CN105181109A - Wire ice-shedding skip trajectory binocular measurement method - Google Patents

Abstract

The invention relates to a wire ice-shedding skip trajectory binocular measurement method, which comprises the steps of (1) system calibration: erecting a binocular measurement device at a location parallel to a to-be-measured lead, connecting the measurement device with a host computer to form a measurement system in the two-camera-based measurement manner, and acquiring the internal parameters and the structural parameters of the measurement system through the computation of the host computer; (2) picture shooting: continuously shooting a ice-shedding lead by means of the measurement device in the look-up gesture; (3) data processing: conducting the data processing operation by means of the host computer, completing the stereo matching process of the two cameras with lead picture matching points and calculating the three-dimensional coordinates of the matching points; (4) trajectory displaying: conducting the coordinate transformation for the calculation result of the data processing operation, and displaying the lateral motion trajectory, the longitudinal motion trajectory and the axis motion trajectory of the ice-shedding lead. According to the technical scheme of the invention, the method integrates the binocular vision technology, the digital image processing technology, the computer software technology and the like, thus being advanced in technology.

Description

A kind of wire ice-shedding track binocular measuring method

Technical field

The invention belongs to transmission route survey apparatus field, especially a kind of wire ice-shedding track binocular measuring method.

Background technology

In the disaster that transmission system faces, wire ice-shedding phenomenon is the most common, and the safety of transmission system in its serious threat.In recent years, problem is deiced about wire, carry out large quantifier elimination both at home and abroad, be mostly to adopt the method for numerical simulation to calculate the Dynamic tension wire jump height deiced in process, vertical load, steel tower rod member and conductor spacer internal force and basic moment of flexure etc.Small part scholar deices wire and has carried out simulation experiment study, has carried out actual measurement to contiguous items.For the measurement of the data such as tension force, insulator chain axle power, the measurement difficulty of wire ice-shedding track is comparatively large, at present, not yet has document to provide the ripe scheme of wire ice-shedding trajectory measurement.

At present, the maturation method of the wire ice-shedding trajectory measurement of integrated structure parameter is not also had.In order to solve the problem of current wire ice-shedding trajectory measurement difficulty, the structure of the present invention to biocular systems designs, provide the optimization acquiring method of structural parameters, adopt the polar curve characteristic crossing with wire to carry out Stereo matching, realize the noncontact of wire ice-shedding track, accurate, real-time measurement.

Summary of the invention

The object of the invention is to overcome prior art deficiency, a kind of wire ice-shedding track binocular measuring method that solve current wire ice-shedding trajectory measurement difficulty, that consider structural parameters is provided.

The technical solution used in the present invention is:

Consider a wire ice-shedding track binocular measuring method for structural parameters, the steps include:

(1) system calibrating: measurement mechanism is erected at the position parallel with wire to be measured, measurement mechanism adopts binocular, i.e. the mode of two camera measurements, is connected to form measuring system by measurement mechanism and host computer, calculated by host computer, obtain intrinsic parameter and the structural parameters of measuring system;

(2) picture shooting: measurement mechanism adopts to be looked up attitude and take continuously and deice wire;

(3) data processing: carry out data processing by host computer, the three-dimensional coordinate of the Stereo matching and match point that complete two cameras and wire images match point calculates;

(4) show track: coordinate transform is carried out to the result of calculation after data processing, demonstrate and deice wire transverse movement track, lengthwise movement track and axes movement locus.

And, the (1) described measurement mechanism of step comprises support, left camera, right camera, data line, support is " T " font, both sides, top are symmetrically installed with a camera respectively, the camera on the left side is left camera, the right camera be right camera, adjust two camera positions, make the line between two cameras perpendicular to wire direction and between two cameras between mid point be positioned at below wire.

And the (1) described measuring system intrinsic parameter of step adopts plane checkerboard pattern method to solve, and scaling board used is of a size of 600mm*400mm, lattice point size 50mm*50mm, and demarcating picture is 20 width.

