WO2016176838A1

WO2016176838A1 - Oral cavity detection and data processing device and method, and oral cavity detection system

Info

Publication number: WO2016176838A1
Application number: PCT/CN2015/078372
Authority: WO
Inventors: 孙玉春; 吕培军; 王勇; 赵慧洁; 崔海华; 姜宏志; 戴宁; 李虹
Original assignee: 北京大学口腔医院; 北京航空航天大学; 南京航空航天大学
Priority date: 2015-05-06
Filing date: 2015-05-06
Publication date: 2016-11-10
Also published as: CN107072762B; CN107072762A

Abstract

Disclosed are an oral cavity detection and data processing device and method, and an oral cavity detection system. The oral cavity detection device comprises: a probe (101), a projector (102) and a camera (103). The probe (101) is used to detect a tooth surface image in an oral cavity. The probe (101) comprises a reflector group, the reflector group comprises a plurality of mirrors which are arranged according to a predetermined positional relationship, and the mirrors are used to reflect a stripe image projected toward the reflector group onto the surfaces of teeth in the oral cavity and reflect the tooth surface image out of the oral cavity. The projector (102) is used to project a stripe image toward the reflector group of the probe (101), and the reflector group reflects the stripe image toward the surfaces of teeth. The camera (103) is used to acquire a reflected tooth surface image containing a stripe pattern from the reflector group of the probe (101). By means of such a device, tooth surface images of a plurality of surfaces and an accurate relative position relationship therebetween can be obtained, and errors in the process of image mosaicking can be reduced, while the frequency of moving the probe (101) is reduced.

Description

Oral detection, data processing device and method, and oral detection system

Technical field

The present invention relates to the field of stomatology, and more particularly to an oral probing, data processing apparatus and method, and an oral probing system.

Background technique

Dental impressions are an important source of information storage during clinical dental care and repair. At present, digital impression technology that is easy to operate, high in accuracy, and convenient to store is increasingly valued by the dental profession. The methods of digital impression acquisition are divided into two categories according to the location of the detection, namely, extraoral detection and intraoral detection. Among them, intraoral detection is a new detection method that has appeared abroad in recent years. Intraoral detection allows the detector to be inserted into the patient's mouth to directly detect the tooth and obtain a digital impression in real time. It has more advantages than oral detection. First, the patient's satisfaction is improved. Secondly, for the doctor, the quality of the impression is further improved, the steps of the chair side operation are reduced, precious time is saved, and the material and labor consumption is greatly reduced. It also establishes a platform for communication between patients and doctors. Doctors and patients can discuss digital impressions obtained, allowing patients to understand their condition and the doctor's intention to repair, or customize the correction according to the patient's needs. The treatment plan makes the doctor-patient relationship more harmonious. Due to the special nature of intraoral detection, in addition to requiring a small detector, it is also required to detect as quickly as possible.

The German Cerec system realizes intraoral data measurement based on the principle of surface structure optical vision measurement. As the world's most successful commercial intraoral detection system, it has been in a monopoly since the 1970s, and it has not emerged until the beginning of the 20th century, such as 2008. Laser-based 3M company's Lava Chairside Oral Scanner intraoral detection system, 2011 based on confocal measurement principle of Denmark's 3Shape company's TRIOS intraoral detection system, 2011 based on binocular stereo vision measurement principle of Germany's Hint-Els company's DirectScan intra-oral detection The system and the 2011 Iter intraoral 3D measurement system of CADDENT.

An overview of the typical products and systems of the intraoral detection system is as follows:

The Cerec series of intraoral dental measurement systems from Sirona Dental System, Germany, enable the combination of 3D digital detectors and dental restoration systems for online restoration of teeth. The measurement system uses the basic principle of active triangle and confocal microscopy, a blue light source. This version of the system is capable of measuring the full mouth of the teeth, while the previous version only allowed the measurement of some of the teeth at a certain angle. However, the system requires a coating on the surface of the tooth prior to detection to suppress strong reflection of the tooth surface to form a uniform reflective surface. The measurement accuracy is 0.019 mm.

The iTero digital imaging system of Cadent LTD of the United States adopts the principle of parallel confocal microscopy, which enables the surface of different materials to be imaged, so there is no need to add a coating on the tooth surface. Separate detection is performed by three monochromatic light sources of red, green and blue, and the three results are combined to obtain color texture information.

The E4D system of D4D Technologies LLC in the United States uses optical coherence tomography and confocal microscopy. In general, the system does not need to be probed after the surface of the tooth is added. The desired source is a laser source.

The 3M ESPE company in the United States developed the Lava Chairside Oral Scanner system, and the imaging part consists of a lens system and blue LEDs. The probe is 13.2 mm wide and weighs 390 g. The system uses the principle of active wavefront sampling to record three-dimensional data in video form and build a data model in real time. The system requires the addition of a coating to the tooth surface prior to detection.

The US IOS Technologies INC developed the IOS FastScan system. The system uses the active triangle principle and the Schleimpflug principle, and the light source is a laser light source. The detection rod design of the system has great advantages. The laser light source can be automatically moved on the detecting rod, and the operator can automatically detect the three parts of the buccal side, the tongue and the occlusal surface to achieve the measurement of the full mouth teeth. The system requires the addition of a coating to the tooth surface prior to detection.

The DENSYS 3D system from Densys LTD of Israel uses the principle of active stereo vision measurement, and the light source is visible light. The system takes only a few milliseconds to acquire images, enabling real-time measurements. Compared with other products, this system has a very light probe, weighing only 100g, and measuring accuracy of 0.03mm.

DPI-3D Department of Dimensional Photonics International INC, USA The principle of stripe interference is adopted, and the required laser source band is 350-500mm. The advantages of this system compared to systems using visible light sources are: greater resistance to changes in ambient light intensity and noise, higher accuracy, greater depth of projection, enhanced contrast and reflective surfaces. Measuring ability. The system uses a small, hand-held, real-time detection system to capture digital images, and the system does not require detection after adding a coating to the tooth surface. The system is in the development stage and there are no commercial products.

The 3D Progress system of MHT Optic Research AG in Switzerland uses confocal microscopy. The system's single-angle detection time is 0.1s, which can reach 14 detections per second, enabling the measurement of full-mouth teeth within 3 minutes. In general, the system does not require detection after adding a coating to the tooth surface. The system is in the development stage and has not yet been commercialized.

The Direct Scan system of HINT-ELS in Germany uses the stereo vision principle based on projection grating. The system is capable of 12-15 micron accuracy and the optical detector images a single angle every 200ms.

Denmark's 3Shape's TRIOS system uses confocal microscopy to directly perform three-dimensional detection of intraoral teeth without spraying.

The above systems still have different defects, including: all are single-reflection lenses, which have low detection efficiency. Continuous detection of occlusal surface, lip and cheek surface, tongue and palate surface, near-neighbor and adjacent middle surface when detecting a single tooth And the connection area between the faces; the single field of view detection range is small, usually only 1/2 single tooth surface; the multi-view data software splicing too many times, resulting in poor accuracy of multi-dental detection.

Summary of the invention

It is an object of the present invention to provide an apparatus and method for high accuracy intraoral detection.

According to an aspect of the invention, an oral detection device is provided, comprising: a probe for detecting an image of a tooth surface in an oral cavity, the probe comprising a mirror group, the mirror group comprising a plurality of mirrors arranged in a predetermined positional relationship, Reflecting the image of the fringe projected onto the mirror set onto the surface of the tooth in the mouth and reflecting the image of the surface of the tooth out of the mouth; projector, A stripe image is projected onto the mirror set of the probe and reflected by the mirror set onto the tooth surface; and a camera is configured to obtain a reflected image of the tooth surface including the stripe pattern from the mirror set of the probe.

Optionally, the mirror group of the probe comprises two or more mirrors, and the two or more mirrors are arranged in a predetermined positional relationship.

Optionally, the projector is a stripe projector and the stripe image is a sinusoidal stripe image.

Optionally, the probe length is 60-85 mm, the probe cross-sectional dimension is no more than 30 mm*20 mm; the projector and the camera are located at one end of the probe away from the mirror group; the projection baseline length of the projector and the camera is not less than the lens of the projector and the camera The sum of the radii.

With such a device, the projector can project a stripe image onto the tooth surface of the oral cavity through the mirror group of the probe, and the tooth surface image including the stripe pattern is reflected out of the oral cavity by the mirror group, thereby enabling the camera to obtain the tooth surface image. Since the mirror group can reflect the surface images of the plurality of faces of the tooth, and the plurality of mirror faces are arranged in a predetermined relative positional relationship, an accurate relative positional relationship of the tooth surface images of the plurality of faces can be obtained, and the stitching using the image can be reduced. The error in the data splicing process increases the accuracy of the detection.

