CN105091748A - Measuring system for tolerance dimensions of railway vehicle - Google Patents

Abstract

The invention discloses a measuring system for tolerance dimensions of a railway vehicle. The system includes a hand-held measuring device and a three-dimensional data processing unit. The hand-held measuring device includes a DLP projector, two CCD cameras, a timing sequence control board, an image acquisition card and a data processor. A three-dimensional data processing unit interface includes a point cloud display area, a point cloud information area, an instruction editing area and a camera display area. The timing sequence control board is separately connected to the data processor, the two CCD cameras, and the DLP projector. The DLP projector is connected to the data processor. The two CCD cameras are connected via the image acquisition card and the data processor. The system is stable, efficient, and can be better fit for detecting the dimensions of the internal decorative objects of the railway vehicle, which prevents the defects of traditional methods in obtaining the tolerance dimensions of the internal decorative objects, such as low precision, slow speed, and human influence and meets the requirements in the production of modern railway vehicles.

Description

Rail vehicle tolerance dimension measuring system

Technical field

The present invention relates to a kind of rail vehicle tolerance dimension measuring system, by introducing advanced measuring method and repair means, improving packing quality in rail vehicle, belonging to rail vehicle three-dimensional measurement technical field.

Background technology

The assembling of rail vehicle inside gadget is that the important quality that rail vehicle is produced controls corner, its level of control concentrated expression rail vehicle product development and quality control level, therefore becomes the focus that rail traffic vehicles manufacturing enterprise pays close attention to.In the rail vehicle manufacturing cycle, product design, process exploitation, production phase all can produce considerable influence to interior trim fitted position.

Flourish along with track traffic industry, the manufacture requirements of all circles to rail vehicle is also more and more higher.Abroad, the Caroliner company of gondola Spanesi company, Sweden develops body of a motor car electronic measurement system and have certain advantage in measuring accuracy, operability, utilize laser, measurement that noctovisor scan technology can realize vehicle body three-dimensional dimension, meet the new demand of Modern Vehicle Repair industry to detection technique.But at home, the application of producing at vehicle body of railway vehicle of intelligent scanning measuring system is also fewer.And traditional rail vehicle interior trim measuring method is by measuring manually, measurement result is affected by human factors more, and the efficiency measured and precision etc. are all difficult to the requirement meeting the modern production cycle.

Summary of the invention

Object: in order to overcome the deficiencies in the prior art, the invention provides a kind of rail vehicle tolerance dimension measuring system, can measure rail vehicle interior trim tolerance dimension quickly and efficiently.System possesses bulk test technique automatic, can simultaneously multimetering, can the three-dimensional data of the measured automobile interior of acquisition rapidly and efficiently, thus improves efficiency of assembling, significant to the assembling of whole rail traffic vehicles.

Technical scheme: for solving the problems of the technologies described above, the technical solution used in the present invention is:

A kind of rail vehicle tolerance dimension measuring system, is characterized in that: comprise hand-held measuring equipment and three-dimensional data processing unit;

Hand-held measuring equipment, comprises DLP projector, two CCD camera, sequential control plate, image pick-up card and data processors;

Three-dimensional data processing unit interface, comprises a cloud viewing area, some cloud block of information, commard editor district and camera viewing area.

Described rail vehicle tolerance dimension measuring system, it is characterized in that: sequential control plate is connected with data processor, two CCD camera, DLP projector respectively, DLP projector is connected with data processor, and two ccd video cameras are all connected with data processor by image pick-up card;

Described rail vehicle tolerance dimension measuring system, is characterized in that: the photocentre axle of described two CCD camera and the angle of DLP projector light axle all between 20 degree to 60 degree, and keep the relative position of DLP projector and two CCD camera constant when measuring.

Described rail vehicle tolerance dimension measuring system, is characterized in that: described rail vehicle tolerance dimension measuring system is provided with the USB interface for being connected with computing machine.

