Method and system for generating a three-dimensional image of at least a partial section of a row of teeth, computer program product and mirror element

Abstract

The invention relates to a method and system for generating a three-dimensional image (A, A2) of at least a partial section of a row of teeth (Z) with a stereo camera system (1) of a surgical microscope (2) and a computer program product, wherein the stereo camera system (1) comprises a first image capture device (5a) and a further image capture device (5b), wherein a. in an image (I5a) generated by the first image capture device (5a) and an image (I5b) generated by the further image capture device (5b), a mirror image (S5a, S5b) is detected, which is provided by a mirror element (3) arranged in a detection area (4) of the stereo camera system (1) such that a mirror image of at least one partial section of the row of teeth (Z) can be imaged by the first and further image capture device (5a, 5b), b. a pose (P) of the mirror element (3) is determined, c. the three-dimensional image (A, A2) is determined as a function of at least the imaged mirror images (S5a, S5b) and the pose (P) of the mirror element (3).

Description

[0001] The invention relates to a method and a system for generating a three-dimensional image of at least a partial section of a row of teeth using a stereo camera system of a surgical microscope, as well as a computer program product and a mirror element for generating the three-dimensional image.

[0002] In dentistry, especially during restorative treatment with implants or orthodontic treatment, dental impressions are often taken. These serve, for example, to determine the bite position and / or jaw status, particularly for documenting this at various points in time during treatment. Dental impressions are also important as a basis for the planning and fabrication of implants and braces / aligners.

[0003] As is well known, to make a dental impression a malleable mass is pressed onto the row of teeth. After a waiting time of usually a few minutes the then dried mass can be removed from the row of teeth to provide a negative model of the row of teeth. This negative can then be used to create a model of the row of teeth, e.g. a plaster model, which can be used for the subsequent applications described above. The problem is that when the mass is removed from the row of teeth slight changes can occur in the negative, which reduces the precision of the model created. This deterioration may require the impression to be repeated. The time required to make the impression, in particular the waiting time explained above, and the material consumption are also generally undesirable aspects.

[0004] Also known are so-called intraoral scanners, which are used as an alternative to the described process of creating an impression with a moldable material. Such intraoral scanners use various measurement methods to create a reconstruction of the surface of a row of teeth and also a digital model, which can then serve as the basis for the described applications. For example, it is conceivable that a dental impression could be created based on the model using additive manufacturing processes, particularly 3D printers. An implant could also be produced in this way, which could then be implanted promptly.

[0005] Intraoral scanners typically use sensors to generate signals based on a specific physical measurement principle, which are then processed to reconstruct or create a model. Common measurement principles used by intraoral scanners include confocal laser scanning, triangulation, or tactile measurement. To create the most complete and accurate model possible, it is necessary to capture all sides of a row of teeth or a tooth using the appropriate sensor(s). A disadvantage of conventional intraoral scanners is that a sensor carrier with the appropriate sensor(s) must be inserted into the mouth to capture the back side of a row of teeth, i.e., the side facing the inside of the mouth.This insertion can be very uncomfortable for the patient, as it may result in uncomfortable contact with the lips or other areas of the mouth. The potentially required deep insertion into the mouth can also be uncomfortable for the patient. Another disadvantage is that the sensor carrier, or the part of the sensor carrier inserted into the mouth, must be cleaned after use, which undesirably delays reuse.

[0006] For the person taking the dental impression, such as the dentist or orthodontist, such an intraoral scanner represents an additional device in the treatment room, which, due to its size, adversely reduces the available space in the treatment room. Furthermore, it can be difficult to integrate such an additional device into established processes, such as the process for taking a dental impression.

[0007] Surgical microscopes are also known. Such surgical microscopes are used by a user during treatment to provide a particularly magnified image of a treatment area, in particular of the site. So-called stereo surgical microscopes generally comprise two separate optical channels for beam guidance and can provide the user with an impression of depth of the examination area. For this purpose, the beams guided in the two channels can be captured by the user's eyes via eyepieces. Digital surgical microscopes alternatively or additionally comprise two image capture devices, each of which captures the beams in one of the optical channels to generate an image. Based on the two images, which are also referred to below as corresponding images, a three-dimensional image is then provided to the user via a suitable display device.Furthermore, other surgical microscopes are also known that can optically capture depth information and provide three-dimensional images of a treatment area. For this purpose, optical acquisition systems that enable interferometry, triangulation, time-of-flight (TOF), or microlens array-based provision of depth information can also be used in the surgical microscope, particularly as an alternative to a stereoscopic system.

[0008] To ensure correct representation, precise calibration of the stereo camera system is required. Known calibration methods are used to determine intrinsic and extrinsic camera parameters, which are then used by image processing to ensure correct representation. Intrinsic camera parameters describe parameters that affect the respective camera / image capture device itself, such as its distortion. Extrinsic camera parameters describe, in particular, a spatial relationship between the image capture devices and thus the camera images. Such intrinsic and extrinsic camera parameters are known to those skilled in the art.

[0009] Known from the prior art is DE 10 2020 133 627 A1, which discloses a method and an intraoral scanner for detecting the topography of the surface of a translucent, in particular dental, object.

[0010] Also known is DE 10 2019 008 510 A1, which also discloses an intraoral scanner, in particular for three-dimensional scanning of a dentate or edentulous upper or lower jaw, including jaw parts, in connection with implant prosthetics.

[0011] Also known is DE 10 2016 121 687 A1, which also discloses an intraoral scanner for digital dental impressions in the dental field and a method for producing a digital dental impression using an intraoral scanner.

[0012] Also known is EP 3 689 295 A1, which discloses a dental observation device using a so-called dental microscope.

[0013] Against this background, the technical problem arises of creating a method and a system for generating a three-dimensional image of at least a partial section of a dental arch, as well as a computer program product and a mirror element for generating this three-dimensional image, which enable the three-dimensional image to be generated quickly, accurately, and in a way that is comfortable for the patient. In addition, improved integration into existing application processes and treatment rooms should be enabled, which in turn reduces the deployment costs for such a system.

[0014] The solution to the technical problem is achieved by the subject matter having the features of the independent claims. Further advantageous embodiments of the invention are set forth in the dependent claims.

[0015] A method is proposed for generating a three-dimensional image of at least a partial section of a dental arch using a stereo camera system of a surgical microscope. The surgical microscope can be used to magnify examination objects or areas, particularly in medical applications. Thus, in particular, partial areas of the mouth or partial areas within the mouth can be displayed.

