DE102016000415A1 - Device for the three-dimensional detection of a surface structure - Google Patents

Abstract

The invention relates to a device (10) for three-dimensionally detecting a surface structure (11, 11 ', 11'), in particular for detecting teeth (13), with a chromatic-confocal optic (40, 42, 44) having an optical axis (41), and a light source (16) having a grid of individual light sources (34) spaced laterally from the optical axis. In this case, the optics images the individual light sources (34) onto the surface structure (11, 11 ', 11 ") in such a way that a grid of measuring points (49) is produced there, which are laterally spaced from one another. An image pickup unit (50) is optically equivalent to the individual light sources (34) arranged to detect the grid of measurement points (49). Furthermore, a control and evaluation device (70) is set up to control the light source (16) in such a way that the individual light sources (34) emit light with different spectral composition in chronological order, and with the aid of the image recording unit (50) the grid of measuring points ( 49) correlates with the time sequence of the light emitted by the individual light sources (34).

Description

BACKGROUND OF THE INVENTION

1. Field of the invention

• a) mit einer chromatisch-konfokalen Optik, die eine optische Achse umfasst, und
• b) mit einer Lichtquelle, die ein Raster von lateral zur optischen Achse beabstandeten Einzellichtquellen aufweist,
• c) wobei die Optik die Einzellichtquellen derart auf die Oberflächenstruktur abbildet, dass dort ein Raster von Messpunkten erzeugt wird, die voneinander lateral beabstandet sind,
• d) wobei eine Bildaufnahmeeinheit optisch äquivalent zu den Einzellichtquellen angeordnet ist, um das Raster von Messpunkten zu erfassen.

The invention relates to a device for the three-dimensional detection of a surface structure, in particular for the detection of teeth,

• a) with a chromatic-confocal optical system comprising an optical axis, and
• b) with a light source having a grid of individual light sources spaced laterally from the optical axis,
• c) wherein the optics images the individual light sources onto the surface structure in such a way that a grid of measuring points is generated there, which are laterally spaced from one another,
• d) wherein an image pickup unit is arranged optically equivalent to the individual light sources to detect the grid of measurement points.

2. Description of the Related Art

From the DE 10 2006 007 172 B4 For example, a device is known in which a chromatic confocal measuring arrangement is used to detect a surface structure three-dimensionally.

The chromatic-confocal measuring arrangement is based on the knowledge that the refractive power of a lens is generally dependent on the wavelength of the light passing through. As a result, depending on the wavelength, a lens has spatially differently extending focus areas and thus also image areas. This effect is known in optical systems as chromatic aberration and is regarded as an undesirable image defect, which should be corrected as much as possible, for example, by the use of achromats or apochromats.

In the case of the chromatic-confocal measuring arrangement, however, the so-called longitudinal chromatic aberration (CLA) is utilized by imaging white light onto a surface via optics with the largest possible CLA. In this way, the short-wave, blue portion of the light is focused closer to the optics than the long-wave, red portion.

The backscattered light from the surface is confocal imaged on a spectrometer via the chromatic optics and a beam splitter.

If the surface is now, for example, at the place where the blue light is focused, the backscattered blue light almost completely returns to the optics. For the blue light thus the chromatic-confocal condition is met. The red light, on the other hand, due to the blurred focus on the surface, is mostly scattered in directions that do not fall within the confocal optical path of the optic. Overall, significantly less red light will fall on the spectrometer.

By determining the spectral components of the backscattered light with the aid of the spectrometer, therefore, the distance of the surface from the measuring arrangement can be detected.

In order to obtain a three-dimensional image of the surface, the known chromatic-confocal measuring arrangements are scanned over the surface to be detected.

To avoid the scanning movement is in the device according to the DE 10 2006 007 172 B4 generates a grid of laterally spaced measuring points with the aid of a grid arrangement of individual light sources on the surface, the color spectrum of which is determined with the aid of a high-resolution image converter with imaging optics and an upstream prism or grating. In this way one obtains the topography of the surface without having to move the measuring arrangement with respect to the object.

A disadvantage of the devices used hitherto is the relatively high cost and expense for the combination of prism or grating and image converter with imaging optics as a spectrometer.

SUMMARY OF THE INVENTION

The object of the invention is therefore to provide a device for three-dimensional detection of a surface structure, in particular for detecting teeth, in which the image-recording unit can be better, in particular smaller and less expensive, realized.

• e) eine Steuer- und Auswerteeinrichtung vorgesehen ist, die dazu eingerichtet ist,
• – die Lichtquelle derart anzusteuern, dass die Einzellichtquellen in zeitlicher Abfolge Licht mit unterschiedlicher spektraler Zusammensetzung abgeben, und
• – mit Hilfe der Bildaufnahmeeinheit das Raster von Messpunkten korreliert mit der zeitlichen Abfolge des von den Einzellichtquellen abgegebenen Lichts zu erfassen.

According to the invention the above object is achieved with a device of the type mentioned, in the

• e) a control and evaluation device is provided, which is set up to
• To control the light source in such a way that the individual light sources emit light of different spectral composition in chronological order, and
• - To detect the grid of measuring points correlated with the temporal sequence of the light emitted from the individual light sources with the aid of the image recording unit.

