CN103017653B

CN103017653B - Registration and measurement method of spherical panoramic image and three-dimensional laser scanning point cloud

Info

Publication number: CN103017653B
Application number: CN201210490922.6A
Authority: CN
Inventors: 刘守军; 龚书林
Original assignee: WUHAN HAIDASHU CLOUD TECHNOLOGY Co Ltd
Current assignee: WUHAN HAIDASHU CLOUD TECHNOLOGY Co Ltd
Priority date: 2012-11-27
Filing date: 2012-11-27
Publication date: 2015-06-24
Anticipated expiration: 2032-11-27
Also published as: CN103017653A

Abstract

The invention provides a registration and measurement method of a spherical panoramic image and a three-dimensional laser scanning point cloud. The method comprises the following steps: firstly, using a three-dimensional structure simulation spherical panoramic image of a unit sphere, converting a pixel coordinate of the spherical panoramic image into a three-dimensional rectangular coordinate, then calculating an attitude parameter of the spherical panoramic image in point cloud data and a position coordinate of the optical center of the spherical panoramic image in the point cloud data to carry out the registration to the spherical panoramic image in the point cloud data; obtaining a depth map of the spherical panoramic image through the calculation; and finally moving a mouse to a target point position to be measured when the measurement is required to be carried out on the spherical panoramic image to calculate a three-dimensional coordinate of the target point. The registration and measurement method provided by the invention not only can carry out the height registration on the spherical panoramic image and the three-dimensional laser scanning point cloud, and also can realize fast measurement on a target point simultaneously.

Description

Registration and measurement method of spherical panoramic image and three-dimensional laser scanning point cloud

Technical Field

The invention relates to a method for registering a spherical panoramic image and a three-dimensional laser scanning point cloud and then quickly measuring according to an image, and belongs to the field of photoelectric mapping.

Background

The three-dimensional laser point cloud is a high-precision three-dimensional coordinate of the surface of an object, and the coordinates and characteristic information of the ground object can be directly obtained through the point cloud, but texture and color information does not exist, and the performance is not visual. The spherical panoramic image data has vivid texture and color information and can be used as texture supplement of three-dimensional laser point cloud data. The three-dimensional laser point cloud data and the spherical panoramic image data have advantages and characteristics, if the three-dimensional laser point cloud data and the spherical panoramic image data are used in a combined mode, corresponding image pixels can be found through the three-dimensional laser point cloud to conduct point cloud coloring, otherwise, the spherical panoramic image serves as a display carrier, the corresponding three-dimensional laser point cloud can be found through image pixel coordinates, and ground object measurement is conducted. However, to achieve the above purpose, the problem of registration between the two must be solved, which is currently a hot spot of research. Meanwhile, how to quickly acquire point cloud coordinates according to any pixel in the registered spherical panoramic view is also an urgent problem to be solved, and the problem of point cloud data storage must be solved to solve the problem of quick measurement of the spherical panoramic image because massive point cloud data is not suitable for application of the internet.

Disclosure of Invention

The invention provides a method for registering and measuring a spherical panoramic image and three-dimensional laser scanning point cloud, which solves the problems in the background technology.

The technical scheme adopted for realizing the above purpose of the invention is as follows:

a registration and measurement method for a spherical panoramic image and a three-dimensional laser scanning point cloud comprises the following steps:

(1) simulating a spherical panoramic image by using the three-dimensional structure of the unit sphere, and converting the pixel coordinates (u, v) of the spherical panoramic image into spherical polar coordinatesThen, the spherical polar coordinates of the spherical panoramic image are converted into three-dimensional rectangular coordinates (x)₀，y₀，z₀)；

(2) Selecting more than three points which are respectively positioned on different planes from the spherical panoramic image as control points, selecting the control points from point cloud data corresponding to the control points, and calculating the attitude parameters (phi, omega, kappa) of the spherical panoramic image in the point cloud data and the position coordinates (Xc, Yc, Zc) of the optical center of the spherical panoramic image in the point cloud data according to the coordinates of the control points in the spherical panoramic image and the coordinates in the point cloud data, thereby registering the spherical panoramic image in the point cloud data;

(3) searching point clouds in a space range by taking optical center position coordinates (Xc, Yc and Zc) of the spherical panoramic image as an origin, converting a distance value d from each searched point to the optical center into an RGB value, and storing the converted RGB values in a picture format to obtain a depth map of the spherical panoramic image;

(4) when the measurement is needed on the spherical panoramic image, the mouse is moved to the position of the target point to be measured, and the pixel coordinates (u, v) and the spherical polar coordinates of the target point are read at the momentCalculating the pixel coordinate (u ', v') of the target point in the depth map of the spherical panoramic image according to the attitude parameter (phi, omega, kappa) and the pixel coordinate (u, v), reading the RGB value in the pixel, converting the RGB value into a distance value d, and finally converting the distance value d and the angle coordinate into a distance value dThe three-dimensional coordinates (x, y, z) of the target point can be calculated.

