Image-based transparent object defect detection

Abstract

A defect detection system for a transparent three-dimensional (3D) object includes a transparent platform configured to support the transparent 3D object, a plurality of light sources disposed below the transparent platform to illuminate the transparent 3D object through the transparent platform, a plurality of cameras, and a computing device. The plurality of cameras are disposed around and angled relative to the transparent platform above a plane of the transparent platform and are configured to generate a plurality of images of the transparent 3D object from a plurality of directions when the transparent 3D object is illuminated by one or more of the plurality of light sources. The computing device is configured to process the plurality of images to determine whether the transparent 3D object includes a defect and output an indication of whether the transparent 3D object includes a defect.

Description

Technical Field

[0001] This disclosure relates to the field of manufacturing transparent products, and more specifically to detecting defects in or on transparent (3D) printed products. Background Technology

[0002] Image-based defect detection is problematic for transparent objects because it is difficult to image transparent objects effectively. Any defect detection performed on a transparent object without a high-quality image for evaluation will be inaccurate. Attached Figure Description

[0003] The present disclosure is illustrated by way of example and not limitation in the accompanying drawings, in which the same reference numerals indicate similar elements. It should be noted that different references to "a" or "one" embodiment in this disclosure do not necessarily refer to the same embodiment, and such references refer to at least one.

[0004] Figure 1A A perspective view of one embodiment of an image-based quality control system according to an embodiment of the present disclosure is shown, which performs automatic defect detection on transparent three-dimensional (3D) objects.

[0005] Figure 1B An embodiment according to this disclosure is shown. Figure 1A A side view of an image-based quality control system.

[0006] Figure 2A It is an optical image of an image-based quality control system according to an embodiment of the present disclosure.

[0007] Figure 2B This is a schematic diagram of a side-view camera according to an embodiment of the present disclosure.

[0008] Figure 3 An illumination assembly for a side lamp of an image-based quality control system according to an embodiment of the present disclosure is shown.

[0009] Figure 4 This is a light propagation diagram of an image-based quality control system according to an embodiment of the present disclosure.

[0010] Figure 5A These are example images captured by a top-down camera of an image-based quality control system according to embodiments of the present disclosure.

[0011] Figures 5B to 5E These are example images captured by different side-view cameras of an image-based quality control system according to embodiments of the present disclosure.

[0012] Figure 6A flowchart is shown of a method for performing defect detection on a transparent 3D object according to an embodiment of the present disclosure.

[0013] Figure 7 A flowchart illustrating a method for determining the identifier of a transparent 3D object according to an embodiment of the present disclosure is shown.

[0014] Figure 8 A flowchart is shown of a method for processing an image captured by an imaging system to detect defects in a transparent 3D object according to an embodiment of the present disclosure.

[0015] Figure 9 A flowchart is shown of a method for processing an image captured by an imaging system to detect defects in a transparent 3D object according to an embodiment of the present disclosure.

[0016] Figure 10 A block diagram of an example computing device according to an embodiment of the present disclosure is shown.

[0017] Figure 11 A tooth repositioning appliance according to an embodiment of the present disclosure is shown.

[0018] Figure 12 A tooth repositioning system according to an embodiment of the present disclosure is shown.

[0019] Figure 13 An orthodontic treatment method using multiple appliances according to an embodiment of the present disclosure is illustrated.

[0020] Figure 14 A method for designing orthodontic appliances to be manufactured directly is shown according to embodiments of the present disclosure.

[0021] Figure 15 A method for digitally planning orthodontic treatment according to embodiments of the present disclosure is shown. Detailed Implementation

[0022] The embodiments described herein cover systems, methods, and / or computer-readable media suitable for image-based quality control (IBQC) of transparent custom-made products. Transparent custom-made products can be customized medical devices. For example, in some embodiments, image-based quality control systems and methods can be implemented in the post-manufacturing inspection of transparent orthodontic aligners. Quality control of transparent custom-made products is particularly challenging, especially in the manufacture of orthodontic aligners that must be individually customized for each patient. Furthermore, each aligner in a series of aligners used to treat a single patient is unique compared to other aligners in the same series because each aligner is specific to a different treatment phase. Each patient receives pairs of aligners for each treatment phase, one unique aligner for treating the upper dental arch and one unique aligner for treating the lower dental arch. In some cases, a single treatment may involve 50-60 phases to treat a complex case, meaning that there are 100-120 aligners uniquely manufactured for a single patient. Capturing high-quality images of transparent objects for automated image-based quality control is particularly difficult, which reduces the accuracy of such IBCQ systems when used with transparent 3D objects.