The inventor has recognized that no complex combination of prism or grid and Image converter with imaging optics is necessary as a spectrometer when the differently colored spectral components impinge instead of simultaneously on the image acquisition unit in time sequence. As a result, the time axis replaces a dispersion direction which is usually necessary for detecting the spectral components.

Upon detecting the intensities originating from the measurement points in the image acquisition unit, the temporal sequence of the illumination is then taken into account in order to determine for which color the confocal condition is best fulfilled. From this, the distance to the surface structure can then be determined for each measuring point. Decisive in each case is the sequence within a pair of individual light source and associated measuring point. The time sequence of the spectral composition at the individual light sources can be different among one another.

Under different spectral composition of the light emitted from the individual light sources in time sequence light is to be understood here especially a different color of the light, in particular with a respective narrowest wavelength band. Preferably, the time sequence of the spectral compositions is selected so that the center of the wavelength band changes continuously. Although, depending on the desired resolution and spaced-apart wavelength bands are conceivable, the wavelength bands follow advantageous but spectrally close to each other.

As a light source that can produce such a temporal sequence of lighting, for example, different colored LEDs or the like come into question, which are mapped to the same measuring point. Preferably, however, the light source in front of a raster element for generating the individual light sources, a dispersive element with a dispersion direction, in particular an optical grating with a diffraction direction, and at least one aligned at least partially in the dispersion direction row of individually controllable broadband light sources, preferably white light sources, in particular LEDs on. Broadband light sources are understood here in particular as meaning that the change of the light sources without the use of the dispersive element would not lead to a sufficient resolution of the chromatic-confocal condition. By using the dispersive element, however, when the broadband light sources are switched, light of different wavelengths successively falls on the raster element, as a result of which the individual light sources in turn emit color light of sufficiently different color in a corresponding chronological sequence.

As a raster element, the light source may comprise, for example, a perforated screen with a grid of apertures. In this case, the hole screen aperture may be preceded by a microlens array. In this course, a condenser lens can be used to focus as much light intensity in the apertures. Alternatively, the individual light sources can also be represented by a split optical fiber bundle.

Preferably, the control and evaluation device is set up such that the broadband light sources are activated and deactivated in any desired but fixed sequence.

In this case, the broadband light sources clocked at the same frequency can be activated and deactivated, wherein the timings of the individual white light sources are mutually out of phase. The sequence of colored light in the measuring points is thus at least offset in time.

Preferably, one applies to a direction of dispersion correlating, for example, a proportional, at least one monotonically increasing or monotonically decreasing phase shift, wherein the amount of phase shift is adapted to the desired dispersion range. Assuming that the light passing into the apertures via the microlens optics emanates from at least two adjacent LEDs, which can be achieved using very small LEDs and a corresponding dimensioning of the microlens optics, a practically continuous change in the central wavelength of the Light which passes through the aperture. The frequency with which the light color is passed through periodically, continuously, and in steps, corresponds to the clock frequency of the broadband light sources. The image recording unit preferably also comprises a raster element whose raster is optically equivalent to the raster of the individual light sources. In the simplest case, this is realized by an exclusive reading, or taking into account of the corresponding pixels of the image sensor, and in a less simple but possibly necessary case, this is a hole grid aperture having a grid of apertures.

Preferably, the image acquisition unit comprises a planar image acquisition sensor which, at least at the positions at which the measurement points are imaged, permits temporal discrimination of the intensity backscattered by the measurement points.

This is preferably made possible by a lock-in-camera sensor. Such lock-in-camera sensors provide at the output, in principle, in addition to the amplitude signal for each pixel at least also nor a phase signal for an optionally charged to the photon current alternating component.

An elaborate version of such a lock-in-camera sensor with increased sensitivity and recording speed and better dynamics is available as a pOCT (parallel optical low coherence tomography sensor) or fast lock-in sensor on the market. A simpler version of such a lock-in-camera sensor is now available on the market as a time-of-flight image sensor (TOF sensor), also known as a distance aerea image sensor (DAI sensor) , In such a ToF or DAI sensor, at least two storage elements are used for each pixel for storing a detected photon intensity, wherein a synchronization device and / or a trigger connection is used to determine in which of the storage elements the instantaneous incident photon intensity is integrated.

Such a behavior that the incident photon intensity of a measuring point is alternately integrated in different memory elements can also be determined by a full-frame CCD sensor (FF-CCD) or by a frame transfer CCD sensor (FT-CCD). , or realized by a time-delayed integration sensor (TDI). These sensors all operate on the principle that the charges separated by the incident photons are passed through other charged-coupled devices (CCDs) for readout. In order to realize the functionality required here, sufficiently unexposed CCD cells must still be available between the measuring points imaged on the sensor in the direction of the charge coupling of the sensor which themselves must not receive any light, ie, for example. B. are covered by a matching to the measuring points grid aperture. In this case, another fixed microlens array could be helpful to reduce the usually more than 10 microns aperture apertures of the hole screen aperture on the pixel dimension, which is usually less than 10 microns to zoom out. If it is possible to arrange the screen diaphragm matrix and the microlens array so that each measurement point has its own charge chain, without being influenced by other measurement points, advantageously the charge chain of an FF-CCD sensor or TDI sensor can be continuously read out and even within the Readout the integration time can be changed continuously by changing the clock frequency. The read-out images would then each carry a light-dark signal in the columns whose phase shift relative to the read-out clock the time information is included.

