is the strict precise geometrical model that each pixel is changed one by one from the object plane to image planes that adopts.This model physics needs sizable calculated amount.

Summary of the invention:

The objective of the invention is to overcome the above-mentioned deficiency of prior art, a kind of formation method of the optical sensor based on the image planes drive point is provided, this method is by adopting the object point of some small samples, in conjunction with s internal and external orientation information, the process terrestrial coordinate is to a spot of image planes drive point that changes between the imaging surface coordinate system.Utilize these drive points then, carry out fitting of a polynomial, form piece image fast, significantly reduced calculated amount, and kept degree of precision as interpolation.

Above-mentioned purpose of the present invention is achieved by following technical solution:

A kind of formation method of the optical sensor based on the image planes drive point comprises the steps:

(1) set up optical sensor over the ground in the imaging process object plane to the Topological Mapping relation of image planes:

$\left[\begin{array}{c}{x}_p\\ y_p\\ z_p\\ 1\end{array}\right]=\left[M_{\mathrm{ground}}^{\mathrm{orbit}}\right]\left[M_{\mathrm{ordit}}^{\mathrm{platform}}\right]\left[M_{\mathrm{platrorm}}^{\mathrm{camera}}\right]\left[\begin{array}{c}{x}_g\\ y_g\\ z_g\\ 1\end{array}\right]$

$=\left[\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{12}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right]\left[\begin{array}{c}{x}_g\\ y_g\\ z_g\\ 1\end{array}\right]$

Wherein: x _p, y _p, z _pBe the coordinate of following of image coordinates system, x _g, y _g, z _gBe the coordinate of following of earth axes, z _p=z _g=0,

For being tied to the transition matrix of orbital coordinate system from ground coordinate,

For orbit coordinate is tied to the transition matrix of satellite platform coordinate system,

Be tied to the transition matrix of image coordinates system for the satellite platform coordinate;

(2) on image planes, choose n small sample point coordinate (x _Pn, y _Pn) as the image planes drive point, described n small sample point coordinate (x _Pn, y _Pn) with object plane on n corresponding point coordinate (x _Gn, y _Gn) transformational relation as follows:

$\left(\begin{array}{c}{x}_{p1}\\ y_{p1}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\·\left(\begin{array}{c}{x}_{g1}\\ y_{g1}\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p2}\\ {\mathrm{<figure-callout\; id="2"\; label="y\; p"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_p\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g2}\\ {\mathrm{<figure-callout\; id="2"\; label="y\; g"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_g\\ 0\\ 1\end{array}\right),$

$\left(\begin{array}{c}{x}_{p3}\\ {\mathrm{<figure-callout\; id="3"\; label="y\; p"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_p\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g3}\\ {\mathrm{<figure-callout\; id="3"\; label="y\; g"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_g\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p4}\\ y_{p4}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g4}\\ y_{g4}\\ 0\\ 1\end{array}\right)$

.

.

.

$\left(\begin{array}{c}{x}_{\mathrm{pn}}\\ y_{\mathrm{<figure-callout\; id="0"\; label="pn"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">pn</figure-callout>}}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{23}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{\mathrm{<figure-callout\; id="0"\; label="gn\; y\; gn"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">gn</figure-callout>}}& \\ y_{\mathrm{gn}}{}_{}\\ 0\\ 1\end{array}\right),$

N is a positive integer, and n 〉=4;

(3) n the image planes drive point that obtains in the step (2) carried out interpolation, obtain the pixel between the drive point, form complete image.

In the formation method of above-mentioned optical sensor based on the image planes drive point, on image planes, choose n small sample point coordinate (x in the step (2) _Pn, y _Pn) method be: image planes evenly are divided into the n equal portions, and the central point of choosing each equal portions obtains n small sample point altogether as a small sample point.

