DE102008024559A1 - Three dimensional geometry recording device has light source for emitting light in light level and camera arranged externally to light level - Google Patents

Abstract

The invention relates to a 3-D geometry detection device with a light source (12) which is set up for outputting light (20) in a light plane (E), a camera (14) arranged outside the light plane (E1), a first mirror ( 16) arranged to reflect a portion of the light (20) perpendicular to the light plane (E1) on an object (22), and a second mirror (18) arranged to reflect a portion of the light (20) perpendicular to the object Light plane (E1) is arranged on the object (22) and with the first mirror (16) forms an angle (alpha) of less than 180 °, so that the object (22) over its circumference (U1) in the plane of light (E) is illuminated; wherein the camera (14) is arranged to take an image of the complete periphery of the object (22) in the light plane (E1).

Description

The The invention relates to a 3D geometry detection device. According to one Second aspect, the invention relates to a 3D geometry detection method.

3D geometry sensing devices can Also known as 3D scanners and serve to the outer dimensions to measure an object. An important application of 3D geometry acquisition devices and method is the measurement of persons, so person geometry data to obtain. These person geometry data can then be used, for example, for Making tailored clothing, custom-made shoes, Prostheses or orthoses are used.

adversely in known 3D geometry detection devices are the high Measuring time and the high hardware costs. Especially when can move the object to be measured, for example because the If the person to be measured is breathing, measuring errors can occur during long measuring times come.

Of the Invention is based on the object, the measuring time in the 3D geometry detection reduce and reduce the cost of hardware.

The Invention solves the problem of a 3D geometry acquisition device with (a) at least one light source adapted to deliver Light in each case in one light plane, (b) one outside the respective light plane arranged camera, (c) a first mirror for reflecting a portion of the light perpendicular to the plane of light on an object and (d) a second mirror for reflecting a portion of the light perpendicular to the plane of light disposed on the object and forms an angle of less than 180 ° with the first mirror, so that the object is over its full extent in the picture plane is illuminable, whereby the Camera is arranged to take a picture of the full extent of the object to record in the light plane.

According to one second aspect triggers the invention by a 3D geometry detection method with the Steps (a) emitting light in at least one light plane, so that an object to be measured directly, indirectly via a first mirror, which is perpendicular for reflecting a part of the light is arranged to the light plane on the object, and indirectly via a second mirror which is perpendicular for reflecting a part of the light is arranged to the light plane on the object and with the first mirror forms an angle of less than 180 °, is irradiated so that the object is in full over its full extent the light plane is illuminated and (b) taking camera pictures of the irradiated object with an outside of the light plane arranged Camera, giving the camera an image of the full scope of the object in the light plane.

Advantageous The invention is characterized in that the measured value recording are carried out very quickly can. In particular, the measured value recording time for a recording all needed Images are less than 2 seconds. It is another advantage that the invention can be implemented with little effort. So are one Camera and a light source sufficient to record the readings.

Advantageous is also that objects of different sizes can be measured. For special size or especially small objects need only suitable mirrors and a suitable lens for the camera chosen become.

in the The scope of the present description is under a light source, which is set up to emit light in a light plane, in particular, understood a device that emits light that is essentially exclusively spreading in one plane. For example, the light plane has a thickness less than 1 cm, preferably less than 2 mm. The thinner the Light level is the higher becomes the measurement accuracy. It is therefore particularly favorable if the Light source is a laser. With increasing distance of the camera from The light level also increases the maximum possible measurement accuracy. It is therefore cheap, when the distance between the camera and the light plane is larger than half a diameter with respect to the plane of the light to be measured Object. For example, this diameter may be estimated in advance or roughly measured.

Under the feature that the mirror for reflecting a part of the light is arranged perpendicular to the plane of light, it should be understood that the light does not leave the light plane due to the reflection at the mirror. The Light plane is thus imaged in itself. It is not necessary that the light plane is a complete plane in the mathematical Meaning is. It is sufficient if the light plane is a conical cutout the full level is, the opening angle For example, be chosen smaller than 90 ° can.