According to another aspect of the present invention, an oral image data processing apparatus is provided, comprising: a parameter storage module for storing a relative positional relationship of mirrors in a mirror group of the probe; and an image acquisition module for acquiring a tooth surface image from the camera The image of the tooth surface is a specularly reflected image of the mirror group of the probe in a predetermined positional relationship, and includes image information of the plurality of faces of the tooth; and an image processing module for splicing according to the relative positional relationship of the mirrors in the mirror group The three-dimensional data obtained from the image of the tooth surface obtained by the camera acquires three-dimensional data of the tooth.

Optionally, the tooth surface image comprises sinusoidal fringes that are projected by the projector through the mirror set onto the tooth surface, and the mirror set reflects the tooth surface image containing the sinusoidal fringes to the camera.

Optionally, the image processing module includes: an image segmentation unit, configured to segment the tooth surface image according to the relative positional relationship of the mirrors in the mirror group of the probe in the parameter storage module, to obtain a mirror-reflected single-mirror tooth surface image, wherein The relative positional relationship of the mirrors in the mirror group includes the image area of each specular reflection image in the image of the tooth surface; a measuring unit for obtaining three-dimensional data of a single-mirror tooth surface in a single-mirror tooth surface image based on a sinusoidal stripe image of a single-mirror tooth surface image; and a three-dimensional stitching unit for reflecting the probe according to the parameter in the parameter storage module The relative positional relationship of the mirrors in the mirror group is unified into the same coordinate system by the three-dimensional data of the single mirror surface, and the three-dimensional data of the single tooth surface is obtained. The relative positional relationship of the mirrors in the mirror group includes the specular reflection images of the mirror group. The coordinate conversion relationship.

Optionally, the image processing module further includes: an adjacent splicing unit, configured to splicing three-dimensional data of a single tooth surface of the adjacent tooth according to a common area of the three-dimensional data of the single tooth surface of the adjacent tooth, and obtaining three-dimensional data of the plurality of tooth surfaces. .

Optionally, the oral image data processing apparatus further includes: a calibration module, configured to calibrate a relative positional relationship of the mirrors in the mirror group of the probe, comprising: an area calibration unit, configured to calibrate the image of the tooth surface acquired by the camera, each mirror surface An image area of the reflected image; a coordinate calibration unit configured to acquire a coordinate conversion relationship of the specular reflection image according to three or more corresponding points in a common area of the three-dimensional data of the single mirror tooth surface acquired from different mirrors; the parameter storage module, Obtaining an image region of each specular reflection image in the tooth surface image from the region calibration unit, and acquiring a coordinate transformation relationship of the specular reflection image from the coordinate calibration unit.

With such a device, the three-dimensional data obtained by three-dimensionally reconstructing the tooth surface image reflected by the mirror group acquired by the camera can be spliced according to the relative positional relationship of the mirror surfaces in the mirror group, and the three-dimensional data of the teeth can be obtained. Since the mirror group can reflect the image of the tooth surface of the plurality of faces, and the relative positional relationship of the mirror surface is fixed, the relative positional relationship of each face of the tooth surface image can be obtained during the splicing process, thereby reducing the error in the image splicing process. Increase the accuracy of the detection. Further, since the plurality of surface three-dimensional data of the teeth are simultaneously acquired, the detection efficiency can be improved.

According to still another aspect of the present invention, there is provided an oral cavity detecting system comprising: any one of the above oral detecting devices for projecting a stripe image onto a tooth surface in the oral cavity, and acquiring a tooth surface image including a stripe pattern by a camera; Any of the oral image data processing apparatuses for acquiring three-dimensional data of teeth based on an image of a tooth surface acquired by the oral detecting device.

With such a system, an image of a tooth surface including a plurality of faces of the tooth can be acquired by the oral cavity detecting device, and the image of the tooth surface of the plurality of faces is stitched by the oral image data processing device according to the relative positional relationship of the mirror faces of the mirror group in the oral detecting device , to obtain three-dimensional data of teeth, thereby reducing errors in image stitching process and increasing the accuracy of detection.

According to still another aspect of the present invention, an oral probe is provided, comprising a mirror group including a plurality of mirror surfaces arranged in a predetermined positional relationship for reflecting a stripe image projected to the mirror group to the oral cavity The inner tooth surface reflects an image of the tooth surface containing the stripe pattern out of the mouth to detect an image of the tooth surface containing images of the plurality of faces of the tooth.

Optionally, the mirror group comprises two mirrors on both sides of the central axis of the probe detection area for detecting 2 to 4 images of the tooth surface. Optionally, the angle between the mirror surface and the probe axis is in the range of [43°, 47°], and the rotation angle of the normal vector around the probe detection area is in the range of [41°, 49°].

Optionally, the mirror group comprises three mirrors respectively located on the central axis and the central axis of the probe detection area for detecting 3 to 4 images of the tooth surface. Optionally, the mirror elevation angle of the central axis is in the range of [45°, 60°]; the mirror surface on both sides of the central axis is at an angle of [43°, 47°] to the probe axis, and the probe detection area is around The rotation angle of the vector is in the range of [41°, 49°].

Optionally, the mirror group comprises four mirrors for detecting 4 to 5 images of the tooth surface, wherein the two mirrors are located at the proximal end and the distal end of the central axis of the probe detection area; the other two mirrors are located at the probe detection Both sides of the central axis of the area.

Optionally, the specular elevation angle of the proximal end is in the range of [30°, 45°], and the specular elevation angle of the distal end is in the range of [45°, 60°]; the mirror surface on both sides of the central axis of the probe detection area The angle with the axis of the probe is in the range of [43°, 47°], and the rotation angle of the normal vector around the probe detection area is in the range of [41°, 49°].

Such a probe includes a mirror group composed of a plurality of mirrors arranged in a predetermined positional relationship, capable of detecting an image of a tooth surface including images of a plurality of faces of the tooth, and capable of acquiring a relative positional relationship of images of the plurality of faces of the tooth on the surface of the tooth Image processing and stitching can reduce errors and increase oral detection accuracy.

According to still another aspect of the present invention, an oral image detecting method is provided, comprising: a projector projecting a stripe image to a mirror group of a probe, and reflecting the lens image to a tooth surface, wherein the probe comprises a mirror group, the mirror group A plurality of mirrors arranged in a predetermined positional relationship are included; the mirror set of the probe reflects the image of the tooth surface containing the stripe pattern to the outside of the mouth; and the camera acquires an image of the surface of the tooth reflected from the set of mirrors.

In this way, the projector can project a stripe image onto the tooth surface of the oral cavity through the mirror group of the probe, and the tooth surface image including the stripe pattern is reflected out of the oral cavity by the mirror group, thereby enabling the camera to obtain the tooth surface image. Since the mirror group can reflect the surface image of the plurality of faces of the tooth, and the plurality of mirror faces are arranged in a predetermined relative positional relationship, an accurate relative positional relationship of the tooth surface images of the plurality of faces can be obtained, and the image can be reduced by using the image. The error in the splicing process increases the accuracy of the detection.

According to still another aspect of the present invention, an oral image data processing method is provided, comprising: acquiring a tooth surface image from a camera, wherein the tooth surface image is an image reflected by a mirror group of mirrors in a predetermined positional relationship of the probe, including teeth Image information of multiple faces; according to the relative positional relationship of the mirrors in the mirror group, three-dimensional data obtained from the image of the tooth surface acquired by the camera is spliced to obtain three-dimensional data of the teeth.

Optionally, the relative positional relationship of the mirrors in the mirror group includes an image region of each specularly reflected image in the image of the tooth surface, and a coordinate conversion relationship of each specularly reflected image of the mirror group.

Optionally, according to the relative positional relationship of the mirrors in the mirror group, the three-dimensional data of the tooth surface acquired by the camera is spliced, and the three-dimensional data of the teeth is obtained, including: segmenting the image of the tooth surface according to the image region of each specular reflection image in the image of the tooth surface, obtaining each a mirror-reflected single-mirror tooth surface image; based on a sinusoidal fringe of a single-mirror tooth surface image, based on Phase measurement profilometry obtains three-dimensional data of single-mirror tooth surface in single-mirror tooth surface image; unified three-dimensional data of single-mirror tooth surface into the same coordinate system according to the coordinate conversion relationship of each specular reflection image of the mirror group, and obtains a single tooth Surface 3D data.

