CN104299266A - Scanning device and three-dimensional reconstruction method of false tooth cavity preparation body with inverted recess - Google Patents

Abstract

The invention discloses a scanning device of a false tooth cavity preparation body with an inverted recess. The scanning device comprises a box body, a laser scanning mechanism, a power-driven translation platform, a power-driven goniometer stage, a dental crown clamp base platform and a dental crown clamp. The power-driven translation platform is arranged on the bottom surface in the box body, the power-driven goniometer stage is arranged on the power-driven goniometer stage, an arc recess is formed in the upper surface of the power-driven goniometer stage, a rotary platform support is arranged on the arc recess, the dental crown clamp base platform is connected with the upper surface of the rotary platform support through a power-driven rotary platform, and the dental crown clamp is arranged on the dental crown clamp platform. An inverted L-shaped support is arranged at the rear portion of the power-driven translation platform, a fixing surface plate is arranged at the upper end portion of the inverted L-shaped support, and the laser scanning mechanism is arranged on the fixing surface plate. The invention further discloses a three-dimensional reconstruction method of the false tooth cavity preparation body with the inverted recess. By means of the scanning device and three-dimensional reconstruction method of the false tooth cavity preparation body with the inverted recess, the inverted recess in the false tooth cavity preparation body can be scanned, and scanning precision is high and smaller than 0.1mm.

Description

With scanister and the three-dimensional rebuilding method thereof of recessed artificial tooth cavity preparation body

Technical field

The invention belongs to the scanning technique field of artificial tooth, be specifically related to a kind of scanister with recessed artificial tooth cavity preparation body and three-dimensional rebuilding method thereof.

Background technology

The recessed appearance profile by tooth body or other tissue, and relative position relationship embodies; Formed and recessedly should possess two conditions, one is that the end area of object and floorage are unequal, two is object major axis and reference plane out of plumb, namely there is the difference of scope axially, or be in heeling condition, is the two from trapezoidal top by the saying of mathematics, do one perpendicular to following straight line, result in formation of two triangles in left and right, and a middle rectangle, so the triangle on both sides is just called Dao Ao district.

KaVoPREPassistant system of the prior art is consuming time longer in the reconstruction of artificial tooth cavity preparation swept-volume and parameter measurement, and due to the limitation of device structure, projection line cannot enter the recessed of the artificial tooth cavity preparation body of inclination, make to rebuild image disappearance concave portion, in practical application, have obvious limitation.

Scanning at present for artificial tooth cavity preparation body adopts contactless raster scanning, and blocked by the tooth body of outside in tooth body inside owing to falling recess in artificial tooth cavity preparation body, use grating horizontal scanning, laser rays cannot arrive inside.

Summary of the invention

For solving the technical matters of existing existence, the embodiment of the present invention provides a kind of scanister with recessed artificial tooth cavity preparation body and three-dimensional rebuilding method thereof.

For achieving the above object, the technical scheme of the embodiment of the present invention is achieved in that

The embodiment of the present invention provides a kind of scanister with recessed artificial tooth cavity preparation body, comprise casing, laser scanning mechanism, motorized precision translation stage, electric angle position platform, corona fixture base station, corona fixture, described motorized precision translation stage is arranged on the bottom surface of box house, described electric angle position platform is arranged on motorized precision translation stage, described electric angle position platform upper surface is provided with arc-shaped recess, described arc-shaped recess is provided with universal stage support, be connected with corona fixture base station by electric rotary table above described universal stage support, described corona fixture is arranged on above corona fixture base station, the rear portion of described motorized precision translation stage is provided with inverted L shape support, described inverted L shape support upper end is provided with fixed panel, described laser scanning mechanism is arranged on fixed panel, and described laser scanning mechanism comprises angled symmetrically arranged a pair CCD camera, is arranged on the laser instrument at fixed panel center.

In such scheme, described a pair CCD camera is connected with computing machine by USB interface, and described laser instrument is connected with computing machine by RS232 interface.

The embodiment of the present invention also provides a kind of three-dimensional rebuilding method with recessed artificial tooth cavity preparation body, the method comprises: the horizontal direction of scanister scanning artificial tooth cavity preparation body and positive and negative 45 degree of directions, obtain the three-dimensional point data with recessed artificial tooth cavity preparation body by view data split recessed in the view data of the artificial tooth cavity preparation body horizontal direction of scanning acquisition and cavity preparation body.