And the computing method of the structural parameters of the measuring system that step is (1) described are:

If P is an angle point on scaling board, so in left and right camera coordinates system, P point meets:

P _c1＝R _lP _w+T _l(1)

P _c2＝R _rP _w+T _r(2)

P _c2＝R'P _c1+T'(3)

P _c1＝R”P _c2+T”(4)

Easily obtained by formula (1), (2), (3)

R'＝R _rR _l ^-1(5)

T'＝T _r-R'T _l(6)

In like manner, can be obtained by formula (1), (2), (4)

R"＝R _lR _r ^-1(7)

T"＝T _l-R"T _r(8)

Adopt less scaling board, be of a size of 300mm*300mm, lattice point size 50mm*50mm, take several and demarcate picture, consider re-projection error function:

$\begin{array}{c}E=\underset{i}{\Σ}\[\underset{j}{\Σ}{||P_{lij}-\widetilde{P_{lij}}(R_{li},T_{li})||}^2+\\ \underset{j}{\Σ}{||P_{rij}-\widetilde{P_{rij}}(R_{ri},T_{ri})||}^2\]\end{array}---\left(9\right)$

In formula, P _lijand P _rijrepresent jth the angle point of P point in the i-th width demarcation picture captured by the camera of left and right respectively. with represent the jth subpoint calculated by the rotation matrix of camera internal reference, the i-th width demarcation picture and translation vector in left and right camera, bring formula (5) (6) formula into formula (9), can obtain:

$\begin{array}{c}{E}_1=\underset{i}{\Σ}\[\underset{j}{\Σ}{||P_{lij}-\widetilde{P_{lij}}(R_{li},T_{li})||}^2+\\ \underset{j}{\Σ}{||p_{rij}-\widetilde{p_{rij}}(R_{li},T_{li},R^{\′},T^{\′})||}^2\]\end{array}---\left(10\right)$

Utilize LM (Levenberg-Marquard) optimized algorithm to solve above-mentioned equation, draw the exact value of R' and T'.In above-mentioned solution procedure, the initial value of R' and T' can select certain values of the structural parameters directly calculated by formula (5) (6).

Bring formula (7) (8) formula into formula (9), obtain:

$\begin{array}{c}{E}_2=\underset{i}{\Σ}\[\underset{j}{\Σ}{||p_{lij}-\widetilde{p_{lij}}(R_{ri},T_{ri},R^{\′\′},T^{\′\′})||}^2+\\ \underset{j}{\Σ}{||p_{rij}-\widetilde{p_{rij}}(R_{ri},T_{ri})||}^2\]\end{array}---\left(11\right)$

Relatively solve rear re-projection error E ₁and E ₂size, get the value of the structural parameters corresponding to smaller value.

And the solid matching method of two cameras that step is (3) described and wire images match point is:

First, utilize greyscale transformation, binaryzation, morphologic filtering image processing method from the wire image captured by left and right two cameras, extract wire, and least square fitting is carried out to wire center line, then, gets certain some p on left camera wire matching center line _l, its pixel homogeneous coordinates are m _l, then its polar curve on right image is l _r=Fm _l, its match point corresponding on right image is l _rwith the intersection point p of right image upper conductor matching center line _r, wherein, matrix based on F, sets world coordinate system in the present system and overlaps with left camera coordinates system, then antisymmetric matrix.

And the three-dimensional coordinate computing method of step (3) described match point are:

Set up world coordinate system O respectively _wx _wy _wz _w, be positioned at the camera coordinates system O of camera photocentre _c1x _c1y _c1z _c1and O _c2x _c2y _c2z _c2, be positioned at principal point for camera (u ₀, v ₀) image physical coordinates system O ₁x ₁y ₁and O ₂x ₂y ₂, be positioned at the image pixel coordinates system O in the imaging plane upper left corner _uv1u ₁v ₁and O _uv2u ₂v ₂.By camera projection theory, can obtain:

$\begin{array}{c}s=\left[\begin{array}{c}u\\ v\\ 1\end{array}\right]=\left[\begin{array}{cccc}{f}_x& 0& u_0& 0\\ 0& f_y& v_0& 0\\ 0& 0& 1& 0\end{array}\right]\left[\begin{array}{cc}R& T\\ 0^T& 1\end{array}\right]\left[\begin{array}{c}{x}_w\\ y_w\\ z_w\\ 1\end{array}\right]\\ =M_1{M}_2{P}_w={\mathrm{MP}}_w\end{array}---\left(12\right)$