[0023] When manufacturing orthodontic appliances for patients worldwide, hundreds of thousands of completely unique and custom-made appliances may need to be produced every day. Therefore, quality control of custom-made products is a particularly challenging task. Quality control can be performed on manufactured appliances to ensure they are defect-free or that defects are within acceptable thresholds. The quality control process may aim to detect one or more of the following quality issues: arch misalignment, curvature, cutting line deviation, debris, webbing, improper trimmed attachments, missing attachments, burrs, eversion, dynamic ridge problems, material breakage, hooks that are too short, air bubbles, etc. Furthermore, some transparent 3D objects may be 3D printed objects. In this case, additional types of defects may occur. Examples of additional defects that may be encountered with transparent 3D printed objects include surface defects, internal volume defects, interface defects, and delamination defects. For example, gaps may exist between one or more thin layers of a transparent 3D object due to the manufacturing process, causing air to become trapped in those gaps. This type of defect is referred to as an "internal volume defect" in this paper. In another example, particles (e.g., debris) may form or aggregate on the surface of a transparent 3D object. This paper refers to this type of defect as a "surface defect." In another example, holes (e.g., pits) may form at the interface between the internal volume and the surface of the transparent 3D object. This paper refers to this type of defect as an "interface defect." This paper may refer to internal volume defects, surface defects, interface defects, and other defects caused during manufacturing (e.g., during the manufacture of a 3D printed object) as "delamination defects." Delamination defects may also include printed layers with abnormal thickness (e.g., layers exceeding a layer thickness threshold) and delamination between printed layers.

[0024] Typically, technicians manually perform quality control processes to inspect manufactured orthodontic appliances. However, this manual quality control process can be very time-consuming and prone to errors due to the inherent subjectivity of technicians. Therefore, embodiments of the present invention can provide a more scalable, automated, and / or objective orthodontic appliance quality control process.

[0025] In use cases of clear dental appliances (e.g., orthodontic appliances, retainers), detecting quality problems enables the repair of clear dental appliances to eliminate quality defects, thereby preventing the delivery of deformed or defective clear dental appliances, and / or reworking deformed clear dental appliances before delivery. In some embodiments, the identification of quality problems may be based on comparing an image of the clear dental appliance with a digitally generated model of the clear dental appliance. In some embodiments, a digital model of each clear dental appliance may be included in a digital file associated with the orthodontic appliance. Optionally, the digital file associated with the manufactured clear dental appliance may provide a digital approximation of the manufactured clear dental appliance's characteristics (e.g., the outer surface of the clear dental appliance, a two-dimensional projection of the outer surface onto a plane, etc.).

[0026] In some embodiments, the digitally generated model of the clear dental appliance and / or the digitally approximate characteristics of the clear dental appliance may be based on manipulation of a digital model of the dental arch associated with a treatment phase in the treatment plan. In some embodiments, the identification of quality problems may be based on a comparison of an approximate first characteristic of the clear dental appliance with a determined second characteristic of the clear dental appliance (e.g., determined by one or more images of the clear dental appliance). Embodiments can improve detection results by eliminating human error (e.g., false positives and false negatives). Furthermore, embodiments can reduce the amount of time spent performing quality control, thereby shortening the delivery cycle of orthodontic appliances so that clear dental appliances can be distributed to customers on schedule.

[0027] In some embodiments, the machine-based defect detection system and method can be implemented in the inspection of molds for orthodontic appliances prior to manufacturing the appliance. In other embodiments, the machine-based defect detection system and method can be implemented in the inspection of orthodontic appliances and / or other dental appliances manufactured by direct manufacturing. In other embodiments, the machine-based defect detection system and method can be used to inspect orthodontic appliances formed on 3D-printed objects (e.g., on 3D-printed molds). Defect detection can also be performed on other dental appliances such as palatal expanders, dental attachments, prefabricated attachment templates for placing attachments on teeth, sleep apnea devices, mouthguards, retainers, etc.

[0028] The disclosed embodiments can be implemented using various software and / or hardware components. For example, the software components may include computer instructions stored in a tangible, non-transitory computer-readable medium, which are executed by one or more processing devices to perform image-based quality control on customized clear dental appliances (e.g., orthodontic appliances). The software may set up and calibrate a camera included in the hardware components, capture images of the clear dental appliance from various angles and / or directions using the camera, generate a digital model of the clear dental appliance, perform an analysis comparing the digital model of the clear dental appliance with the images of the clear dental appliance to detect one or more quality problems (e.g., deformation, cutting line deviation, etc.), and classify the clear dental appliance based on the analysis results and / or machine learning-based applications.

