CN102889882B - Three-dimensional reconstruction method based on bundle adjustment - Google Patents

Abstract

The invention relates to a three-dimensional reconstruction method based on bundle adjustment. The three-dimensional reconstruction method comprises the following steps of: calibrating parameters of the interior orientation of a digital camera in a control field through a first iterative process; calibrating parameters of the exterior orientation of the digital camera through a space resection method with a second iterative process in a measuring field by using the parameters of the interior orientation as a given value; and acquiring three-dimensional spatial values of a plurality of measuring points through a bundle adjustment method based on a collinearity condition equation by using the parameters of the interior orientation and the parameters of the exterior orientation as initial values and using image point coordinates of a plurality of control points and a plurality of measuring points in the measuring field as observed values. By the three-dimensional reconstruction method based on the bundle adjustment in an embodiment of the invention, the positioning accuracy of spatial measuring points can be improved in a three-dimensional reconstruction process.

Description

A kind of three-dimensional rebuilding method based on light beam adjustment

Technical field

The present invention relates to large scale photogrammetric technology field, particularly a kind of three-dimensional rebuilding method based on light beam adjustment.

Background technology

Along with the fast development of Chinese national economy, in the production in the fields such as Aeronautics and Astronautics, electronics, automobile, weaponry and processing, all large-scale metrology precision is proposed to clear and definite expectation and requirement.As airship propelling module is measured (3m), large-sized water turbine field ring is measured (12m), Large Radar Antenna measurement (10m), large aircraft wing measurement etc., in these large-scale workpiece Assembling Production processes, all want the position of accurate positioning workpieces.

Along with digitizing, the widespread use of automatic technology in large-scale workpiece assembling process, the location technology of large-scale workpiece is developed to digitizing, robotization and flexibility orientation by traditional craft location.The Assembling Production of large-scale workpiece is a complicated systems engineering, and a complete assembling system should be in official hour, to assemble and adjust according to unified standard, to assemble out qualified product.The online production process of large scale industry product comprises: assembling, and adjust and detect.Assembling is one of them link, in assembling process, get by some measurement means the position location of large-scale workpiece.Therefore, the measuring accuracy height of the measuring accuracy to position location and workpiece attitude, directly has influence on the mass effect that workpiece assembles.

Summary of the invention

The object of the present invention is to provide a kind of three-dimensional rebuilding method based on light beam adjustment, improve the three-dimensional measurement precision to space measurement point.

The embodiment of the present invention provides a kind of three-dimensional rebuilding method based on light beam adjustment, comprising:

At controlling filed, demarcate the interior orientation parameter of digital camera by the first iterative process;

In measurement field, using described interior orientation parameter as given value, demarcate the outer orientation parameter of described digital camera by thering is space resection's method of secondary iteration process;

Using described interior orientation parameter and described outer orientation parameter as initial value, using the picpointed coordinate of multiple measurement points of the multiple reference mark in described measurement field and described measurement field as observed reading, obtain the three dimensional space coordinate value of described multiple measurement points by the light beam method of adjustment based on collinear equation.

Three-dimensional rebuilding method based on light beam adjustment provided by the invention, obtain respectively interior orientation parameter and the outer orientation parameter of digital camera by the first iterative process and secondary iteration process, and using the interior orientation parameter obtaining by the first iterative process as initial value parameter to the secondary iteration process of demarcating outer orientation parameter, realize the substep of interior orientation parameter to digital camera and outer orientation parameter and demarcated, improved the stated accuracy of digital camera; Using interior orientation parameter and outer orientation parameter as initial value, obtain the three dimensional space coordinate value of space measurement point by the light beam error compensation method based on collinear equation, thereby further improved the positioning precision of space measurement point in three-dimensional reconstruction process.

Brief description of the drawings

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is the schematic flow sheet that the present invention is based on an embodiment of three-dimensional rebuilding method of light beam adjustment;

Fig. 2 is the schematic flow sheet that the present invention is based on another embodiment of three-dimensional rebuilding method of light beam adjustment;

Fig. 3 is the three-dimensional reconstruction schematic diagram to space measurement point in the embodiment of the present invention;

Fig. 4 is the schematic diagram that the reconstructed results of the embodiment of the present invention and the reconstructed results of reference instrument are compared.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.