[0016] The surgical microscope comprises a stereo camera system with a first image capture device and a further image capture device. These can each be designed to generate a two-dimensional image. The images can be generated with a predetermined number of pixels and thus resolution. An image sensor of an image capture device can be, for example, a CMOS sensor or a CCD sensor. Of course, other sensor types can also be used. As explained above, the surgical microscope can comprise two optically separate beam paths, wherein the first image capture device is arranged and / or designed such that an image can be generated based on the beams guided in a first beam path.The additional image capture device can be arranged and / or configured such that a further image can be generated based on the rays guided in the additional beam path. The images can, in particular, be generated simultaneously. Furthermore, the images generated by the first and additional image capture devices, which are then used to generate the three-dimensional image, can be referred to as corresponding images. Also described is a surgical microscope which, in addition to or as an alternative to the stereo camera system, uses a further optical capture system to generate the three-dimensional image, in particular to provide depth information. Such optical capture systems have already been described in the introduction.

[0017] Furthermore, the surgical microscope can comprise at least one optical element for beam guidance and / or shaping, which can be designed in particular as a lens element. The at least one optical element can serve, for example, to generate a magnified image. Optical properties of the surgical microscope, such as magnification, focal point, zoom, exposure time, and the size of a detection area, can be adjustable.

[0018] The stereo camera system can be a calibrated stereo camera system. In particular, the previously explained intrinsic and extrinsic parameters of the stereo camera system can thus be predetermined, in particular via a calibration method known to those skilled in the art. Preferably, the aforementioned parameters are determined for all or predetermined operating states of the surgical microscope, with an operating state being characterized by the set (adjustable) parameters of the surgical microscope (e.g., zoom, focal point, detection range).

[0019] It is possible for the surgical microscope to additionally comprise at least one eyepiece through or into which a user can look to visually capture the image generated by the surgical microscope. In particular, the user can also capture the examination area three-dimensionally through the eyepiece. The surgical microscope can comprise at least one objective or objective system, which includes at least one optical element for beam guidance and / or beam shaping. The eyepiece can be or become optically connected to the objective.

[0020] The surgical microscope can be part of a microscopy system, which can include not only the surgical microscope but also a stand for holding the surgical microscope. The stand can be designed in such a way that it enables movement of the surgical microscope in space, in particular with at least one degree of freedom, preferably with six degrees of freedom, where a degree of freedom can be a translational or a rotational degree of freedom. The degrees of freedom can relate to a reference coordinate system. A vertical axis (z-axis) of this reference coordinate system can be oriented parallel to the gravitational force and opposite to it. A longitudinal axis (x-axis) and a transverse axis (y-axis) of this reference coordinate system can span a plane that is oriented perpendicular to the vertical axis. Furthermore, the longitudinal and transverse axes can also be oriented orthogonally to one another.Furthermore, the stand can comprise at least one drive device for moving the surgical microscope, for example, a servomotor. The stand can also comprise means for force / torque transmission, e.g., gear and / or clutch units. The surgical microscope can thus be movably mounted or supported. This allows a user, among other things, to change the pose, i.e., the position and / or orientation, of the surgical microscope, for example, to change the viewing angle of an examination area or to view other examination areas.

[0021] The surgical microscope can in particular be a dental surgical microscope which is designed to generate images in dental applications.

[0022] A partial section of a row of teeth can, in particular, comprise at least one tooth or part of a tooth. For the purposes of this invention, a tooth can also refer to a dental prosthesis.

[0023] According to the invention, the proposed method comprises the following steps: In a first step, a mirror image is detected in an image generated by the first image capture device of the stereo camera system and in an image generated by the further image capture device of the stereo camera system. This mirror image is provided by a mirror element arranged in a detection area of the stereo camera system.

[0024] In particular, the aforementioned mirror element can be arranged in the detection range of the stereo camera system prior to the detection of the imaged mirror image, whereby a mirror image of at least one partial section of the row of teeth can be imaged by the first and further image capture devices. Furthermore, images can be generated by the image capture devices prior to detection, in particular corresponding images, which are then used to determine the three-dimensional image. This will be explained in more detail below.

[0025] The mirror element can in particular be a dental mirror or a part thereof. The mirror element can in particular have or form a mirror surface which reflects radiation. The image of the mirror surface generated by an image capture device, i.e. the imaged mirror image, is generated by detecting this reflected radiation. The mirror image generated / provided by the mirror element is therefore perceived by detecting the reflected radiation. By detecting by the image capture devices, an imaged mirror image is generated. In addition to the imaged mirror image, the image generated by an image capture device can, however, also comprise further regions which do not depict the mirror image. In other words, the mirror image can be imaged in a partial region of the image generated by an image capture device. At least one partial section of the row of teeth is imaged in the imaged mirror image.The mirror element can further comprise a frame portion that encloses the mirror surface. The mirror element can also comprise a handle portion so that a user can position the mirror element in the room. The mirror element is preferably a mirror element with a non-curved mirror surface. Preferably, the mirror surface is round. However, it is also conceivable to use square mirror surfaces.

[0026] In a second step, a pose of the mirror element, in particular of the mirror surface, is determined. The pose can include a translational and a rotational component. For example, a position of a reference point of the mirror element, for example a center point of the mirror surface, and an orientation of the mirror element, for example the orientation of a normal of the mirror surface, can be determined as the pose of the mirror element. Of course, positions of multiple reference points or orientations of multiple subsections of the mirror element can also be determined as the pose, particularly in the case of a curved mirror surface. Example methods for determining the pose are explained in more detail below. The pose can be determined in a reference coordinate system. This can be, for example, a reference coordinate system of the stereo camera system, the surgical microscope, or the previously explained reference coordinate system.

[0027] In a third step, the three-dimensional image is determined depending on at least the imaged mirror images detected in the (corresponding) images generated by the two image capture devices, and the pose of the mirror element. For this purpose, a stereo reconstruction method can be used, for example, with the imaged mirror images forming input images for this method. Such methods are known to those skilled in the art. In particular, corresponding pixels in the two input images can be determined in such methods. Such corresponding pixels or pixel sets can be determined, for example, using a feature matching method. Corresponding methods and features are known to those skilled in the art. Exemplary features are so-called SIFT features, i.e., features (for) a scale-invariant feature transformation.Of course, other methods can also be used for determination, for example variational methods or AI-based methods. Three-dimensional coordinates can then be determined in the previously explained reference coordinate system for an object point or section that is mapped into corresponding pixels or pixel sets. This can also be referred to as reconstruction, whereby a corresponding reconstruction method is carried out depending on the pose of the mirror element. In particular, at least one method step of the reconstruction method can be carried out depending on the pose. In particular, the pose can be represented by at least one parameter, whereby the at least one method step is carried out depending on the parameter or takes the parameter into account during implementation.Preferably, a stereo triangulation reconstruction method is performed to determine the three-dimensional image. Stereo triangulation reconstruction methods are known to those skilled in the art. In this case, a projection matrix used in the reconstruction, which describes a perspective transformation of three-dimensional object coordinates in a reference coordinate system into two-dimensional image coordinates, can be determined depending on the pose of the mirror element and (known) reflection laws. In other words, the pose of the mirror element influences the projection matrix of both image capture devices and thus also the stereo triangulation reconstruction performed based on or depending on these projection matrices. For example, a so-called homogeneous solution method or a so-called inhomogeneous solution method can be used to determine the three-dimensional coordinates.It is possible that, before determining the three-dimensional image, a rectification procedure is carried out to compensate for or eliminate, in particular, non-linear distortions of the images.