Optionally, according to the relative positional relationship of the mirrors in the mirror group, splicing the image of the tooth surface acquired by the camera to obtain the three-dimensional data of the tooth, and further comprising: splicing the adjacent teeth according to a common area of the three-dimensional data of the single tooth surface of the adjacent tooth Three-dimensional data on the surface of a single tooth, obtaining three-dimensional data on multiple tooth surfaces.

Optionally, the method further includes: calibrating a relative positional relationship of the mirrors in the mirror group of the probe, comprising: calibrating an image area of each specular reflection image in the image of the tooth surface acquired by the camera; and obtaining a single mirror tooth surface obtained from different mirror surfaces. The coordinate conversion relationship of the specular reflection image is obtained by three or more corresponding points in the common region of the three-dimensional data.

By such a method, the tooth surface image reflected by the mirror group acquired by the camera can be spliced according to the relative positional relationship of the mirror surfaces in the mirror group, and the three-dimensional data of the teeth can be acquired. Since the mirror group can reflect the image of the tooth surface of the plurality of faces, and the relative positional relationship of the mirror surface is fixed, the relative positional relationship of each face of the tooth surface image can be obtained during the splicing process, thereby reducing the error in the image splicing process. Increase the accuracy of the detection.

In addition, according to still another aspect of the present invention, an oral detection method is provided, comprising: any of the above oral image detecting methods, projecting a stripe image onto a tooth surface in an oral cavity, and acquiring a tooth surface image including a stripe pattern by a camera; Any of the above oral image data processing methods for acquiring three-dimensional data of teeth based on a tooth surface image.

By such a method, the tooth surface image including the plurality of faces can be acquired by the oral cavity detecting device, and the tooth surface image of the plurality of faces can be spliced by the oral image data processing device according to the relative positional relationship of the mirror faces of the mirror group in the oral detecting device. Obtain three-dimensional data of teeth, thereby reducing errors in image stitching and increasing the accuracy of detection.

DRAWINGS

The drawings described herein are provided to provide a further understanding of the invention, and constitute a part of this application. The pair is improperly defined by the present invention. In the drawing:

1 is a schematic view of one embodiment of an oral cavity detecting device of the present invention.

2 is a schematic diagram of one embodiment of an oral cavity detecting device of the present invention.

3 is a schematic illustration of one embodiment of a probe mirror assembly of the present invention.

4 is a schematic illustration of another embodiment of a probe mirror assembly of the present invention.

Figure 5 is a schematic illustration of yet another embodiment of a probe mirror assembly of the present invention.

Figure 6 is a schematic view showing the overall structure of an embodiment of the probe of the present invention.

Figure 7 is a schematic diagram of one embodiment of a mirror assembly of a probe of the present invention.

FIG. 8 is a schematic diagram showing a geometric representation of a plane mirror vector.

Figure 9 is a schematic illustration of one embodiment of an oral image data processing apparatus of the present invention.

Figure 10 is a schematic illustration of another embodiment of an oral image data processing apparatus of the present invention.

Figure 11 is a schematic view showing still another embodiment of the oral image data processing apparatus of the present invention.

Figure 12 is a schematic illustration of still another embodiment of the oral image data processing apparatus of the present invention.

Figure 13 is a schematic illustration of one embodiment of an oral probing system of the present invention.

Figure 14 is a flow chart of one embodiment of an oral image detecting method of the present invention.

Figure 15 is a flow chart showing an embodiment of the oral image data processing method of the present invention.

16 is a flow chart of one embodiment of a portion of an oral image data processing method of the present invention.

Figure 17 is a flow chart showing still another embodiment of the oral image data processing method of the present invention.

Figure 18 is a flow chart of one embodiment of the oral probing method of the present invention.

detailed description

The technical solution of the present invention will be further described in detail below through the accompanying drawings and embodiments.

A schematic view of one embodiment of the oral cavity detecting device of the present invention is shown in FIG. among them, 101 is a probe for detecting an image of a tooth surface in the oral cavity. The end of the probe includes a mirror set that includes a plurality of mirrors that are arranged at predetermined relative positions. There is a projector and camera at the end of the probe away from the mirror set. The probe is a hollow structure that ensures that the projector's light is projected onto the mirror group, and the image reflected by the mirror group can be captured by the camera. 102 is a projector for projecting a fringe image to a mirror group of the probe. The fringe image is reflected by the mirror mirror and then projected onto the tooth surface; the image of the tooth surface is then mirrored out of the mouth by the probe mirror. 103 is a camera that acquires a reflected image of the tooth surface including the stripe pattern from the mirror group of the probe. Since the plurality of mirrors are arranged at a predetermined relative position, the image acquired by the camera may include the image of the tooth surface of the plurality of faces while acquiring the relative positional relationship of the plurality of faces of the tooth, so that when the image of the tooth surface of the plurality of faces is spliced, It is better to restore the relative positional relationship of multiple faces of the tooth, reduce the image mosaic error and increase the detection accuracy. Since the probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, the number of movements of the probe can be reduced during use, thereby improving detection efficiency.

In one embodiment, the projector is a fringe projector and the projected fringe image is a sinusoidal fringe image. Based on the phase measurement profilometry based on the change in phase in the tooth surface image containing the sinusoidal stripe pattern acquired by the camera, the height information can be acquired, thereby enabling the tooth three-dimensional data to be acquired from the tooth surface image containing the sinusoidal stripe pattern.

In one embodiment, the schematic of the oral cavity sensing device of the present invention is illustrated in FIG. In Figure 2, the screen on the left is the camera and the screen on the right is the projector. According to the pinhole imaging model of the camera, the world coordinate system (P[X, Y, Z] ^T ) is between the camera and the projector's coordinate system (m _L [x _L , y _L ] ^T , m _R [x _R , The relationship of y _R ] ^T ) can be expressed as follows:

s _L [m _L ^T 1] ^T =A _L [R _L |T _L ][P ^T 1] ^T

s _R [m _R ^T 1] ^T =A _R [R _R |T _R ][P ^T 1] ^T

Where s _L and s _R are scale factors, A _L and [R _L |T _L ] represent the internal and external parameters of the camera, and A _R and [R _R |T _R ] represent the internal and external parameters of the projector. In this measurement system, the camera coordinate system is defined as the world coordinate system, and the relationship between the camera coordinate system (O _L -X _L Y _L Z _L ) and the projector coordinate system (O _R -X _R Y _R Z _R ) It can be expressed as:

Where R and T are external parameters of the system, R is the twiddle factor, and T is the translation matrix.

According to the basic principle of 3D measurement, under the condition of the structural parameters of the system and the internal parameters of the camera and the projector, the matching point between the camera and the projector is known, and the three-dimensional reconstruction of the object can be realized.

The camera image point m _L [x _L , y _L ] ^T can be directly obtained from the image, and the point coordinate m _R [x _R , y _R ] ^T in the projector coordinate system matching the image point can be obtained by phase measurement profilometry obtain.

Knowing the structural parameters of the system and the parameters within the sensor, the camera image point m _L [x _L , y _L ] ^T and the point coordinate m _R [x _R , y _R ] ^T corresponding to the point coordinate in the projector coordinate system P[X , Y, Z] ^T can be obtained by the following formula:

Through such a method, coordinate transformation can be performed on the tooth surface image acquired by the camera, and the acquired three-dimensional data of the tooth surface is unified to the world coordinate system according to the relative positional relationship of the projector, the mirror group, and the camera coordinate system, thereby making the tooth surface Three-dimensional data can be applied to oral restoration.

Since the probe needs to reach into the patient's mouth, the size is required to be small. However, in order to maximize the measurement field of view, it is necessary to control the size of the probe and the position of the projector and camera. In one embodiment, the parameters of the probe and the projector and camera are as shown in Table 1.

Table 1

The working distance L of the system is the sum of the length of the probe and the distance from the outer end surface of the camera lens to the optical center of the lens; the length B of the photographing baseline is the distance between the optical center of the projector and the optical center of the camera; the angle of intersection of the optical axis is the projector and the camera. The optical axis intersection angle; the system working range is the range of the probe single detection.

In addition, the point cloud resolution is 40 μm, and the single field of view accuracy is 30 μm (2σ). From the front and back depth of field formula:

Where ΔL ₁ is the foreground depth, ΔL ₂ is the back depth of field, and the parameters include the aperture number F _c (which may be 8), the allowable circle diameter δ (which may be 1.5 pixels in size), and the working distance L (100 mm).

At this time, the depth of field of the system is 5.16 mm, which is able to obtain a clear image of the tooth surface within a depth of 5.16 mm.