By the feature that the object can be illuminated over its circumference in the light plane, it is understood in particular that an envelope ellipse lying in the light plane can be illuminated around the object. It is not necessary for every single point of the object to be illuminated over its circumference. For example, the object can be attached to man chen bodies have undercuts that are not reached by the light. Although this leads to the fact that no measured value can be generated for this point, this disadvantage is compensated by the significant improvement in the measuring time. When the 3D geometry detection device is a 3D body scanner for detecting the geometry of a person, the light planes are preferably horizontal.

In a preferred embodiment the 3D geometry acquisition device comprises a computing unit, which is designed to triangulate a circumferential curve to calculate that describes the scope of the object in the light plane. These Arithmetic unit can the light source and the mirrors directly be assigned, d. H. in spatial Close to these are located. For example, the arithmetic unit is in the camera is arranged. But it is also possible that the computing unit is physically separated is such that the data collected by the measuring, for example, via a Data network to a spatially be sent separate arithmetic unit. In this case includes the 3D geometry acquisition device a data exchange unit for Send and receive data.

Especially Preferably, the light source is longitudinal slidably mounted on a light source receptacle that is parallel runs to the mirrors. That is the case when for the first mirror and the second mirror each have a straight line, which lies in the mirror plane of the mirror and parallel to the direction runs, in which the light source is longitudinally displaceable is mounted. For example, the light source holder is a straight Guide rail, at the the light source over a drive is arranged automatically or manually movable. This allows the 3D geometry acquisition device to operate automatically become. To perform For a measurement, the light source only has to pass along the light source be moved while taking pictures with the camera become.

The Measurement time can be significantly reduced if the 3D geometry acquisition device a second light source or other light sources that set up are for giving off light of other light planes that are the first Light plane of the first light source run parallel. Here is it especially cheap, if the other light sources light in different colors emit. Alternatively to the shortening of the measuring time allowed the provision of a second light source or even a third, fourth or further light source, an increase in measurement accuracy same measuring time.

In a preferred embodiment There is a second camera on the other side first light plane is arranged. For example, a link between the two cameras then perpendicular to the light plane or the light planes stand. Through a second camera or even a third, fourth or more cameras, the measurement accuracy or the number of measurable surface points increase.

Prefers the mirrors are arranged so that essentially all of the Light source to the object current light beams at most once reflected by a mirror. This is how measuring errors become avoided. Under the feature that essentially all light rays at the most Once reflected, it is understood that a small part, for example less than 15% or 10% of the luminous flux, are reflected several times can.

It an object receptacle can be provided to which the one to be measured Object is positioned. The object shot is then preferred arranged in an interior of a triangle, which consists of a projection the camera and the mirror to the light plane arises.

Around To be able to determine the position of the light plane, it is advantageous if the 3D geometry detection device includes at least one non-reflecting object, wherein the light source is arranged so that this non-specular object irradiated during operation becomes. runs For example, the light plane horizontally, so must with the camera only the height of the point where the plane of light is not measured reflecting object falls.

Prefers the arithmetic unit is designed to carry out a method with the steps (i) determining a position of the light plane, (ii) reading in a Image from the camera, (iii) determining perimeter pixels of the Pictures that belong to the illuminated perimeter of the object and (iv) for a plurality of the peripheral pixels detected an intersection of the light beam which belongs to the respective pixel the light plane. Preferably, all peripheral pixels are determined which belong to the illuminated periphery of the object.

The The entirety of the peripheral pixels is already a 2D representation of the object according to a Cut in the light plane. It is possible to use these perimeter pixels For example, to interpolate with a compensation curve. Prefers These steps are for repeated a variety of light planes, resulting in a 3D representation of the object results. For example, this 3D representation may be in a representation through triangular surfaces or a similar one Representation to be converted.