In above formula, s is scale factor, f _xwith f _ybe respectively the scale factor in both direction, R and T is respectively world coordinates

And in the three-dimensional coordinate computation process of step (3) described match point, carry out subpoint compensation calculation to the radial distortion of camera lens and tangential distortion, its computing method are:

In camera coordinates system, after lens distortion, the coordinate of subpoint is:

$\begin{array}{c}\left[\begin{array}{c}{\stackrel{-}{x}}_c\\ {\stackrel{-}{y}}_c\end{array}\right]=(1+p_1{r}^2+p_2{r}^4+p_5{r}^6)\left[\begin{array}{c}{x}_c^{\′}\\ y_c^{\′}\end{array}\right]+\\ \left[\begin{array}{c}2p_3{x}_c^{\′}{y}_c^{\′}+p_4(r^2+2x_c^{\′2})\\ p_3(r^2+2y_c^{\′2})+2p_4{x}_c^{\′}{y}_c^{\′}\end{array}\right]\end{array}---\left(13\right)$

In above formula, P=[p ₁p ₂p ₃p ₄p ₅], be distortion factor, $\left[\begin{array}{c}{x}_c^{\′}\\ y_c^{\′}\end{array}\right]=\left[\begin{array}{c}{x}_c/z_c\\ y_c/z_c\end{array}\right],r^2=x_c^{\′2}+y_c^{\′2}.$

For arbitrary measurement point, two, left and right imaging plane obtains two image pixel coordinates (u _l, v _l) and (u _r, v _r), bring formula (12) into, can system of equations be obtained:

$\left\{\begin{array}{c}(u_l{M}_{31}^1-M_{11}^1)x_w+(u_l{M}_{32}^1-M_{12}^1)y_w\\ +(u_l{M}_{33}^1-M_{13}^1)z_w=M_{14}^1-u_l{M}_{34}^1\\ (v_l{M}_{31}^1-M_{21}^1)x_w+(v_l{M}_{32}^1-M_{22}^1)y_w\\ +(v_l{M}_{33}^1-M_{23}^1)z_w=M_{24}^1-v_l{M}_{34}^1\\ (u_r{M}_{31}^2-M_{11}^2)x_w+(u_r{M}_{32}^2-M_{12}^2)y_w\\ +(u_r{M}_{33}^2-M_{13}^2)z_w=M_{14}^2-u_r{M}_{34}^2\\ (v_r{M}_{31}^2-M_{21}^2)x_w+(v_r{M}_{32}^2-M_{22}^2)y_w\\ +(v_r{M}_{33}^2-M_{23}^2)z_w=M_{24}^2-v_r{M}_{34}^2\end{array}\right.---\left(14\right)$

By above-mentioned system of equations referred to as AP' _w=B, utilizes least square method to solve, and can obtain P' _w=(A ^ta ^-1) A ^tb.

And, step (4) in coordinate transformation method be:

X is positioned at by measuring the wire ice-shedding track obtained _woZ _w(X _c1oZ _c1) in face, in order to weigh wire motion conditions in the horizontal and vertical directions, define new coordinate system X' _woZ' _w, wherein OX' _waxle is parallel to baseline, OZ' _waxle is perpendicular to baseline, and transformation for mula is therebetween $\left[\begin{array}{c}{X}_w^{\′}\\ Z_w^{\′}\end{array}\right]=\left[\begin{array}{cc}sin\mathrm{\α}& cos\mathrm{\α}\\ -cos\mathrm{\α}& sin\mathrm{\α}\end{array}\right]\left[\begin{array}{c}{X}_w\\ Z_w\end{array}\right].$

Advantage of the present invention and good effect are:

Wire ice-shedding track binocular measuring method provided by the invention can realize to wire deice ice-shedding track noncontact, measure in real time, accurately, the wire that can be applied to from now on deices in experiment and path monitoring, obtains wire ice-shedding raw data.

Wire ice-shedding track binocular measuring method provided by the invention utilizes re-projection error function to be optimized structural parameters, effective raising structural parameters solving precision, by comparing the re-projection error value of left and right camera in biocular systems, realize values of the structural parameters further preferred.; On the other hand, native system utilizes wire center line to have the characteristic of intersecting with biocular systems polar curve, realize match point to extract fast and accurately, matching is carried out to wire center line, effective raising matching precision, combine binocular vision technology, digital image processing techniques, computer software technology etc., advanced technology.