Light beam adjustment described in the embodiment of the present invention (bundle adjustment) method, refer to the picpointed coordinate of the picpointed coordinate at reference mark, point to be located (can be also space measurement point) and part or all of other measurement data be all considered as to observed reading, with integrally side by side solve them or the calculation method of value (being also net result) and point to be located volume coordinate.The parameter of optimizing by light beam method of adjustment in the embodiment of the present invention comprises: the interior orientation parameter (x of digital camera ₀, y ₀, f, k ₁, k ₂, k ₃, p ₁, p ₂), join (X outside the outer orientation of camera _s, Y _s, Z _s, Az, El, Roll), space measurement puts corresponding object space coordinate (X, Y, Z); Wherein, in interior orientation parameter, x ₀, y ₀represent principal point coordinate, f represents that main distance (comprises the axial main apart from f of the first coordinate axis _xaxial main apart from f with the second coordinate axis _y), k ₁, k ₂, k ₃, p ₁, p ₂the distortion factor of the camera lens of representative digit camera, outside in direction parameter, X _s, Y _s, Z _sthree vertical elements of representative digit camera, three position angle elements of Az, El, Roll representative digit camera.

In addition, it will be understood by those skilled in the art that, the interior orientation parameter that controlling filed in the embodiment of the present invention has been used to calibrating camera arranges, and is provided with multiple coded targets in controlling filed, and multiple coded targets are specially the artificial target with highly reflective energy; Measurement field in the embodiment of the present invention is the measurement place forming in order to measure measured target, is provided with the measurand (measurand can be for example planar antenna) that need to carry out three-dimensional reconstruction in measurement field, is provided with multiple measurement points in measurand.

Fig. 1 is the schematic flow sheet that the present invention is based on an embodiment of three-dimensional rebuilding method of light beam adjustment; As shown in Figure 1, the embodiment of the present invention specifically comprises the steps:

Step 101, at controlling filed, demarcate and obtain the interior orientation parameter of digital camera by the first iterative process;

Step 102, in measurement field, using interior orientation parameter as given value, the outer orientation parameter of demarcating digital camera by thering is the space resection method of secondary iteration process;

Step 103, using interior orientation parameter and outer orientation parameter as initial value, using the picpointed coordinate of the multiple measurement points in the multiple reference mark in measurement field and measurement field as observed reading, obtain the three dimensional space coordinate value of the plurality of measurement point by the light beam error compensation method based on collinear equation.

The three-dimensional rebuilding method based on light beam adjustment that the embodiment of the present invention provides, obtain respectively interior orientation parameter and the outer orientation parameter of digital camera by the first iterative process and secondary iteration process, and participate in secondary iteration process using the interior orientation parameter obtaining by the first iterative process as initial value and demarcate the outer orientation parameter of digital camera, the interior orientation parameter and the outer orientation parameter that have realized digital camera are carried out substep demarcation, have improved the stated accuracy of digital camera; Again further using interior orientation parameter and outer orientation parameter as initial value, obtain the three dimensional space coordinate value of space measurement point by the light beam error compensation method based on collinear equation, further improved the positioning precision of space measurement point in three-dimensional reconstruction process.

Fig. 2 is the schematic flow sheet that the present invention is based on another embodiment of three-dimensional rebuilding method of light beam adjustment, and the embodiment of the present invention specifically comprises that taking constraint condition orthogonality restriction condition and engineer's scale uniform condition carry out exemplary illustration as example; As shown in Figure 2, the embodiment of the present invention specifically can comprise the steps:

Step 201, at controlling filed, the first coordinate axis and second coordinate axis of the image coordinate system of digital camera are set to vertical relation, make camera model meet orthogonality restriction condition;

The pel spacing axially and between the axial direction of the second coordinate axis of the first coordinate axis of step 202, digital camera is set to consistent and the first coordinate axis is identical with the ratio scale of the second coordinate axis, makes camera model meet engineer's scale uniform condition;

Step 203, the multiple coded targets and the corresponding volume coordinate of the plurality of coded target that are arranged in controlling filed are demarcated to the interior orientation parameter initial value of this camera model by direct linear transformation's method;

Step 204, known quantity using interior orientation parameter initial value and camera model as direct linear transformation's method, the interior orientation parameter of demarcating digital camera by the first iterative process;

