CN103473756B - Autoregistration merges the method for Aerial Images and close shot image - Google Patents

Abstract

The present invention proposes a kind of method that autoregistration merges Aerial Images and close shot image, comprise: to the Aerial Images collected with arrange that identification plate close shot image is smoothing, sharpening, abatement Aerial Images and the noise that identification plate close shot image is set, strengthen Aerial Images and the edge that identification plate close shot image is set, then carry out the Aerial Images after coordinate conversion process acquisition image enhaucament and identification plate close shot image vertical view is set; Indicating board size determined mathematics conversion proportion relation according to Aerial Images with arranging in identification plate close shot image vertical view, being transformed in the coordinate system of Aerial Images by arranging identification plate close shot image, complete uniform coordinate conversion; In Aerial Images, the described correspondence position arranging identification plate close shot image in search location, determines the registration transformation relation between two width images according to cross-correlation similarity measure; Image pixel rgb value weighted method is adopted to realize Aerial Images and the fusion of identification plate close shot image overlapping region is set.

Description

Autoregistration merges the method for Aerial Images and close shot image

Technical field

The present invention relates to scene of a traffic accident field of information processing, be specifically related to a kind of method that autoregistration merges Aerial Images and close shot image.

Background technology

Aerial Images has vital role in scene of a traffic accident exploration, and it can reduce each key element in the scene of the accident efficiently, but Aerial Images is in order to obtain global information, and camera site is often higher, therefore not enough to the more tiny mark information resolution characteristic in ground.Close shot image provides each key element detailed information with higher resolution in scene of a traffic accident exploration, but cannot reflect its situation in whole scene.In order to better carry out crash analysis, the comprehensive Aerial Images of registration integration technology and close shot image information is utilized to be following development trends, multiple site survey of traffic accident image is combined, make full use of the feature that difference reconnoitres image, piece image is expressed the multi-aspect information from the scene of the accident simultaneously, many-sided situations such as the key element position in the scene of the accident, key element details are reflected by the image after registration merges, thus scene of a traffic accident information is provided more intuitively.

Can adopt the feature object in manual type search location Aerial Images and close shot image at present, and then the artificial registration utilizing feature object to realize Aerial Images and close shot image merges.But when close shot image is more, manual type just seems to waste time and energy, and accuracy is not high.Therefore, a kind of merge the mark of Aerial Images and close shot image for autoregistration and implementation method very necessary.

Summary of the invention

The present invention is intended at least solve the technical matters existed in prior art, especially innovatively proposes a kind of method that autoregistration merges Aerial Images and close shot image.

In order to realize above-mentioned purpose of the present invention, the invention provides a kind of method that autoregistration merges Aerial Images and close shot image, its key is, comprises the steps:

Step 1, to the Aerial Images collected with arrange that identification plate close shot image is smoothing, sharpening, abatement Aerial Images and the noise that identification plate close shot image is set, strengthen Aerial Images and the edge that identification plate close shot image is set, then carry out the Aerial Images after coordinate conversion process acquisition image enhaucament and identification plate close shot image vertical view is set;

Step 2, indicating board size determined mathematics conversion proportion relation according to Aerial Images with arranging in identification plate close shot image vertical view, being transformed in the coordinate system of Aerial Images by arranging identification plate close shot image, completes uniform coordinate conversion;

Step 3, in Aerial Images, the described correspondence position arranging identification plate close shot image in search location, determines the registration transformation relation between two width images according to cross-correlation similarity measure;

Step 4, adopts image pixel rgb value weighted method to realize Aerial Images and arrange the fusion of identification plate close shot image overlapping region.