With CCD and TDI sensors, however, the same effect as in the DAI / ToF sensor, although at a reduced speed, can also be represented by a specific synchronized forward and / or backward charge shift during image integration.

With all the mentioned lock-in-camera sensors, it is possible to obtain an amplitude signal and a phase signal from the image data read. This allows a temporal discrimination of the backscattered from the measurement points intensity. The temporal discrimination takes place with a sufficiently high speed for the three-dimensional detection of the surface structure.

Alternatively, the image acquisition unit may comprise a CMOS image converter with fast readout capability. Due to their read-out strategy, CMOS image converters are in principle suitable for reading only small subranges in a correspondingly higher speed. Since in the image recording unit required here, the total area of all measuring points on the sensor is only less than 3%, in extreme cases, even only 0.1% of the total image area, the CMOS sensor with suitable control and suitable design of the measuring point grid, which may also by optical mapping or by redirecting via optical fibers can be read out with up to 1000 times frame rate. In principle, sufficient temporal discrimination or determination of the phase position of the alternating component of the incoming photon current is then possible as well. In a diversion of the readout light by optical fibers can also be particularly advantageous to any cost or fast sensor formats, or on a simple line sensor, or, since TDI sensors are usually built very narrow and elongated, also resorted to such lower-cost standard TDI sensors become.

Another way to achieve temporal discrimination with a conventional ordinary frame rate camera sensor is by applying a fast relative movement of the light exiting the aperture over the camera sensor surface. Preferably, for this purpose, a circular movement of the light of the aperture points, with a maximum diameter slightly smaller than the minimum distance of the metering orifices, carried out. For this purpose, z. B. the camera sensor, but advantageously also only an approximately 1: 1 to 1: 2 imaging microlens array, circular with a radius of, for example, 30-60 microns under the Aufnahmelochrasteblende with an integer multiple of the frame rate, and an integer fraction of Light modulation frequency oscillate around the light axis. The same effect could be caused by a slanted fiber plate rotating between the pickup hole stopper and the camera sensor Fibers, or via a rotating or vibrating mirror plate, or via a rotating or vibrating transparent wedge plate, or via a micromechanical mirror array.

A suitable temporal discrimination of the photon current in the measuring points is not only possible with a periodically wave-shaped modulated photon current in sine, triangular or rectangular form, but also with pulsed modulated photon currents. Discrimination of a pulsed signal is preferably possible with all mentioned CCD, TDI and CMOS sensors. For temporal discrimination of pulse-shaped photon currents and surface detectors from the field of X-ray technology (timepix detectors) are suitable, which determine the direct entry time of individual photon quanta with very high temporal resolution.

If the incoming light amount is too low, all of the above sensors may be supplemented to increase the sensitivity by using advanced light amplification systems such as microchannel image intensifiers. In the case of the CCD sensors, EMCCD sensors are suitable for this, which already carry out a charge multiplication internally within the charge chain.

The trigger connection of the above-mentioned image recording units is preferably connected to the control and evaluation unit and configured to control the trigger connection with the same frequency as the broadband light sources. But it is also possible that the image pickup unit sends a trigger signal for driving the light sources and the microlens array. As a result, a correlation of the time sequence within the measuring device can be established. The frequency of the illumination device can be between 1 times and up to several million times the desired measuring point frame rate and is essentially determined by the image converter.

Preferably, the control and evaluation unit is further configured to determine the distance of the surface structure to the optics from the measurement point data supplied by the image acquisition unit. For example, in the case of a DAI sensor, the control and evaluation unit can be set up to determine the distance from the ratio of the intensities detected in the two memory elements of a pixel.

Chromatic optics in the present application is understood to mean an optic (even a single lens) which has a longitudinal chromatic aberration CLA sufficient to cover the desired measuring range / resolution of the depth information about the surface structure with the wavelengths of the light used. Preferably, therefore, the optics is a hyperchromatic optics, whereby a larger depth measurement range is achieved. A hyperchromatic optic has an equivalent Abbe number smaller than the Abbe number of all individual elements, d. H. she has a particularly high CLA. Preferably, the chromatic optics could be supplemented by an electrically adjustable lens, preferably a liquid lens, to ensure a sufficient continuous depth measurement range.