According to a preferred embodiment of the method according to the invention, the camera is above or below the light plane (s) of the arranged measuring object.

in the The invention will be explained in more detail with reference to the accompanying drawings. It shows

1 a schematic, perspective view of a 3D geometry detection device according to the invention,

2 a top view of the 3D geometry detection device according to 1 and

3 a sketch of how in the context of a 3D geometry detection method according to the invention, the geometry of the object is measured.

1 shows a 3D geometry detection device 10 which is a first light source 12 , a first camera 14 , a first mirror 16 and a second mirror 18 includes. The two mirrors 16 . 18 include an opening angle α of less than 180 °. The opening angle is also greater than 90 ° and in the present case is about 120 °.

The first light source 12 emits light 20 in a light plane E ₁ , which in the present case is a horizontal plane. The light 20 meets a first part, which is indicated by a first arrow P1, on an object to be measured 22 , in the present case, to one person. A second part of the light 20 , which is indicated by an arrow P2, falls on the first mirror 16 and from there on to the object 22 reflected. An indicated by an arrow P3 portion of the light 20 falls to the second mirror 18 and from there is also on the object 22 reflected. The opening angle α is in relation to a position of the first camera 14 chosen so that the object 22 is illuminated in the light plane E ₁ over a complete circumference U ₁ .

Because the mirrors 16 . 18 are arranged so that the light 20 is always reflected back into the first light plane E ₁ , the circumference U _{1 is} a planar curve, which can be described mathematically as Kuve u = (u _x (t), u _y (t)) with a running parameter t. In the present case, this plane curve is horizontal, because the first light plane E ₁ also extends horizontally.

The 3D geometry acquisition device 10 also includes a second light source 24 and a second camera 26 , The second light source emits light in a second light plane E ₂ , which runs parallel to the first light plane E ₁ . The two cameras 14 . 26 are arranged so that they include the two light planes E ₁ , E ₂ between them. A connection between the two cameras 14 . 26 is also approximately perpendicular to E ₁ and E ₂ .

The two light sources 12 . 24 are longitudinally displaceable on a holder 28 attached, the longitudinally displaceable at a light source receptacle 30 is attached. The two cameras 14 . 16 are also at the light source receptacle 30 attached. By means of the holder 28 are the two cameras 14 . 26 displaceable in a direction of displacement R, wherein the displacement direction R can be perpendicular to the light planes E ₁ , E ₂ .

The light source recording 30 is arranged on an angle bisector of the opening angle α, so that the entire structure is mirror-symmetrical with respect to the bisector of the opening angle α. The camera 14 is mounted at a height H ₁₄ which is larger than the object 22 so that she can record the illuminated circumference U ₁ with her visual field. At least one of the two cameras 14 . 26 is also arranged so that it can receive a second circumference U ₂ , which of the light in the second light plane E ₂ on the object 22 is produced. In the case of multiple cameras and multiple light levels, however, each camera can capture all the perimeters.

2 shows a view of the structure according to 1 from above. It can be seen that the first camera 14 and the second camera 26 that from the first camera 14 is concealed, arranged so that all light rays from the light sources 12 . 24 , in the 2 from the first camera 14 are hidden, either directly the object 22 meet (light fan 32.1 ), indirectly via the first mirror 16 on the object 22 arrive (light fan 32.2 ) or over the second mirror 18 on the object 22 arrive (light fan 32.3 ). All other rays on the mirror 16 . 18 can not meet in one of the two cameras 14 . 26 reach. The cameras 14 . 26 , the mirror 16 . 18 and the object 22 are also arranged so that the three light fans 32.1 . 32.2 and 32.3 disjoint are.

In the first camera 14 is on a CCD chip 34 , which is schematically drawn, an image of the object 22 added. Because of the light fans 32.1 . 32.2 . 32.2 disjoint, every pixel on the CCD chip can 34 which images the illuminated circumference U ₁ , exactly one of the three light fans 32 be assigned.