Step 205, in measurement field, obtain the multiple image in the measurement field of setting quantity different azimuth by digital camera, on this multiple image, include the picture point at multiple reference mark and the picture point of multiple measurement points;

Step 206, according to multiple picpointed coordinates at corresponding multiple reference mark on the three dimensional space coordinate at the plurality of reference mark and multiple image by the picture point error equation of the corresponding collinearity condition equation of secondary iteration Procedure Acquisition;

Step 207, obtain the outer orientation parameter of camera model by picture point error equation;

Step 208, interior orientation parameter that step 204 is obtained are as internal reference initial value, and the outer orientation parameter that step 207 is obtained, as outer ginseng initial value, is calculated the initial three dimensional space coordinate of multiple measurement points by multi-disc space intersection method;

Step 209, obtain the second error equation matrix of light beam adjustment according to picture point error equation;

Step 210, by the initial three dimensional space coordinate of internal reference initial value, outer ginseng initial value and multiple measurement points, the second error equation matrix is carried out to adjustment processing, determine whether to stop this adjustment computing according to observed reading, and obtain the three dimensional space coordinate value of the plurality of measurement point.

What one of ordinary skill in the art will appreciate that is, at direct linear transformation (Direct Linear Transformation, be called for short DLT) in method, can be drawn the image coordinate (x of picture point by collinearity equation, y) direct linear relation and between the object space coordinate (X, Y, Z) of corresponding measurement point, this linear relation can be determined by the matrix L of 11 lambda coefficients, wherein matrix $L=\left[\begin{array}{cccc}{l}_1& l_2& l_3& l_4\\ l_5& l_6& l_7& l_8\\ l_9& l_{10}& l_{11}& 1\end{array}\right],$ Wherein, (l ₁, l ₂, l ₃) and (l ₅, l ₆, l ₇) dimensionless, (l ₄, l ₈) be long measure, (l ₉, l ₁₀, l ₁₁) unit be the inverse of length; It will be appreciated by persons skilled in the art that corresponding 11 the independent parameter (X of 11 lambda coefficients _s, Y _s, Z _s, (φ, ω, κ), x ₀, y ₀, f, ds, d β), wherein, ds is the inconsistent coefficient of engineer's scale, d β is non-orthogonal property angle.

In above-mentioned steps 201 and step 202, the first coordinate axis of the image coordinate system of the digital camera in digital Photogrammetric System (being specifically as follows x axle) and the second coordinate axis (being specifically as follows y axle) are set to vertical relation, thereby make camera model meet orthogonality restriction condition, i.e. d β=0; The pel spacing axially and between the axial direction of the second coordinate axis of the first coordinate axis of digital camera is set to unanimously, and the first coordinate axis is identical with the ratio scale of the second coordinate axis, thereby makes camera model meet engineer's scale uniform condition, i.e. ds=0.

In above-mentioned steps 203, by the picpointed coordinate at the reference mark of measurement field, (x, y) and corresponding object space coordinate (X, Y, Z) draws matrix by DLT method $L=\left[\begin{array}{cccc}{l}_1& l_2& l_3& l_4\\ l_5& l_6& l_7& l_8\\ l_9& l_{10}& l_{11}& 1\end{array}\right]$ In 11 lambda coefficients draw the relational expression of the interior orientation parameter of digital camera, shown in following equation (1-1):

$\left\{\begin{array}{c}{x}_0=-(l_1{l}_9+l_2{l}_{10}+l_3{l}_{11})/(l_9^2+l_{10}^2+l_{11}^2)\\ y_0=-(l_5{l}_9+l_6{l}_{10}+l_7{l}_{11})/(l_9^2+l_{10}^2+l_{11}^2)\\ f_x=\sqrt{(l_1^2+l_2^2+l_3^2)/(l_9^2+l_{10}^2+l_{11}^2)-x_0^2}\\ f_y=f_x\end{array}\right.---(1-1)$

By above-mentioned equation (1-1), can draw by the multiple reference mark in measurement field the principal point coordinate (x of digital camera ₀, y ₀) and captured image the first coordinate axis is axial main apart from f _xaxial main apart from f with the second coordinate axis _yinitial value.