In such scheme, view data split acquisition recessed in the described view data of artificial tooth cavity preparation body horizontal direction that scanning obtained and artificial tooth cavity preparation body with the three-dimensional point data of recessed artificial tooth cavity preparation body is: carry out Image semantic classification, image registration, image co-registration successively to recessed view data in the view data of artificial tooth cavity preparation body horizontal direction and nest hole.

In such scheme, described Image semantic classification is: random noise in cloud data recessed in nest hole and impulse disturbances are reduced or removal according to median filtering algorithm.

In such scheme, described image registration is: scanister is carried out artificial tooth cavity preparation swept-volume from different vertical level positions and obtain that view data recessed in the artificial tooth cavity preparation body of different angles is complete splices, the view data unification of different angles under the same coordinate system.

In such scheme, described image registration is specially: set under a certain position with recessed artificial tooth cavity preparation body as with reference to unified coordinate system, the coordinate system of other positions by transformation matrix of coordinates unification under unified coordinate system, theoretical according to Screw, space vector substep solves transformation matrix, and transformation matrix of coordinates can be analyzed to rotation matrix and translation vector: X ₀for uniform coordinate, R _1,0for rotation matrix, X ₁for the coordinate under another location, T _1,0for translation vector

X ₀＝R _1,0·X ₁+T _1,0

R _1,0and T _1,0solve by following realization:

The solution procedure of rotation matrix R is as follows:

Suppose P ₁and P ₂for given two space characteristics points, its line constitutes an amount of space

According to the character that space vector and its position have nothing to do, can X and X' be expressed as under two coordinate system W and W', when not considering the affecting of displacement, have:

X＝R·X'

By the rotational transform relational equation of Screw theory, there is a turning axle u=(u _x, u _y, u _z) rotation matrix that is made up of turning axle is U, and X=RX' is met

X-X'＝U·(X+X')

The pass that wherein U and u is formed is

$U=\mathrm{Screw}\left(u\right)\left[\begin{array}{ccc}0& -u_x& u_y\\ u_x& 0& -u_z\\ -u_y& u_z& 0\end{array}\right];$

By Cayley theorem, the pass of U and R is

R＝[1+U]·[1-U] ^-1；

From mathematical analysis, when another space vector known, be set to under W and W', be expressed as Y and Y', X-X'=U (X+X') formula can be rewritten as

$\left[\begin{array}{c}X-X^{,}\\ Y-Y^{,}\end{array}\right]=\left[\begin{array}{c}U\·(X+X^{,})\\ U\·(Y+Y^{,})\end{array}\right]$

As long as vector (X+X') and (Y+Y') not conllinear, then

Σ u=Δ certainly exists, wherein:

$\mathrm{\Σ}=\left[\begin{array}{c}\mathrm{Screw}(X+X^{,})\\ \mathrm{Screw}(Y+Y^{,})\end{array}\right],\mathrm{\Δ}=\left[\begin{array}{c}X-X^{,}\\ Y-Y^{,}\end{array}\right]$

Can in the hope of rotation matrix R by above formula;

Translation vector T solves:

X ₀=R _1,0x ₁+ T _1,0t in formula _1,0, be in fact two coordinate system W ₀and W ₁displacement between initial point, when being determined a P by stereoscopic vision at W ₀and W ₁under expression X ₀and X ₁time, displacement T _1,0can directly be calculated by the difference of the two

T _1,0＝X ₀-R _1,0·X ₁

Try to achieve rotation matrix R and translation vector T according to formula above, view data so just can be made unified to unified coordinate system W ₀under.