With such a device, the mirror group of the probe can reflect the image of the fringe projected by the projector and project onto the surface of the tooth; the camera can acquire the surface of the tooth reflected from the mirror group. The size of the probe ensures that the entrance cavity can be extended, and the image of the tooth surface of a single tooth can be measured at one time, which is convenient to use and improves work efficiency.

A schematic diagram of one embodiment of the mirror assembly portion of the probe of the present invention is shown in FIG. The mirror group of the probe is arranged by a mirror arranged in a predetermined position, including a left mirror 301 and a right mirror 302. The bottom of FIG. 3 is a hollow structure, and the light projected by the projector can be projected from the bottom space to the mirror surface and reflected to the tooth surface. . The left mirror 301 and the right mirror 302 are arranged at predetermined positions, and the tooth surface images of at least two faces can be measured in a single shot. The single tooth includes the occlusal surface, the proximal side, the distal side, the buccal surface, and the lingual surface. According to the relative positional relationship between the probe and the tooth during use, the 2-4 surface image of the tooth surface can be obtained. Obtaining two faces at the same time, such as obtaining the occlusal surface and the buccal surface, the occlusal surface and the lingual surface, the lingual surface and the distal side, the occlusal surface, and the near middle side, etc., while obtaining three surfaces, such as obtaining a part of the occlusal surface, A part of the distal side and the lingual surface; and a case where four faces are simultaneously obtained, such as obtaining a part of the occlusal surface, a distal side, a part of the lingual surface, and a part of the buccal surface. Such a probe mirror group can acquire a single tooth surface image of 2-4 faces at a time, and can obtain the relative positional relationship of the image of the tooth surface of a plurality of face faces with the relative positional relationship of the mirror surface, thereby obtaining more information for image stitching. The amount of splicing of the image of a single tooth surface is more accurate. Since the probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, the number of movements of the probe can be reduced during use, thereby improving detection efficiency.

In one embodiment, a schematic of the probe mirror assembly is shown in FIG. The probe mirror set includes a left mirror 401, a right mirror 402, and a mid mirror 403. As in Fig. 3, the bottom of Fig. 4 is a hollow structure, and the light projected by the projector can be projected from the bottom space to the mirror surface and reflected to the tooth surface. The left mirror 401, the right mirror 402, and the middle mirror 403 are arranged at predetermined positions, and the tooth surface images of at least three faces can be measured in a single time. According to the relative positional relationship between the probe and the tooth during use, it is possible to obtain an image of the tooth surface of 3-4 faces. At the same time, the three faces are obtained, such as obtaining the occlusal surface, the distal side and the lingual surface, and the four faces are obtained, such as obtaining a part of the occlusal surface, a distal side, a part of the lingual surface, and a part of the buccal surface. Such a probe mirror group can obtain a single tooth surface image of 3-4 faces at a time, and the relative positional relationship of the mirror surface can obtain the relative positional relationship of the image of the plurality of face tooth surfaces, thereby obtaining more information for image stitching. Amount to splicing the image of a single tooth surface more acurrate. In addition, obtaining the tooth surface image of at least 3 faces at a time can improve the detection efficiency, and can also reduce the number of stitching and further improve the accuracy.

In one embodiment, a schematic of the probe mirror assembly is shown in FIG. The probe mirror set includes a left mirror 501, a right mirror 502, a lower mirror 503, and an upper mirror 504. The left mirror 501, the right mirror 502, the lower mirror 503, and the upper mirror 504 are arranged at predetermined positions, and the tooth surface images of the five teeth occlusal surface, the proximal side, the distal side, the buccal surface, and the lingual surface can be acquired. Such a probe mirror group can acquire five faces of a single tooth at a time, that is, all the tooth surface images, and the relative positional relationship of the mirror faces can obtain the relative positional relationship of the image of the tooth surface of each face at a time, thereby improving the detection accuracy. At the same time, the detection efficiency is further improved.

In one embodiment, the overall structure of the probe is shown in FIG. The left mirror 601, the right mirror 602, the lower mirror 603 and the upper mirror 604 are located in the detection area of the probe to form a mirror group. The light projected by the projector is projected through the probe cavity 605 to the mirror set, and the image of the tooth surface is projected onto the mirror set, which is acquired by the camera via the probe cavity 605 to the camera. In one embodiment, the camera acquires the buccal and partial occlusal image from the left mirror 601, acquires the tooth lingual surface and the partial occlusal image from the right mirror 602, and obtains the occlusal surface and the near-middle image from the lower mirror 603. The mirror 604 acquires an image of the occlusal surface and the distal side of the tooth. Such a probe can maintain the relative positional relationship of the mirrors in the mirror group and is suitable for extending into the human mouth for ease of use.

In one embodiment, in order to ensure the detection efficiency of the three-dimensional detector in the mouth, it is necessary to reasonably design the posture of the mirror group in the probe, so that multiple mirrors can simultaneously measure the various faces of the tooth to obtain the largest measurement area.

To select and optimize the relative position and attitude parameters of the probe group of the probe, it is necessary to establish a space rectangular coordinate system on the mirror group. As shown in Fig. 7, the mirror angle of the mirror group is used as a variable to represent the plane mirror. Plane equation, optical path simulation based on the principle of light reflection, can be realized by Matlab. Observe the measured area of the light and obtain the theoretical optimal attitude of the mirror group by comparison.

A, mathematical model

To optimize the attitude structure parameters of the mirror group, you first need to establish a plane on the mirror group. Cartesian coordinate system, as shown in Figure 7.

The quantitative result of optimizing the structural parameters is to give the attitude parameters of each mirror in the mirror group, so it is necessary to determine the plane equation of the four mirrors in the established coordinate system.

The normal vector of a plane is an important element for solving a one-sided surface. The normal vector and any point on the plane can form a plane equation. Take the mirror group structure in Fig. 6 as an example. For the left mirror 601, the mirror normal vector solution process is as follows:

The normal vector of the left and right mirrors in the mirror group has two degrees of freedom, as shown in Figure 8. Suppose a plane passes through the x-axis and is at an angle α with the xoz plane (as shown at 801 in Figure 8), which is rotated by an angle β around the z-axis (shown as 802 in Figure 8), and then, according to the mirror Dimensions, the excess portion is cropped, and the form of the left mirror is visible as the geometry of the plane mirror is shown as 803 in FIG.

For the two mirrors in the middle, the specular normal vector is determined only by its pitch angle, so the two normal vectors are: (tan(θ ₁ ), 0, 1) and (tan(θ ₂ ), 0, 1), θ ₁ , θ ₂ are plane normal vectors of the two mirrors in the middle, respectively. The mirror 604 is set to pass (x ₁ , 10, 0), and the lower mirror 603 is over (x ₂ , 10, 0).

α, β, θ ₁ , x ₁ , θ ₂ , and x ₂ are six parameters that determine the pose relationship of the mirror group.

B, optical path tracking model

The main means of optimization of the mirror group is optical path simulation. The simulation is based on the principle of light reflection.

In the spatial Cartesian coordinate system based on the mirror group (Fig. 7), the position of the optical center of the stripe projector is assumed to be Xop = (100, 0, 10). According to the structural design of the system, the position of the optical center of the camera is Xoc. = (100, 30, 10).

Light is emitted from the optical center of the projector, projected onto the mirror group, and reflected by the mirror group to the object within its measurement range, and then reflected by the mirror group into the camera's optical center. A light completes the complete path of "projector-mirror group-measured object-mirror group-camera", that is, a light is reflected by the projector through the mirror group to a certain point of the object to be measured, and the camera can pass through the mirror group The reflection gets the image of the point, and the light is considered "effective light." Observe the "effective light" covering the surface of the object and adjust the positional relationship between the mirrors.

C, the simulation process

Covering the surface of the measured object with effective light as an observation index, in the mirror group The positional relationship between the lenses is optimized. The following six parameters are determined for the position and posture of the mirror group of the oral detector:

α, β: determine the posture of the left mirror and the right mirror;

θ ₁ , x ₁ , θ ₂ , x ₂ : determines the posture of the upper mirror surface and the lower mirror surface.

In the simulation process, the teeth are replaced by a rectangular parallelepiped having a length of 10 mm, a width of 10 mm and a height of 5 mm. The bottom surface is parallel to the xoz plane of the coordinate system established by the oral probe, and the center is at coordinates (2, 1.5, -2.5).

Constantly adjust the six parameters and observe the area of the surface covered by the reflected light to select the best solution.

The attitude of the two plane mirrors in the middle is determined first. Considering that the measurement accuracy is affected by the angle between the incident light and the surface of the object, it is determined that θ ₁ =60°, θ ₂ =30°; considering the common area of the two plane mirror reflection images in the middle Area, determine x ₁ = -9, x ₂ = -14.