[0028] The proposed method advantageously enables a simple, accurate and patient-friendly generation of a three-dimensional image, in particular an intraoral scan, since, in comparison to the existing methods explained above, usually only a small mirror element needs to be arranged in the mouth in order to capture a partial section of a row of teeth and to image it three-dimensionally.

[0029] The proposed method is used in particular for generating a three-dimensional image of the back of a row of teeth and / or of biting or chewing surfaces of the row of teeth. The back of the row of teeth can in particular refer to a side of the row of teeth facing the inside of the mouth. A three-dimensional image of the front of a row of teeth or of a partial section of this row of teeth can be determined without the explained first and second steps. In particular, the three-dimensional image can thus be determined as a function of corresponding images generated by the image capture devices, wherein known methods for stereo reconstruction can be applied in particular. The image capture devices for generating the image can be arranged outside the oral cavity, i.e., extraorally. The method can thus also be referred to as an extraoral scan.

[0030] The three-dimensional image can be created, for example, in the form of a CAD data set, for example, in STL format. Such a three-dimensional image can then be visualized, for example, by output via a display device. The three-dimensional image can also be used in further processes, for example, in CAD / CAM processes, e.g., to create a dental impression.

[0031] In a further embodiment, the three-dimensional image is additionally determined as a function of at least one optical property of the mirror element. The at least one optical property can be predetermined. Alternatively, the at least one optical property can also be determined in a further step of the proposed method, in particular image-based, i.e. by evaluating at least one image of the mirror element. An optical property of the mirror element can in particular be a magnification or reduction property. Of course, other optical properties of the mirror element that affect the representation of the mirror image can also be taken into account. For example, the optical property can be represented by the projection matrix explained. This advantageously results in a high level of accuracy in the generated three-dimensional image.

[0032] Alternatively or cumulatively, the three-dimensional image is additionally determined depending on at least one imaging property of the surgical microscope. An imaging property can be, in particular, a set magnification (zoom), a set focal point, or another imaging property that affects the imaged mirror image. This imaging property can also be represented by the projection matrix described above. This also advantageously results in the generation of a three-dimensional image with high accuracy.

[0033] In a further embodiment, (corresponding) images of virtual image capture devices are determined at least as a function of the pose of the mirror element, wherein the at least one corresponding section of the row of teeth reflected in a catadioptric system is imaged in an image of such a virtual image capture device. Furthermore, the three-dimensional image is determined as a function of the images, in particular of the corresponding section in the respective images, of the virtual image capture devices.

[0034] A catadioptric system refers to an optical transmission system that includes at least the mirror element and optical elements of an image capture device, e.g., an objective. In particular, the catadioptric system can include optical elements of the objective of the surgical microscope. A beam path through the catadioptric system can be determined based on known or determinable optical properties of the optical elements in the catadioptric system, as well as the pose of the mirror element and known optical principles.

[0035] The virtual image capture device is a mathematical or physical model of an image capture device that can be evaluated using computer support. Depending on the model, a virtual image generated by the virtual image capture device can be generated or calculated, in particular by a computer-implemented calculation of the pixels. This virtual image depends, among other things, on parameters of the (modeled) image capture device and a pose of the (modeled) image capture device. These parameters, in particular extrinsic and / or intrinsic parameters, of the virtual image capture device can be dependent on the optical properties of the mirror element. If the mirror surface is, for example, non-curved and has no magnification properties, the intrinsic parameters of the virtual image capture device can be the same as the intrinsic parameters of the modeled image capture device.

[0036] In particular, depending on the pose of the mirror element, a pose of the virtual image capture device can be determined such that the virtual image of the virtual image capture device in this pose depicts the non-reflected section of the row of teeth that is mirrored by the catadioptric system and thus also by the mirror element. This refers to the corresponding section. In addition to the pose of the mirror element, the generation of such a virtual image also depends on the (other) properties of the catadioptric system, e.g., the zoom setting of a lens.

[0037] The three-dimensional image can then be determined using a method known to those skilled in the art, depending on the virtual image, in particular the corresponding section. Exemplary methods have been explained previously. In such a method, properties of the catadioptric system, and thus also the pose of the mirror element, can be taken into account. In other words, the reflected image can be transformed into an unreflected image, and the three-dimensional image can then be generated based on the unreflected image. This advantageously results in a precise and computationally simple determination of the three-dimensional image, which can be generated, in particular, with high accuracy.

[0038] In a further embodiment, the pose is determined by evaluating at least one property of the imaged mirror image or the imaged mirror element (or a section thereof). The at least one property can be determined image-based, in particular by evaluating the image generated by the respective image capture device. In particular, the imaged mirror surface or the imaged mirror element can be recognized in such an image, for example by object recognition methods known to those skilled in the art. Object recognition methods can be segmentation methods, for example. For example, the imaged mirror surface, the imaged frame section, or an imaged handle section can be recognized image-based. It is possible, for example, to produce a section, e.g. a frame section, of the mirror element from material with predetermined optical properties, e.g.made of matte material, in particular to enable reliable detection of this section in the image. Alternatively, detection can also be achieved by a user selecting an image area in which the mirror image or the section to be detected is depicted, for example via a suitable input device.

[0039] A property of the imaged mirror image or the imaged mirror element can be a geometric property of the imaged mirror image, e.g., a dimensional property such as a dimensional size. A dimensional size can be a width, a height, a diameter, or another dimensional size. A property can also be a shape property, e.g., a geometric shape such as a circle, an ellipse, a rectangle, or another geometric shape. In particular, a shape factor can be determined that represents a relationship between an imaged shape and an actual shape, with the pose being determined as a function of the shape factor.