The beneficial effect of technique scheme is: through merging the image after registration, it has abundanter comprehensive field data, can react scene of the accident entirety and crucial local scene situation better, be convenient to analysis and the utilization of scene of a traffic accident information.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, described step 3 comprises:

Step 3-1, identification plate color according to arranging identification plate close shot image has rotational invariance and scale invariability, described identification plate structural color is carried out search location as the described index feature arranging identification plate close shot image to it in Aerial Images, obtains and the band of position of identification plate close shot image in Aerial Images is set;

Step 3-2, by adopting Harris Corner Detection Algorithm using the center of circle of described identification plate sector structure place circle as the registration control points arranging identification plate close shot image and Aerial Images;

Step 3-3, according to Aerial Images with arrange identification plate close shot image and be RGB color image, selects to adopt image rgb value cross-correlation registration, obtains optimal registration position.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, described registration control points algorithm is:

E(u,v)＝∑ _x,yw(x,y)[I(x+u,y+v)-I(x,y)] ²

Wherein, w (x, y) is window function, [I (x+u, y+v)-I (x, y)] ²for the Grad of image, w (x, y) is rectangular window or Gaussian window, and for each displacement (u, v), in above formula, bilinear approximation is expressed as,

$E(u,v)\≅[u,v]M\left[\begin{array}{c}u\\ v\end{array}\right],$

M is the matrix of 2*2, and concrete formula is,

$\left[\begin{array}{cc}A& C\\ C& B\end{array}\right]=e^{\frac{x^2+y^2}{2{\mathrm{\σ}}^2}}\⊗\left[\begin{array}{cc}{\mathrm{Ix}}^2& \mathrm{IxIy}\\ \mathrm{IxIy}& {\mathrm{Iy}}^2\end{array}\right],$

Ix, Iy are respectively the Grad in image x, y direction; E is approximate as local cross correlation function, and M describes shape in this, if λ ₁, λ ₂two eigenwerts, then λ of matrix M ₁, λ ₂can represent the curvature of local autocorrelation function, due to isotropy, so M keeps rotational invariance, wherein A, B, C represent the element value of matrix M.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, described cross-correlation registration Algorithm is:

Aerial Images is selected a contingent window W (l, m), the upper searching of identification plate close shot image S (i, j) the most similar window corresponding with it is being set , it is rotational transform window, with cross-correlation similarity measure is, wherein l, m are positive integer,

$R(i,j)=\frac{\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}W(l,m)S_M^{i,j}(l,m)}{\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left[S^{i,j}(l,m)\right]}^2}(i^{*},j^{*}),1\≤i,j\≤L-M+1$

The wherein video in window of W (i, j) to be size be M × M, be the contingent window in Aerial Images, video in window W (i, j) to arrange in identification plate close shot image S with scan mode automatic search in search, finds correlation function value R in this process _nposition (the i of maximum of points in (i, j) ^*, j ^*), and think to be exactly optimal registration position herein.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, described step 4 comprises:

Step 4-1, by the rgb value R of Aerial Images ₁(i, j), G ₁(i, j), B ₁(i, j) and the rgb value R of identification plate close shot image is set ₂(i, j), G ₂(i, j), B ₂(i, j) is multiplied by a weight coefficient separately, merges and forms new image F (i, j),

F(i,j)＝aI ₁(i,j)+(1-a)I ₂(i,j),

Wherein, a is weight factor, and 0≤a≤1, I ₁for Aerial Images, I ₂for arranging identification plate close shot image.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, described identification plate is right angle sector structure.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, the color of described identification plate is different.

In sum, owing to have employed technique scheme, the invention has the beneficial effects as follows:

Through merging the image after registration, it has abundanter comprehensive field data, can react scene of the accident entirety and crucial local scene situation better, be convenient to analysis and the utilization of scene of a traffic accident information;

Utilize the comprehensive Aerial Images of registration integration technology and identification plate close shot image information is set, multiple site survey of traffic accident image is combined, make full use of the feature that difference reconnoitres image, piece image is expressed the multi-aspect information from the scene of the accident simultaneously;

Many-sided situations such as the key element position in the scene of the accident, key element details are reflected by the image after registration merges, thus scene of a traffic accident information is provided more intuitively.

Additional aspect of the present invention and advantage will part provide in the following description, and part will become obvious from the following description, or be recognized by practice of the present invention.