It is possible the first mirror 16 , the second mirror 18 or both mirrors together relative to the camera 14 so pivotable or slidable to design that the corresponding positions are adjustable, if the light fans should not be disjoint or if the object 22 should not be illuminatable over its full extent. 2 also shows two mirror images 36.1 . 36.2 of the object 22 ,

3 shows a transverse view of the 3D geometry detection device 10 , Schematically is an electrical processing unit 38 drawn with the first camera 14 and the second camera 26 (not shown) is wireless or wired in communication. The arithmetic unit 38 is programmed to automatically perform the procedure described below.

First, a position of the light plane E _{1 is} determined. This happens, for example, that a not shown position detection device, the position of the holder 28 relative to the light source recording 30 is detected. This may be, for example, an optical or electrical displacement meter. Alternatively, the first camera detects 14 the position of the light plane E ₁ , for example, the height H above the ground, by a spot on a non-reflecting object 40 (please refer 2 ), for example a frame of the mirror 16 , is recorded.

Then the camera captures 14 with her CCD chip 34 all pixels or pixels r _i that belong to the illuminated circumference U ₁ . This happens, for example, in that the first light source 12 Light of a certain wavelength λ emits and the image on the CCD chip 34 from the arithmetic unit 38 it then searches which pixels in the corresponding wavelength λ exceed a given intensity value.

For example, the pixels r ₁ and r _{2 are} peripheral pixels. After reading the picture from the camera 14 are determined by this the circumferential pixels r ₁ , r ₂ . For all these peripheral pixels r is calculated by the arithmetic unit 38 first determined to which light fan 32 (see. 2 ) the corresponding pixel belongs, and then the associated light beam is detected. For example, for the peripheral pixel r _1, the associated light beam S _{1 is determined} in the form of a straight line. In the intersection of the beam S ₁ with the light plane E ₁ is the point p ₁ , which lies on the illuminated circumference U ₁ . By calculating the corresponding intersection point, the coordinates of the point p ₁ in the coordinate system of the 3D geometry detection device become 10 determined.

While the beam S ₁ to the first fan of light 32.1 (see. 2 ), is determined by the position of the peripheral pixel r ₁ on the CCD chip 34 determines that the beam S ₂ to the third light fan 32.3 ( 2 ) belongs. From the predetermined distance A ₁₈ between a focus of the first camera 14 and the second mirror 18 first, the point p ' ₂ and from this the point p ₂ on the circumference U _{1 of} the object 22 be determined. In this way, the associated points p on the circumference U _{1 are} determined for all circumferential pixels r.

A method according to the invention is carried out by first of all the object 22 between the first mirror 16 , second mirror 18 and camera 14 is arranged. After that, the first light source 12 moving from a first position to a second position while keeping the object 22 constantly lit. At the same time are from the first camera 14 constantly taken pictures and to the arithmetic unit 38 Posted. If a second camera is available, this is also with the arithmetic unit 38 connected and sends the pictures taken by her.

This is the first light source 12 for example, moving at a speed which is smaller than the quotient of a thickness of the light plane E and a read-out time, which is needed to the CCD chip 34 the first camera 14 read. Basically, the speed is freely selectable. The images are either first in a digital memory of the camera 14 or the arithmetic unit 38 digitally stored or processed in real time as described above.

10

3D geometry detection device

12

first light source

14

first camera

16

first mirror

18

second mirror

20

light

22

object

24

second light source

26

second camera

28

holder

30

Light Source Record

32

light fan

34

CCD chip

36

reflection

38

computer unit

40

Not reflecting object

α

opening angle

e

light plane

p

Point

P

arrow

U

scope

R

displacement direction

H

height

λ

wavelength

r

Peripheral pixel

S

beam of light

i

running Index

t

control variable

Claims (12)