A kind of rail vehicle tolerance dimension measuring method, the rail vehicle tolerance dimension measuring system described in employing, comprises the following steps:

(1) before measuring, hand-held measuring equipment and computing machine adopt USB interface to be connected, and supporting three-dimensional data processing unit is installed in a computer; During measurement, use DLP projector to project to testee the raster image that one group of light intensity is dextrorotation distribution, and use two CCD camera to take the raster image be out of shape through testee surface modulation simultaneously;

(2) absolute phase values of raster image is obtained according to phase shift algorithm and multifrequency heterodyne solution phase method; According to the systematic parameter of demarcating in advance or phase height mapping relation, calculate the three dimensional point cloud on testee surface from absolute phase values;

(3) after obtaining the cloud data of testee, select the menu item on corresponding software, calculate required automobile interior tolerance dimension, with the form display measurement result of message box.

Described rail vehicle tolerance dimension measuring method, is characterized in that: automobile interior tolerance dimension comprises: flatness, angle, length, cylinder circularity.

Beneficial effect: rail vehicle tolerance dimension measuring system provided by the invention, utilize and project to measured target object the raster image that one group of light intensity is Sine distribution, the three dimensional point cloud on testee surface is obtained by correlation computations, related algorithm process is carried out to three dimensional point cloud, obtain the data needed, wherein relate to the angle of object, flatness, length, circularity, also relate to the three-dimensional values etc. of miniature parts.Compared with the method for existing measuring vehicle interior trim tolerance, the invention has the advantages that: (1) is affected by human factors little, greatly simplified the operation steps of rail wheel dimension tolerance measurement, easy and simple to handle, testing result can show on the computer screen intuitively; (2) data that cannot collect when can collect traditional measurement method also measure local, image data efficiency is high; (3) measurement size precision is high, reproducible.

Accompanying drawing explanation

Fig. 1 is the schematic diagram of hand-held measuring equipment in the present invention;

Fig. 2 is three-dimensional data processing unit interface schematic diagram in the present invention;

Fig. 3 is measuring method of the present invention;

Fig. 4 is phase measurement consistency profiles measurement of angle schematic diagram;

Fig. 5 is that corner detection approach surveys object length schematic diagram;

Fig. 6 is observation station and axis geometric relationship figure.

In figure: DLP projector 1, CCD camera 2,3, sequential control plate 4, image pick-up card 5, data processor 6, tested rail vehicle 7.

Embodiment

Below in conjunction with accompanying drawing, the present invention is further described.

As shown in Figure 1, a kind of rail vehicle tolerance dimension measuring system, measuring system is divided into two parts, hand-held measuring equipment and three-dimensional data process generation unit.

Hand-held measuring equipment comprises: DLP projector 1, left and right CCD camera 2,3, sequential control plate 4, data processor 5 and image pick-up card 6.Wherein, the photocentre axle of two CCD camera 2,3 and the angle of DLP projector 1 photocentre axle all between 20 degree to 60 degree, and need the strict relative position of DLP projector 1 and two CCD camera that keeps constant when measuring.In Fig. 1,7 represent tested rail vehicle.

Sequential control plate 4 respectively with DLP projector 1, two CCD camera 2,3, data processor 5 is connected, and two CCD camera 2,3 are connected with data processor 5 by image pick-up card 6.

Wherein DLP projector 1 is the raster image of Sine distribution to the testee one group of light intensity that projects, the synchronizing signal that two CCD camera 2,3 are transmitted according to sequential control plate 6 carries out image acquisition, and the view data of collection is sent on data processor 5 by image pick-up card 6 and stores.

The system architecture of hand-held measuring equipment shown in Fig. 1 solves the acquisition problem of destination object three-dimensional data, and measuring equipment and computing machine adopt USB interface to be connected, and supporting three-dimensional data processing unit is installed in a computer, and computing machine is with figure video card.Accompanying drawing 2 represents three-dimensional data processing unit interface schematic diagram.

As shown in Figure 3, rail vehicle tolerance dimension measuring method, concrete steps are as follows:

(1) use DLP projector 1 to project to the tested rail vehicle 7 of testee the raster image that one group of light intensity is dextrorotation distribution when measuring, and use CCD camera 2,3 to take the raster image be out of shape through testee surface modulation simultaneously;

(2) absolute phase values of raster image is obtained according to phase shift algorithm and multifrequency heterodyne solution phase method; According to the systematic parameter of demarcating in advance or phase height mapping relation, calculate the three dimensional point cloud on testee surface from absolute phase values; (concrete implementation step is see Li Zhongwei. based on structural light three-dimensional measuring technique and systematic study [D] [D] of digital fringe projection. and Wuhan: the Central China University of Science and Technology, 2009).