Converting the pixel coordinates (u, v) into spherical polar coordinates in the step (1)The formula of (1) is:

，

in the formula, dx is the longitudinal pixel size of the spherical panoramic image, and dy is the transverse pixel size of the spherical panoramic image; polar coordinates of sphereConversion to three-dimensional rectangular coordinates (x)₀，y₀，z₀) The calculation formula of (2) is as follows:

。

in the step (2), the calculation method of the attitude parameters (phi, omega, kappa) and the optical center position coordinates (Xc, Yc, Zc) of the spherical panoramic image comprises the following steps: obtaining the following relation according to the point in the point cloud data, the point in the spherical panoramic image and the three-point collinearity of the spherical center of the spherical panoramic image:

wherein R is a rotation matrix, and R is a rotation matrix,

obtaining after conversion:

wherein:

，

wherein a is₁、a₂、a₃、b₁、b₂、b₃、c₁、c₂、c₃An element that is a rotational matrix;

and expanding the converted formula to obtain:

setting: ,

then: ,

,

and is also provided with: ，

wherein alpha and beta are observed values, and the corresponding correction number is v_α、v_β(ii) a The above equation is developed into a linear form by using a taylor formula, and a linearized error equation can be obtained as follows:

，

wherein, (alpha), (beta) are calculated values of alpha and beta obtained by substituting approximate values of parameters to be determined into a formula, and delta Xc, delta Yc, delta Zc, delta phi, delta omega and delta kappa are correction numbers of unknown numbers Xc, Yc, Zc, phi, omega and kappa;

then, the partial derivatives of α, β for each unknown are calculated using the following formula:

and then converting the error equation into a matrix form:

wherein,

according to the least square adjustment principle, the following equation is listed:

the solution of the normal equation is:

thus, 6 elements delta Xc, delta Yc, delta Zc, delta phi, delta omega and delta kappa of the correction matrix H of the approximate value of the unknown number to be measured are solved;

and finally, carrying out repeated iterative computation on the computation result, taking the sum of the unknown number approximate value and the correction number of the last iterative computation as a new approximate value during each iteration, repeating the computation process, solving a new correction number, and repeating the computation process until the correction number is smaller than a set threshold value, and finally obtaining the solution of the unknown number, wherein the solution is as follows:

wherein Xc⁰As an approximation of Xc, Δ Xc¹For the first calculated correction, Δ Xc²The number of corrections for the second calculation is equal to the other data.

The specific steps in the step (3) are as follows: firstly, presetting a depth map of a spherical panoramic image, dividing the preset depth map into grids according to an angle coordinate and a pixel coordinate, and opening up a data set for storing a point cloud distance stored in each grid; then, the optical center position coordinates (Xc, Yc, Zc) of the spherical panoramic image are used as the origin to search the point cloud in the space range, and the polar coordinates of the points are calculated according to the searched points and the optical center coordinatesAnd a value of distance to the optical center d; according to the polar coordinates of the pointsCalculating a specific grid in the depth map corresponding to the determined point, and updating the distance value d of the point in the grid; and finally, converting all the distance values d into RGB values, and storing the converted RGB values in corresponding grids one by one to obtain the depth map of the spherical panoramic image.

When a plurality of points are simultaneously corresponding to a grid, the minimum distance value among the plurality of points is selected.

The registration and measurement method of the spherical panoramic image and the three-dimensional laser scanning point cloud provided by the invention has the following advantages: the three-dimensional laser point cloud and the panoramic image data are combined after registration, so that corresponding image pixels can be found through the three-dimensional laser point cloud to perform point cloud coloring, otherwise, the spherical panoramic image is used as a display carrier, and the corresponding three-dimensional laser point cloud can be found through image pixel coordinates to perform ground feature measurement. Meanwhile, a large amount of point cloud data are converted into a depth map and stored in a picture mode, so that the data volume is greatly reduced, and rapid measurement is realized.