Accompanying drawing explanation

Above-mentioned and/or additional aspect of the present invention and advantage will become obvious and easy understand from accompanying drawing below combining to the description of embodiment, wherein:

Fig. 1 is the method flow diagram that autoregistration of the present invention merges Aerial Images and close shot image;

Fig. 2 is that autoregistration of the present invention to merge in the method for Aerial Images and close shot image Aerial Images and arranges the image registration of identification plate close shot and merges schematic diagram;

Fig. 3 is that autoregistration of the present invention to merge in the method for Aerial Images and close shot image Aerial Images and arranges identification plate close shot automatic image registration process flow diagram;

Fig. 4 is method that autoregistration of the present invention merges Aerial Images and close shot image merges Aerial Images and close shot image identification plate for autoregistration.

Embodiment

Be described below in detail embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Being exemplary below by the embodiment be described with reference to the drawings, only for explaining the present invention, and can not limitation of the present invention being interpreted as.

In describing the invention, it will be appreciated that, term " longitudinal direction ", " transverse direction ", " on ", D score, "front", "rear", "left", "right", " vertically ", " level ", " top ", " end " " interior ", the orientation of the instruction such as " outward " or position relationship be based on orientation shown in the drawings or position relationship, only the present invention for convenience of description and simplified characterization, instead of indicate or imply that the device of indication or element must have specific orientation, with specific azimuth configuration and operation, therefore can not be interpreted as limitation of the present invention.

In describing the invention, unless otherwise prescribed and limit, it should be noted that, term " installation ", " being connected ", " connection " should be interpreted broadly, such as, can be mechanical connection or electrical connection, also can be the connection of two element internals, can be directly be connected, also indirectly can be connected by intermediary, for the ordinary skill in the art, the concrete meaning of above-mentioned term can be understood as the case may be.

Technical solution of the present invention carries out pattern bamboo product to the identification plate that the scene of the accident is placed, proposition identification plate color is as the index feature arranging identification plate close shot image, image Corner Detection Algorithm is utilized to determine to arrange the correspondence position of identification plate close shot image in Aerial Images, determined the registration transformation relation arranged between identification plate close shot image and Aerial Images by cross-correlation similarity measure, finally adopt image pixel rgb value weighted method to realize arranging the fusion of identification plate close shot image and Aerial Images.

As shown in Figure 1, the invention provides a kind of method that autoregistration merges Aerial Images and close shot image, its key is, comprises the steps:

Step 1, to the Aerial Images collected with arrange that identification plate close shot image is smoothing, sharpening, abatement Aerial Images and the noise that identification plate close shot image is set, strengthen Aerial Images and the edge that identification plate close shot image is set, then carry out the Aerial Images after coordinate conversion process acquisition image enhaucament and identification plate close shot image vertical view is set;

Step 2, indicating board size determined mathematics conversion proportion relation according to Aerial Images with arranging in identification plate close shot image vertical view, being transformed in the coordinate system of Aerial Images by arranging identification plate close shot image, completes uniform coordinate conversion;

Step 3, in Aerial Images, the described correspondence position arranging identification plate close shot image in search location, determines the registration transformation relation between two width images according to cross-correlation similarity measure;

Step 4, adopts image pixel rgb value weighted method to realize Aerial Images and arrange the fusion of identification plate close shot image overlapping region.

The beneficial effect of technique scheme is: through merging the image after registration, it has abundanter comprehensive field data, can react scene of the accident entirety and crucial local scene situation better, be convenient to analysis and the utilization of scene of a traffic accident information.

Described autoregistration merges the method for Aerial Images and close shot image, and preferably, described step 3 comprises:

Step 3-1, identification plate color according to arranging identification plate close shot image has rotational invariance and scale invariability, described identification plate structural color is carried out search location as the described index feature arranging identification plate close shot image to it in Aerial Images, obtains and the band of position of identification plate close shot image in Aerial Images is set;

Step 3-2, by adopting Harris Corner Detection Algorithm using the center of circle of described identification plate sector structure place circle as the registration control points arranging identification plate close shot image and Aerial Images;

Step 3-3, according to Aerial Images with arrange identification plate close shot image and be RGB color image, selects to adopt image rgb value cross-correlation registration, obtains optimal registration position.