In such scheme, described image co-registration is: according to method of weighted mean, two width images are merged, and first carries out weight assignment to recessed view data in the view data of the horizontal direction of overlapping region and artificial tooth cavity preparation body; Then carry out superposed average according to the weighted value composed and obtain pixel value; Suppose that the weighted value of horizontal direction and recessed view data is taken as ɑ ₁and ɑ ₂, ɑ ₁and ɑ ₂all satisfy condition ∈ (0,1), and ɑ ₁+ ɑ ₂=1; Suppose f ₁and f ₂be the recessed image to be spliced with horizontal direction respectively, f is the pixel after merging, and so, the expression formula of the grey scale pixel value at overlapping region place is:

$f(x,y,z)=\begin{array}{c}\left\{\begin{array}{cc}{f}_1(x,y,z)& (x,y,z)\∈f_1\\ ({\mathrm{\α}}_1{f}_1(x,y,z)+{\mathrm{\α}}_2{f}_2(x,y,z))/2& (x,y,z)\∈f_1\∩f_2\text{.}\\ f_2(x,y,z)& (x,y,z)\∈f_2\end{array}\right.\end{array}$

Compared with prior art, beneficial effect of the present invention:

Physical construction of the present invention have employed electric angle position table apparatus, and it is 45 degree and electric rotary table rotating 360 degrees that this device can make motorized precision translation stage positive and negative, and motorized precision translation stage is positive and negative is 45 degree, can to the recessed line scanning in artificial tooth cavity preparation body.

The scanning accuracy of the present invention to artificial tooth cavity preparation body is higher, and scanning accuracy is less than 0.1mm.

Accompanying drawing explanation

A kind of structural representation with recessed artificial tooth cavity preparation swept-volume device that Fig. 1 provides for the embodiment of the present invention;

Fig. 2 is the side view of Fig. 1;

Fig. 3 is the schematic diagram from multiple position detection artificial tooth cavity preparation body;

Fig. 4 is space vector and the expression graph of a relation under different coordinates thereof.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.

The embodiment of the present invention provides a kind of with recessed artificial tooth cavity preparation swept-volume device, as Fig. 1, shown in 2, this artificial tooth cavity preparation swept-volume device comprises casing 11, laser scanning mechanism 10, motorized precision translation stage 1, electric angle position platform 2, corona fixture base station 5, corona fixture 6, described motorized precision translation stage 1 is arranged on the bottom surface of casing 11 inside, described electric angle position platform 2 is arranged on motorized precision translation stage 1, described electric angle position platform 2 upper surface is provided with arc-shaped recess, described arc-shaped recess is provided with universal stage support 3, be connected with corona fixture base station 5 by electric rotary table 4 above described universal stage support 3, described corona fixture 6 is arranged on above corona fixture base station 5, the rear portion of described motorized precision translation stage 1 is provided with inverted L shape support 9, described inverted L shape support 9 upper end is provided with fixed panel, described laser scanning mechanism 10 is arranged on fixed panel, and described laser scanning mechanism 10 comprises a pair CCD camera 7 arranged in 30 degree of angle symmetrical, the laser instrument 8 being arranged on fixed panel center.

Described a pair CCD camera 7 is connected with computing machine by USB interface, and described laser instrument 8 is connected with computing machine by RS232 interface.

The embodiment of the present invention also provides a kind of with recessed artificial tooth cavity preparation body three-dimensional rebuilding method, the method comprises: the horizontal direction of artificial tooth cavity preparation swept-volume device scanning artificial tooth cavity preparation body and positive and negative 45 degree of directions, obtain the three-dimensional point data of artificial tooth by view data split recessed in the view data of the artificial tooth cavity preparation body horizontal direction of scanning acquisition and artificial tooth cavity preparation body.

View data split acquisition recessed in the described view data of artificial tooth cavity preparation body horizontal direction that scanning obtained and artificial tooth cavity preparation body with the three-dimensional point data of recessed artificial tooth cavity preparation body is: carry out Image semantic classification, image registration, image co-registration.

The method realizes especially by following steps:

Step 101: with horizontal direction and positive and negative 45 degree of directions of recessed artificial tooth cavity preparation swept-volume device scanning artificial tooth cavity preparation body.

Concrete, with the artificial tooth cavity preparation body that a pair CCD camera 7 and laser instrument 8 collection of recessed artificial tooth cavity preparation swept-volume device are placed on corona fixture 6, and then by adjustment electric angle position platform 2, universal stage support 3 is moved left and right in the horizontal direction to become positive and negative miter angle respectively, at this moment, described a pair CCD camera 7 and laser instrument 8 gather data recessed in artificial tooth cavity preparation body.