All the above considerations regarding the image recording unit, regardless of the design of the light source, also apply to the case in which individual light sources are otherwise produced instead of the broadband light sources with a dispersive element whose spectral composition can be varied over time. Such a light source may consist for example of a plurality of colored LEDs or of a plurality of colored laser light sources, or of white light sources with downstream bandpass color filters, which are provided with suitable optical means z. B. prisms or dichroic mirrors are mapped to the measuring point grid and are individually controlled in time. In this case, it would be advantageous to overlap the wavelength bands of these light sources, wherein narrow gaps between the wavelength bands of these light sources can also be compensated by a corresponding confocal blur, or by the electrically adjustable lens.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be explained with reference to several embodiments with reference to the drawings. It shows:

1 a schematic longitudinal section through an intraoral camera for three-dimensional detection of a surface structure;

2 a perspective view of a lighting and image pickup unit of the intraoral camera;

3 a schematic representation of the intraoral camera, in which the image pickup unit has a lock-in sensor and the light source has a white light LED line;

4 a perspective view of another illumination and image recording unit of the intraoral camera;

5 a perspective view of another illumination and image recording unit of the intraoral camera;

6 a schematic representation of the intraoral camera, in which the image pickup unit has a common camera sensor and the light source consists of a few special colored LEDs;

7 a schematic representation of the detector pixel and the drive signals of the intraoral camera, in which the image pickup unit has a timepix detector and the light source emits short flashes of light;

8th a schematic representation of the images obtained an intraoral camera, in which the image pickup unit has a CCD or a TDI camera sensor.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

1 shows an example of a device for three-dimensional detection of a surface structure of an intraoral camera 10 for use in dental medicine, in particular for detecting teeth 13 and their surface structure 11 ,

The intraoral camera 10 has an elongated housing 12 on which the essential components of the intraoral camera 10 receives.

Inside the case 12 is at its proximal end 14 a light source 16 arranged, their detailed structure better 2 is apparent.

The light source 16 has to generate the illumination first a line 18 of several (here ten) juxtaposed white light LEDs 20.n ( 20.1 to 20:10 ) as individually controllable white light sources.

Furthermore, the light source points 16 as a dispersive element a grid 22 with a grid plane 24 and a dispersion direction 26 on.

The line 18 and the grid 22 are offset from one another in such a way that for each white light LED 20.n the zeroth diffraction order of the lattice 22 not below 0 ° vertically out of the grid 22 exits, but already has a deflection angle. In particular, the line 18 doing so arranged a vertical projection of the line 18 on the grid level 24 at least one directional component in the dispersion direction 26 having. Preferably, the projection direction is parallel to the dispersion direction 26 , In the embodiments, the line is 18 each carried out in a straight line for cost reasons. The theoretically preferred shape of the line 18 however, would be a circular arc whose center is in the center of the grid 22 lies.

Due to the diffraction effect of the grid 22 become the spectral components of the white light LEDs 20.n coming illumination light in the dispersion direction 26 separated from each other, depending on which white light LED 20.n is activated, in the beam path after the grid 22 each different colored illumination light is generated.

In the beam path after the grid 22 are also a microlens array 30 and a lighting hole screen 32 so arranged that from the grid 22 next, diffracted in 1st order, colored illumination light on apertures 34 the illumination hole screen 32 is focused.

All white light LEDs 20.n the line 18 are below the beam direction 17 the 0th diffraction order of the lowest grid lines of the grid 22 arranged so that no undiffracted white light through the apertures 34 occurs. For better utilization of the white light LEDs 20.n emitted light is located in this beam path also has a condenser lens 28 , which is designed depending on the position as a normal convex lens or as a convex cylindrical lens.

In the further course of the beam path follows a chromatic-confocal optics 40 with an optical axis 41 along which a polarizing filter 43 , a beam splitter cube 42 , a delay plate 47 and a hyperchromatic lens group 44 are arranged. By another, in the returning beam path in front of the image pickup unit 50 built-in polarization filter 53 , Sturgeon light, which has not passed through the intended optical path, can be reduced. For this purpose, as a beam splitter cube 42 preferably also a polarizing beam splitter cube 42 used. a beam splitter cube 42 and a hypochromatic lens group 44 is arranged. Furthermore, the look includes 40 at the distal end 15 the intraoral camera 10 a deflecting mirror 46 with which the beam path on one side of the housing 12 arranged viewing window 48 is diverted.

The optics 40 is designed such that the apertures 34 the illumination hole screen 32 on the surface to be detected 11 be imaged. The apertures 34 thus represent individual light sources that are on the surface structure 11 a grid of illuminated measuring points 49 produce. The illumination hole screen 32 and the optics 40 are coordinated so that the measuring points 49 are laterally spaced approximately between 100 .mu.m and 500 .mu.m apart from one another in order to crosstalk between the measuring points 49 to prevent.

The intraoral camera 10 also has an image pickup unit 50 on, the optically equivalent to light source 16 next to the beam splitter cube 42 is arranged.

In the embodiment of 1 to 3 includes the image capture unit 50 a picture converter 52 and a recording grid 54 which also has a grid of apertures 56 includes. In this case, an optically equivalent arrangement means the light source 16 and the image capture unit 50 that the optics 40 the apertures 34 the illumination hole screen 32 and the apertures 56 the recording grid aperture 54 to the same measuring points 49 on the surface texture 11 maps.

On the recording grid aperture 54 is preferably omitted if the image converter is not a CCD sensor and the pixels of the image converter at least slightly smaller than the intended apertures 56 are. The function of the recording aperture can then simply by neglecting the consideration of the normally accepted by the recording grid 54 covered recording pixels 58 of the image converter 52 respectively.