(3), after obtaining the cloud data of testee, select the menu item on corresponding software, required automobile interior tolerance dimension can be calculated, as flatness, angle, length, cylinder circularity etc., with the form display measurement result of message box.Wherein relevant interior trim size calculation step is as follows:

3.1 flatnesses calculate

Flatness refers to the variation of tested real surface to its ideal plane.Flatness error is compared tested real surface and ideal plane, and line value distance is between the two flatness error value.Ignore measuring error, the cloud data of plane to be measured is considered as the True Data of tested real surface, the plane simulated is considered as ideal plane.

3.11st step adopts the proper vector estimation technique (EVE) to determine the initial parameter values of fit Plane; If treat that the equation of fit Plane is: ax+by+cz=d; Wherein: a, b, c are the unit normal vector of plane, that is: a ²+ b ²+ c ²=1, wherein: d is the distance of true origin to plane, d>=0;

3.12nd step is established and is scanned a certain plane, obtains n data point, then three-dimensional coordinate (the x of any data point _i, y _i, z _i) to the distance of this plane be:

d _i＝|ax _i+by _i+cz _i-d|(1)

Obtain best-fitting plane, then should a under conditions ²+ b ²+ c ²=1, meet:

$e={\mathrm{\Σ}}_i{d}_i^2={\mathrm{\Σ}}_i{({\mathrm{ax}}_i+{\mathrm{by}}_i+{\mathrm{cz}}_i-d)}^2\→\min ---\left(2\right)$

3.4th step obtains objective function by lagrange's method of multipliers:

$f={\mathrm{\Σ}}_i{d}_i^2-\mathrm{\λ}(a^2+b^2+c^2-1)---\left(3\right)$

Wherein λ is Lagrange's multiplier;

Ask local derviation final to d, a, b, c respectively formula (3):

$\left[\begin{array}{ccc}{\mathrm{\Σ}}_i\mathrm{\Δ}{x}_i\mathrm{\Δ}{x}_i& {\mathrm{\Σ}}_i\mathrm{\Δ}{x}_i\mathrm{\Δ}{y}_i& {\mathrm{\Σ}}_i\mathrm{\Δ}{x}_i\mathrm{\Δ}{z}_i\\ {\mathrm{\Σ}}_i\mathrm{\Δ}{x}_i\mathrm{\Δ}{y}_i& {\mathrm{\Σ}}_i\mathrm{\Δ}{y}_i\mathrm{\Δ}{y}_i& {\mathrm{\Σ}}_i\mathrm{\Δ}{y}_i\mathrm{\Δ}{z}_i\\ {\mathrm{\Σ}}_i\mathrm{\Δ}{x}_i\mathrm{\Δ}{z}_i& {\mathrm{\Σ}}_i\mathrm{\Δ}{y}_i\mathrm{\Δ}{z}_i& {\mathrm{\Σ}}_i\mathrm{\Δ}{z}_i\mathrm{\Δ}{z}_i\end{array}\right]\left[\begin{array}{c}a\\ b\\ c\end{array}\right]=\mathrm{\λ}\left[\begin{array}{c}a\\ b\\ c\end{array}\right]---\left(4\right)$

Wherein: $\mathrm{\Δ}{x}_i=x_i-\stackrel{\‾}{x},\mathrm{\Δ}{y}_i-\stackrel{\‾}{y},\mathrm{\Δ}{z}_i=z_i-\stackrel{\‾}{z};$ the mean value of cloud data three-dimensional coordinate.

3.14th step solves above-mentioned matrix, obtains initial value a, b, c of plane parameter.