For α, β, an exhaustive comparison is made from 41° to 49°, respectively, and the parameters to be compared include the following:

1) Whether there is a loop on the upper surface of the tooth;

2) the total area S of the four mirrors facing the teeth;

3) coverage of the tooth side surface P;

4) The overlapping area SL of the lower mirror surface and the left mirror surface on the upper surface of the tooth.

5) The overlapping area SR of the mirror surface and the right mirror surface on the upper surface of the tooth.

In one embodiment, it is preferred that the upper surface of the tooth is covered with no holes, and the parameters of S, P, SL, and SR are large, and the preliminary results obtained by simulation and judgment are: when α∈[43°, 47°], β When ∈[41°, 49°], θ ₁ ∈ [45°, 60°], θ ₂ ∈ [30°, 45°], the above five requirements are satisfied.

Among them, the better posture is:

With such a probe, there is a mirror group that satisfies the above five requirements, and it is possible to obtain an image of the tooth surface of each face of the entire tooth at a time, thereby improving the detection efficiency.

In one embodiment, if the probe mirror set is the two mirror structures in Figure 3, then It is necessary to satisfy α∈[43°, 47°], β∈ [41°, 49°]. Such a probe can acquire more image of the tooth surface as much as possible, and acquire the relative positional relationship of the image of the tooth surface of the plurality of faces, thereby reducing the error when the image of the tooth surface is spliced.

In one embodiment, if the probe mirror group is the three mirror structures in FIG. 4, it is necessary to satisfy α∈[43°, 47°], β∈[41°, 49°], and the medium mirror pitch angle θ∈ [45°, 60°]. Such a probe can acquire more image of the tooth surface as much as possible, so that the relative positional relationship of each surface can be obtained when the image is spliced, and the error of the image mosaic on the tooth surface can be reduced, and the number of detections can be reduced, and the efficiency can be improved.

A schematic diagram of one embodiment of the oral image data processing apparatus of the present invention is shown in FIG. 901 is an image acquisition module, 902 is a parameter storage module, and 903 is an image processing module. The parameter storage module 902 stores the relative positional relationship of the mirrors in the mirror group of the probe. The relative positional relationship of the mirrors may include image regions of respective specularly reflected images in the image of the tooth surface, and coordinate transformation relationships of the respective specularly reflected images of the set of mirrors. The image acquisition module 901 is configured to acquire a tooth surface image from a camera, and the tooth surface image is reflected by the mirror group of the probe to the camera. The mirror group has a plurality of mirror surfaces in a predetermined position relationship, and can reflect the tooth surface image of the plurality of faces of the tooth. The image acquisition module 901 passes the image of the tooth surface to the image processing module 903. The image processing module 903 splices the image of the tooth surface according to the relative positional relationship of the mirrors in the mirror group stored in the parameter storage module 902, and obtains three-dimensional data of the teeth and three-dimensional data of the teeth. Can be made up of point clouds. In one embodiment, the tooth surface image contains sinusoidal stripes that are projected by the projector to the mirror set, reflected to the tooth surface, and then reflected by the mirror set to the camera. Due to the different heights of the tooth surface, the sine fringe phase of the tooth surface acquired by the camera will be different. Through the phase feature, three-dimensional reconstruction of the tooth surface image can be achieved, thereby acquiring three-dimensional data of the tooth.

With such a device, it is possible to splicing the image of the tooth surface reflected by the mirror group acquired by the camera according to the relative positional relationship of the mirror surfaces in the mirror group, and acquiring the three-dimensional data of the teeth. Since the mirror group can reflect the image of the tooth surface of the plurality of faces, and the relative positional relationship of the mirror surface is fixed, the relative positional relationship of each face of the tooth surface image can be obtained during the splicing process, thereby reducing the error in the image splicing process. Increase the accuracy of the detection. Since the probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, in use The detection process can reduce the number of probe movements, thereby improving detection efficiency.

A schematic diagram of another embodiment of the oral image data processing apparatus of the present invention is shown in FIG. 1001, 1002, and 1003 are respectively an image acquisition module, a parameter storage module, and an image processing module. The image acquisition module 1001 and the parameter storage module 1002 are similar to those in the embodiment of FIG. The image processing module 1003 includes an image segmentation unit 1013, a phase measurement unit 1023, and a three-dimensional splicing unit 1033. Since the mirror group is composed of a plurality of mirrors arranged in a predetermined relative positional relationship, the acquired tooth surface image is composed of a plurality of specularly reflected images. The image dividing unit 1013 divides the tooth surface image based on the image regions of the respective specular reflection images in the tooth surface image stored in the parameter storage module 1002, and acquires each mirror-reflected single-mirror tooth surface image. The surface area of the tooth reflected by the image of the single mirror tooth surface may overlap. The phase measuring unit 1023 performs three-dimensional reconstruction of the single-mirror tooth surface image based on the phase-measuring profilometry based on the single-specular tooth surface image acquired by the image segmentation unit 1013, and obtains the multi-specular tooth surface by using the sinusoidal stripe information on the single-mirror tooth surface image. image. In one embodiment, if the single-mirror tooth surface image includes the tooth surface image of the five faces of the tooth, the single tooth surface three-dimensional data can be obtained by splicing, and the single tooth surface three-dimensional data can be point cloud data. Such a device can realize segmentation, three-dimensional reconstruction and splicing of the tooth surface image, thereby obtaining three-dimensional data of the teeth of the whole tooth, improving the efficiency of data processing and reducing the error of multiple measurement splicing. Since the probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, the number of movements of the probe can be reduced during use, thereby improving detection efficiency.

A schematic diagram of still another embodiment of the oral image data processing apparatus of the present invention is shown in FIG. 1101, 1102, and 1103 are respectively an image acquisition module, a parameter storage module, and an image processing module. The image processing module 1103 includes an image segmentation unit 1113, a phase measurement unit 1123, and a three-dimensional splicing unit 1133, and the modules and units are the same as those of FIG. Similar in the examples. The three-dimensional splicing unit 1133 is capable of acquiring three-dimensional data of a single tooth surface. At the same time, since the teeth in the oral cavity are adjacent to each other, three-dimensional data of a part of the tooth surface adjacent to the tooth is obtained, and therefore, the detection of the adjacent tooth has a certain common area. The image processing module 1103 further includes an adjacent splicing unit 1143, and acquires three-dimensional data of a single tooth surface obtained by splicing the three-dimensional splicing unit 1133, and the common area acquired according to the adjacent tooth detection, Three-dimensional data of the surface of a single tooth of adjacent teeth is spliced to obtain three-dimensional data of the tooth surface of a plurality of teeth. In one embodiment, the corresponding point can be manually calibrated in the three-dimensional data of the single tooth surface of the adjacent tooth, and the corresponding point can be a point with a prominent feature, and by calibrating three or more corresponding points, the adjacent tooth can be obtained. The coordinate transformation relationship between the three-dimensional data of a single tooth surface realizes the splicing of the three-dimensional coordinates of each point. In one embodiment, three-dimensional data of the teeth of the full oral cavity can be obtained by multiple stitching.

With such a device, three-dimensional data of adjacent single tooth surfaces can be spliced to obtain three-dimensional data of a plurality of tooth surfaces. Since the splicing according to the common area of the three-dimensional data of the adjacent single tooth surface, that is, the relative positional relationship of the acquired adjacent teeth, the splicing process error of the three-dimensional data of the adjacent tooth surface is reduced, and the three-dimensional data of the plurality of tooth surfaces is made. More in line with the real state, improving the accuracy of image stitching. Since the probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, the number of movements of the probe and the number of image stitching can be reduced during use, thereby improving detection efficiency.

A schematic diagram of still another embodiment of the oral image data processing apparatus of the present invention is shown in FIG. The image acquisition module 1203 includes an image segmentation unit 1213, a phase measurement unit 1223, a three-dimensional splicing unit 1233, and an adjacent splicing unit 1243. The modules and units are respectively included in the image acquisition module, the parameter storage module, and the image processing module. Both are similar to those in the embodiment of FIG. A calibration module 1204 is also included. The calibration module 1204 is used to calibrate the relative positional relationship of the mirrors in the mirror group of the probe. The calibration module 1204 includes a region calibration unit 1214 and a coordinate calibration unit 1224. The area calibration unit 1214 calibrates the image area of each of the specular reflection images; the coordinate calibration unit 1224 calibrates the coordinate conversion relationship of the specular reflection image.

In one embodiment, since the accuracy required for oral measurement is high, even if the probe is manufactured in accordance with a predetermined positional relationship, a certain error is generated, and therefore calibration is required at the first use.