As an image converter 52 a lock-in-camera sensor is provided, it depends only on the required measurement speed and resolution and the existing stray light whether a fast-lock-in-sensor (pOCT) or a time-of-flight sensor (DAI Sensor) or a correspondingly operated CCD sensor or even a correspondingly evaluated fast CMOS sensor is used. The here essential property of the image acquisition unit 50 or the imager is that each pixel, the light at the position of the apertures 56 receives, includes a lock-in processing mechanism, and thus in addition to the amplitude signal A also indicates the phase angle φ of the alternating component of the photon current (see. 3 ). Depending on the type of lock-in-camera sensor used, the determination of the phase angle φ, however, also takes place only in the control and evaluation unit 70 ,

Finally, the intraoral camera includes 10 another control and evaluation unit 70 , in the 1 for purposes of better displayability outside the housing 12 is arranged. Of course, the control and evaluation unit 70 but also completely or partially within the housing 12 be arranged.

The control and evaluation unit 70 is for controlling the white light LEDs 20.n with the line 18 the light source 16 connected. Although such a connection on the real system is probably realized via a digital connection, a multiplex connection or the like, are in 3 For better understanding in each case individual lines 72.n to the corresponding white light LEDs 20.n shown. This generally applies to connections shown here.

On the input side is the control and evaluation unit 70 via a readout line 73 with the image converter 52 connected, in this readout line 73 the phase signal φ and the amplitude signal A, or at least raw signals which are suitable for calculating the phase and amplitude signal, is transmitted.

Furthermore, the control and evaluation unit 70 via a trigger line 74 with the image acquisition unit 50a connected, which transmits the zero point of the phase angle.

Not shown is a connection of the control and evaluation unit 70 to a commonly available display device, usually a computer with a monitor, the same time a calculation of the complete surface structure of the individual recordings, a storage and a metrological evaluation of the images.

The intraoral camera 10 works as follows:
With the help of the control and evaluation unit 70 become the white light LEDs 20.n over the wires 72.n acted upon with control signals Sn.

Preferably, drive signals are used here, which lead to a sinusoidal, triangular or rectangular light emission. In 3 For example, rectangular signals are shown which, as can be seen, are each phase-shifted relative to one another in such a way that each white-light LED 20.n approximately a phase shift of a maximum of φ / m is shifted to the next LED. φ here is the principle-related phase measurement range of the image sensor used which is usually at least π (180 ° or +/- 90 °) and a maximum of 2π (360 ° or +/- 180 °), and m is the number (see. 20.m in 2 ) of the LEDs necessary to represent the wavelength range necessary for the depth measurement range in the dispersion device. The total number n of the LEDs is thus greater than m, since the illumination hole grid aperture 32 also in dispersion direction 26 is extended, and also the marginal apertures 34 to transmit the entire wavelength range. The LED 20.x So could the same drive signal as the LED 20. (x + m) and possibly 20. (x + 2m) receive. The frequency of the drive signals Sn, however, is for all LEDs 20.n is the same and is considerably determined by the imaging sensor used. For ToF, or DAI sensors, it is in the range of 500 kHz to 40 MHz. For pOCT sensors the usual working range is between 1 kHz and 100 kHz. If the lock-in function is represented by one of the mentioned suitable CCD sensors, frequencies in the range can in principle be represented from 1 kHz to 50 MHz. If the phase signal is obtained using a CMOS sensor, the frequency is preferably at most a fraction of the maximum image readout rate of the sensor, that is less than 1 kHz. However, the frequency should be at least a multiple of the desired measuring point sampling rate, ie at least 100 Hz.

Due to the fact that every white light LED 20.n at a different angle to the grid 22 with respect to the dispersion direction 26 stands and the different spectral components of the incident light at the grid 22 each aperture is diffracted differently 34 the illumination hole screen 32 illuminated in time sequence by differently colored illumination light. As a result, the apertures appear 34 as individual light sources that change their color in chronological order.

Because at the apertures 34 the emergent light has a defined phase position for each light color and by the confocal discrimination in the recording aperture grid 54 only photons of a narrow band of wavelengths in the respective pixels 58 passes correlates the phase angle of the pixel signal of the image pickup unit 50a directly with the distance of the measuring point from the objective lens.

For the timing of any fundamental patterns are conceivable. However, the order within the sequence must be fixed in order to allow correlation with the data of the image acquisition unit.

About the beam splitter 42 and the hyperchromatic lens group 44 become the individual light sources on the surface structure 11 of the tooth 13 displayed.

Due to the increased chromatic longitudinal aberration of the hyperchromatic lens group 44 However, the individual light sources, depending on the current color at different distances from the lens group 44 focused. Therefore, only one individual light source becomes sharp in one measurement point at a time 49 on the surface texture 11 Shown when removing the surface texture 11 coincides with the focusing distance of the respective color. The other colors are smeared over a larger lateral area, leaving them only a small part through the optics 40 back, through the optionally associated orifice plate 56 , in the associated recording pixels 58 get the image converter.

For example, in typical hypochromatic lens groups 44 , as in 3 shown, blue light B closer to the lens group focused as red light R. Runs the surface structure 11 so as in 3 as a dashed line 11 ' As shown, the illuminated spot of light appears sharp at that time, to which blue light B is incident from the single light source. Conversely, the dashed line would be shown 11 '' the time of the red light R decisive.