[0040] As previously explained, the pose of the mirror element can influence the imaging by the image capture devices. Thus, the pose can also influence how an actual property of the mirror image or mirror element is mapped into a property of the imaged mirror image or element. If a relationship between the actual property and the property of the imaged mirror image can be described by a transformation matrix that depends on the pose, the pose can be determined as a function of the actual property and the property of the imaged mirror image or imaged mirror element. The actual property can be known in advance, e.g., from a model, in particular a CAD model, of the mirror element, or it can be determinable.

[0041] If the mirror element or a part thereof, in particular the mirror surface, is circular, and the imaged mirror image is elliptical, the pose can be determined depending on the properties of the ellipse, for example, the orientation and lengths of the axes of the ellipse, as well as the known properties of the circular mirror element, such that the known properties are transformed into the properties of the imaged mirror image. If polygonal mirror elements, in particular equilateral polygonal mirror elements, are used, at least part of the pose can be determined via a ratio of the edge lengths in the image and the relative position of the edges in the image to one another.

[0042] The property can also be a pose of the imaged mirror image or the imaged mirror element, or a part thereof, in the image coordinate system. The position can be determined, for example, as the position of a reference point, e.g., a geometric center. The orientation can be determined, for example, as the orientation of an axis of a reference section. If the mirror element includes a handle section, for example, this can be recognized in the image and its position and / or orientation determined. For example, the orientation of a longitudinal axis of the handle section can be determined.

[0043] If the pose is determined by evaluating at least one property of the imaged mirror image or the imaged mirror element, this advantageously results in a simple determination of the pose, since the images already generated can be evaluated to determine the pose.

[0044] Alternatively, the pose can be determined based on markers. For this purpose, the mirror element can have or form at least one marker element for determining the pose of the mirror element. Of course, it is also possible for the mirror element to have or form multiple markers, whereby the pose of the mirror element can be determined depending on a previously known relative position and / or dimensions of these markers.

[0045] The marker element can be an active or, preferably, a passive marker element. It can be designed to be detected by a detection device. The detection device can, in particular, be an image detection device. In this case, the marker element can therefore be an optically detectable marker element. It is conceivable, for example, for an optically detectable marker to have a predetermined pattern which enables the determination of the pose of the marker and thus of the mirror element. Such optically detectable patterns can, for example, be in the form of QR codes. Such optically detectable markers can also be reflective marker elements, which are designed to be reflective, for example, for radiation from a predetermined wavelength range, such as the infrared wavelength range.The image capture device for optically capturing the marker element can be an image capture device of the stereo camera system or a different image capture device. If the mirror element comprises multiple marker elements, the pose of the mirror element can also be determined depending on the relative position of the imaged marker elements in the image. The pose of the mirror element can also be determined at least partially by stereoscopically determining the pose of at least one marker element.

[0046] Using the image capture device and the marker element, a so-called monoscopic pose determination can be performed. The pose can be determined by evaluating a two-dimensional image, in particular, precisely one two-dimensional image from precisely one image capture device. In particular, the intensity values of pixels of the two-dimensional image can be evaluated to determine the position. Such methods for image-based position detection with precisely one image capture device and / or based on precisely one two-dimensional image are known to those skilled in the art. If a stereo camera system is used, the pose can also be determined by evaluating the corresponding images from the image capture devices.

[0047] In particular, the pose can be determined using optical tracking methods, in which an image capture device is used to determine the pose. Such optical tracking can be marker-based tracking, in which special visually or optically detectable marker elements, such as QR codes or other optical patterns, are used to determine the pose.

[0048] Alternatively, in particular in the previously explained determination of the pose by evaluating at least one property of the imaged mirror image or the imaged mirror element, markerless tracking methods can also be used, whereby in these methods features are recorded and used to determine the pose.

[0049] As an alternative to an optically detectable marker element, a marker element that can be detected in another way can also be used to determine the pose of the mirror element, for example, a marker element that can be detected magnetically, capacitively, inductively, or radio-based. For example, a marker element can be designed as an RFID tag.

[0050] It is also possible for the mirror element to comprise a pose sensor, for example, an initial sensor or a GNSS sensor, wherein the pose is determined based on output signals from such a sensor. In such an embodiment, the surgical microscope can comprise a receiving device for the output signals generated by the pose sensor or can be connected to such a receiving device.

[0051] Of course, it is also possible to determine the pose using a hybrid method, whereby such a hybrid method combines at least two of the previously explained methods for determining a pose.

[0052] In the case of marker-based pose determination, a very precise pose determination is advantageously achieved, which in turn leads to the generation of an accurate three-dimensional image.

[0053] A marker element can also be identifiable, in particular uniquely. For example, a pattern of an optically detectable marker can encode an identity of the marker. Thus, the marker element or the mirror element can be identified by detecting the marker element. Properties of the mirror element, in particular the optical properties explained above, can be assigned to the identity thus determined. This assignment, as well as the identity and the properties, can be stored, for example, in a memory device in a retrievable or readable manner.

[0054] This advantageously results in a simple determination of the optical properties of the mirror element.

[0055] In a further embodiment, for marker-based determination of the pose, at least one marker element is imaged by at least one image capture device of the stereo camera system or by a further image capture device. The at least one marker element is arranged on the mirror element or is formed by it. The pose is then determined depending on at least one property of the imaged marker element. This has already been explained above. The further image capture device can, in particular, be a tracking camera or an environment camera of a microscopy system, which is designed differently from the image capture devices of the stereo camera system. This tracking or environment camera can, in particular, serve for marker-based tracking of further instruments.In both cases, the integration of an optical determination of the pose when using the surgical microscope or a microscopy system with the surgical microscope is advantageously as simple as possible.

[0056] In a further embodiment, a focus position of the surgical microscope is adjusted depending on the pose of the mirror element. This makes it possible, in particular, to set a focus position to a point on the mirror surface or to a point that is no more than a predetermined distance from the mirror surface. The predetermined distance can be dependent, in particular, on a depth of field of the surgical microscope, in particular smaller than the depth of field. The depth of field is known in advance or can be determined. This makes it possible to achieve a high imaging quality for the mirror image, which in turn advantageously increases the accuracy of the generated three-dimensional image. In particular, this also makes it easier and more reliable to detect the imaged mirror image in the image of an image capture device.

[0057] In a further embodiment, the radiation captured to generate the images of the stereo camera system is filtered. In a preferred embodiment, the filtering is a polarization filter. However, other radiation filters can of course also be used. This advantageously suppresses unwanted reflections on a tooth surface in the image, which in turn increases the accuracy of the generated three-dimensional image. Polarization filtering enables the greatest possible or even complete suppression of reflections.