In the formation method of above-mentioned optical sensor based on the image planes drive point, the method for in the step (3) n image planes drive point being carried out interpolation is a polynomial interpolation, and concrete grammar is as follows:

$I\left(x\right)=a_M{x}_{\mathrm{pn}}^n+a_{M-1}{x}_{\mathrm{pn}}^{n-1}+\&CenterDot;\&CenterDot;\&CenterDot;+a_2{x}_{\mathrm{<figure-callout\; id="2"\; label="pn"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">pn</figure-callout>}}^2+a_1{x}_{\mathrm{pn}}+a_0$

$I\left(y\right)=b_M{y}_{\mathrm{pn}}^n+b_{N-1}{y}_{\mathrm{pn}}^{n-1}+\&CenterDot;\&CenterDot;\&CenterDot;+b_2{y}_{\mathrm{<figure-callout\; id="2"\; label="pn"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">pn</figure-callout>}}^2+b_1{y}_{\mathrm{pn}}+b_0$

Wherein, M, N are the size that obtains image after the interpolation, (x _Pn, y _Pn) be image planes drive point coordinate, a ₀～a _MFor n time with x _PnBe the equation multinomial coefficient of independent variable, b ₀～b _NFor n time with y _PnEquation multinomial coefficient for independent variable.

The present invention compared with prior art has following advantage:

(1) the present invention is according to the geometric model from earth coordinates-image coordinates system transforms, set up platform optics useful load geometric distortion model, and utilize polynomial interpolation that a spot of image planes drive point is interpolated to piece image, this method has been done effective conversion between object plane and image planes, significantly reduce calculated amount, and kept degree of precision.

(2) the inventive method is finished the imaging under the different attitudes of satellite platform, reflected the geometric distortion of the image that causes because of attitude, and embodied the remote sensing images geometric distortion with the mapping relations between the inside and outside element of orientation of satellite platform, finally realized the conversion from the object plane to image planes;

(3) optical remote sensor imaging geometric model of the present invention is considered all multifactorly, comprises elements of interior orientation: detector size, focal length etc.; Elements of exterior orientation: platform track height, attitude, track six roots of sensation number etc., thought in conjunction with polynomial interpolation, to becoming piece image to a small amount of point interpolation that the imaging surface coordinate transformation obtains through terrestrial coordinate, both solve traditional precise geometrical model because of the excessive problem of computation complexity, remedied the shortcoming that rational function model lacks physical meaning again.

Description of drawings

Fig. 1 is how much topological relation figure of optical remote sensor imaging;

Fig. 2 is each coordinate system corresponding relation synoptic diagram among the present invention;

Fig. 3 is an optical remote sensor imaging method flow diagram of the present invention;

The ideal image of Fig. 4 for adopting the inventive method to obtain;

Fig. 5 30 ° of ideal image of driftage for adopting the inventive method to obtain.

Embodiment

The present invention is described in further detail below in conjunction with the drawings and specific embodiments:

Be illustrated in figure 1 as how much topological relation figure of optical remote sensor imaging, the theoretical foundation of linear optics remote sensor model is as follows: can use the pin-hole model approximate representation for the image space of any one some P on image planes on the space, be that the projected position of some p ' on image planes on any object plane is p, p is video camera photocentre O and the P line OP of ordering and the intersection point of image planes.

Among Fig. 1, (x, y z) are TDICCD optical sensor coordinate, and O is the optical sensor coordinate origin, and blue arrow is the satellite flight direction, Ψ _yBe the side-sway angle, Ψ _xIt is the angle of pitch.Wherein Δ p ' is the ground error deviation distance that causes because of attitude error.

Linear optics remote sensor imaging model is a form of the imaging geometry of picture point and object point being write as perspective projection matrix:

$P=M(p\&prime;+\mathrm{\&Delta;p}\&prime;)\&DoubleRightArrow;Z_c=\left[\begin{array}{c}u\\ v\\ 1\end{array}\right]=M\left[\begin{array}{c}{x}_g+X_{\mathrm{\&Delta;p}\&prime;}\\ y_g+Y_{\mathrm{\&Delta;p}\&prime;}\\ z_g+Z_{\mathrm{\&Delta;p}\&prime;}\\ 1\end{array}\right]---\left(1\right)$

(x wherein _p, y _p) be the coordinate of the point under the image pixel coordinate system, (x _g, y _g, z _g) be the coordinates of spatial points under the terrestrial coordinate system, the perspective projection matrix that M is, Z _cBe the coordinate of p under the camera coordinate system, wherein z _p, z _gBe 0.