That of the surface structure 11 backscattered light of a sharp measuring point 49 then falls again as far as possible in the optics 40 the intraoral camera 10 that the measuring point 49 due to the beam splitter cube 42 in the associated aperture 56 or the associated recording pixel 58 the image acquisition unit 50a is shown.

To the required depth resolution perpendicular to the surface structure 11 to get, now comes the lock-in-camera sensor as an image converter 52 for use.

As with the solid line 11 in 3 is recognizable, has the surface texture 11 in the measuring point 49 which is the pixel under consideration 58 of the lock-in-camera sensor, a distance to the optics 40 in which the chromatic confocal condition is most likely to be satisfied for the red light color, which for that lateral measuring point position is due to the diffraction 1 , Order at the grid 22 only in the light beam L2 from the LED 20.2 can come. The detected light signal in the pixel 58 thus has the phase of the LED 20.2 , For the measuring point 49 ' On the other hand, this light color originates from the light beam L1 from the light-emitting diode 20.1 , and thus carries its phase signal. Would be the surface 11 however, by the line 11 ' represented, then the phase signal of the pixel 58 through the light beam L2 + m from the LED 20.m + 2 determined, since only the blue light component is diffracted by the lattice in 1st order in this direction, and the chromatic confocal condition is met in the point B. At first, theoretically only the measurement of m depth steps seems to be possible by this mode of operation, since the light beams L which cover the depth measurement range are indeed only from LEDs which lie in the angular range of the 1st diffraction order between the light wavelengths of the confocal depth range of the hyperchromatic lens group. Due to the finite smallness of the panels 34 and 56 However, there is a smooth transition between these theoretical depths. This becomes the case of the measuring point 49 also a small amount of light from the LED 20.3 over the light beam L3 for illuminating the pixel 58 contribute, because even this light even over the same diffraction angle at the edge of the aperture 34 another identical wavelength to the measuring point 49 sends. In addition, due to the conical shape of the colored light rays and the finite smallness of the measuring aperture 56 also still a significant proportion of the light with adjacent wavelengths in the recording pixels 58 , This light comes then also to a small extent from the neighboring LEDs. The behind the orifice plate 56 , or in the recording pixel 58 measured phase thus corresponds to a superposition of the phase signal from at least 2 adjacent LEDs of the LED line 18 and changes continuously with the distance of the measuring point 49 ,

Analogous to this is in 7 represented as the light sources ( 16 . 16a . 16b . 16c ) for operation with superimposed pulsed light flashes by the signals S _n , S _m, for example, can be controlled to a timepix detector as an image sensor 52 in the image capture unit 50a to use. In 7 is a single recording pixel 58 of the timepix detector (T-pixel). The incoming superimposed flash of light carries the depth information in its relative arrival time T. The photodiode PD and the amplifier V of the T-pixel generate a voltage pulse U _P which is compared in the comparator K with the threshold voltage U _S. The counter started by the trigger signal S _T counts the pulses of the time signal S _Z until it is stopped by the comparator signal U _K and thus supplies as a result the relative arrival time T of the superposed flash of light. If the T-pixel may not simultaneously detect the amplitude or the time when the threshold voltage U _P is undershot, it may be necessary to carry out a second measuring operation with reversed impulse sequence of the signals S _n , S _m so as not to affect the depth measurement too much by amplitude differences in FIG the measuring points 49 distorted light flashes. Preferably, the control pulses S _n , S _{m are} sinusoidal and shifted to the nearest neighbor by 2 / 3π in their phase. With suitable dimensioning of the panels 34 and 54 and with suitable positioning and dimensioning of the LEDs, as well as the use of symmetrical control, it is then also possible to determine distances which do not correspond to an exact focus position.

4 shows another chromatically modulated light source at the in 1 illustrated intraoral 3D camera 10 can be used. In this light source 16b becomes a concentrator 27 used only necessary to those already in the light sources 21.1 to 21.3 to merge a limited, different colors, wide, wavelength band limited color spectrum and centrally through the apertures 34 of the aperture grid 32 respectively. The concentrator 27 is then preferably a dispersive element and designed here as a prism. For deflecting and extending the beam path is preferably still a deflection mirror 19 and a panel 23 to suppress the direct in the direction of radiant light installed. The concentrator 27 , as well as the deflection mirror 19 and the aperture 23 can be omitted altogether if the light sources 21.1 to 21.m Extremely small, so that the image of each light source is almost completely through the apertures 34 into each associated measuring point, and back through the apertures 56 enters the image recording unit. Such a light source 16c is in 5 shown. Preferably, in the light sources 16b and 16c only a few, preferably between 3 and 7, individual light sources 21.1 to 21.3 respectively. 21.4 used, whose wavelength spectrum have at least one sinusoidal or triangular-like shape. In addition, these wavelength spectra must overlap each other up to approximately their mean wavelength. In 6 For example, those preferably used wavelength _spectra I _{21.1 (λ)} to I _{21.4 (λ)} for a light source 16b or 16c with 4 individual light sources and the associated control signals S ₁ to S ₄ shown. These individual light sources are also controlled by a periodic signal of the same frequency, which leads to a sinusoidal or triangular-like photon intensity modulation. The modulation is also phase-shifted, as described in the embodiment 1. According to the smaller number of z. B. m = 4 light sources 21.1 to 21.4 is the shift angle of the signals S ₁ to S ₄ significantly larger. It preferably has the value 2π / (m + 1). That through the apertures 34 the illumination hole screen 32 reaching light is then in principle almost white light, since all wavelengths are emitted over a drive period. Exactly considered, however, each wavelength has its own phase position due to superposition of the light sources involved, which behaves continuously to the wavelength.