[0058] In a further embodiment, a filter element is arranged in an illumination beam path of the surgical microscope and / or in an imaging beam path. In particular, a filter element can be arranged in each imaging beam path. In both cases, a good structural integration of a filter element into the surgical microscope or into a microscopy system is achieved, enabling the generation of a three-dimensional image with high accuracy.

[0059] Alternatively or additionally, a filter element is arranged on the mirror element. For example, the filter element can be arranged on a mirror surface of the mirror element. This advantageously results in the fact that no additional filter element needs to be integrated into the surgical microscope or microscopy system to generate a three-dimensional image with high accuracy, since the mirror element provides the desired filter properties.

[0060] In a further embodiment, at least one partial section of the row of teeth is illuminated with radiation having predetermined radiation properties. Such radiation properties can, for example, be (a) predetermined wavelength(s) of the radiation used for illumination, predetermined intensities, or, in a preferred embodiment, predetermined polarization properties or other properties of the radiation used for illumination. This also advantageously reduces reflections on the tooth surface, which in turn has a beneficial effect on the accuracy of the generated three-dimensional image.

[0061] In a further embodiment, the images generated by the stereo camera system are filtered, with the three-dimensional image being determined as a function of at least the filtered images. In particular, filtering can be performed to reduce imaged unwanted reflections. However, filtering can also generally serve to improve image quality, which in turn advantageously increases the accuracy of the three-dimensional image. An exemplary method for filtering for reflection suppression is the so-called tone mapping method.

[0062] For example, it is possible to temporarily color the teeth during image acquisition. Color filtering can then be performed, reducing, for example, the color components in the image that differ from the color of the stain.

[0063] In a further embodiment, at least one quality measure of the three-dimensional image is determined. Furthermore, user information is generated if the quality measure is less than a predetermined threshold. In this case, the quality measure is selected such that it is proportional to the quality of the three-dimensional image. The user information can be output to the user via an output device, for example, via an optical or acoustic output device. This can be part of the surgical microscope or a microscopy system.

[0064] For example, it can be determined whether a point density in a predetermined section of the generated three-dimensional image is less than the predetermined threshold. If this is the case, user information can be generated, since it may be desirable to perform a more accurate reconstruction with a higher density, at least in this sub-area. In particular, the determined quality measure, e.g., the point density, can be generated as part of the user information.

[0065] If the quality measure is determined for a specific sub-area, information about the sub-area, in particular its pose, can also be generated as part of the user information.

[0066] This advantageously ensures that a high level of accuracy of the generated three-dimensional image is ensured, since in the event of an undesirably low quality, user information is generated and the process is, for example, carried out again, in particular for the partial area for which the quality measure was determined to be too low.

[0067] Further proposed is a system for generating a three-dimensional image of at least a partial section of a row of teeth, wherein the system comprises a stereo camera system and at least one evaluation device. The stereo camera system comprises a first and a further image capture device. The system is configured to carry out a method according to one of the embodiments described in this disclosure, in particular the steps: a) detecting a mirror image in an image generated by the first image capture device and in an image generated by the further image capture device, wherein the mirror image is provided by a mirror element which is arranged in a detection area of the stereo camera system and which reflects the at least one partial section of the row of teeth, b) Determination of a pose of the mirror element, c) Determination of the three-dimensional image depending at least on the imaged mirror images and the pose of the mirror element.

[0068] The evaluation device can be designed as a computing device or comprise such a device. A computing device, in turn, can comprise a microcontroller or an integrated circuit or be designed as such. The evaluation device can execute at least one, but preferably all, of steps a), b), and c).

[0069] The system can be part of a surgical microscope or a microscopy system, wherein the surgical microscope or the microscopy system can comprise the stereo camera system and the evaluation device. Furthermore, the system can comprise a detection device for detecting a marker element. Furthermore, the system can comprise a filter element for filtering the radiation used to generate the images of the stereo camera system. Furthermore, the system can comprise an illumination device for illuminating the partial section with predetermined radiation properties. Furthermore, the system can comprise an output device for user information.

[0070] The system advantageously enables the execution of a method according to one of the embodiments described in this disclosure with the advantages also already described.

[0071] In a further embodiment, the system comprises a mirror element for arrangement in the detection area of the stereo camera system.

[0072] Further proposed is a computer program product comprising a computer program, wherein the computer program comprises software means for executing one, several, or all steps, in particular the set of first step, second step, and third step, of a method according to one of the embodiments described in this disclosure, when the computer program is executed by or in a computer or in an automation system. In other words, the method can be a computer-implemented method.

[0073] Further proposed is a mirror element for generating a three-dimensional image of at least a partial section of a row of teeth. According to the invention, the mirror element has or forms at least one marker element for determining the pose of the mirror element. Alternatively or cumulatively, the mirror element has or forms at least one filter element for filtering the reflected radiation. This and corresponding advantages have already been explained above.

[0074] The invention is explained in more detail using exemplary embodiments. The figures show: Fig. 1 a schematic flow diagram of a method according to the invention, Fig. 2 a schematic flow diagram of a method according to the invention in a further embodiment, Fig. 3 a schematic block diagram of a system according to the invention, Fig. 4a a mirror element in a first pose, Fig. 4b that in Fig. 4a shown mirror element in another pose, Fig. 4c that in Fig. 4a shown mirror element in another pose, Fig. 4d that in Fig. 4a shown mirror element in another pose, Fig. 5 a schematic representation of a mirror element according to the invention, Fig. 6 a schematic block diagram of a virtual image capture device, Fig. 7a a schematic view of a stereoscopic recording of a row of teeth without a mirror element and Fig. 7b a schematic view of a stereoscopic recording of a row of teeth with a mirror element.

[0075] In the following, the same reference symbols designate elements with the same or similar technical features.

[0076] Fig. 1 shows a schematic flow diagram of a method according to the invention for generating a three-dimensional image A of at least a partial section of a row of teeth Z (see Fig. 3) with a stereo camera system 1 of a surgical microscope 2. Before a first step S1 of the method, a mirror element 3 is arranged in a detection area 4 of the stereo camera system 1 such that a mirror image of at least the partial section of the row of teeth Z can be imaged by a first image capture device 5a and by a further image capture device 5b of the stereo camera system 1 (arrangement step S1). After this arrangement, an image I5a, I5b is generated with the first image capture device 5a as well as with the further image capture device 5b, wherein these images I5a, I5b can be referred to as corresponding images (image generation step S1). These images are generated from different positions and / or with different orientations. Furthermore, they can be generated simultaneously or with a predetermined maximum time offset.In a first step S1 of the method, an imaged mirror image S5a is then detected in the image I5a generated by the first image capture device 5a, for example using a method for object recognition. The image I5a generated by the first image capture device 5a can, in addition to a partial area in which the mirror image provided by the mirror element 3 is imaged, comprise further partial areas which, for example, image other partial sections of the row of teeth Z, in particular sections not mirrored by the mirror element 3. Accordingly, an imaged mirror image S5b is also detected in the image I5b generated by the further image capture device 5b. The imaged mirror images S5a, S5b detected in this way form input variables for determining the three-dimensional image A. In a second step S2 of the method, a pose P, i.e. a position and / or orientation of the mirror element 3, is determined in a reference coordinate system.The reference coordinate system can be a reference coordinate system of the surgical microscope 2 or a reference coordinate system of the stereo camera system 1. Of course, other reference coordinate systems are also conceivable. In addition to the imaged mirror images S5a, S5b, the pose P represents another input variable for determining the three-dimensional image A.