In above-mentioned steps 204, the camera model of the digital camera of the DLT with two restriction conditions is as shown in system of equations (1-2):

$\left\{\begin{array}{c}{f}_1=A(x+\mathrm{\Δx})+(l_{1}X+l_{2}Y+l_{3}Z+l_4)\\ f_2=A(y+\mathrm{\Δy})+(l_{5}X+l_{6}Y+l_{7}Z+l_8)\\ g_1=B(l_1{l}_5+l_2{l}_6+l_3{l}_7)-(l_1{l}_9+l_2{l}_{10}+l_3+l_{11})(l_5{l}_9+l_6{l}_{10}+l_7{l}_{11})\\ g_2=B[(l_1^2+l_2^2+l_3^2)-(l_5^2+l_6^2+l_7^2)]+{(l_5{l}_9+l_6{l}_{10}+l_7{l}_{11})}^2+{(l_1{l}_9+l_2{l}_{10}+l_3{l}_{11})}^2\end{array}---(1-2)\right.$

Wherein, A=l ₉x+l ₁₀y+l ₁₁z+1, B=l ₉ ²+ l ₁₀ ²+ l ₁₁ ².In system of equations (1-2), the first two equation is the DLT equation expression formula transformation results after distortion correction, and latter two equation is orthogonality restriction and the equational transformation results of engineer's scale restriction condition.It will be understood by those skilled in the art that, the first pair of equation in above-mentioned system of equations (1-2) and the number at reference mark, space have relation, if there be n reference mark, can list 2n equation, add second pair of equation in system of equations (1-2), so can list altogether 2n+2 equation.The parameter of camera model optimization comprises 11 lambda coefficients and 5 distortion factor (k ₁, k ₂, k ₃, p ₁, p ₂), 16 parameters altogether, so need at least 7 reference mark to participate in computing.Because system of equations (1-2) is nonlinear, thus can be to the then adjustment computing of the advanced line linearityization of system of equations (1-2), obtain the first error equation matrix corresponding to camera model, shown in following equation (1-3):

V=MX-L （1-3）

Wherein:

$V={\left[\begin{array}{ccccccc}{v}_{f_1}& v_{f_2}& K& v_{f_{2n-1}}& v_{f_{2n}}& v_{g_1}& v_{g_2}\end{array}\right]}^T$

X=[Δl ₁ Δl ₂ Δl ₃ Δl ₄ Δl ₅ Δl ₆ Δl ₇ Δl ₈ Δl ₉ Δl ₁₀ Δl ₁₁ Δk ₁ Δk ₂ Δk ₃ Δp ₁ Δp ₂] ^T

L=[Δf ₁ Δf ₂ K Δf _2n-1 Δf _2n Δg ₁ Δg ₂] ^T

This linearization procedure is first iterative process, is the increment of institute's Optimal Parameters in X matrix, the initial value that the initial value of parameter is obtained by above-mentioned steps 203, and in above formula, Δ f _i=f _i-(f _i), (f _i) be the f after last iteration _ivalue, the Criterion of Iterative of this iterative process is be not more than the first predetermined threshold value, this first predetermined threshold value for example can be set to 10 ^-9.

In above-mentioned steps 205～step 207, because the resection of single image space is based on collinearity equation formula, the picpointed coordinate at the reference mark of the some that piece image is covered is used as observed reading and is solved the interior orientation parameter of this image and the photogrammetric process of outer orientation parameter.In embodiments of the present invention, obtain the multiple image at the reference mark in the measurement field of setting quantity different azimuth by digital camera.In the case of the interior orientation parameter that gets digital camera, only need further calibration for cameras outer orientation parameter, therefore, the embodiment of the present invention is a simplification of single image space resection.In the embodiment of the present invention, particularly, thus the general type of picture point error equation that can row collinearity condition equation, as shown in equation (1-4):

V=At-L （1-4）

Wherein:

V=[v _x v _y] ^T

L=[x-(x)y-(y)] ^T

In equation (1-4), (x, y) be picture coordinate survey value, (V _x, V _y) be corresponding correction, t matrix is representing the increment of outer orientation parameter, this computation process is a secondary iteration process; In addition,, about the initial value of the outer orientation parameter of digital camera, in shooting process, the summary value of the camera attitude can be using digital camera vertical photography time is as initial value.This picture point error equation (1-4) meets picture point residual error quadratic sum minimum, that is: V ^tv=min.In addition, due to this secondary iteration process solution procedure that is a least square, therefore secondary iteration process is an optimizing process that progressively approaches true value, Camera extrinsic is counted to initial value for inner to picture point normal equation (this picture point normal equation is specially the local derviation equation of picture point error equation), the small quantity for the treatment of solution parameter with LM algorithm (Levenberg-Marquardt algorithm) solves successively, afterwards, the parameter to be measured that iteration is obtained is as initial value substitution picture point normal equation again, so iteration is carried out always, until picture point error equation convergence, the root-mean-square error that convergence criterion in the embodiment of the present invention can be decided to be the micro-deviation of parameter to be measured is less than the second predetermined threshold value, this second predetermined threshold value for example can be set to 10 ^-10.In one embodiment, digital camera is apart from shooting distance general 5 meters (m) left and right at reference mark, and the initial value of the digital camera outer orientation parameter starting is thus decided to be: displacement of the lines X _s=5000mm, Y _s/ Z _sdirection is 0 millimeter (mm), and the initial value (φ, ω, κ) of angle element is all made as 0 degree.

In above-mentioned steps 208～step 210, the interior orientation parameter that step 204 is obtained is as internal reference initial value, the outer orientation parameter that step 207 is obtained is as outer ginseng initial value, calculate the initial three dimensional space coordinate of multiple measurement points with multi-disc space intersection method, initial value using this initial three dimensional space coordinate as unknown point coordinate, then using the picpointed coordinate of the multiple reference mark in measurement field and multiple measurement points as observed reading, entirety is treated and is asked parameter optimization.Particularly, according to picture point error equation (1-4), the second error equation matrix that can row light beam method of adjustment is as shown in equation (1-5):

V=At+BX ₁+CX ₂+DX ₃-L （1-5）

Wherein:

V＝[v _x v _y] ^T

$B=\left[\begin{array}{ccc}\frac{\∂x}{\∂X}& \frac{\∂x}{\∂Y}& \frac{\∂x}{\∂Z}\\ \frac{\∂y}{\∂X}& \frac{\∂y}{\∂Y}& \frac{\∂y}{\∂Z}\end{array}\right]$

X ₁=[ΔX ΔY ΔZ] ^T

$C=\left[\begin{array}{ccc}\frac{\∂x}{\∂x_0}& \frac{\∂x}{\∂y_0}& \frac{\∂x}{\∂f}\\ \frac{\∂y}{\∂x_0}& \frac{\∂y}{\∂y_0}& \frac{\∂y}{\∂f}\end{array}\right]$

X ₂=[Δx ₀ Δy ₀ Δf] ^T

$D=\left[\begin{array}{ccccc}\frac{\∂x}{\∂k_1}& \frac{\∂x}{\∂k_2}& \frac{\∂x}{\∂k_3}& \frac{\∂x}{\∂p_1}& \frac{\∂x}{\∂p_2}\\ \frac{\∂y}{\∂k_1}& \frac{\∂y}{\∂k_2}& \frac{\∂y}{\∂k_3}& \frac{\∂y}{\∂p_1}& \frac{\∂y}{\∂p_2}\end{array}\right]$

X ₃=[ΔK ₁ ΔK ₂ ΔK ₃ ΔP ₁ ΔP ₂] ^T

L=[x-(x)y-(y)] ^T

In equation (1-5), the modified value that V is picpointed coordinate, A is the local derviation matrix of outer orientation parameter, t is the increment of outer orientation parameter, the volume coordinate local derviation matrix that B is measurement point, X ₁for the volume coordinate increment of measurement point, C is the local derviation matrix of interior orientation parameter, X ₂for the increment of interior orientation parameter, the local derviation matrix that D is distortion parameter, X ₃for the increment of distortion parameter, L is the difference of picture point iteration.

Further, in adjustment processing process, in the time that the D coordinates value of the interior orientation parameter to digital camera, outer orientation parameter and multiple measurement points is optimized, because the partial parameters order of magnitude differs greatly, thereby cause causing solution to be made mistakes in the time solving system of linear equations, because for fear of or reduce system of equations number of coefficients level gap, need to represent by " dividing " magnitude the angle element of the outer orientation parameter of digital camera, distortion factor in the interior orientation parameter of digital camera is carried out to multiple and amplify processing, thereby reduce the error that solves of linear equation.