That in the image acquisition unit 50 Signal detected now has a phase determined solely by superposition of the narrow color components filtered by the confocal condition from at least 2 of the overlapping spectra _{21.m (λ)} , which also correlates with the depth of the measurement point as described in Example 1.

A single light source of the embodiment 2 with a preferably used wavelength _spectrum I _{21.m (λ)} , for example, from a colored LED, which already emits a sufficient wavelength band, or, for example, from a blue emitting LED or laser diode with a downstream fluorescent dye which emits a corresponding wavelength band, or from a broadband emitting white LED with a downstream color filter, which passes a corresponding wavelength band can be produced.

As with the available wavelength spectra of light emitting diodes and the available Floureszenzfarbstoffen, and also with color filters in practice, however, not every wavelength spectrum can be represented, and also the objects could not equally well backscatter each wavelength, especially with such light sources 21.m , poorly measurable zones or gaps in the depth measuring range arise. To remedy this deficiency is in the chromatic lens group 44 an electrically adjustable lens 45 inserted, which by a corresponding control by the control unit 70 causes a proper shift of the depth measuring range. Preferably, in this case, a stage control, which with the image recording cycle of the camera sensor 52 synchronized is used. However, it is also possible the lens 45 However, the control frequency of the light sources must be controlled periodically 20.n . 21.m an integer multiple this periodic signal amount.

For detection, the lock-in-camera sensor principles can also be used here. A pulse-shaped modulation of the photon intensity is also possible in principle here, but not meaningful, since then at least also the intensity ratios of the incoming photon stream would have to be evaluated.

In 5 and 6 However, for the embodiment 2 is an image pickup unit 50b shown in which a common camera sensor 52 can be used. For temporal discrimination in the measuring orifice 56 incoming photon currents are the measuring apertures on a corresponding microlens array 80 which by a drive 82 , Preferably a piezoelectric vibrator or a magnetic oscillating system, in a circular swinging motion 88 is placed on the ordinary camera sensor 52 displayed. At a preferably used slightly magnifying magnification of about 1: 1 to 1: 2 is formed on the camera sensor 52 at each measuring point position a small annular exposure 89 whose radius is about twice to 3 times as large as the swinging motion 88 of the microlens array 80 , The frequency for the light source signals S ₁ to S _m must be an integer multiple of the oscillation frequency of the microlens array S _ML . The annular exposure 89 is then a modulation according to the phase of the incoming confocal discriminated light color imprinted. From the angle φ of this ring pattern can thus determine the depth of the associated measuring point. Im in 5 In the example shown, the control frequency for the light source signals S was twice the frequency of the oscillation frequency of the microlens array S _ML .

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102006007172 B4 [0002, 0009]

Claims (18)