[0077] In a third step S3, the three-dimensional image A is then determined depending on at least the imaged mirror images S5a, S5b and the pose P of the mirror element 3, for example, via or with a stereo reconstruction. This has already been explained above.

[0078] It should be mentioned that the Fig. The sequence of the first and second steps S1, S2 shown in Figure 1 is not mandatory. In particular, it is possible to determine the pose P simultaneously with or before the detection of the imaged mirror images S5a, S5b.

[0079] Furthermore, it is possible for the three-dimensional image A to be additionally determined as a function of at least one optical property of the mirror element 3. This optical property then forms a further input variable for the third step S3. The optical property can be predetermined. In this case, it is particularly possible to identify the mirror element, wherein the preferably unique identity of the mirror element 3 can be assigned to the at least one optical property. Depending on this assignment, the optical property can then be determined with the determined identity of the mirror element 3, for example, retrieved from a database that can be stored in a storage device and represents the assignment of identity to the at least one optical property. Such an identification step, which in Fig. 1 is not shown, can be performed prior to the third step S3. Preferably, an image-based identification of the mirror element can be performed, for which purpose at least one of the generated images I5a, I5b, in which the mirror element 3 is imaged, is evaluated. For example, but not necessarily, the identification can be performed simultaneously with the determination of the pose in the second step S2. The identification can be performed by evaluating at least one property of the imaged mirror element or marker-based. This will be explained in more detail below with regard to the determination of the pose P.

[0080] Alternatively or cumulatively, an imaging property of the surgical microscope 2, for example, a currently set zoom, can also be determined before the third step S3. This imaging property of the surgical microscope 2 can then form a further input variable for the third step S3. Thus, the three-dimensional image A can additionally be determined depending on the at least one optical property of the mirror element 3 and / or the at least one imaging property of the surgical microscope 2.

[0081] The determination of the pose P of the mirror element 3 in the second step S2 can be image-based, in particular by evaluating at least one property of the imaged mirror image S5a, S5b or of the imaged mirror element 3. The property can be a dimensional property. The property can also be a shape property. Such properties were explained previously. In particular, the properties in the image S5a, S5b can be determined and compared with previously known properties of the mirror image or the mirror element 3 in a reference position (see e.g. Fig. 4a), in particular with a computationally determinable image of the mirror image or of the mirror element 3 in this reference position. The pose P of the mirror element 3 can then be determined from a deviation between the actual property in the image and the previously determined property. For example, a transformation can be determined with which the image of the mirror image or of the mirror element in the reference position is transformed into the actual image, wherein this transformation contains information about the current pose P of the mirror element 3.

[0082] Alternatively, the pose P can be determined marker-based. For this purpose, marker elements, which in an exemplary embodiment are Fig. 5 as optically detectable marker elements 6. Such marker elements 6 can be active marker elements, i.e. marker elements that generate detectable signals while consuming energy, or passive marker elements that can be detected without energy consumption by the marker element. An example of a passive marker element is the Fig. 5, which are designed in the form of barcodes or optical patterns. Of course, optically detectable markers can also be active markers, which, for example, generate optically detectable signals while consuming energy. As an alternative to optically detectable marker elements 6, however, marker elements that can be detected in other forms, such as magnetically detectable marker elements, can also be used. A marker element can be arranged in or on the mirror element 3, in which case the pose P of the mirror element can be determined depending on the detected marker element.

[0083] If the at least one marker element is an optically detectable marker element 6, e.g. the one in Fig. 5, the at least one marker element can be imaged by at least one image capture device 5a, 5b of the stereo camera system 1 or by another (not shown) image capture device to determine the pose P, and the pose P can then be determined depending on at least one property of the imaged marker element. In this case, the determination of the pose P can only be carried out after the respective image 15a, 15b has been captured.

[0084] It is also conceivable that a focus position of the surgical microscope 2 is adjusted depending on the pose P of the mirror element 3. In this case, the generation of the images I5a, I5b, which are evaluated to determine the three-dimensional image A, can take place after the determination of the pose P. If the pose P is determined image-based, images can be generated before the determination of the pose P, which serve to determine the pose P, wherein the focus position is then adjusted depending on the pose P and then the images I5a, I5b are generated, which serve to detect the imaged mirror image S5a, S5b.

[0085] Fig. Figure 2 shows a schematic flow diagram of a method according to the invention in a further embodiment. In contrast to the Fig. 1, images I5a, I5b, i.e. corresponding images, of a front side 7 of the row of teeth Z are additionally generated in a further image generation step SBV, wherein, depending on these images I5a, I5b, a three-dimensional partial image A1 of the front side is generated in a reconstruction step SRV. Following this, the image shown in Fig. 1 is carried out, wherein the image A generated thereby is a partial image A2 of a rear side 8 of the row of teeth Z and of chewing surfaces 9 of the teeth of the row of teeth Z. The partial images A1, A2 are fused / combined in a fusion step FS to form a resulting image A of the row of teeth Z. For this purpose, in particular, corresponding points in the three-dimensional partial images A1, A2 can be detected in order to fuse the partial images A1, A2. For this purpose, it may be necessary to scale the structures imaged in at least one of the partial images A1, A2, in particular those in the partial image A2 of the rear side 8. Such scaling can be carried out in particular as a function of the distance between the mirror element 13 and the mirrored section of the row of teeth Z. This distance can be determined, in particular as a function of the pose P of the mirror element 13.The distance can also be determined depending on different focus positions of the stereo camera system 1, which will be explained in more detail below. The partial images A1, A2 and the resulting image A can preferably be provided in an STL data format, which advantageously enables their use in further processes, particularly in CAD / CAM processes. Likewise, the resulting three-dimensional image A or the partial images A1, A2 can be displayed to a user via a suitable display device. In this case, it is possible to overlay information, for example, color information, on the displayed image.