To show further below in more details of the present invention and feature interpretation:

1, Image semantic classification: to the Aerial Images collected with arrange that identification plate close shot image is smoothing, sharpening, abatement Aerial Images and the noise that identification plate close shot image is set, strengthen Aerial Images and the edge that identification plate close shot image is set, then carry out the Aerial Images after coordinate conversion process acquisition image enhaucament and identification plate close shot image vertical view is set.Utilize homogeneous coordinate transformation by described Aerial Images and arrange identification plate close shot image conversion to overhead view image, the homogeneous coordinate transformation equation of employing is:

$\left(\begin{array}{c}{x}^{\′}\\ y^{\′}\end{array}\right)=\frac{\left(\begin{array}{cc}a& b\\ c& d\end{array}\right)\left(\begin{array}{c}x\\ y\end{array}\right)+\left(\begin{array}{c}e\\ f\end{array}\right)}{1+\left(\begin{array}{cc}u& v\end{array}\right)\left(\begin{array}{c}x\\ y\end{array}\right)},$

Wherein, a, b, c, d, e, f, u, v are coordinate transformation parameter, and x, y are coordinate vector, x ', y ' and be coordinate vector to be solved;

2, uniform coordinate: indicating board size determined mathematics conversion proportion relation according to Aerial Images with arranging in identification plate close shot image vertical view, being transformed in the coordinate system of Aerial Images by arranging identification plate close shot image, completes uniform coordinate conversion.

3, image registration: search location arranges the correspondence position of identification plate in Aerial Images of identification plate close shot image, determines the registration transformation relation between two width images according to cross-correlation similarity measure.

3.1 have rotational invariance and scale invariability according to identification plate color characteristic, identification plate sector structure color is carried out search location as the index feature arranging identification plate close shot image to it in Aerial Images, obtains and the band of position of identification plate close shot image in Aerial Images is set;

3.2 should be comparatively clear and definite, stable according to registration control points, and for Aerial Images with to arrange on identification plate close shot image all easily identification and the less prominent feature point of target, by adopting Harris Corner Detection Algorithm using the center of circle of identification plate sector structure place circle as the registration control points arranging identification plate close shot image and Aerial Images.The mathematical formulae of its algorithm is:

E (u, v)=∑ _x,yw (x, y) [I (x+u, y+v)-I (x, y)] ²formula (1)

Wherein, w (x, y) is window function, [I (x+u, y+v)-I (x, y)] ²for the Grad of image.W (x, y) is rectangular window or Gaussian window, and for each little displacement (u, v), u, v represent displacement respectively, and formula (1) can be expressed as by bilinear approximation

$E(u,v)\≅[u,v]M\left[\begin{array}{c}u\\ v\end{array}\right]$ Formula (2)

M is the matrix of 2*2:

$\left[\begin{array}{cc}A& C\\ C& B\end{array}\right]=e^{\frac{x^2+y^2}{2{\mathrm{\σ}}^2}}\⊗\left[\begin{array}{cc}{\mathrm{Ix}}^2& \mathrm{IxIy}\\ \mathrm{IxIy}& {\mathrm{Iy}}^2\end{array}\right]$ Formula (3)

Ix, Iy are respectively the Grad in image x, y direction; E can be similar to as local cross correlation function, and M describes shape in this.If λ ₁, λ ₂two eigenwerts, then λ of matrix M ₁, λ ₂the curvature of local autocorrelation function can be represented.Due to isotropy, so M keeps rotational invariance, wherein A, B, C represent the element value of matrix M.

3.3 according to Aerial Images with arrange identification plate close shot image and be RGB color image, selects to adopt image rgb value cross-correlation registration.Aerial Images is selected a contingent window W (l, m), the upper searching of identification plate close shot image S (i, j) the most similar window corresponding with it is being set (rotational transform window), with cross-correlation similarity measure is:

$R(i,j)=\frac{\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}W(l,m)S_M^{i,j}(l,m)}{\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{m=1}{\overset{M}{\mathrm{\Σ}}}{\left[S^{i,j}(l,m)\right]}^2}(i^{*},j^{*}),1\≤i,j\≤L-M+1$ Formula (4)

The wherein video in window of W (i, j) to be size be M × M, it is the contingent window in Aerial Images.Video in window W (i, j) to arrange in identification plate close shot image S with scan mode automatic search in search, finds correlation function value R in this process _nposition (the i of maximum of points in (i, j) ^*, j ^*), and think to be exactly optimal registration position herein, wherein l, m are positive integer, and subscript N is positive integer.