3.15th step calculates the standard deviation δ arriving a little the distance of fit Plane, and rejects fit Plane using 2 δ as threshold value apart from excessive interference noise point;

31.6th step utilizes remaining significant figure strong point to recalculate parameter a, b, c of fit Plane; Repeat above step until arrive a little your distance of fit Plane be all less than the threshold value that this time calculate, finally obtain best fit Plane, as the ideal plane of benchmark;

The coordinate of each available point is brought in plane equation by the 3.17th step, judges that this point is positioned at plane upper side or downside; Calculate each point in plane to be measured, to the distance d (computing method are see formula (1)) of ideal plane, to try to achieve a little to the maximal value d of plan range _max, be the flatness of plane to be measured.

3.2 measurement of angle

Measurement of angle, namely measures the angle between two planes.Phase measurement consistency profiles takes measurement of an angle principle, as shown in Figure 4.

3.21st step carries out plane fitting to the three dimensional point cloud obtained, and plane fitting process to 3.6, simulates P1, P2 two planes see step 3.1;

The plane equation of 3.22nd step Calculation Plane P1 and plane P2 under world coordinate system, and obtain the intersection L of P1 and P2;

3.23rd step appoints 1 O got on L, calculate plane P 1 crosses O point and perpendicular on straight line L1 and P2 of L O point and the equation in coordinates formula of straight line L2 perpendicular to L;

The angle theta that 3.24th step calculates L1 and L2 is the angle between necessary requirement two plane.

3.3 linear measure longimetry

The method measuring length mainly contains following two kinds:

Corner detection approach

It is generally acknowledged that angle point is that two dimensional image brightness changes violent point, the maximum point of curvature on the curve of image border, or two, two or more linear edge is with the point of certain angle of intersection.Corner detection approach principle as shown in Figure 5.

3.31st step adopts Harris Corner Detection Algorithm, detects angle point in pictures taken and marks, as figure is labeled as 1,2,3,4;

A local window in Harris Corner Detection Algorithm research image after different directions carries out a small amount of skew, the average of the image brightness values in window.Harris corner detection operator can simply be described as: in certain neighborhood of angle point, and the change of brightness is all very large on the straight line that any is passed through this point.The pixel to be detected to each gets window, calculates the non-regularization autocorrelation value of this pixel from all directions, and selects minimum value as the angle point response function of this pixel.

3.32nd step calculates the distance L between angle point, is the length of required object.

(2) edge extracting method

The edge of image refers to the part that image local area brightness is changed significantly, and namely changes to the larger gray-scale value of another gray scale difference from the play of having to go to the toilet in very little buffer area of a gray-scale value.Realize the rim detection of image, find the gray scale transition position of gradation of image matrix exactly with discretize gradient approximating function according to two dimensional gray matrix gradient vector, then in the picture the point of these positions is linked up and just constitute so-called image border.

The step of rim detection:

1) common filtering method mainly contains gaussian filtering, namely adopts the Gaussian function of discretize to produce one group of normalized gaussian kernel, is then weighted summation based on the every bit of gaussian kernel function to gradation of image matrix.

2) strengthen: the basis strengthening edge is the changing value determining each vertex neighborhood intensity of image.Strengthen algorithm the point that gradation of image vertex neighborhood intensity level has significant change to be highlighted.

3) detect: through the image strengthened, often have the Grad of a lot of point larger in neighborhood, and in specific applications, these points are not the marginal point that will look for, and come someway to accept or reject these points so should adopt.Conventional method is detected by thresholding method.

The image of 3.33rd step to shooting carries out edge extracting, extracts the profile of the object that will measure;

3.34th step matching edge contour point obtains the equation of edge line under world coordinate system;

Distance between the straight line of the 3.35th step digital simulation, required by being.

3.4 cylinder roundness calculation

Cylinder circularity refers to the difference of distance between point on the face of cylinder to cylindrical center's axis and cylindrical radius.The set of the point equaling a constant R to the distance of a central axis can be thought in the face of cylinder, and just uniquely can determine a cylinder by this feature is known by 7 parameters, these 7 parameters are the direction vector (a of this central axis respectively ₁, b ₁, c ₁) and straight line on the coordinate (x of certain a starting point ₀, y ₀, z ₀), and the radius R of cylinder.Observation station and axis geometric relationship figure, observation station and axis geometric relationship figure, as shown in Figure 6.