The optical remote sensor imaging geometric model will be considered all multifactor, comprising detector size, focal length, platform track height, attitude, track six roots of sensation number etc.; Generally adopt GPS to determine the current location of satellite at the rail platform, star is quick determines one group of seasonal effect in time series attitude information with gyroscope.

Object plane in the remote sensing satellite imaging process/image planes transforming relationship is as follows:

$\left[\begin{array}{c}{x}_p\\ y_p\\ z_p\\ 1\end{array}\right]=\left[M_{\mathrm{ground}}^{\mathrm{orbit}}\right]\left[M_{\mathrm{orbit}}^{\mathrm{platform}}\right]\left[M_{\mathrm{platform}}^{\mathrm{camera}}\right]\left[\begin{array}{c}{x}_g\\ y_g\\ z_g\\ 1\end{array}\right]$

$=\left[\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right]\left[\begin{array}{c}{x}_g\\ y_g\\ z_g\\ 1\end{array}\right]---\left(2\right)$

Wherein: x _p, y _pBe the coordinate of following of image coordinates system, x _g, y _gBe the coordinate of following of earth axes, wherein z _p=z _g=0,

For being tied to the transition matrix of orbital coordinate system from ground coordinate,

For orbit coordinate is tied to the transition matrix of satellite platform coordinate system,

Be tied to the transition matrix of image coordinates system for the satellite platform coordinate.Be illustrated in figure 2 as each coordinate system corresponding relation synoptic diagram among the present invention, showed among Fig. 2 in the conversion process from the thing to the picture, complex mapping relation between each coordinate system, modeling method of the present invention takes into full account the physics overall process, calculating is accurate, error is little, is than the Geometric Modeling Method near truth.Wherein

To embody formula as follows:

$\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\left(\begin{array}{cccc}\frac{-\mathrm{fo}}{r_s}& 0& 0& 0\\ 0& \frac{-\mathrm{fo}}{r_s}& 0& 0\\ 0& 0& \frac{-\mathrm{fo}}{r_s}& -\mathrm{<figure-callout\; id="0"\; label="fo"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">fo</figure-callout>}\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& \cos {\mathrm{\&psi;}}_y& \sin {\mathrm{\&psi;}}_{\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">y</figure-callout>}}& 0\\ 0& -\sin {\mathrm{\&psi;}}_y& \cos {\mathrm{\&psi;}}_{\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">y</figure-callout>}}& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}\cos {\mathrm{\&Psi;}}_x& 0& -\sin {\mathrm{\&psi;}}_x& 0\\ 0& 1& 0& 0\\ {\sin \mathrm{\&psi;}}_x& 0& \cos {\mathrm{\&psi;}}_x& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}\cos (\mathrm{\&psi;}+\stackrel{\&CenterDot;}{\mathrm{\&psi;t}})& \sin (\mathrm{\&psi;}+\stackrel{\&CenterDot;}{\mathrm{\&psi;}}t)& 0& 0\\ -\sin (\mathrm{\&psi;}+\stackrel{\&CenterDot;}{\mathrm{\&psi;t}})& \cos (\mathrm{\&psi;}+\stackrel{\&CenterDot;}{\mathrm{\&psi;}}t)& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)$