The calibration is divided into two aspects. One is to calibrate the image area of each specular reflection image, to segment the image of the tooth surface, and to segment the image of different specular reflection, that is, to obtain the image of the surface of the single mirror surface. It can be achieved by manually marking the image of the tooth surface obtained. Regional calibration unit by manual segmentation of the first acquired tooth surface image An image area of each specular reflection image is acquired and the result is sent to parameter storage module 1202. Since the relative positions of the mirror group and the camera are unchanged, the image area of each specular reflection image is unchanged in the subsequent image segmentation, and the image of the tooth surface image can be performed by using the image region of each specular reflection image stored by the parameter storage module 1202. Segmentation to obtain 3D data on a single mirrored tooth surface. Such calibration can eliminate the impact of manufacturing process and precision on data processing, and more accurate segmentation of the tooth surface image, increasing the accuracy of data processing.

Another aspect of calibration is to calibrate the coordinate transformation relationship of the specularly reflected image to obtain the rotation factor R and translation matrix T required to unify the different specularly reflected single-specular tooth surface images into the same coordinate system. In one embodiment, the coordinate transformation relationship between the other three mirrors and the three-dimensional coordinates of the reflected image of the lower mirror 603 is obtained by using the three-dimensional coordinates of the image reflected by the lower mirror 603 in FIG. 6 as a reference.

The precise stitching of the map is an image obtained by measuring the structured light, and is stitched by matching the feature points in the image.

Since the relative positional relationship of the mirror groups of the mirror group is constant, the coordinate conversion relationship between the mirror surfaces is unchanged. In the first measurement, the image of the tooth surface obtained by different mirrors can be compared, and the single-mirror tooth surface image of the calibration has a corresponding area of the common area of the two mirrors of the common area, and the number of corresponding points is at least three pairs. You can select the point with obvious features as the corresponding point and bring the coordinates of the corresponding point into the formula:

The transformation matrices R, T are obtained. The coordinate calibration unit 1204 sends the acquired transformation matrix R, T to the parameter storage module 1202. In the subsequent use, the image processing module 1203 performs the three-dimensional reconstruction of the single-mirror tooth surface according to the R and T stored by the parameter storage module 1202. Three-dimensional data is spliced to obtain three-dimensional data of a single tooth surface. Such calibration can eliminate the impact of the manufacturing process and accuracy on the data processing, so that the three-dimensional data of the single tooth surface obtained by the splicing is more accurate.

A schematic diagram of one embodiment of the oral cavity detection system of the present invention is shown in FIG. mouth The cavity detecting system may be composed of any one of the above-mentioned oral detecting device and the oral image data processing device. As shown in FIG. 13, the projector 1311, the probe 1312 and the camera 1313 constitute an oral detecting device for acquiring an image of the oral tooth surface; The acquisition module 1301, the parameter storage module 1302, and the image processing module 1303 constitute an oral image data processing device for acquiring dental three-dimensional data according to the acquired tooth surface image. In one embodiment, the oral image data processing device may further include a calibration module 1304. Used to calibrate the parameters of the data processing to eliminate errors caused by the manufacturing process of the device. The projector 1311 projects a stripe pattern to the mirror group of the probe 1312, which may be a sinusoidal stripe pattern. The mirror group reflects the stripe pattern onto the tooth surface, and then acquires the tooth surface image, and reflects the tooth surface image out of the oral cavity, which is acquired by the camera 1313. The tooth surface image contains sinusoidal stripes, and the phase of the sinusoidal stripe pattern occurs due to the different tooth surface heights. Variety. The camera 1313 transmits the tooth surface image to the image acquisition module 1301, and the image acquisition module 1301 transmits the tooth surface image to the image processing module 1303. The image processing module 1303 processes the tooth surface image according to the relative positional relationship parameter of the mirror in the mirror group stored by the parameter storage module 1302, and may include a single-mirror image area to the tooth surface image according to the tooth surface image stored by the parameter storage module 1302. Segmentation is performed to obtain a single-mirror tooth surface image; based on the sinusoidal stripe on the three-dimensional data of the single-mirror tooth surface, three-dimensional reconstruction is performed according to phase measurement profilometry to obtain three-dimensional data of the single-mirror tooth surface; and then the specular reflection image stored according to the parameter storage module 1302 The coordinate transformation relationship between the three-dimensional data of the single-mirror tooth surface is unified into the same coordinate system to obtain three-dimensional data of the multi-specular tooth surface; when the three-dimensional data of the single-mirror surface of the tooth surface is obtained, a single piece can be obtained. The three-dimensional data of the tooth surface; in addition, the three-dimensional data of the single tooth surface of the adjacent tooth may be spliced according to the common area of the three-dimensional data of the adjacent tooth surface, and the three-dimensional data of the tooth surface may be obtained, and the whole three-dimensional data of the tooth surface may be obtained. Oral teeth 3D data.

With such a system, an image of a tooth surface including a plurality of faces can be acquired by the oral cavity detecting device, and the image of the tooth surface image of the plurality of faces can be spliced by the oral image data processing device according to the relative positional relationship of the mirror faces of the mirror group in the oral detecting device. , to obtain three-dimensional data of teeth, thereby reducing errors in image stitching process and increasing the accuracy of detection. Due to The probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, and can reduce the number of movements of the probe and the number of image stitching during use, thereby improving detection efficiency.

A flowchart of one embodiment of the oral image detecting method of the present invention is shown in FIG.

In step 1401, the projector projects a fringe image onto the mirror set of the probe and reflects it onto the tooth surface via the mirror set. The probe includes a mirror group, and the mirror group is composed of a plurality of mirrors arranged in a predetermined positional relationship. There are at least two mirrors that reflect the 2-4 faces of the tooth surface; it can be 3 faces that reflect the 3-4 faces of the tooth surface; it can also be 4 faces that reflect the 5 faces of the tooth. In one embodiment, the projector may be a stripe projector and the projected stripe image is a sinusoidal stripe image. Since the sinusoidal fringe image changes its phase when projected onto a surface of a different height, a three-dimensional reconstruction can be performed according to the sinusoidal stripe pattern of the tooth surface.

In step 1402, the mirror set of the probe reflects the image of the tooth surface containing the stripe pattern out of the mouth. The stripe pattern may be a sinusoidal stripe pattern, and the phase information of the reflected sinusoidal stripe pattern can be reduced to height information of the tooth surface due to the difference in height of the tooth surface. Since the mirror group includes a plurality of lenses in a predetermined positional relationship, the reflected tooth surface image is composed of a plurality of specularly reflected images.

In step 1403, the camera acquires an image of the tooth surface reflected from the mirror set.

In this way, the projector can project a stripe pattern onto the tooth surface of the oral cavity through the mirror group of the probe, and the tooth surface image including the stripe pattern is reflected out of the oral cavity by the mirror group, thereby enabling the camera to obtain an image of the tooth surface. Since the mirror group can reflect the surface image of the plurality of faces of the tooth, and the plurality of mirror faces are arranged in a predetermined relative positional relationship, an accurate relative positional relationship of the tooth surface images of the plurality of faces can be obtained, and the image can be reduced by using the image. The error in the splicing process increases the accuracy of the detection.

A flowchart of one embodiment of the oral image data processing method of the present invention is shown in FIG.

In step 1501, a tooth surface image is acquired from the camera. Since the tooth surface image is reflected by the mirror group to the camera, and the mirror group is composed of a plurality of mirror surfaces in a predetermined positional relationship, the tooth surface image is composed of a plurality of lens reflection images in a predetermined positional relationship.

In step 1502, the tooth surface image acquired by the camera is spliced according to the relative positional relationship of the mirrors in the mirror group, and the tooth three-dimensional data is acquired. Since the relative positional relationship of the lenses is constant, the tooth surface images reflected by different specular surfaces can be obtained according to the relative positional relationship of the lenses, thereby obtaining three-dimensional data of the teeth.

In one embodiment, a flow chart for stitching a tooth surface image to obtain three-dimensional data of the teeth is shown in FIG.

In step 1601, the tooth surface image is segmented according to the image area of each specularly reflected single mirror tooth surface image in the tooth surface image to obtain a single mirror tooth surface image.

In step 1602, since the tooth surface image includes a stripe pattern, the stripe pattern may be a sinusoidal stripe pattern. According to the phase measurement profilometry, the single-mirror tooth surface image is three-dimensionally reconstructed based on the phase information of the sinusoidal stripe pattern to obtain three-dimensional data of the single-mirror tooth surface.