4, image co-registration: adopt image pixel rgb value weighted method to realize Aerial Images and the fusion of identification plate close shot image overlapping region is set.Method is by Aerial Images I ₁rgb value R ₁(i, j), G ₁(i, j), B ₁(i, j) and identification plate close shot image I is set ₂rgb value R ₂(i, j), G ₂(i, j), B ₂(i, j) is multiplied by a weight coefficient separately, merges and forms new image F (i, j).

F (i, j)=aI ₁(i, j)+(1-a) I ₂(i, j) formula (5)

Wherein: a is weight factor, and 0≤a≤1, the size of a can be regulated as required.

As shown in Figure 2, for using the specific embodiment of said method, arranging identification plate close shot image 1 is the fusion process using blue logo plate, and arranging identification plate close shot image 2 is the fusion process using red marker plate, in specific implementation process, be not limited to the amalgamation mode shown in Fig. 2.

As shown in Figure 3, the method for registration is,

S1, first pre-service identification plate close shot image I ₂; Determine that registration control points initial transformation value is R;

S2, rotational transform initial transformation value R, the image after its conversion is I ^*=R (I ₂);

S3, to arranging identification plate close shot image and Aerial Images I ₁the determined image window of registration control points is analyzed, and draws similarity measure evaluation;

S4, determines that whether transformed value R is optimum, if optimum, performs end, if do not reach optimum, continues to upgrade transformed value R, until reach optimum R.

As shown in Figure 4,1 be blue right angle sector structure, 2 be green right angle sector structure, 3 be red right angle sector structure, 4 for yellow right angle sector structure, but Fig. 4 is exemplary illustration, be not limited to the kind of above-mentioned color in practical operation, right angle sector structure is preferred sector structure.Fig. 4 arranges a set of identification plate of identification plate close shot image co-registration to Aerial Images by 4.Described identification plate color is different, can arrange a set of 4 identification plates incessantly according to scene of the accident scene situation, can also arrange N and overlap identification plate.

The invention has the beneficial effects as follows:

Through merging the image after registration, it has abundanter comprehensive field data, can react scene of the accident entirety and crucial local scene situation better, be convenient to analysis and the utilization of scene of a traffic accident information;

Utilize the comprehensive Aerial Images of registration integration technology and identification plate close shot image information is set, multiple site survey of traffic accident image is combined, make full use of the feature that difference reconnoitres image, piece image is expressed the multi-aspect information from the scene of the accident simultaneously;

Many-sided situations such as the key element position in the scene of the accident, key element details are reflected by the image after registration merges, thus scene of a traffic accident information is provided more intuitively.

In the description of this instructions, specific features, structure, material or feature that the description of reference term " embodiment ", " some embodiments ", " example ", " concrete example " or " some examples " etc. means to describe in conjunction with this embodiment or example are contained at least one embodiment of the present invention or example.In this manual, identical embodiment or example are not necessarily referred to the schematic representation of above-mentioned term.And the specific features of description, structure, material or feature can combine in an appropriate manner in any one or more embodiment or example.

Although illustrate and describe embodiments of the invention, those having ordinary skill in the art will appreciate that: can carry out multiple change, amendment, replacement and modification to these embodiments when not departing from principle of the present invention and aim, scope of the present invention is by claim and equivalents thereof.

Claims (7)

1. autoregistration merges a method for Aerial Images and close shot image, it is characterized in that, comprises the steps:

Step 1, to the Aerial Images collected with arrange that identification plate close shot image is smoothing, sharpening, abatement Aerial Images and the noise that identification plate close shot image is set, strengthen Aerial Images and the edge that identification plate close shot image is set, then carry out the Aerial Images after coordinate conversion process acquisition image enhaucament and identification plate close shot image vertical view is set;

Step 2, indicating board size determined mathematics conversion proportion relation according to Aerial Images with arranging in identification plate close shot image vertical view, being transformed in the coordinate system of Aerial Images by arranging identification plate close shot image, completes uniform coordinate conversion;

Step 3, in Aerial Images, the described correspondence position arranging identification plate close shot image in search location, determines the registration transformation relation between two width images according to cross-correlation similarity measure;

Step 4, adopts image pixel rgb value weighted method to realize Aerial Images and arrange the fusion of identification plate close shot image overlapping region.