[0086] Fig. Figure 3 shows a schematic block diagram of a system according to the invention for generating a three-dimensional image of at least a partial section of a row of teeth Z, wherein the system comprises a stereo camera system 1 and at least one evaluation device 10. The system is configured to at least Fig. 1 and Fig. 2 to perform steps S1, S2, S3. Of course, the system can also be configured to Fig. 2. These steps, or at least parts thereof, can be performed by the evaluation device 10.

[0087] In Fig. Figure 3 schematically shows detection areas EB of the image capture devices 5a, 5b and optically separated beam paths 11a, 11b of the surgical microscope 2. Not shown is an illumination device of the surgical microscope 2, which can illuminate the row of teeth Z. The illumination device can generate radiation with predetermined radiation properties, in particular predetermined polarization properties. The radiation reflected by the row of teeth Z reaches the image sensors of the image capture devices 5a, 5b via the beam paths 11a, 11b, whereby an image 15a, 15b of the row of teeth Z can be generated. This image can be evaluated by the evaluation device 10.Also shown is a mirror element 3, which also reflects rays from the row of teeth Z, in particular its rear side 8. This reflected radiation also passes through the beam paths 11a, 11b to the image sensors, where it is imaged as a mirror image. This image can then be detected by the evaluation device 10 as an imaged mirror image.

[0088] It is possible for the radiation captured to generate the images I5a, I5b of the stereo camera system 1 to be filtered. This can be achieved by filter elements that are arranged, for example, in the beam paths 11a, 11b. A filter element can also be arranged in an illumination beam path of an illumination device (not shown) of the surgical microscope 1. A filter element can also be arranged on / at the mirror element 3. Such a filter element can, in particular, be a polarization filter element. Furthermore, it is possible for the evaluation device 10 to filter the image I5a, I5b generated by the image capture devices 5a, 5b, for example, to suppress reflections.

[0089] Fig. Figure 4a shows an image of a mirror element 3 in a reference pose. The mirror element 3 comprises a handle section 12 and a round mirror section 13, which in turn comprises a mirror surface 14. A reference point P3 of the mirror element 3 is, for example, a center point of this mirror surface 14. A mirror-fixed coordinate system with a longitudinal axis x3 and a transverse axis y3 as well as a vertical axis z3 is shown (see Fig. 4b).

[0090] Fig. 4b shows an image of the mirror element 3 in a pose P, which is established when the mirror element 3 is moved from the Fig. 4a is rotated around the longitudinal axis x3a. It can be seen that the Fig. 4a is then imaged as an ellipse. Depending on the orientation and the length of the major and minor axes of this ellipse, which can be detected, for example, using an object recognition method, it is then possible to determine the angle of rotation with which the mirror element 3 was rotated about the longitudinal axis x3a, whereby the current pose P of the Fig. 4b shown mirror element 3 can be determined.

[0091] In an analogous manner, Fig. 4c and Fig. 4d depicted mirror elements 3, which, compared to the one in Fig. 4a shown reference position around the transverse axis y3 ( Fig. 4c) or around the vertical axis z3 ( Fig. 4d). The corresponding rotation angles can, for example, depend on the orientation of the longitudinal axis x3 ( Fig. 4d) and / or depending on the orientation and length of the axes of an elliptical image of the mirror surface 14.

[0092] From the Fig. 4a to 4d it can be seen that a shape-based determination of the pose P of the mirror element 3 can be carried out, whereby shape properties of the imaged mirror element 3 are determined and the pose P can then be determined depending on these properties.

[0093] It can also be seen that a center point of the mirror surface 14 is detectable. If, in addition to a point reflected on the mirror surface 14, e.g., the center point, the focus is also on a non-reflective edge of the mirror surface 14, the distance of the mirror surface 14, in particular the point of reflection, from the tooth row Z can be determined by a difference in the focus positions.

[0094] In particular, a difference can be determined between the focus position when focusing the stereo camera system 1 or the surgical microscope 2 on a point of the non-reflective edge, e.g., a point on the frame section 13, and the focus position when focusing on a point of the object projected on the mirror surface 14, e.g., a point reflected in the center of the mirror surface 14. This difference in the focus positions can represent a distance of the mirror element 13 from the object, here a point on the row of teeth Z, whereby the distance can therefore be determined as a function of this difference. In addition, the distance can also be determined as a function of the pose P of the mirror element 13. This distance information can be used for scaling in the described stereo reconstruction, in particular to achieve a magnification when reconstructing a projected section, e.g.,the back 8 of the Z row of teeth, to be adapted to a magnification during a reconstruction of an unmirrored section, e.g. the front 7 of the Z row of teeth, i.e. to carry out a scaling.

[0095] Fig. 5 shows a schematic representation of a mirror element 3 according to the invention. This has a marker element 6 designed as an optically detectable barcode on a handle section 12 or forms this. The mirror element 3 also has further optically detectable marker elements 6 designed as barcodes on a frame section 13 of the mirror surface 14. These can be detected in an image of the mirror element 3, wherein these marker elements in particular enable an identification of the mirror element 3 as well as a determination of the pose P of the mirror element 3. In particular, in the Fig. Each marker element 6 can be detected by the mirror element 3 shown in Figure 5, whereby the pose P can then be determined via a relative arrangement of the marker elements in the image.

[0096] Fig. 6 shows a schematic representation of an image capture device 5 of a stereo camera system 1 (see Fig. 3) and a mirror element 3, which is arranged in the detection area EB of the image capture device 5. Also shown is an object point OP to be imaged, for example a point on the surface of a row of teeth Z (see Fig. 3). Also shown is a normal to a mirror surface 14 of the mirror element 3.

[0097] Also shown is a virtual image capture device 15. A (virtual) image of this virtual image capture device 15 can be determined by evaluating a mathematical or physical model. The model is determined in particular in such a way that a virtual image is generated which images a section of the tooth row Z, in particular the object point OP, which corresponds to the section mirrored in the catadioptric system, in an unmirrored manner, but taking into account the properties of the catadioptric system. The three-dimensional image A can then be determined using a method known to those skilled in the art as a function of the virtual image, for example, via or with a stereo reconstruction. This has already been explained above.

[0098] Fig. 7a shows a schematic view of a stereoscopic detection of a row of teeth Z by two image capture devices 5a, 5b of a stereo camera system 1 without a mirror element 3. Shown is a detection of a front side 7 of the row of teeth Z as it is, for example, in the further image generation step SBV (see Fig. 2) to reconstruct a three-dimensional image A of the front side 7.