Step 102: with recessed artificial tooth cavity preparation swept-volume device by the three-dimensional point data of view data split acquisition recessed in the view data of the artificial tooth cavity preparation body horizontal direction of scanning acquisition and artificial tooth cavity preparation body with recessed artificial tooth cavity preparation body.

Concrete, obtain view data recessed in the view data of the horizontal direction of artificial tooth cavity preparation body and artificial tooth cavity preparation body respectively from a pair CCD camera 7 and laser instrument 8 with the computing machine of recessed artificial tooth cavity preparation swept-volume device, then view data split is obtained the three-dimensional point data with recessed artificial tooth cavity preparation body.

The split of the view data of described artificial tooth cavity preparation body horizontal direction and the interior recessed view data of artificial tooth cavity preparation body is realized by following steps:

Step 201: Image semantic classification.

Concrete, use median filtering algorithm, the random noise in cloud data recessed in the artificial tooth cavity preparation body obtained and impulse disturbances are reduced or removes, the information of image border can be retained simultaneously preferably.

Step 202: image registration.

Concrete, because Tooth preparation can tilt mobile 45 °, therefore, be equivalent to laser scanning device 10 and carry out Tooth preparation scanning from different angles.

Splice complete for Tooth preparation, need the Tooth preparation data unification of different angles under the same coordinate system.

If position 1 (as Fig. 3) W ₀for reference unified coordinate system, M _{i, i-1}for i place, position coordinate system W _ito i-1 place, position coordinate system W _i-1transformation matrix of coordinates, i=1 ..., n.

The coordinate system of other positions is by transformation matrix of coordinates M _{i, i-1}unified to reference unified coordinate system W ₀under, that is: X ₀=M _1,0x ₁

Theoretical by Screw, space vector substep solves transformation matrix, and transformation matrix of coordinates can be analyzed to rotation matrix and translation vector: X ₀for uniform coordinate, R _1,0for rotation matrix, X ₁for the coordinate under another location, T _1,0for translation vector

X ₀＝R _1,0·X ₁+T _1,0

The solution procedure of rotation matrix R is as follows:

Suppose P ₁and P ₂for given two space characteristics points, its line constitutes an amount of space

According to the character that space vector and its position have nothing to do, x and X' can be expressed as, as shown in Figure 4 under two coordinate system W and W'.When not considering the affecting of displacement, have:

X＝R·X'

By the rotational transform relational equation of Screw theory, there is a turning axle u=(u _x, u _y, u _z) rotation matrix that is made up of turning axle is U, and X=RX' is met

X-X'＝U·(X+X')

The pass that wherein U and u is formed is

$U=\mathrm{Screw}\left(u\right)\left[\begin{array}{ccc}0& -u_x& u_y\\ u_x& 0& -u_z\\ -u_y& u_z& 0\end{array}\right];$

By Cayley theorem, the pass of U and R is

R＝[1+U]·[1-U] ^-1

Because U has the character of singular matrix, X=RX' formula cannot directly solve.From mathematical analysis, when another space vector known, be set to under W and W ', be expressed as Y and Y', X-X'=U (X+X') formula can be rewritten as

$\left[\begin{array}{c}X-X^{,}\\ Y-Y^{,}\end{array}\right]=\left[\begin{array}{c}U\·(X+X^{,})\\ U\·(Y+Y^{,})\end{array}\right]$

Can prove, as long as vector (X+X') and (Y+Y') not conllinear, then

Σ u=Δ certainly exists, wherein:

$\mathrm{\Σ}=\left[\begin{array}{c}\mathrm{Screw}(X+X^{,})\\ \mathrm{Screw}(Y+Y^{,})\end{array}\right],\mathrm{\Δ}=\left[\begin{array}{c}X-X^{,}\\ Y-Y^{,}\end{array}\right]$

Can in the hope of rotation matrix R by above formula.

Translation vector T solves:

X ₀=R _1,0x ₁+ T _1,0t in formula _1,0, be in fact two coordinate system W ₀and W ₁displacement between initial point.When being determined a P by stereoscopic vision at W ₀and W ₁under expression X ₀and X ₁time, displacement T _1,0can directly be calculated by the difference of the two

T _1,0＝X ₀-R _1,0·X ₁

Step 203: image co-registration.