In step 1603, three-dimensional data of the single-mirror tooth surface is unified into the same coordinate system according to the coordinate conversion relationship of each specular reflection image, thereby obtaining three-dimensional data of the multi-specular tooth surface. When the tooth surface image contains a surface image of the entire surface of the tooth, it is possible to acquire three-dimensional data of the single tooth surface.

Through such a method, the segmentation, three-dimensional reconstruction and splicing of the tooth surface image can be realized, thereby obtaining three-dimensional data of the single tooth surface of the entire tooth, thereby improving the efficiency of data processing and reducing the error of multiple measurement splicing.

In one embodiment, the three-dimensional data of the single tooth surface of the adjacent tooth may be spliced according to the common area of the three-dimensional data of the single tooth surface of the adjacent tooth to obtain three-dimensional data of the plurality of tooth surfaces. In one embodiment, a single piece of adjacent teeth can be manually The corresponding point is calibrated in the three-dimensional data of the tooth surface, and the corresponding point may be a point with prominent features. By calibrating three or more corresponding points, the coordinate conversion relationship between the three-dimensional data point cloud of the single tooth surface of the adjacent tooth can be obtained. splice. In one embodiment, full oral tooth three-dimensional data can also be obtained.

In this way, since the splicing according to the common area of the three-dimensional data of the adjacent single tooth surface, that is, the relative positional relationship of the adjacent teeth is obtained, the splicing process error of the three-dimensional data of the adjacent tooth surface is reduced, so that a plurality of The three-dimensional data of the tooth surface is more in line with the real state, which improves the accuracy of image stitching.

A flowchart of still another embodiment of the oral image data processing method of the present invention is shown in FIG.

In step 1701, the relative positional relationship of the mirrors in the mirror group of the probe is calibrated. The relative positional relationship of the mirrors in the mirror group includes the image area of each specular reflection image and the coordinate transformation relationship of the specular reflection image.

The image area of each specular reflection image is calibrated to obtain a single-mirror surface image, which can be realized by manually marking the surface image of the tooth obtained. The image area of each specular reflection image is acquired and stored by manual segmentation calibration of the first acquired tooth surface image. Since the relative positional relationship between the mirror group and the camera is unchanged, the image area of each specular reflection image is unchanged in the subsequent image segmentation, and the image region of each specular reflection image stored is used to segment and acquire the tooth surface image. Single mirrored tooth surface image.

To calibrate the coordinate transformation relationship of the specular reflection image, it is necessary to obtain the rotation factor R and the translation matrix T required to unify the different specularly reflected single-mirror tooth surface images into the same coordinate system. Since the relative positional relationship of the mirror groups of the mirror group is constant, the coordinate conversion relationship between the mirror surfaces is unchanged. In the first measurement, the image of the tooth surface obtained by different mirrors can be compared, and the single-mirror surface image of the refraction of the calibration has two mirrors of a common area, and the corresponding points of the common area of the reflected image have at least three pairs of corresponding points. The points with obvious features can be selected as corresponding points to obtain the transformation matrices R, T.

In step 1702, a tooth surface image is acquired from the camera. Since the tooth surface image is reflected by the mirror group to the camera, and the mirror group is composed of a plurality of mirrors in a predetermined position relationship The face is constructed such that the tooth surface image is composed of a plurality of lens reflection images in a predetermined positional relationship.

In step 1703, the tooth surface image acquired by the camera is spliced according to the relative positional relationship of the mirrors in the mirror group, and the tooth three-dimensional data is acquired. The relative positional relationship of the mirrors in the mirror group is obtained by step 1701. Since the relative positional relationship of the lenses is certain, the surface images of the different specularly reflected teeth can be segmented, coordinate-converted and spliced according to the relative positional relationship of the lenses, thereby obtaining three-dimensional data of the multi-specular tooth surface. When a single-mirror tooth surface image of all sides of the tooth is acquired, three-dimensional data of a single tooth surface can be obtained.

In step 1704, three-dimensional data of a single tooth surface of an adjacent tooth is spliced according to a common area of three-dimensional data of a single tooth surface of an adjacent tooth, and three-dimensional data of the plurality of tooth surfaces is obtained.

Such a method can eliminate the error caused by the relative positional relationship between the manufacturing process and the precision of the mirror by the calibration, thereby eliminating the influence of the error on the data processing, and more accurately segmenting and splicing the image of the tooth surface, increasing the data. The accuracy of the processing.

The oral cavity detecting method of the present invention comprises the combination of any one of the above oral image detecting method and the oral image data processing method, and performs image processing in the oral image data processing method by using the tooth surface image acquired in the oral image detecting method, thereby acquiring Three-dimensional data of teeth. In one embodiment, a flow chart of the oral detection method is shown in FIG.

In step 1801, the projector projects a stripe pattern onto the mirror set of the probe and is reflected by the mirror set onto the tooth surface. The mirror group is composed of a plurality of mirrors arranged in a predetermined positional relationship. There are at least two mirrors that reflect the 2-4 faces of the tooth surface; it can be 3 faces that reflect the 3-4 faces of the tooth surface; it can also be 4 faces that reflect the 5 faces of the tooth. In one embodiment, the projector may be a stripe projector and the projected stripe pattern is a sinusoidal stripe pattern. Since the sinusoidal stripe pattern changes its phase when projected onto a surface of a different height object, three-dimensional reconstruction can be performed according to the sinusoidal stripe pattern of the tooth surface.

In step 1802, the mirror set of the probe will include a tooth surface map of the stripe pattern Like reflection to the outside of the mouth. The stripe pattern may be a sinusoidal stripe pattern, and the phase information of the reflected sinusoidal stripe pattern can be reduced to height information of the tooth surface due to the difference in height of the tooth surface. Since the mirror group includes a plurality of lenses in a predetermined positional relationship, the reflected tooth surface image is composed of a plurality of specularly reflected images.

In step 1803, the camera acquires an image of the tooth surface reflected from the mirror set.

In step 1804, a tooth surface image is acquired from the camera. Since the tooth surface image is reflected by the mirror group to the camera, and the mirror group is composed of a plurality of mirror surfaces in a predetermined positional relationship, the tooth surface image is composed of a plurality of lens reflection images in a predetermined positional relationship. In one embodiment, the tooth surface image comprises a surface image of 2 to 5 faces of the tooth.

In step 1805, the tooth surface image acquired by the camera is spliced according to the relative positional relationship of the mirrors in the mirror group, and the tooth three-dimensional data is acquired. Since the relative positional relationship of the lenses is constant, the tooth surface images reflected by different specular surfaces can be obtained according to the relative positional relationship of the lenses, thereby obtaining three-dimensional data of the multi-specular tooth surface. If the tooth surface image contains a surface image of the entire surface of the tooth, it is possible to acquire three-dimensional data of the tooth surface of a single tooth. In one embodiment, in the detection of adjacent teeth, the image of the tooth surface of the adjacent tooth may be controlled to have a certain common area, so that after acquiring the three-dimensional data of the single tooth surface of the adjacent tooth, three pairs are calibrated according to the common area. The corresponding points above achieve the splicing of the three-dimensional data of the single tooth surface of the adjacent teeth, and obtain three-dimensional data of the plurality of tooth surfaces. In one embodiment, full oral tooth three dimensional data can be obtained.

By such a method, the tooth surface image including the plurality of faces can be acquired by the oral cavity detecting device, and the tooth surface image of the plurality of faces can be spliced by the oral image data processing device according to the relative positional relationship of the mirror faces of the mirror group in the oral detecting device. Obtain three-dimensional data of teeth, thereby reducing errors in image stitching and increasing the accuracy of detection. Since the probe mirror group can simultaneously reflect images of multiple surfaces of the tooth, the number of movements of the probe and the number of image stitching can be reduced during use, thereby improving detection efficiency.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not the alignment limitations; although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that Specific implementation The modifications and equivalents of some of the technical features are included in the scope of the technical solutions claimed in the present invention without departing from the spirit of the present invention.