Detailed Description

The present invention will be described in detail with reference to specific examples.

In this embodiment, a method is adopted in which a spherical panoramic image is first simulated using a three-dimensional structure of a unit sphere, and pixel coordinates (u, v) of the spherical panoramic image are first converted into spherical polar coordinatesThe conversion formula is as follows:

，

in the formula, dx is the longitudinal pixel size of the spherical panoramic image, and dy is the transverse pixel size of the spherical panoramic image;

then converting the spherical polar coordinate of the spherical panoramic image into a three-dimensional rectangular coordinate (x)₀，y₀，z₀)；

Then selecting more than three points on different planes from the spherical panoramic image as control points, selecting the control points from the point cloud data corresponding to the control points, and calculating the attitude parameters of the spherical panoramic image in the point cloud data according to the coordinates of the control points in the spherical panoramic image and the coordinates in the point cloud data (the attitude parameters of the spherical panoramic image in the point cloud data are obtained by calculation: (ω, κ) and the position coordinates (Xc, Yc, Zc) of the optical center of the spherical panorama image in the point cloud dataSo as to register the spherical panoramic image in the point cloud data; the specific calculation steps are as follows: firstly, according to the three collinear points of the point cloud data, the point in the spherical panoramic image and the spherical center of the spherical panoramic image, the following relational expression is obtained:

wherein R is a rotation matrix, and R is a rotation matrix,

obtaining after conversion:

wherein:

，

and expanding the converted formula to obtain:

setting: ,

then: ,

,

and is also provided with: ，

wherein alpha and beta are observed values, and the corresponding correction number is v_α、v_β； Xc、Yc、Zc、And omega and kappa are undetermined unknown numbers, and because the observed values and the unknown numbers in the above formula are in a nonlinear function relationship, in order to facilitate calculation, a nonlinear function expression needs to be expanded into a linear form by using a Taylor formula. After the above formula is developed by taylor formula, the linearized error equation can be obtained as follows:

，

wherein (alpha), (beta) are calculated values of alpha and beta obtained by substituting approximate values of the parameters to be determined into a formula, and delta Xc, delta Yc, delta Zc,Δ ω and Δ κ are unknowns Xc, Yc, Zc,The number of corrections of ω, κ;

the partial derivatives of the obtained alpha and beta with respect to the respective unknowns are shown as follows, ，。

and then converting the error equation into a matrix form:

wherein,

the solution of the normal equation is:

since each coefficient in the linearization process is taken from a first term of the taylor series expansion, and the initial value of the unknown number is generally rough, the calculation needs to be performed iteratively. The sum of the unknown number approximate value and the correction number of the last iteration calculation is used as a new approximate value during each iteration, the calculation process is repeated, the new correction number is obtained, the new correction number is repeatedly approached until the correction number is smaller than a set threshold value, and finally the solution of the unknown number is obtained, wherein the solution is as follows:

After the registration, a depth map of the spherical panoramic image needs to be generated, and the specific method is as follows: firstly, presetting a depth map of a spherical panoramic image, dividing the preset depth map into grids according to an angle coordinate and a pixel coordinate, and opening up a data set for storing a point cloud distance stored in each grid; then, the optical center position coordinates (Xc, Yc, Zc) of the spherical panoramic image are used as the origin to search the point cloud in the space range, and the polar coordinates of the points are calculated according to the searched points and the optical center coordinatesAnd a value of distance to the optical center d; according to the polar coordinates of the pointsCalculating a specific grid in the depth map corresponding to the determined point, and updating the distance value d of the point in the grid; and finally, converting all the distance values d into RGB values, and storing the converted RGB values in corresponding grids one by one to obtain the depth map of the spherical panoramic image.

Finally, when the measurement is needed on the spherical panoramic image, the mouse is moved to the position of the target point to be measured, and the pixel coordinates (u, v) and the spherical polar coordinates of the target point are readAccording to the attitude parameters (ω, κ) and pixel coordinates (u, v) of the objectReading the RGB value of the pixel according to the pixel coordinate (u ', v') of the point in the depth map of the spherical panoramic image, converting the RGB value into a distance value d, and finally obtaining the distance value d and the angle coordinateThe three-dimensional coordinates (x, y, z) of the target point can be calculated.