The three-dimensional rebuilding method based on light beam adjustment that the embodiment of the present invention provides, obtain respectively interior orientation parameter and the outer orientation parameter of digital camera by the first iterative process and secondary iteration process, and using the interior orientation parameter obtaining by the first iterative process as initial value parameter to the secondary iteration process of demarcating outer orientation parameter, realize the substep of interior orientation parameter to digital camera and outer orientation parameter and demarcated, improved the stated accuracy of digital camera; Using interior orientation parameter and outer orientation parameter as initial value, obtain the three dimensional space coordinate value of space measurement point by the light beam error compensation method based on collinear equation, thereby further improved the positioning precision of space measurement point in three-dimensional reconstruction process.

Further, above-mentioned embodiment illustrated in fig. 2 in, using the three dimensional space coordinate value of multiple measurement points as observed reading, be optimized by the internal direction parameter of adjustment computing and outer orientation parameter; By light beam method of adjustment, all known or unknown parameters are carried out to global optimization correction, repeatedly optimize and estimate, thereby finally draw the measurement result of the accurate three-dimensional reconstruction of best interior orientation parameter, outer orientation parameter and multiple measurement points of the demarcation of digital camera.

Further, for the technique effect of the clearer explanation embodiment of the present invention, can also obtain by the maximum length of the measuring error of the measuring error of the embodiment of the present invention, reference instrument, measurement field the measuring accuracy value of the embodiment of the present invention.Particularly, in order to assess the relative accuracy in the embodiment of the present invention, can pass through following equation (1-6):

$\mathrm{Es}=1000000*\sqrt{{\mathrm{RMS}}^2-{\mathrm{Sta}}^2}/\mathrm{MDL}---(1-6)$

In above formula (1-6), Es represents the precision in the embodiment of the present invention, and RMS represents the measuring error of the embodiment of the present invention, and Sta represents the measuring error of reference instrument, and MDL represents the maximum length of measurement field.

Fig. 3 is the schematic diagram that in the embodiment of the present invention, multiple measurement points is carried out three-dimensional reconstruction, the schematic diagram that the reconstructed results that Fig. 4 is the embodiment of the present invention is compared with the reconstructed results of reference instrument; For what be illustrated more clearly in the embodiment of the present invention, multiple measurement points are carried out to the technique effect of three-dimensional reconstruction, the embodiment of the present invention adopts light beam method of adjustment to carry out three-dimensional reconstruction to 90 measurement points that are arranged in the measurand in measurement field, rebuild effect as shown in Figure 3, compared with the measurement result of the three-dimensional coordinate of reconstruction and V-STARS system as shown in Figure 4.

As shown in Figure 3, " circle marker " represents the three-dimensional reconstruction effect of reference instrument (embodiment of the present invention adopts the U.S. V-STARS of GSI company Digital Photogrammetric System), and " target cross " represents the three-dimensional reconstruction effect of the embodiment of the present invention; The measuring error of the relative V-STARS system of measuring error of the 3 d space coordinate to 90 measurement points is added up, and statistics is as shown in table 1:

The measuring error of the relative V-STARS system of measuring error in table 1 embodiment of the present invention

By the error of 90 measurement points is added up, and the measuring error of the measuring error in the embodiment of the present invention and V-STARS system is compared, can find out that according to above-mentioned table 1 measuring error of the relative V-STARS system of the embodiment of the present invention is RMS=0.1750mm; Further, the maximum measurement range of the measurement field of the embodiment of the present invention is MDL=5000mm, V-STARS system is that 5000mm, measuring error is Sta=0.042mm in the maximum length of measurement field, therefore can show that according to equation (1-6) the relative accuracy error of the embodiment of the present invention is 3/100000ths.

One of ordinary skill in the art will appreciate that: all or part of step that realizes above-described embodiment can complete by the relevant hardware of programmed instruction, aforesaid program can be stored in a computer read/write memory medium, this program, in the time carrying out, is carried out the step that comprises said method embodiment; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CDs.