Claims

An oral detecting device, comprising:

a probe for detecting an image of a tooth surface in the oral cavity, the probe comprising a mirror group, the mirror group comprising a plurality of mirrors arranged in a predetermined positional relationship for reflecting a stripe image projected to the mirror group Go to the surface of the tooth in the mouth and reflect the image of the surface of the tooth to the outside of the mouth;

a projector for projecting a stripe image to the mirror group of the probe, and reflecting the surface of the tooth through the mirror group;

a camera for acquiring a reflected image of the tooth surface comprising a stripe pattern from a mirror set of the probe.
The apparatus according to claim 1, wherein the mirror group of the probe comprises two or more mirrors, and the two or more mirrors are arranged in a predetermined positional relationship.
The apparatus of claim 1 wherein said projector is a fringe projector and said fringe image is a sinusoidal fringe image.
The device of claim 1 wherein:

The length of the probe is 60-85 mm, and the cross-sectional dimension of the probe is not more than 30 mm*20 mm;

The projector and the camera are located at an end of the probe away from the mirror group;

The photographic baseline length of the projector and the camera is not less than the sum of the lens radii of the projector and the camera.
An oral image data processing device, comprising:

a parameter storage module for storing a relative positional relationship of the mirrors in the mirror group of the probe;

An image acquisition module, configured to acquire a tooth surface image from a camera, wherein the tooth surface image is a specularly reflected image of a mirror group of the probe in a predetermined positional relationship, and includes image information of the plurality of faces of the tooth;

An image processing module, configured to splicing three-dimensional data obtained according to the image of the tooth surface acquired by the camera according to a relative positional relationship of the mirrors in the mirror group, and acquiring the tooth Three-dimensional data of teeth.
The device according to claim 5, wherein said tooth surface image comprises a sinusoidal stripe projected by said projector through said mirror group to a tooth surface, said mirror group comprising said sinusoidal stripe The tooth surface image is to the camera.
The apparatus according to claim 6, wherein the image processing module comprises:

An image segmentation unit, configured to segment the tooth surface image according to a relative positional relationship of mirrors in a mirror group of the probe in the parameter storage module, to obtain a mirror-reflected single-mirror tooth surface image, wherein the mirror group The relative positional relationship of the mirror surface includes an image area of each specularly reflected image in the image of the tooth surface;

a phase measuring unit, configured to acquire, according to the sinusoidal stripe of the single mirror tooth surface image, three-dimensional data of a single mirror tooth surface in a single mirror tooth surface image based on phase measurement profilometry;

a three-dimensional tiling unit, configured to unify the three-dimensional data of the single-mirror tooth surface into the same coordinate system according to the relative positional relationship of the mirrors in the mirror group of the probe in the parameter storage module, to obtain three-dimensional data of a single tooth surface, wherein The relative positional relationship of the mirrors in the mirror group includes a coordinate conversion relationship of each specular reflection image of the mirror group.
The apparatus according to claim 7, wherein the image processing module further comprises:

Adjacent splicing unit, for arranging the three-dimensional data of the single tooth surface of the adjacent tooth according to a common area of the three-dimensional data of the single tooth surface of the adjacent tooth, and obtaining three-dimensional data of the plurality of tooth surfaces.
The device according to claim 5, characterized in that

The device also includes:

a calibration module, configured to calibrate a relative positional relationship of the mirrors in the mirror group of the probe, comprising: an area calibration unit, configured to calibrate an image area of each specular reflection image in the image of the tooth surface acquired by the camera; coordinate calibration a unit for acquiring a coordinate conversion relationship of the specular reflection image according to three or more corresponding points in a common area of the three-dimensional data of the single-mirror tooth surface acquired from different mirrors;

The parameter storage module is configured to acquire an image region of each specular reflection image in the tooth surface image from the region calibration unit, and acquire a coordinate conversion relationship of the specular reflection image from the coordinate calibration unit.
An oral detection system comprising:

A mouth detecting device according to any one of claims 1 to 4, for projecting a stripe image onto a tooth surface in the oral cavity, and acquiring a tooth surface image including a stripe pattern by a camera;

A dental image data processing apparatus according to any of claims 5-9 for acquiring three-dimensional data of teeth based on a tooth surface image acquired by said oral detecting means.
An oral probe characterized by comprising a mirror group, the mirror group comprising a plurality of mirrors arranged in a predetermined positional relationship for reflecting a stripe image projected to the mirror group to The tooth surface in the oral cavity reflects the image of the tooth surface containing the stripe pattern out of the mouth to detect an image of the tooth surface containing images of the plurality of faces of the tooth.
The probe according to claim 11, wherein the mirror group comprises two mirrors on both sides of the central axis of the probe detecting area for detecting 2 to 4 images of the tooth surface.
The probe according to claim 12, wherein the angle between the mirror surface and the probe axis is in the range of [43°, 47°], and the rotation angle of the normal vector around the probe detection area is [41°, 49°]. Within the scope.
The probe according to claim 11, wherein the mirror group comprises three mirrors respectively located on the central axis and the central axis of the probe detecting area for detecting the image of the surface of the tooth surface 3-4.
The probe according to claim 14, wherein the mirror pitch angle of said central axis is in the range of [45°, 60°]; the angle between the mirror surfaces on both sides of said central axis and the axis of the probe is [43] In the range of °, 47°], the rotation angle of the normal vector around the probe detection area is in the range of [41°, 49°].
The probe according to claim 11, wherein said mirror group comprises four mirrors for detecting 4 to 5 face images of the tooth surface, wherein the two mirror faces are located at the proximal end of the axis of the probe detecting region And the distal end; the other two mirrors are located at the The probe detects the sides of the central axis of the area.
The probe according to claim 16, wherein the specular elevation angle of the proximal portion is in the range of [30°, 45°], and the specular elevation angle of the distal end portion is [45°, 60°] Within the range; the angle between the mirrors on both sides of the axis of the probe detection area and the axis of the probe is in the range of [43°, 47°], and the rotation angle of the normal vector around the probe detection area is [41°, 49° ] within the scope.
An oral image detecting method, comprising:

The projector projects a fringe image to the mirror group of the probe, and is reflected by the mirror group to the tooth surface, wherein the probe includes a mirror group, and the mirror group includes a plurality of mirror surfaces arranged in a predetermined positional relationship;

a mirror set of the probe reflects an image of the tooth surface including the stripe pattern to the outside of the mouth;

A camera acquires an image of the tooth surface reflected from the set of mirrors.
The method according to claim 18, wherein the mirror group of the probe comprises two or more mirrors, and the two or more mirrors are arranged in a predetermined positional relationship.
The method of claim 18 wherein said projector is a fringe projector and said fringe image is a sinusoidal fringe image.
An oral image data processing method, comprising:

Acquiring a tooth surface image from the camera, wherein the tooth surface image is an image reflected by a mirror group formed by a mirror surface in a predetermined positional relationship of the probe, and includes image information of the plurality of faces of the tooth;

And acquiring three-dimensional data of the teeth according to the three-dimensional data obtained from the image of the tooth surface acquired by the camera according to the relative positional relationship of the mirrors in the mirror group.
The method according to claim 21, wherein the relative positional relationship of the mirrors in the mirror group comprises an image region of each specularly reflected image in the image of the tooth surface, and a coordinate conversion relationship of each specularly reflected image of the mirror group.
The method of claim 22 wherein said tooth surface image comprises sinusoidal fringes projected by said projector through said set of mirrors to a tooth surface, said mirror group comprising said sinusoidal fringes Image of the tooth surface to the phase machine.
The method according to claim 23, wherein the framing the image of the tooth surface acquired by the camera according to the relative positional relationship of the mirrors in the mirror group to obtain the three-dimensional data of the tooth comprises:

Dividing the tooth surface image according to an image region of each specular reflection image in the tooth surface image to obtain a mirror-reflected single-mirror tooth surface image;

Obtaining three-dimensional data of the single-mirror tooth surface in the single-mirror tooth surface image based on the phase measurement profilometry according to the sinusoidal fringe of the single-mirror tooth surface image;

According to the coordinate conversion relationship of each specular reflection image of the mirror group, the three-dimensional data of the single mirror tooth surface is unified into the same coordinate system, and three-dimensional data of the single tooth surface is obtained.
The method according to claim 24, wherein the splicing the image of the tooth surface acquired by the camera according to the relative positional relationship of the mirrors in the mirror group to obtain the three-dimensional data of the tooth further comprises:

According to a common area of the three-dimensional data of the single tooth surface of the adjacent teeth, the three-dimensional data of the single tooth surface of the adjacent teeth is spliced to obtain three-dimensional data of the plurality of tooth surfaces.
The method of claim 21, further comprising:

Calibrating the relative positional relationship of the mirrors in the mirror group of the probe, including:

Aligning an image area of each specularly reflected image in the image of the tooth surface acquired by the camera;

Obtaining a coordinate conversion relationship of the specular reflection image according to three or more corresponding points in a common region of the three-dimensional data of the single-mirror tooth surface acquired from different mirror surfaces.
An oral detection method comprising:

The oral image detecting method according to any one of claims 18 to 20, wherein a stripe pattern is projected onto a tooth surface in the oral cavity, and an image of the tooth surface including the stripe pattern is obtained by a camera;

The oral image data processing method according to any one of claims 21 to 26, wherein the three-dimensional data of the teeth is acquired based on the image of the tooth surface.