Accompanying drawing explanation

Fig. 1 is the structural representation of measurement mechanism in the present invention;

Fig. 2 is the schematic diagram that in the present invention, three-dimensional coordinate calculates.

Embodiment

Below by accompanying drawing, the invention will be further described in conjunction with specific embodiments, and following examples are descriptive, is not determinate, can not limit protection scope of the present invention with this.

Consider a wire ice-shedding track binocular measuring method for structural parameters, the steps include:

(1) system calibrating: measurement mechanism is erected at the position parallel with wire to be measured, measurement mechanism adopts binocular, the i.e. mode of two camera measurements, see Fig. 1, measurement mechanism and host computer are connected to form measuring system, calculated by host computer, obtain intrinsic parameter and the structural parameters of measuring system, measurement mechanism comprises support, left camera, right camera, data line, support is " T " font, both sides, top are symmetrically installed with a camera respectively, the camera on the left side is left camera, adjust two camera positions, make the line between two cameras perpendicular to wire direction and between two cameras between mid point be positioned at below wire,

See Fig. 2, the measurement range of selected wire ice-shedding height is 4.0m;

The value selecting baseline distance B is 1.5m;

In order to ensure that the value of horizontal projection angle β in measuring process can not be excessive, the value of horizontal projection angle α is set to 67 °;

For increasing the focal length of camera, under the requirement of satisfied measurement range, reduce the field angle to 28 ° of camera, now visual range is about 1.0-5.0m;

In wire ice-shedding trajectory measurement process, adopt and look up attitude, and adjustment System position level and symmetrical about wire, ensure structural symmetry in measuring process.

The (1) described measuring system intrinsic parameter of step adopts plane checkerboard pattern method to solve, and scaling board used is of a size of 600mm*400mm, lattice point size 50mm*50mm, and demarcating picture is 20 width, and the computing method of the structural parameters of measuring system are:

If P is an angle point on scaling board, so in left and right camera coordinates system, P point meets:

P _c1＝R _lP _w+T _l(1)

P _c2＝R _rP _w+T _r(2)

P _c2＝R'P _c1+T'(3)

P _c1＝R”P _c2+T”(4)

Easily obtained by formula (1), (2), (3)

R'＝R _rR _l ^-1(5)

T'＝T _r-R'T _l(6)

In like manner, can be obtained by formula (1), (2), (4)

R"＝R _lR _r ^-1(7)

T"＝T _l-R"T _r(8)

Adopt less scaling board, be of a size of 300mm*300mm, lattice point size 50mm*50mm, take several and demarcate picture, consider re-projection error function:

$\begin{array}{c}E=\underset{i}{\Σ}\[\underset{j}{\Σ}{||P_{lij}-\widetilde{P_{lij}}(R_{li},T_{li})||}^2+\\ \underset{j}{\Σ}{||P_{rij}-\widetilde{P_{rij}}(R_{ri},T_{ri})||}^2\]\end{array}---\left(9\right)$

In formula, P _lijand P _rijrepresent jth the angle point of P point in the i-th width demarcation picture captured by the camera of left and right respectively. with represent the jth subpoint calculated by the rotation matrix of camera internal reference, the i-th width demarcation picture and translation vector in left and right camera, bring formula (5) (6) formula into formula (9), can obtain:

$\begin{array}{c}{E}_1=\underset{i}{\Σ}\[\underset{j}{\Σ}{||P_{lij}-\widetilde{P_{lij}}(R_{li},T_{li})||}^2+\\ \underset{j}{\Σ}{||p_{rij}-\widetilde{p_{rij}}(R_{li},T_{li},R^{\′},T^{\′})||}^2\]\end{array}---\left(10\right)$

Utilize LM (Levenberg-Marquard) optimized algorithm to solve above-mentioned equation, draw the exact value of R' and T'.In above-mentioned solution procedure, the initial value of R' and T' can select certain values of the structural parameters directly calculated by formula (5) (6).