2. autoregistration according to claim 1 merges the method for Aerial Images and close shot image, and it is characterized in that, described step 3 comprises:

Step 3-1, identification plate color according to arranging identification plate close shot image has rotational invariance and scale invariability, described identification plate structural color is carried out search location as the described index feature arranging identification plate close shot image to close shot image, obtains the band of position of close shot image in Aerial Images;

Step 3-2, by adopting Harris Corner Detection Algorithm using the center of circle of described identification plate sector structure place circle as the registration control points arranging identification plate close shot image and Aerial Images;

Step 3-3, according to Aerial Images with arrange identification plate close shot image and be RGB color image, selects to adopt image pixel rgb value cross-correlation registration, obtains optimal registration position.

3. autoregistration according to claim 2 merges the method for Aerial Images and close shot image, and it is characterized in that, described registration control points algorithm is:

E(u,v)＝Σ _x,yw(x,y)[I(x+u,y+v)-I(x,y)] ²

Wherein, w (x, y) is window function, [I (x+u, y+v)-I (x, y)] ²for the Grad of image, w (x, y) is rectangular window or Gaussian window, and for each displacement (u, v), in above formula, bilinear approximation is expressed as,

$E(u,v)\widetilde{=}\[u,v\]M\left[\begin{array}{c}u\\ v\end{array}\right],$

M is the matrix of 2*2, and concrete formula is,

$M=\left[\begin{array}{cc}A& C\\ C& B\end{array}\right]=e^{\frac{x^2+y^2}{2{\mathrm{\σ}}^2}}\⊗\left[\begin{array}{cc}{\mathrm{Ix}}^2& IxIy\\ IxIy& {\mathrm{Iy}}^2\end{array}\right]$

Ix, Iy are respectively the Grad in image x, y direction; E is approximate as local cross correlation function, and M describes shape in this, if λ ₁, λ ₂two eigenwerts, then λ of matrix M ₁, λ ₂can represent the curvature of local autocorrelation function, due to isotropy, so M keeps rotational invariance, wherein A, B, C represent the element value of matrix M.

4. autoregistration according to claim 2 merges the method for Aerial Images and close shot image, and it is characterized in that, described cross-correlation registration Algorithm is:

Aerial Images is selected a contingent window W (l, m), the upper searching of identification plate close shot image S (i, j) the most similar window corresponding with it is being set it is rotational transform window, with cross-correlation similarity measure is, wherein l, m are positive integer,

$R(i,j)=\frac{\underset{i=1}{\overset{M}{\Σ}}\underset{m=1}{\overset{M}{\Σ}}W(l,m)S_M^{i,j}(l,m)}{\underset{i=1}{\overset{M}{\Σ}}\underset{m=1}{\overset{M}{\Σ}}{\[S^{i,j}(l,m)\]}^2},1\≤i,j\≤L-M+1$

Video in window W (i, j) with scan mode automatic search, finds the position (i of maximum of points in correlation function value R (i, j) in this process in search close shot image S ^*, j ^*), and think to be exactly optimal registration position herein.

5. autoregistration according to claim 1 merges the method for Aerial Images and close shot image, and it is characterized in that, described step 4 comprises:

Step 4-1, by the rgb value R of Aerial Images ₁(i, j), G ₁(i, j), B ₁(i, j) and the rgb value R of identification plate close shot image is set ₂(i, j), G ₂(i, j), B ₂(i, j) is multiplied by a weight coefficient separately, merges and forms new image F (i, j),

F(i,j)＝aI ₁(i,j)+(1-a)I ₂(i,j)

Wherein, a is weight factor, and 0≤a≤1, I ₁for Aerial Images, I ₂for arranging identification plate close shot image.

6. autoregistration according to claim 2 merges the method for Aerial Images and close shot image, and it is characterized in that, described identification plate is right angle sector structure.

7. autoregistration according to claim 2 merges the method for Aerial Images and close shot image, and it is characterized in that, the color of described identification plate is different.