Concrete, recessed data are made up of two aspects, the data in unenhanced and recessed.Image registration achieves the image conversion of two aspects to a coordinate system.By method of weighted mean, two width images are merged.Method of weighted mean is implemented as follows:

According to method of weighted mean, two width images are merged, described two width images refer to the recessed interior figure of unenhanced figure (figure except recessed) and electric angle position platform motion scan, image co-registration to the same coordinate system, is completed piece image by image conversion by registration.

First with the image in recessed, weight assignment is carried out to the unenhanced image of overlapping region;

Then carry out superposed average according to the weighted value composed and obtain pixel value;

Suppose recessed in and unenhanced weighted value be taken as ɑ ₁and ɑ ₂, ɑ ₁and ɑ ₂all satisfy condition ∈ (0,1), and ɑ ₁+ ɑ ₂=1;

Suppose f ₁and f ₂be recessed interior and unenhanced image to be spliced respectively, f is the pixel after merging, and so, the expression formula of the grey scale pixel value at overlapping region place is:

$f(x,y,z)=\begin{array}{c}\left\{\begin{array}{cc}{f}_1(x,y,z)& (x,y,z)\∈f_1\\ ({\mathrm{\α}}_1{f}_1(x,y,z)+{\mathrm{\α}}_2{f}_2(x,y,z))/2& (x,y,z)\∈f_1\∩f_2\text{.}\\ f_2(x,y,z)& (x,y,z)\∈f_2\end{array}\right.\end{array}$

Claims (8)

1. the scanister with recessed artificial tooth cavity preparation body, it is characterized in that: comprise casing (11), laser scanning mechanism (10), motorized precision translation stage (1), electric angle position platform (2), corona fixture base station (5), corona fixture (6), described motorized precision translation stage (1) is arranged on the inner bottom surface of casing (11), described electric angle position platform (2) is arranged on motorized precision translation stage (1), described electric angle position platform (2) upper surface is provided with arc-shaped recess, described arc-shaped recess is provided with universal stage support (3), described universal stage support (3) top is connected with corona fixture base station (5) by electric rotary table (4), described corona fixture (6) is arranged on corona fixture base station (5) top, the rear portion of described motorized precision translation stage (1) is provided with inverted L shape support (9), described inverted L shape support (9) upper end is provided with fixed panel, described laser scanning mechanism (10) is arranged on fixed panel, and described laser scanning mechanism (10) comprises angled symmetrically arranged a pair CCD camera (7), is arranged on the laser instrument (8) at fixed panel center.

2. the scanister with recessed artificial tooth cavity preparation body according to claim 1, it is characterized in that: described a pair CCD camera (7) is connected with computing machine by USB interface, and described laser instrument (8) is connected with computing machine by RS232 interface.

3. the three-dimensional rebuilding method with recessed artificial tooth cavity preparation body, it is characterized in that, the method comprises: the horizontal direction of scanister scanning artificial tooth cavity preparation body and positive and negative 45 degree of directions, obtain the three-dimensional point data with recessed artificial tooth cavity preparation body by view data split recessed in the view data of the artificial tooth cavity preparation body horizontal direction of scanning acquisition and cavity preparation body.

4. the three-dimensional rebuilding method with recessed artificial tooth cavity preparation body according to claim 3, it is characterized in that, view data split acquisition recessed in the described view data of artificial tooth cavity preparation body horizontal direction that scanning obtained and artificial tooth cavity preparation body with the three-dimensional point data of recessed artificial tooth cavity preparation body is: carry out Image semantic classification, image registration, image co-registration successively to recessed view data in the view data of artificial tooth cavity preparation body horizontal direction and nest hole.

5. the three-dimensional rebuilding method with recessed artificial tooth cavity preparation body according to claim 4, it is characterized in that, described Image semantic classification is: random noise in cloud data recessed in nest hole and impulse disturbances are reduced or removal according to median filtering algorithm.

6. the three-dimensional rebuilding method with recessed artificial tooth cavity preparation body according to claim 4, it is characterized in that, described image registration is: scanister is carried out artificial tooth cavity preparation swept-volume from different vertical level positions and obtain that view data recessed in the artificial tooth cavity preparation body of different angles is complete splices, the view data unification of different angles under the same coordinate system.