$\left(\begin{array}{cccc}\cos (\mathrm{\&Phi;}+\stackrel{\&CenterDot;}{\mathrm{\&Phi;t}})& 0& -\sin \left(\mathrm{\&Phi;}\stackrel{\&CenterDot;}{+\mathrm{\&Phi;}}t\right)& 0\\ 0& 1& 0& 0\\ \sin (\mathrm{\&Phi;}+\stackrel{\&CenterDot;}{\mathrm{\&Phi;}}t)& 0& \cos \left(\mathrm{\&Phi;}\stackrel{\&CenterDot;}{+\mathrm{\&Phi;t}}\right)& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& \cos (\mathrm{\&theta;}+\stackrel{\&CenterDot;}{\mathrm{\&theta;t}})& \sin (\mathrm{\&theta;}+\stackrel{\&CenterDot;}{\mathrm{\&theta;t}})& 0\\ 0& -\sin (\mathrm{\&theta;}+\stackrel{\&CenterDot;}{\mathrm{\&theta;t}})& \cos (\mathrm{\&theta;}+\stackrel{\&CenterDot;}{\mathrm{\&theta;t}})& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}\cos ({\mathrm{\&gamma;}}_0+\mathrm{\&omega;t})& 0& \sin ({\mathrm{\&gamma;}}_0+\mathrm{\&omega;t})& 0\\ 0& 1& 0& 0\\ -\sin ({\mathrm{\&gamma;}}_0+\mathrm{\&omega;t})& 0& \cos ({\mathrm{\&gamma;}}_0+\mathrm{\&omega;t})& 0\\ 0& 0& 0& 1\end{array}\right)$

$\left(\begin{array}{cccc}\cos (\mathrm{\&pi;}-i_0)& \sin (\mathrm{\&pi;}-i_0)& 0& 0\\ -\sin (\mathrm{\&pi;}-i_0)& \cos (\mathrm{\&pi;}-i_0)& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}\cos {\mathrm{\&omega;}}_e\mathrm{<figure-callout\; id="0"\; label="t"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">t</figure-callout>}& 0& \sin {\mathrm{\&omega;}}_e\mathrm{<figure-callout\; id="0"\; label="t"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">t</figure-callout>}& 0\\ 0& 1& 0& 0\\ -\sin {\mathrm{\&omega;}}_e\mathrm{<figure-callout\; id="0"\; label="t"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">t</figure-callout>}& 0& \cos {\mathrm{\&omega;}}_e\mathrm{<figure-callout\; id="0"\; label="t"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">t</figure-callout>}& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}\cos (\mathrm{\&pi;}-i_0)& -\sin (\mathrm{\&pi;}-i_0)-& 0& 0\\ \sin (\mathrm{\&pi;}-i_0)& \cos (\mathrm{\&pi;}-i_0)& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right)$ (3)

$\left(\begin{array}{cccc}{\cos \mathrm{\&gamma;}}_0& 0& -\sin {\mathrm{\&gamma;}}_0& 0\\ 0& 1& 0& 0\\ \sin {\mathrm{\&gamma;}}_0& 0& \cos {\mathrm{\&gamma;}}_0& 0\\ 0& 0& 0& 1\end{array}\right)\left(\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& R_e+\mathrm{<figure-callout\; id="0"\; label="h"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">h</figure-callout>}\\ 0& 0& 0& 1\end{array}\right)$

Wherein, R _eBe earth radius, h is the height of terrain object, and H is the orbit altitude of satellite, r _sBe oblique distance (satellite is to the distance of being shone between the target), γ ₀Be the earth's core subtended angle of satellite position to ascending node, i ₀Be orbit inclination, ω _eBe rotational-angular velocity of the earth 7.2722 * 10 ^-5Radian per second, ω are the angular velocity that satellite rotates around the earth's core in orbit plane, and t is the time, and θ is a roll angle,

Be rate of roll, Φ is the angle of pitch,

Be rate of pitch, Ψ is a crab angle,

Be yaw rate, fo is the focal length of optical sensor, Ψ _xBe the preceding visual angle of optical sensor, Ψ _yBe optical sensor side-looking angle, fo is the focal length of camera.

Be illustrated in figure 3 as optical remote sensor imaging method flow diagram of the present invention.

More any coordinate in terrestrial coordinate system and the coordinate in image coordinates system can be set up relation one to one by coordinate transform.Get small sample in image planes after utilizing the coordinate corresponding relation, promptly the image planes drive point carries out the fault image of interpolation to obtain wanting then between these points.Between coordinate system, in the conversion process, only analyze the situation of 2 dimensions, do not consider coordinate, i.e. z at optical axis direction _p=0.By formula (2), the mathematic(al) representation of its coordinate transform can be write as:

$\left(\begin{array}{c}{x}_p\\ y_{\mathrm{<figure-callout\; id="0"\; label="p"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">p</figure-callout>}}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_g\\ y_{\mathrm{<figure-callout\; id="0"\; label="g"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">g</figure-callout>}}\\ 0\\ 1\end{array}\right)---\left(4\right)$