[0099] Fig. 7b a schematic view of a stereoscopic recording of a row of teeth Z by two image recording devices 5a, 5b of a stereo camera system 1 with a mirror element 3. Shown is a mirror element 3 which is arranged in the interior of the mouth and which forms a mirror image of a back side 8 (see Fig.3) of the tooth row Z, whereby this mirror image is then imaged by the image capture devices 5a, 5b. Based on the correspondingly imaged mirror images S5a, S5b, a three-dimensional image A of the back 8 is then reconstructed.

[0100] Furthermore, preoperative data can also be used with the images generated by the image acquisition devices 5a, 5b to generate the three-dimensional images A.

[0101] Alternatively and analogously to the stereoscopic acquisition of a treatment area, it may also be possible to optically acquire depth information using a different acquisition system used in the surgical microscope in order to generate a three-dimensional image of at least a partial section of a row of teeth. For this purpose, instead of the image generated by the first image acquisition device and the image generated by the further image acquisition device, an alternative image can be generated with this other acquisition system, in which a mirror image provided by a mirror element arranged in the acquisition range of the surgical microscope and in particular in the acquisition range of the acquisition system is detected. Then, as explained, the pose of the mirror element and the three-dimensional image can be determined as a function of at least the imaged mirror image and the pose of the mirror element.Thus, among other things, the optical depth information of a partial section of the tooth row can be captured by the surgical microscope and in particular by the acquisition system and embedded in a suitable coordinate system. List of reference symbols 1 stereo camera system 2 surgical microscope 3 mirror element 4 Detection range 5, 5a, 5b Image capture device 6 optically detectable marker element 7 Front 8 Back 9 occlusal surface 10 Evaluation device 11a, 11b Beam path 12 Handle section 13 frame section 14 Mirror surface 15 virtual image capture device SA arrangement step SB image generation step S1 first step S2 second step S3 third step SBV image generation step SRV reconstruction step EB detection area I5a, I5b images S5a, S5b depicted mirror image P Pose A1, A2, A image OP object point QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This 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

[0000] DE 10 2020 133 627 A1

[0009] DE 10 2019 008 510 A1

[0010] DE 10 2016 121 687 A1

[0011] EP 3 689 295 A1

[0012]

Claims

[1] Method for generating a three-dimensional image (A, A2) of at least a partial section of a row of teeth (Z) with a stereo camera system (1) of a surgical microscope (2), wherein the stereo camera system (1) comprises a first image capture device (5a) and a further image capture device (5b), wherein a. in an image (I5a) generated by the first image capture device (5a) and an image (I5b) generated by the further image capture device (5b), a mirror image (S5a, S5b) is detected, which is provided by a mirror element (3) arranged in a detection area (4) of the stereo camera system (1) such that a mirror image of at least one partial section of the row of teeth (Z) can be imaged by the first and further image capture device (5a, 5b), b. a pose (P) of the mirror element (3) is determined, c. the three-dimensional image (A, A2) is determined as a function of at least the imaged mirror images (S5a, S5b) and the pose (P) of the mirror element (3). [2] Method according to claim 1, characterized by that the three-dimensional image (A, A2) is additionally determined as a function of at least one optical property of the mirror element (3) and / or at least one imaging property of the surgical microscope (2). [3] Method according to one of the preceding claims, characterized by that images of virtual image capture devices are determined at least as a function of the pose (P) of the mirror element (3), wherein the at least one corresponding section of the row of teeth (Z) mirrored in a catadioptric system is imaged in an image of such a virtual image capture device, wherein the three-dimensional image (A, A2) is determined as a function of at least the images of the virtual image capture devices. [4] Method according to one of the preceding claims, characterized by that the pose (P) is determined by evaluating at least one property of the imaged mirror image or the imaged mirror element (3) or marker-based. [5] Method according to claim 4, characterized by in that for the marker-based determination of the pose (P) at least one marker element (3) is imaged by at least one image capture device (5a, 5b) of the stereo camera system (1) or by a further image capture device, wherein the at least one marker element (6) is arranged on the mirror element (3) or is formed by it, wherein the pose (P) is determined as a function of at least one property of the imaged marker element (6). [6] Method according to one of the preceding claims, characterized by that a focus position of the surgical microscope (2) is adjusted depending on the pose (P) of the mirror element (3). [7] Method according to one of the preceding claims, characterized by that the radiation detected to generate the images (I5a, I5b) of the stereo camera system (1) is filtered. [8] Method according to claim 7, characterized by that the filtering is a polarization filtering. [9] Method according to claim 7 or 8, characterized by that a filter element is arranged in an illumination beam path and / or in an imaging beam path (11a, 11b) and / or on the mirror element (3). [10] Method according to one of the preceding claims, characterized by that at least one partial section of the row of teeth (Z) is illuminated with radiation having predetermined radiation properties. [11] Method according to claim 10, characterized by that the radiation is generated with predetermined polarization properties. [12] Method according to one of the preceding claims, characterized bythat the images (I5a, I5b) generated by the stereo camera system are filtered, wherein the three-dimensional image (A, A2) is determined as a function of at least the filtered images. [13] Method according to one of the preceding claims, characterized by that at least one quality measure of the three-dimensional image (A, A2) is determined, wherein user information is generated if the quality measure is smaller than a predetermined threshold value. [14] System for generating a three-dimensional image (A, A2) of at least a partial section of a row of teeth (Z), comprising a stereo camera system (1) which comprises a first and at least one further image capture device (5a, 5b), and at least one evaluation device (10), characterized by that the system is configured to perform the following steps: a. Detection of a depicted mirror image (S5a, S5b) in an image (I5a) generated by the first image capture device (5a) and in an image (I5b) generated by the further image capture device (5b), wherein the mirror image is generated by a mirror element (3) which is arranged in a detection area (4) of the stereo camera system (1) and which reflects the at least one partial section of the row of teeth (Z), b. Determination of a pose (P) of the mirror element (3), c. Determination of the three-dimensional image (A, A2) as a function of at least the imaged mirror images (S5a, S5b) and the pose (P) of the mirror element (3). [15] System according to claim 14, characterized by that the system comprises a mirror element (3). [16] A computer program product comprising a computer program, the computer program comprising software means for executing one, several or all steps a) to c) of the method according to any one of claims 1 to 14 when the computer program is executed by or in a computer or an automation system. [17] Mirror element for generating a three-dimensional image (A, A2) of at least a partial section of a row of teeth (Z), characterized by that the mirror element (3) has or forms at least one marker element (6) for determining the pose (P) of the mirror element (3) and / or at least one filter element for filtering the reflected radiation.

Images