3D geometry acquisition device with (a) a light source ( 12 ), which is adapted to emit light ( 20 ) in a plane of light (E), (b) a camera arranged outside the light plane (E ₁ ) ( 14 ), (c) a first mirror ( 16 ), which is used to reflect part of the light ( 20 ) perpendicular to the plane of light (E ₁ ) on an object ( 22 ) and (d) a second mirror ( 18 ), which is used to reflect part of the light ( 20 ) perpendicular to the plane of light (E ₁ ) on the object ( 22 ) and with the first mirror ( 16 ) forms an angle (α) of less than 180 °, (e) so that the object ( 22 ) can be illuminated over its circumference (U ₁ ) in the light plane (E); (f) where the camera ( 14 ) is arranged to form an image of the complete circumference of the object ( 22 ) in the light plane (E ₁ ). 3D geometry detection device according to claim 1, characterized by an arithmetic unit ( 38 ), which is designed to calculate from the image by means of triangulation a circumferential curve which defines the circumference (U ₁ ) of the object ( 22 ) in the light plane (E). 3D geometry acquisition device according to one of the preceding claims, characterized in that the light source ( 12 . 24 ) longitudinally displaceable on a light source receptacle ( 30 ) mounted parallel to the mirrors ( 16 . 18 ) runs. 3D geometry detection device according to one of the preceding claims, characterized by a second light source ( 24 ), which is adapted to emit light ( 20 ) in a second light plane (E ₂ ), which is at the first light plane (E ₁ ) of the first light source ( 12 ) runs parallel. 3D geometry acquisition device according to one of the preceding claims, characterized by a second camera ( 26 ) disposed on another side with respect to the first light plane (E ₁ ). 3D geometry detection device according to one of the preceding claims, characterized in that the mirrors ( 16 . 18 ) are arranged so that substantially all of the light source ( 12 ) to the object ( 22 ) running light beams at most once from one of the mirrors ( 16 . 18 ) are reflected. 3D geometry detection device according to one of the preceding claims, characterized by at least one non-reflecting object ( 40 ). 3D geometry detection device according to one of claims 2 to 7, characterized in that the arithmetic unit ( 38 ) is configured to perform a method comprising the steps of: (i) determining a position of the light plane (E ₁ ), (ii) reading an image from the camera ( 14 ), (iii) determining perimeter pixels (r) of the image corresponding to the illuminated perimeter of the object (3) 22 ), (iv) for a plurality of the peripheral pixels (r _i ), determining an intersection of a light beam (S _i ) associated with the respective peripheral pixel (r _i ) with the light plane (E ₁ , E ₂ ) and repeating steps (i) to (iv) for a plurality of light planes. 3D geometry detection method, comprising the steps of (a) emitting light ( 20 ) in a plane of light so that an object ( 22 ) (i) directly, (ii) indirectly via a first mirror ( 16 ), which is used to reflect part of the light ( 20 ) perpendicular to the plane of light (E ₁ , E ₂ ) on the object ( 22 ) and (iii) indirectly via a second mirror ( 18 ), which is used to reflect part of the light ( 20 ) perpendicular to the plane of light (E ₁ , E ₂ ) on the object ( 22 ) and with the first mirror ( 16 ) forms an angle (α) of less than 180 °, is irradiated, so that the object ( 22 ) is illuminated over its circumference (U) in the light plane (E ₁ , E ₂ ), and (b) taking camera images of the irradiated object ( 22 ) with a camera arranged outside the light plane (E ₁ , E ₂ ), so that the camera has an image of the complete, illuminated periphery (U) of the object ( 22 ) in the light plane (E ₁ , E ₂ ) absorbs. 3D geometry detection method according to claim 9, characterized in that the object ( 22 ) is arranged within a triangle, which by the light source ( 12 ), the first mirror ( 16 ) and the second mirror ( 18 ) is formed. 3D geometry detection method according to claim 9 or 10, characterized in that the camera ( 14 ) above or below the object to be measured ( 22 ) is arranged. 3D geometry acquisition device according to one of claims 10 to 11, characterized by the step: - moving the light source ( 12 ) so that the light plane (E ₁ , E ₂ ) changes in position and at the same time captures a plurality of images.