On image planes, choose n small sample point coordinate (x _Pn, y _Pn) as the image planes drive point, described n small sample point coordinate (x _Pn, y _Pn) with object plane on n corresponding point coordinate (x _Gn, y _Gn) transformational relation as follows:

$\left(\begin{array}{c}{x}_{p1}\\ y_{p1}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g1}\\ y_{g1}\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p2}\\ {\mathrm{<figure-callout\; id="2"\; label="y\; p"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_p\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g2}\\ {\mathrm{<figure-callout\; id="2"\; label="y\; g"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_g\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p3}\\ {\mathrm{<figure-callout\; id="3"\; label="y\; p"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_p\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g3}\\ {\mathrm{<figure-callout\; id="3"\; label="y\; g"\; filenames="HSA00000359416400032.png"\; state="\{\{state\}\}">y</figure-callout>}}_g\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p4}\\ y_{p4}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g4}\\ y_{g4}\\ 0\\ 1\end{array}\right)$

.

.

.

$\left(\begin{array}{c}{x}_{\mathrm{pn}}\\ y_{\mathrm{pn}}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{23}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{\mathrm{<figure-callout\; id="0"\; label="gn\; y\; gn"\; filenames="HSA00000359416400011.png,HSA00000359416400021.png"\; state="\{\{state\}\}">gn</figure-callout>}}& \\ y_{\mathrm{gn}}{}_{}\\ 0\\ 1\end{array}\right)---\left(5\right)$

N is a positive integer, and n 〉=4;

Utilize a spot of drive point significantly to reduce from terrestrial coordinates to the image planes data operation quantity.The drive point of some can reduce calculated amount under the prerequisite that guarantees geometric distortion image fitting precision.

N image planes drive point carried out interpolation, obtain the pixel between the drive point, form complete image, with parameter designing and the parameter prediction that instructs satellite platform.

$\left(x\right)=a_M{x}_{\mathrm{pn}}^n+a_{M-1}{x}_{\mathrm{pn}}^{n-1}{a}_2{x}_{\mathrm{pn}}^2{a}_1{x}_{\mathrm{pn}}+a_0$

$I\left(y\right)=b_N{y}_{\mathrm{pn}}^n+b_{N-1}{y}_{\mathrm{pn}}^{n-1}{b}_2{y}_{\mathrm{pn}}^2{b}_1{y}_{\mathrm{pn}}+b_0$ (6)

Wherein, M, N are the size that obtains image after the interpolation, (x _Pn, y _Pn) be image planes drive point coordinate, a ₀～a _MFor n time with x _PnBe the equation multinomial coefficient of independent variable, b ₀～b _NFor n time with y _PnEquation multinomial coefficient for independent variable.

If when N=1, equation becomes:

$\left\{\begin{array}{c}I\left(x\right)=a_{1}x+a_0\\ I\left(y\right)=b_1+b_0\end{array}\right.---\left(7\right)$

Can utilize (x _P1, y _P1), (x _P2, y _P2), (x _P3, y _P3), (x _P4, y _P4) four drive points can calculate multinomial coefficient, thereby utilize the method for extrapolated value in the polynomial expression, can obtain all pixels of image planes successively.

Be illustrated in figure 4 as the ideal image that adopts the inventive method to obtain; Figure 5 shows that the 30 ° of ideal image of driftage that adopt the inventive method to obtain, adopt formation method of the present invention can reflect the piecture geometry fault that causes because of the variation of platform attitude as seen from the figure.

The above; only be the embodiment of the best of the present invention, but protection scope of the present invention is not limited thereto, anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.

The content that is not described in detail in the instructions of the present invention belongs to this area professional and technical personnel's known technology.