Claims

1. A registration and measurement method for a spherical panoramic image and a three-dimensional laser scanning point cloud is characterized by comprising the following steps:

(1) simulating a spherical panoramic image by using the three-dimensional structure of the unit sphere, and converting the pixel coordinates (u, v) of the spherical panoramic image into spherical polar coordinatesThen converting the spherical polar coordinate of the spherical panoramic image into a three-dimensional rectangular coordinate (x)₀，y₀，z₀)；

(2) Selecting more than three points which are respectively positioned on different planes from the spherical panoramic image as control points, selecting the control points from point cloud data corresponding to the control points, and calculating the attitude parameters of the spherical panoramic image in the point cloud data according to the coordinates of the control points in the spherical panoramic image and the coordinates in the point cloud dataAnd the position coordinates (Xc, Yc, Zc) of the optical center of the spherical panoramic image in the point cloud data, thereby registering the spherical panoramic image in the point cloud data;

the specific steps in the step (3) are as follows: firstly, presetting a depth map of a spherical panoramic image, dividing the preset depth map into grids according to an angle coordinate and a pixel coordinate, and opening up a data set for storing a point cloud distance stored in each grid; then, the optical center position coordinates (Xc, Yc, Zc) of the spherical panoramic image are used as the origin to search the point cloud in the space range, and the polar coordinates of the points are calculated according to the searched points and the optical center coordinatesAnd a distance value d to the optical center; according to the polar coordinates of the pointsCalculating a specific grid in the depth map corresponding to the determined point, and updating the distance value d of the point in the grid; finally, all the distance values d are converted into RGB values, and the converted RGB values are stored in corresponding grids one by one, so that a depth map of the spherical panoramic image can be obtained;

(4) when the mouse needs to be measured on the spherical panoramic imageThe target is moved to the position of the target point to be measured, and the pixel coordinates (u, v) and the spherical polar coordinates of the target point are readAccording to the attitude parametersAnd calculating the pixel coordinate (u ', v') of the target point in the depth map of the spherical panoramic image by the pixel coordinate (u, v), reading the RGB value in the pixel, converting the RGB value into a distance value d, and finally, calculating the distance value d and the angle coordinate according to the distance value d and the angle coordinateThe three-dimensional coordinates (x, y, z) of the target point can be calculated.

2. The method of claim 1, wherein the method comprises the steps of: converting the pixel coordinates (u, v) into spherical polar coordinates in the step (1)The formula of (1) is:

3. the method of claim 1, wherein the method comprises the steps of: attitude parameters in step (2)And the calculation method of the optical center position coordinates (Xc, Yc, Zc) of the spherical panoramic image comprises the following steps: obtaining the following relation according to the point in the point cloud data, the point in the spherical panoramic image and the three-point collinearity of the spherical center of the spherical panoramic image:

[\begin{matrix} {X - X}_{c} \\ {Y - Y}_{c} \\ {Z - Z}_{c} \end{matrix}] = R [\begin{matrix} x_{0} \\ y_{0} \\ z_{0} \end{matrix}]

wherein R is a rotation matrix, and R is a rotation matrix,

obtaining after conversion:

[\begin{matrix} x_{0} \\ y_{0} \\ z_{0} \end{matrix}] = {kR}^{T} [\begin{matrix} {X - X}_{c} \\ {Y - Y}_{c} \\ {Z - Z}_{c} \end{matrix}]

wherein:

and expanding the converted formula to obtain:

\begin{matrix} x_{0} = k [a_{1} ({X - X}_{c}) + b_{1} ({Y - Y}_{c}) + c_{1} ({Z - Z}_{c})] \\ y_{0} = k [a_{2} ({X - X}_{c}) + b_{2} ({Y - Y}_{c}) + c_{2} ({Z - Z}_{c})] \\ z_{0} = k [a_{3} ({X - X}_{c}) + b_{3} ({Y - Y}_{c}) + c_{3} ({Z - Z}_{c})] \end{matrix}

setting:

\begin{matrix} X_{0} = a_{1} ({X - X}_{c}) + b_{1} ({Y - Y}_{c}) + c_{1} ({Z - Z}_{c}) \\ Y_{0} = a_{2} ({X - X}_{c}) + b_{2} ({Y - Y}_{c}) + c_{2} ({Z - Z}_{c}) \\ Z_{0} = a_{3} ({X - X}_{c}) + b_{3} ({Y - Y}_{c}) + c_{3} ({Z - Z}_{c}) \end{matrix},