7. the three-dimensional rebuilding method with recessed artificial tooth cavity preparation body according to claim 6, it is characterized in that: described image registration is specially: set under a certain position with recessed artificial tooth cavity preparation body as with reference to unified coordinate system, the coordinate system of other positions by transformation matrix of coordinates unification under unified coordinate system, theoretical according to Screw, space vector substep solves transformation matrix, and transformation matrix of coordinates can be analyzed to rotation matrix and translation vector: X ₀for uniform coordinate, R _1,0for rotation matrix, X ₁for the coordinate under another location, T _1,0for translation vector

X ₀＝R _1,0·X ₁+T _1,0

R _1,0and T _1,0solve by following realization:

The solution procedure of rotation matrix R is as follows:

Suppose P ₁and P ₂for given two space characteristics points, its line constitutes an amount of space

According to the character that space vector and its position have nothing to do, can X and X' be expressed as under two coordinate system W and W', when not considering the affecting of displacement, have:

X＝R·X'

By the rotational transform relational equation of Screw theory, there is a turning axle u=(u _x, u _y, u _z) rotation matrix that is made up of turning axle is U, and X=RX' is met

X-X'＝U·(X+X')

The pass that wherein U and u is formed is

$U=\mathrm{Screw}\left(u\right)\left[\begin{array}{ccc}0& -u_x& u_y\\ u_x& 0& {-u}_z\\ {-u}_y& u_z& 0\end{array}\right];$

By Cayley theorem, the pass of U and R is

R＝[1+U]·[1-U] ^-1；

From mathematical analysis, when another space vector known, be set to under W and W', be expressed as Y and Y', X-X'=U (X+X') formula can be rewritten as

$\left[\begin{array}{c}X-X^{,}\\ Y-Y^{,}\end{array}\right]=\left[\begin{array}{c}U\·(X+X^{,})\\ U\·(Y+Y^{,})\end{array}\right]$

As long as vector (X+X') and (Y+Y') not conllinear, then

Σ u=Δ certainly exists, wherein:

$\mathrm{\Σ}=\left[\begin{array}{c}\mathrm{Screw}(X+X^{,})\\ \mathrm{Screw}(Y+Y^{,})\end{array}\right],\mathrm{\Δ}=\left[\begin{array}{c}X-X^{,}\\ Y-Y^{,}\end{array}\right]$

Can in the hope of rotation matrix R by above formula;

Translation vector T solves:

X ₀=R _1,0x ₁+ T _1,0t in formula _1,0, be in fact two coordinate system W ₀and W ₁displacement between initial point, when being determined a P by stereoscopic vision at W ₀and W ₁under expression X ₀and X ₁time, displacement T _1,0can directly be calculated by the difference of the two

T _1,0＝X ₀-R _1,0·X ₁

Try to achieve rotation matrix R and translation vector T according to formula above, view data so just can be made unified to unified coordinate system W ₀under.

8. the three-dimensional rebuilding method with recessed artificial tooth cavity preparation body according to claim 4, it is characterized in that, described image co-registration is: according to method of weighted mean, two width images are merged, and first carries out weight assignment to recessed view data in the view data of the horizontal direction of overlapping region and artificial tooth cavity preparation body; Then carry out superposed average according to the weighted value composed and obtain pixel value; Suppose that the weighted value of horizontal direction and recessed view data is taken as ɑ ₁and ɑ ₂, ɑ ₁and ɑ ₂all satisfy condition ∈ (0,1), and ɑ ₁+ ɑ ₂=1; Suppose f ₁and f ₂be the recessed image to be spliced with horizontal direction respectively, f is the pixel after merging, and so, the expression formula of the grey scale pixel value at overlapping region place is:

$f(x,y,z)=\left\{\begin{array}{cc}{f}_1(x,y,z)& (x,y,z)\∈f_1\\ ({\mathrm{\α}}_1{f}_1(x,y,z)+{\mathrm{\α}}_2{f}_2(x,y,z))/2& (x,y,z)\∈f_1\∩f_2\\ f_2(x,y,z)& (x,y,z)\∈f_2\end{array}\right..$