Contraption ( 10 ) for the three-dimensional detection of a surface structure ( 11 ; 11 '; 11 ''; 11 ''' ), in particular for detecting teeth ( 13 ), a) with a chromatic-confocal optic ( 40 . 42 . 44 ), which has an optical axis ( 41 ) and b) with a light source ( 16 ; 16a ; 16b ; 16c ), which is a raster from lateral to the optical axis ( 41 ) spaced individual light sources ( 34 ), c) the optics ( 40 . 42 . 44 ) the individual light sources ( 34 ) on the surface structure ( 11 ; 11 '; 11 ''; 11 ''' ) shows that there is a grid of measuring points ( 49 ), which are laterally spaced from each other, d) wherein an image pickup unit ( 50 ; 50a ; 50b ) optically equivalent to the individual light sources ( 34 ) is arranged around the grid of measuring points ( 49 ), characterized in that e) a control and evaluation device ( 70 ), which is designed to: - illuminate the light source ( 16 ; 16a ; 16b ; 16c ) in such a way that the individual light sources ( 34 ) emit light with different spectral composition in chronological order, and - with the aid of the image recording unit ( 50 ; 50a ; 50b ) the grid of measuring points ( 49 ) correlates with the time sequence of the individual light sources ( 34 ) to detect emitted light. Device according to claim 1, characterized in that the light source ( 16a ) in front of a raster element ( 30 . 32 ) for generating the single light source ( 34 ) a dispersive element ( 22 ) with a dispersion direction ( 26 ) and at least one at least partially in the dispersion direction ( 26 ) aligned line ( 18 ) of individually controllable broadband light sources ( 20.n ), so that when driving the broadband light sources ( 20.n ) on the grid element ( 30 . 32 ) successively light with different wavelengths falls. Device according to claim 1, characterized in that the light source ( 16b . 16c ) in front of a raster element ( 30 . 32 ) and a concentrator ( 19 . 27 . 28 ) for generating the individual light sources ( 34 ) at least two broadband, individually controllable light sources ( 21.m ) whose wavelength spectrum in each case has a triangular-like intensity distribution which at least partially overlap, so that a periodic phase-offset sinusoidal activation of the controllable light sources results in a periodic spectrum shift of the light of the individual light sources ( 34 ) by superposition of the light of the broadband, individually controllable light sources ( 21.m ) he follows. Device according to one of the preceding claims, characterized in that the chromatic optics ( 40 ) polarizing components ( 42 . 43 . 47 . 53 ), incorporated in such a way that substantially only that of the surface 11 scattered, and from the light source ( 16 . 16a . 16b ) into the image acquisition unit ( 50 . 50a . 50b ). Device according to one of the preceding claims, characterized in that the chromatic lens group ( 44 ) is a hyperchromatic lens group. Device according to one of the preceding claims, characterized in that the chromatic lens group ( 44 ) an electromechanically adjustable lens ( 45 ), designed such that gaps in the depth measuring range caused by insufficient wavelength bands of the light sources ( 20.n . 21.m ) to be bridged. Apparatus according to claim 2 and 3, characterized in that the individually controllable light sources ( 20.n . 21.m ) can be controlled in any but fixed order. Device according to one of the preceding claims, characterized in that the individually controllable broadband light sources ( 20.n . 21.m ) are driven at the same frequency of an almost temporally symmetrical light-dark control signal, the control signals of the light sources ( 20.n . 21.m ) are out of phase with each other. Device according to one of the preceding claims, characterized in that the electromechanically adjustable lens ( 45 ) is a liquid lens, which by the control unit with the light sources ( 20.n . 21.m ) is correlated. Device according to one of the preceding claims, characterized in that via a synchronization line ( 74 ) the control device ( 70 ) and the image acquisition unit ( 50 . 50a ) are connected to the individually controllable broadband light sources ( 20.n . 21.m ) to synchronize with the sampling frequency of the image pickup unit. Device according to one of Claims 1 to 10, characterized in that the image recording unit has a recording aperture ( 54 ) with optically equivalent to the light sources ( 34 ) arranged pinhole ( 56 ), which is guided by an array of optical fibers to an arbitrary arrangement of the detector pixels (FIG. 58 ) redirects. Device according to one of the preceding claims, characterized in that the image recording unit a receiving aperture ( 54 ) with optically equivalent to the light sources ( 34 ) arranged pinhole ( 56 ), which is guided by a laterally movable microlens array ( 80 ) on any image sensor ( 52 ), wherein the laterally movable microlens array ( 80 ) at least in one direction by a drive ( 82 ) into a periodic movement ( 88 ), wherein the drive frequency (Sn, Sm) of the light sources ( 20.n . 21.m ) is an integer multiple of the drive frequency (S _ML ) of the microlens array, and the drive signals of the light sources (Sn, Sm) with the drive signal (S _ML ) of the microlens array by the control unit ( 70 ) is correlated. Device according to one of claims 1 to 10, characterized in that the image recording unit ( 50 . 50a ) comprises a fast-lock-in-camera sensor (pOCT sensor), in which each pixel ( 58 ) outputs an amplitude signal (A) and a phase signal (φ) of the incoming photon current, which directly with a lock-in evaluation circuit in the pixel 58 is won. Device according to one of claims 1 to 10, characterized in that the image recording unit ( 50 . 50a ) comprises a time-of-flight camera sensor (TOF or DAI sensor) in which each pixel ( 58 ) outputs a plurality of amplitude signals from which the phase signal of the incoming photon current through the control unit ( 70 ) can be calculated. Device according to one of claims 1 to 12, characterized in that the image recording unit ( 50 . 50a . 50b ) comprises a charge-coupled camera sensor (FF-CCD, FT-CCD, TDI sensor), in which the phase signal of the measuring points by evaluation of the amplitude image which by a special control ( 74 ) of the charge coupling was calculated by the control unit. Device according to one of claims 1 to 11, characterized in that the image recording unit ( 50 . 50a . 50b ) comprises a CMOS camera sensor, in which the phase signal of the measuring points by evaluation of the amplitude image which by fast, exclusive reading of the through the measuring points 56 pixel obtained was calculated by the control unit. Device according to one of claims 1 to 10 and 12, characterized in that the image recording unit ( 50b ) comprises a CMOS camera sensor, in which the phase signal of the measuring points by evaluation of the amplitude image ( 89 ) which by a lateral movement ( 88 ) of the microlens array ( 80 ) is calculated by the control unit. Device according to one of claims 1 to 11, characterized in that the image recording unit ( 50 . 50a . 50b ) comprises a pulse-detecting radiation image sensor (time-pixel detector), in which the arrival time of a pulse is measured in each pixel, for which short, intensive control signals (S _m , S _m ) which preferably sinusoidal course of light flashes generate, and which further preferably to the next Neighbor each shifted by 2 / 3π in their phase, are used, and the depth information from the arrival times of the flashes of light is calculated by the control unit.

Images