Finally it should be noted that: above embodiment only, in order to technical scheme of the present invention to be described, is not intended to limit; Although the present invention is had been described in detail with reference to previous embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record aforementioned each embodiment is modified, or part technical characterictic is wherein equal to replacement; And these amendments or replacement do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (6)

1. the three-dimensional rebuilding method based on light beam adjustment, is characterized in that, described method comprises:

At controlling filed, demarcate the interior orientation parameter of digital camera by the first iterative process;

In measurement field, using described interior orientation parameter as given value, demarcate the outer orientation parameter of described digital camera by thering is space resection's method of secondary iteration process;

Using described interior orientation parameter and described outer orientation parameter as initial value, using the picpointed coordinate of multiple measurement points of the multiple reference mark in described measurement field and described measurement field as observed reading, obtain the three dimensional space coordinate value of described multiple measurement points by the light beam method of adjustment based on collinear equation;

Wherein, before the step of described interior orientation parameter of demarcating digital camera by the first iterative process, also comprise:

The first coordinate axis and second coordinate axis of the image coordinate system of digital camera are set to vertical relation, make camera model meet orthogonality restriction condition;

Pel spacing between the axial direction of axial and described second coordinate axis of described first coordinate axis of described digital camera is set to unanimously, and by identical with the ratio scale of described the second coordinate axis described the first coordinate axis, make described camera model meet described engineer's scale uniform condition;

Wherein, the step of the described interior orientation parameter by the first iterative process demarcation digital camera comprises:

The multiple coded targets and the corresponding volume coordinate of described coded target that are arranged in described controlling filed are demarcated to the interior orientation parameter initial value of described camera model by direct linear transformation's method;

Known quantity using described interior orientation parameter initial value and described camera model as direct linear transformation's method, the interior orientation parameter of demarcating described digital camera by the first iterative process.

2. method according to claim 1, is characterized in that, the step of described the first iterative process comprises:

Described camera model is carried out to the computing of linearization adjustment, obtain the first error equation matrix corresponding to described camera model;

Determine whether to stop described adjustment computing by the Criterion of Iterative in described the first error equation matrix.

3. method according to claim 1, is characterized in that, described step of demarcating the outer orientation parameter of described digital camera by having space resection's method of secondary iteration process comprises:

Obtain the multiple image in the described measurement field of setting quantity different azimuth by described digital camera, on described multiple image, include the picture point at multiple reference mark and the picture point of multiple measurement points;

Pass through the picture point error equation of the corresponding collinearity condition equation of secondary iteration Procedure Acquisition according to multiple picpointed coordinates at corresponding multiple reference mark on the three dimensional space coordinate at described multiple reference mark and described multiple image;

Obtain the outer orientation parameter of described camera model by described picture point error equation.

4. method according to claim 3, is characterized in that, the step of described secondary iteration process comprises:

By in the picture point normal equation of picture point error equation described in the outer orientation parameter initial value substitution of described digital camera, calculate the picpointed coordinate at multiple reference mark;

Described in initial value substitution using the picpointed coordinate at described multiple reference mark as iteration again, in picture point normal equation, carry out iteration, until described picture point error equation meets convergence criterion, the root-mean-square error of the deviation that described convergence criterion is described picpointed coordinate is less than the second predetermined threshold value.

5. method according to claim 3, is characterized in that, the described step of obtaining the three dimensional space coordinate value of described multiple measurement points by the light beam error compensation method based on collinear equation comprises:

Using described interior orientation parameter as internal reference initial value, described outer orientation parameter, as outer ginseng initial value, is calculated the initial three dimensional space coordinate of described multiple measurement points by multi-disc space intersection method;

Obtain the second error equation matrix of light beam adjustment according to picture point error equation;

Initial three dimensional space coordinate by described initial value and described multiple measurement points carries out adjustment processing to described the second error equation matrix, determine whether to stop described adjustment computing according to described observed reading, and obtain the three dimensional space coordinate value of described multiple measurement points.

6. method according to claim 5, is characterized in that, describedly after obtaining the three dimensional space coordinate value of described multiple measurement points, also comprises:

Using the three dimensional space coordinate value of described multiple measurement points as observed reading, by described adjustment computing, described interior orientation parameter and described outer orientation parameter are optimized.