Claims (3)

1. the formation method based on the optical sensor of image planes drive point is characterized in that: comprise the steps:

(1) set up the Topological Mapping relation of object plane in the optical remote sensor imaging process to image planes:

$\left[\begin{array}{c}{x}_p\\ y_p\\ z_p\\ 1\end{array}\right]=\left[M_{\mathrm{ground}}^{\mathrm{orbit}}\right]\left[M_{\mathrm{orbit}}^{\mathrm{platform}}\right]\left[M_{\mathrm{platform}}^{\mathrm{camera}}\right]\begin{array}{}\left[\begin{array}{c}{x}_g\\ y_g\\ z_g\\ 1\end{array}\right]\end{array}$

$=\left[\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right]\left[\begin{array}{c}{x}_g\\ y_g\\ z_g\\ 1\end{array}\right]$

Wherein: x _p, y _p, z _pBe the coordinate of following of image coordinates system, x _g, y _g, z _gBe the coordinate of following of earth axes, z _p=z _g=0,

For being tied to the transition matrix of orbital coordinate system from ground coordinate, For orbit coordinate is tied to the transition matrix of satellite platform coordinate system,

Be tied to the transition matrix of image coordinates system for the satellite platform coordinate;

(2) on image planes, choose n small sample point coordinate (x _Pn, y _Pn) as the image planes drive point, described n small sample point coordinate (x _Pn, y _Pn) with object plane on n corresponding point coordinate (x _Gn, y _Gn) transformational relation as follows:

$\left(\begin{array}{c}{x}_{p1}\\ y_{p1}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g1}\\ y_{g1}\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p2}\\ y_{p2}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g2}\\ y_{g2}\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p3}\\ y_{p3}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g3}\\ y_{g3}\\ 0\\ 1\end{array}\right)$

$\left(\begin{array}{c}{x}_{p4}\\ y_{p4}\\ 0\\ 1\end{array}\right)=\left(\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{24}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right)\&CenterDot;\left(\begin{array}{c}{x}_{g4}\\ y_{g4}\\ 0\\ 1\end{array}\right)$

.

.

.

$\left[\begin{array}{c}{x}_{\mathrm{pn}}\\ y_{\mathrm{pn}}\\ 0\\ 1\end{array}\right]=\left[\begin{array}{cccc}{m}_{11}& m_{12}& m_{13}& m_{14}\\ m_{21}& m_{22}& m_{23}& m_{23}\\ m_{31}& m_{32}& m_{33}& m_{34}\\ m_{41}& m_{42}& m_{43}& m_{44}\end{array}\right]\&CenterDot;\left[\begin{array}{c}{x}_{\mathrm{gn}}\\ y_{\mathrm{gn}}\\ 0\\ 1\end{array}\right],$

N is a positive integer, and n 〉=4;

(3) n the image planes drive point that obtains in the step (2) carried out interpolation, obtain the pixel between the drive point, form complete image.

2. the formation method of a kind of optical sensor based on the image planes drive point according to claim 1 is characterized in that: choose n small sample point coordinate (x in the described step (2) on image planes _Pn, y _Pn) method be: image planes evenly are divided into the n equal portions, and the central point of choosing each equal portions obtains n small sample point altogether as a small sample point.

3. the formation method of a kind of optical sensor based on the image planes drive point according to claim 1 is characterized in that: the method for in the described step (3) n image planes drive point being carried out interpolation is a polynomial interpolation, and concrete grammar is as follows:

$I\left(x\right)=a_M{x}_{\mathrm{pn}}^n+a_{M-1}{x}_{\mathrm{pm}}^{n-1}+\&CenterDot;\&CenterDot;\&CenterDot;+a_2{x}_{\mathrm{pn}}^2+a_1{x}_{\mathrm{pn}}+a_0$

$I\left(y\right)=b_N{y}_{\mathrm{pn}}^n+b_{N-1}{y}_{\mathrm{pn}}^{n-1}+\&CenterDot;\&CenterDot;\&CenterDot;+b_2{y}_{\mathrm{pn}}^2+b_1{y}_{\mathrm{pn}}+b_0$

Wherein, M, N are the size that obtains image after the interpolation, x _Pn, y _PnBe image planes drive point coordinate, a ₀～a _nFor n time be the equation multinomial coefficient of independent variable with x, b ₀～b _nFor n time be the equation multinomial coefficient of independent variable with y.

Images