3.41st step sets any observation station coordinate as P _i(x _i, y _i, z _i), so P _ibe the real radius R ' recorded to the vertical range on axis, α is P _ip ₀with the angle of central axis.

$R^{\′}=P_i{P}_0\sqrt{1-{\cos }^2\mathrm{\α}}---\left(5\right)$

Wherein:

$\cos \mathrm{\α}=\frac{a_1(x_i-x_0)+b_1(y_i-y_0)+z_1(z_i-z_0)}{\sqrt{{(x_i-x_0)}^2+{(y_i-y_0)}^2+{(z_i-z_0)}^2}}---\left(6\right)$

Error equation can be classified as:

υ＝R′-R

$\begin{array}{c}{\mathrm{\υ}}_i\\ =\sqrt{{(x_i-x_0)}^2+{(y_i-y_0)}^2+{(z_i-z_0)}^2-{[a_1(x_i-x_0)+b_1(y_i-y_0)+z_i(z_i-z_0)]}^2}\\ -R\end{array}---\left(7\right)$

Wherein υ is obserred coordinate value residual error.

3.42nd step introduces least square constraint v ^tp υ=min (specifically see document: Wang Suihui. theory of errors and measurement adjustment [M]. Shanghai: publishing house of Tongji University, 2010.), resolve system of equations (7); In addition for ensureing the uniqueness of starting point coordinate and axis vector, two equation of condition are introduced:

$a_1^2+b_1^2+c_1^2=1---\left(8\right)$

x ₀＝average(X),y ₀＝average(Y),z ₀＝average(Z)(9)

Wherein x ₀, y ₀, z ₀be respectively x a little _i, y _i, z _ithe mean value of coordinate.

3.43rd step calculates a ₁, b ₁, c ₁, x ₀, y ₀, z ₀and R.

3.44th step calculates cylinder circularity, the υ namely in error equation, as the standard weighing face of cylinder matching quality.

The above is only the preferred embodiment of the present invention; be noted that for those skilled in the art; under the premise without departing from the principles of the invention, can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (6)

1. a rail vehicle tolerance dimension measuring system, is characterized in that: comprise hand-held measuring equipment and three-dimensional data processing unit;

Hand-held measuring equipment, comprises DLP projector, two CCD camera, sequential control plate, image pick-up card and data processors;

Three-dimensional data processing unit interface, comprises a cloud viewing area, some cloud block of information, commard editor district and camera viewing area.

2. rail vehicle tolerance dimension measuring system according to claim 1, it is characterized in that: sequential control plate is connected with data processor, two CCD camera, DLP projector respectively, DLP projector is connected with data processor, and two ccd video cameras are all connected with data processor by image pick-up card.

3. rail vehicle tolerance dimension measuring system according to claim 1, it is characterized in that: the photocentre axle of described two CCD camera and the angle of DLP projector light axle all between 20 degree to 60 degree, and keep the relative position of DLP projector and two CCD camera constant when measuring.

4. rail vehicle tolerance dimension measuring system according to claim 1, is characterized in that: described rail vehicle tolerance dimension measuring system is provided with the USB interface for being connected with computing machine.

5. a rail vehicle tolerance dimension measuring method, adopts the rail vehicle tolerance dimension measuring system described in any one of claim 1-4, comprises the following steps:

(1) before measuring, hand-held measuring equipment and computing machine adopt USB interface to be connected, and supporting three-dimensional data processing unit is installed in a computer; During measurement, use DLP projector to project to testee the raster image that one group of light intensity is dextrorotation distribution, and use two CCD camera to take the raster image be out of shape through testee surface modulation simultaneously;

(2) absolute phase values of raster image is obtained according to phase shift algorithm and multifrequency heterodyne solution phase method; According to the systematic parameter of demarcating in advance or phase height mapping relation, calculate the three dimensional point cloud on testee surface from absolute phase values;

(3) after obtaining the cloud data of testee, select the menu item on corresponding software, calculate required automobile interior tolerance dimension, with the form display measurement result of message box.

6. rail vehicle tolerance dimension measuring method according to claim 5, is characterized in that: automobile interior tolerance dimension comprises: flatness, angle, length, cylinder circularity.