Bring formula (7) (8) formula into formula (9), obtain:

$\begin{array}{c}{E}_2=\underset{i}{\Σ}\[\underset{j}{\Σ}{||p_{lij}-\widetilde{p_{lij}}(R_{ri},T_{ri},R^{\′\′},T^{\′\′})||}^2+\\ \underset{j}{\Σ}{||p_{rij}-\widetilde{p_{rij}}(R_{ri},T_{ri})||}^2\]\end{array}---\left(11\right)$

Relatively solve rear re-projection error E ₁and E ₂size, get the value of the structural parameters corresponding to smaller value.

(2) picture shooting: measurement mechanism adopts to be looked up attitude and take continuously and deice wire.

(3) data processing: carry out data processing by host computer, the three-dimensional coordinate of the Stereo matching and match point that complete two camera wire images match points calculates, and the solid matching method of two cameras and wire images match point is:

First, utilize greyscale transformation, binaryzation, morphologic filtering image processing method from the wire image captured by left and right two cameras, extract wire, and least square fitting is carried out to wire center line, then, gets certain some p on left camera wire matching center line _l, its pixel homogeneous coordinates are m _l, then its polar curve on right image is l _r=Fm _l, its on right image corresponding t ₂=T', when getting R " and T ", R ₂=(R ") ^-1, T ₂=-(R ") ^-1t ", basis matrix [T ₂] _×for T ₂antisymmetric matrix.

The three-dimensional coordinate computing method of described match point are:

Set up world coordinate system O respectively _wx _wy _wz _w, be positioned at the camera coordinates system O of camera photocentre _c1x _c1y _c1z _c1and O _c2x _c2y _c2z _c2, be positioned at principal point for camera (u ₀, v ₀) image physical coordinates system O ₁x ₁y ₁and O ₂x ₂y ₂, be positioned at the image pixel coordinates system O in the imaging plane upper left corner _uv1u ₁v ₁and O _uv2u ₂v ₂.By camera projection theory, can obtain:

$\begin{array}{c}s=\left[\begin{array}{c}u\\ v\\ 1\end{array}\right]=\left[\begin{array}{cccc}{f}_x& 0& u_0& 0\\ 0& f_y& v_0& 0\\ 0& 0& 1& 0\end{array}\right]\left[\begin{array}{cc}R& T\\ 0^T& 1\end{array}\right]\left[\begin{array}{c}{x}_w\\ y_w\\ z_w\\ 1\end{array}\right]\\ =M_1{M}_2{P}_w={\mathrm{MP}}_w\end{array}---\left(12\right)$

be respectively the Intrinsic Matrix of camera, outer parameter matrix and projection matrix.

In the three-dimensional coordinate computation process of step (3) described match point, carry out subpoint compensation calculation to the radial distortion of camera lens and tangential distortion, its computing method are:

In camera coordinates system, after lens distortion, the coordinate of subpoint is:

$\begin{array}{c}\left[\begin{array}{c}{\stackrel{-}{x}}_c\\ {\stackrel{-}{y}}_c\end{array}\right]=(1+p_1{r}^2+p_2{r}^4+p_5{r}^6)\left[\begin{array}{c}{x}_c^{\′}\\ y_c^{\′}\end{array}\right]+\\ \left[\begin{array}{c}2p_3{x}_c^{\′}{y}_c^{\′}+p_4(r^2+2x_c^{\′2})\\ p_3(r^2+2y_c^{\′2})+2p_4{x}_c^{\′}{y}_c^{\′}\end{array}\right]\end{array}---\left(13\right)$

In above formula, P=[p ₁p ₂p ₃p ₄p ₅], be distortion factor, $\left[\begin{array}{c}{x}_c^{\′}\\ y_c^{\′}\end{array}\right]=\left[\begin{array}{c}{x}_c/z_c\\ y_c/z_c\end{array}\right],r^2=x_c^{\′2}+y_c^{\′2}.$

For arbitrary measurement point, two image pixel coordinates (u can be obtained on the imaging plane of two, left and right _l, v _l) and (u _r, v _r), bring formula (12) into, can system of equations be obtained:

$\left\{\begin{array}{c}(u_l{M}_{31}^1-M_{11}^1)x_w+(u_l{M}_{32}^1-M_{12}^1)y_w\\ +(u_l{M}_{33}^1-M_{13}^1)z_w=M_{14}^1-u_l{M}_{34}^1\\ (v_l{M}_{31}^1-M_{21}^1)x_w+(v_l{M}_{32}^1-M_{22}^1)y_w\\ +(v_l{M}_{33}^1-M_{23}^1)z_w=M_{24}^1-v_l{M}_{34}^1\\ (u_r{M}_{31}^2-M_{11}^2)x_w+(u_r{M}_{32}^2-M_{12}^2)y_w\\ +(u_r{M}_{33}^2-M_{13}^2)z_w=M_{14}^2-u_r{M}_{34}^2\\ (v_r{M}_{31}^2-M_{21}^2)x_w+(v_r{M}_{32}^2-M_{22}^2)y_w\\ +(v_r{M}_{33}^2-M_{23}^2)z_w=M_{24}^2-v_r{M}_{34}^2\end{array}\right.---\left(14\right)$

By above-mentioned system of equations referred to as AP' _w=B, utilizes least square method to solve, and can obtain P' _w=(A ^ta ^-1) A ^tb.

(4) show track: carry out coordinate transform to the result of calculation after data processing, demonstrate and deice wire transverse movement track, lengthwise movement track and axes movement locus, coordinate transformation method is:

X is positioned at by measuring the wire ice-shedding track obtained _woZ _w(X _c1oZ _c1) in face, in order to weigh wire motion conditions in the horizontal and vertical directions, define new coordinate system X' _woZ' _w, wherein OX' _waxle is parallel to baseline, OZ' _waxle is perpendicular to baseline, and transformation for mula is therebetween $\left[\begin{array}{c}{X}_w^{\′}\\ Z_w^{\′}\end{array}\right]=\left[\begin{array}{cc}sin\mathrm{\α}& cos\mathrm{\α}\\ -cos\mathrm{\α}& sin\mathrm{\α}\end{array}\right]\left[\begin{array}{c}{X}_w\\ Z_w\end{array}\right].$

Although disclose embodiments of the invention and accompanying drawing for the purpose of illustration, but it will be appreciated by those skilled in the art that: in the spirit and scope not departing from the present invention and claims, various replacement, change and amendment are all possible, therefore, scope of the present invention is not limited to the content disclosed in embodiment and accompanying drawing.

Claims (8)

1. consider a wire ice-shedding track binocular measuring method for structural parameters, the steps include:

(1) system calibrating: measurement mechanism is erected at the position parallel with wire to be measured, measurement mechanism adopts binocular, i.e. the mode of two camera measurements, is connected to form measuring system by measurement mechanism and host computer, calculated by host computer, obtain intrinsic parameter and the structural parameters of measuring system;

(2) picture shooting: measurement mechanism adopts to be looked up attitude and take continuously and deice wire;

(3) data processing: carry out data processing by host computer, the three-dimensional coordinate of the Stereo matching and match point that complete two cameras and wire images match point calculates;

(4) show track: coordinate transform is carried out to the result of calculation after data processing, demonstrate and deice wire transverse movement track, lengthwise movement track and axes movement locus.

2. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 1, it is characterized in that: the (1) described measurement mechanism of step comprises support, left camera, right camera, data line, support is " T " font, both sides, top are symmetrically installed with a camera respectively, the camera on the left side is left camera, the right camera be right camera, adjust two camera positions, make the line between two cameras perpendicular to wire direction and between two cameras between mid point be positioned at below wire.

3. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 2, it is characterized in that: the (1) described measuring system intrinsic parameter of step adopts plane checkerboard pattern method to solve, scaling board used is of a size of 600mm*400mm, lattice point size 50mm*50mm, demarcating picture is 20 width.

4. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 3, is characterized in that: the computing method of the structural parameters of the measuring system that step is (1) described are:

If P is an angle point on scaling board, so in left and right camera coordinates system, P point meets:

P _c1＝R _lP _w+T _l(1)

P _c2＝R _rP _w+T _r(2)

P _c2＝R'P _c1+T'(3)

P _c1＝R”P _c2+T”(4)

Easily obtained by formula (1), (2), (3)

R'＝R _rR _l ^-1(5)

T'＝T _r-R'T _l(6)

In like manner, can be obtained by formula (1), (2), (4)

R"＝R _lR _r ^-1(7)

T"＝T _l-R"T _r(8)

Adopt less scaling board, be of a size of 300mm*300mm, lattice point size 50mm*50mm, take several and demarcate picture, consider re-projection error function:

$\begin{array}{c}E=\underset{i}{\Σ}\[\underset{j}{\Σ}{||P_{lij}-\widetilde{P_{lij}}(R_{li},T_{li})||}^2+\\ \underset{j}{\Σ}{||P_{rij}-\widetilde{P_{rij}}(R_{ri},T_{ri})||}^2\]\end{array}---\left(9\right)$

In formula, P _lijand P _rijrepresent jth the angle point of P point in the i-th width demarcation picture captured by the camera of left and right respectively. with represent the jth subpoint calculated by the rotation matrix of camera internal reference, the i-th width demarcation picture and translation vector in left and right camera, bring formula (5) (6) formula into formula (9), can obtain:

$\begin{array}{c}{E}_1=\underset{i}{\Σ}\[\underset{j}{\Σ}{||P_{lij}-\widetilde{P_{lij}}(R_{li},T_{li})||}^2+\\ \underset{j}{\Σ}{||p_{rij}-\widetilde{p_{rij}}(R_{li},T_{li},R^{\′},T^{\′})||}^2\]\end{array}---\left(10\right)$

Utilize LM (Levenberg-Marquard) optimized algorithm to solve above-mentioned equation, draw the exact value of R' and T'.In above-mentioned solution procedure, the initial value of R' and T' can select certain values of the structural parameters directly calculated by formula (5) (6).

Bring formula (7) (8) formula into formula (9), obtain:

$\begin{array}{c}{E}_2=\underset{i}{\Σ}\[\underset{j}{\Σ}{||p_{lij}-\widetilde{p_{lij}}(R_{ri},T_{ri},R^{\′\′},T^{\′\′})||}^2+\\ \underset{j}{\Σ}{||p_{rij}-\widetilde{p_{rij}}(R_{ri},T_{ri})||}^2\]\end{array}---\left(11\right)$

Relatively solve rear re-projection error E ₁and E ₂size, get the value of the structural parameters corresponding to smaller value.

5. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 4, is characterized in that: the solid matching method of two cameras that step is (3) described and wire images match point is:

First, utilize greyscale transformation, binaryzation, morphologic filtering image processing method from the wire image captured by left and right two cameras, extract wire, and least square fitting is carried out to wire center line, then, gets certain some p on left camera wire matching center line _l, its pixel homogeneous coordinates are m _l, then its polar curve on right image is l _r=Fm _l, its match point corresponding on right image is l _rwith the intersection point p of right image upper conductor matching center line _r, wherein, matrix based on F, sets world coordinate system in the present system and overlaps with left camera coordinates system, then $M_1=\left[\begin{array}{cccc}{f}_x& 0& u_0& 0\\ 0& f_y& v_0& 0\\ 0& 0& 1& 0\end{array}\right],$ T ₁=[000] ^t, when getting R' and T', R ₂=R', T ₂=T', when getting R " and T ", R ₂=(R ") ^-1, T ₂=-(R ") ^-1t ", basis matrix [T ₂] _×for T ₂antisymmetric matrix.

6. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 1, is characterized in that: the three-dimensional coordinate computing method of step (3) described match point are:

Set up world coordinate system O respectively _wx _wy _wz _w, be positioned at the camera coordinates system O of camera photocentre _c1x _c1y _c1z _c1and O _c2x _c2y _c2z _c2, be positioned at principal point for camera (u ₀, v ₀) image physical coordinates system O ₁x ₁y ₁and O ₂x ₂y ₂, be positioned at the image pixel coordinates system O in the imaging plane upper left corner _uv1u ₁v ₁and O _uv2u ₂v ₂.By camera projection theory, can obtain:

$\begin{array}{c}s=\left[\begin{array}{c}u\\ v\\ 1\end{array}\right]=\left[\begin{array}{cccc}{f}_x& 0& u_0& 0\\ 0& f_y& v_0& 0\\ 0& 0& 1& 0\end{array}\right]\left[\begin{array}{cc}R& T\\ 0^T& 1\end{array}\right]\left[\begin{array}{c}{x}_w\\ y_w\\ z_w\\ 1\end{array}\right]\\ =M_1{M}_2{P}_w={\mathrm{MP}}_w\end{array}---\left(12\right)$

In above formula, s is scale factor, f _xwith f _ybe respectively the scale factor in both direction, R and T is respectively rotation matrix and the translation vector that world coordinates is tied to camera coordinates system, $R_1=\left[\begin{array}{ccc}1& 0& 0\\ 0& 1& 0\\ 0& 0& 1\end{array}\right],$ and M=M ₁m ₂be respectively the Intrinsic Matrix of camera, outer parameter matrix and projection matrix.

7. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 6, it is characterized in that: in the three-dimensional coordinate computation process of step (3) described match point, carry out subpoint compensation calculation to the radial distortion of camera lens and tangential distortion, its computing method are:

In camera coordinates system, after lens distortion, the coordinate of subpoint is:

$\begin{array}{c}\left[\begin{array}{c}{\stackrel{-}{x}}_c\\ {\stackrel{-}{y}}_c\end{array}\right]=(1+p_1{r}^2+p_2{r}^4+p_5{r}^6)\left[\begin{array}{c}{x}_c^{\′}\\ y_c^{\′}\end{array}\right]+\\ \left[\begin{array}{c}2p_3{x}_c^{\′}{y}_c^{\′}+p_4(r^2+2x_c^{\′2})\\ p_3(r^2+2y_c^{\′2})+2p_4{x}_c^{\′}{y}_c^{\′}\end{array}\right]\end{array}---\left(13\right)$

In above formula, P=[p ₁p ₂p ₃p ₄p ₅], be distortion factor, $\left[\begin{array}{c}{x}_c^{\′}\\ y_c^{\′}\end{array}\right]=\left[\begin{array}{c}{x}_c/z_c\\ y_c/z_c\end{array}\right],$ r ²＝x' _c ²+y' _c ²。

For arbitrary measurement point, two, left and right imaging plane obtains two image pixel coordinates (u _l, v _l) and (u _r, v _r), bring formula (12) into, can system of equations be obtained:

$\left\{\begin{array}{c}(u_l{M}_{31}^1-M_{11}^1)x_w+(u_l{M}_{32}^1-M_{12}^1)y_w\\ +(u_l{M}_{33}^1-M_{13}^1)z_w=M_{14}^1-u_l{M}_{34}^1\\ (v_l{M}_{31}^1-M_{21}^1)x_w+(v_l{M}_{32}^1-M_{22}^1)y_w\\ +(v_l{M}_{33}^1-M_{23}^1)z_w=M_{24}^1-v_l{M}_{34}^1\\ (u_r{M}_{31}^2-M_{11}^2)x_w+(u_r{M}_{32}^2-M_{12}^2)y_w\\ +(u_r{M}_{33}^2-M_{13}^2)z_w=M_{14}^2-u_r{M}_{34}^2\\ (v_r{M}_{31}^2-M_{21}^2)x_w+(v_r{M}_{32}^2-M_{22}^2)y_w\\ +(v_r{M}_{33}^2-M_{23}^2)z_w=M_{24}^2-v_r{M}_{34}^2\end{array}\right.---\left(14\right)$ By above-mentioned system of equations referred to as AP' _w=B, utilizes least square method to solve, and can obtain P' _w=(A ^ta ^-1) A ^tb.

8. the wire ice-shedding track binocular measuring method of consideration structural parameters according to claim 1, is characterized in that: step (4) in coordinate transformation method be:

X is positioned at by measuring the wire ice-shedding track obtained _woZ _w(X _c1oZ _c1) in face, in order to weigh wire motion conditions in the horizontal and vertical directions, define new coordinate system X' _woZ' _w, wherein OX' _waxle is parallel to baseline, OZ' _waxle is perpendicular to baseline, and transformation for mula is therebetween $\left[\begin{array}{c}{X}_w^{\′}\\ Z_w^{\′}\end{array}\right]=\left[\begin{array}{cc}sin\mathrm{\α}& cos\mathrm{\α}\\ -cos\mathrm{\α}& sin\mathrm{\α}\end{array}\right]\left[\begin{array}{c}{X}_w\\ Z_w\end{array}\right].$