then:

\begin{matrix} x_{0} = {kX}_{0} \\ y_{0} = {kY}_{0} \\ z_{0} = {kZ}_{0} \end{matrix},

\begin{matrix} x_{0} / z_{0} = X_{0} / Z_{0} \\ y_{0} / z_{0} = Y_{0} / Z_{0} \end{matrix},

and is also provided with:

wherein alpha and beta are observed values, and the corresponding correction number is v_α、ν_β(ii) a The above equation is developed into a linear form by using a taylor formula, and a linearized error equation can be obtained as follows:

<math> <mrow> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <msub> <mi>v</mi> <mi>α</mi> </msub> <mo>=</mo> <mrow> <mo>(</mo> <mi>α</mi> <mo>)</mo> </mrow> <mo>-</mo> <mi>α</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>X</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mi>ΔXc</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Y</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mi>ΔYc</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Z</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mi>ΔZc</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>φ</mi> </mrow> </mfrac> <mi>Δφ</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>ω</mi> </mrow> </mfrac> <mi>Δω</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>κ</mi> </mrow> </mfrac> <mi>Δκ</mi> </mtd> </mtr> <mtr> <mtd> <msub> <mi>v</mi> <mi>β</mi> </msub> <mo>=</mo> <mrow> <mo>(</mo> <mi>β</mi> <mo>)</mo> </mrow> <mo>-</mo> <mi>β</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>X</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mi>ΔXc</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Y</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mi>ΔYc</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Z</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mi>ΔZc</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>φ</mi> </mrow> </mfrac> <mi>Δφ</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>ω</mi> </mrow> </mfrac> <mo>+</mo> <mi>Δω</mi> <mo>+</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>κ</mi> </mrow> </mfrac> <mi>Δκ</mi> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math>

wherein (alpha), (beta) are calculated values of alpha and beta obtained by substituting approximate values of the parameters to be determined into a formula, and delta Xc, delta Yc, delta Zc,Δ ω and Δ κ are correction numbers of unknown numbers Xc, Yc, Zc, Φ, ω, κ;

and then converting the error equation into a matrix form: v ═ AH-L

Wherein,

V＝[v_α,v_β]^T

<math> <mrow> <mi>A</mi> <mo>=</mo> <mfenced open='[' close=']'> <mtable> <mtr> <mtd> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>X</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Y</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Z</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>φ</mi> </mrow> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>ω</mi> </mrow> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>α</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>κ</mi> </mrow> </mfrac> </mtd> </mtr> <mtr> <mtd> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>X</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Y</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <msub> <mrow> <mo>&PartialD;</mo> <mi>Z</mi> </mrow> <mi>c</mi> </msub> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>φ</mi> </mrow> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>ω</mi> <mo></mo> </mrow> </mfrac> <mo>,</mo> <mfrac> <mrow> <mo>&PartialD;</mo> <mi>β</mi> </mrow> <mrow> <mo>&PartialD;</mo> <mi>κ</mi> </mrow> </mfrac> </mtd> </mtr> </mtable> </mfenced> </mrow> </math>

H＝[Δx,Δy,Δz,Δφ,Δω,Δκ]^T

L＝[α-(α),β-(β)]^T

A^TAH＝A^TL

the solution of the normal equation is:

H＝(A^TA)^-1A^TL

thereby obtaining 6 elements Delta Xc, Delta Yc, Delta ZC of the correction matrix H of the approximate value of the unknown number to be measured,Δω、Δκ；

X_c＝X_c ⁰+ΔX_c ¹+ΔX_c ²+Λ

Y_c＝Y_c ⁰+ΔY_c ¹+ΔY_c ²+Λ

Z_c＝Z_c ⁰+ΔZ_c ¹+ΔZ_c ²+Λ

φ＝φ⁰+Δφ¹+Δφ²+Λ

ω＝ω⁰+Δω¹+Δω²+Λ

κ＝κ⁰+Δκ¹+Δκ²+Λ

4. The method of claim 1, wherein the method comprises the steps of: when a plurality of points are simultaneously corresponding to a grid, the minimum distance value among the plurality of points is selected.