[0029] In some embodiments, a digital file associated with a transparent 3D dental appliance (e.g., an orthodontic appliance) is received, which is custom-made for a patient's dental arch and is being inspected. In some embodiments, the dental appliance includes identification information, such as a custom barcode or component identification number. A first image of the transparent 3D dental appliance can be generated using one or more imaging devices (e.g., a camera, a blue laser scanner, a confocal microscope, a stereo image sensor, an X-ray device, etc.). The dental appliance identification information can be captured in the first image and interpreted by the IBQC system. Alternatively, a technician can manually input this information into the IBQC system, enabling the IBQC system to retrieve the digital file. In another embodiment, the dental appliance sorting system can sort a series of dental appliances in a known order. The IBQC system can obtain the order of the dental appliances from the dental appliance sorting system to understand which dental appliances are currently being inspected and in what order they arrived at the inspection station. Optionally, dental instruments may arrive at the IBQC system on a tray carrying dental instrument identification information (e.g., RFID tags, barcodes, serial numbers, etc.), which is then read by the inspection system. The IBQC system can then retrieve the digital file associated with the dental instrument based on the identification information.

[0030] In one embodiment, the IBQC system includes an improved design comprising multiple fixed side-view cameras that can operate together, significantly accelerating image-based defect detection compared to previous IBQC systems. In another embodiment, the IBQC system may include a top-view camera and multiple side-view cameras. Furthermore, the IBQC system may include a transparent platform configured to support a transparent 3D object (e.g., a dental instrument) during imaging. The IBQC system may also include a first illumination system for outputting a first illumination during top-view camera imaging, and a second illumination system for outputting a second illumination during imaging by the multiple side-view cameras. During the output of the first illumination, the top-view camera may generate one or more top-view images of the transparent 3D object at a first moment. The top-view images may be used to determine the identifier (ID) of the transparent 3D object and / or to perform defect detection on the transparent 3D object. Furthermore, during the output of the second illumination, the multiple side-view cameras may simultaneously or nearly simultaneously generate corresponding side-view images of the transparent 3D object at a second moment.

[0031] Side-view images can also be used to perform defect detection, with each side-view image targeting a different region or viewpoint of the transparent 3D object. A first illumination system and a second illumination system can be positioned below the transparent platform and configured to illuminate the transparent 3D object through the platform during image capture. Side-view cameras can be positioned above the plane of the transparent platform and at an angle relative to the platform (e.g., each side-view camera can have the same angle). The side-view cameras can be positioned around the transparent platform above its plane and at an angle relative to it, and can be configured to generate multiple images of the transparent 3D object from multiple directions. For example, the side-view cameras can generate a set of side-view images, where the images in the set can be captured simultaneously by different cameras. By using multiple side-view cameras in parallel to simultaneously capture images of multiple different regions or viewpoints of the transparent 3D object, the inspection of transparent 3D objects can be significantly accelerated compared to a system using only a single side-view camera.

[0032] The embodiments discussed herein provide an improved IBQC system that is cheaper and more accurate than previous IBQC systems. Compared to previous IBQC systems, the IBQC system in these embodiments produces brighter images, showing surfaces with higher contrast and greater clarity. In these embodiments, the side-view camera and illumination system are at corresponding angles relative to each other and relative to the transparent platform, these angles being selected to provide optimal depth of focus, uniform illumination, and field of view.

[0033] This document discusses embodiments of orthodontic appliances (also simply referred to as appliances) as a type of transparent 3D object. However, the embodiments are also extended to other types of dental appliances, such as orthodontic retainers, orthodontic splints, sleep appliances inserted into the mouth (e.g., for minimizing snoring, sleep apnea, etc.), palatal expanders, dental attachments, and so on. The embodiments are also applicable to other types of transparent 3D objects. Other transparent 3D objects may include 3D-printed palatal expanders, removable mandibular repositioning devices, and removable surgical fixation devices. Therefore, it should be understood that the embodiments of the appliances referred to herein are also applicable to other types of dental appliances and / or transparent 3D objects, and particularly to other types of transparent dental appliances.

[0034] The dental appliances described in this article can be formed by 3D printing a mold and then thermoforming the dental appliance onto the mold. Alternatively, the dental appliance can be directly printed using 3D printing. In both cases, the same or similar quality control procedures can be performed to detect defects in the dental appliance.

[0035] In the embodiments disclosed herein, each manufactured orthodontic appliance or other dental appliance (e.g., a removable surgical fixation device, a removable mandibular repositioning device, a removable palatal expander) may be sent to an image-based quality control (IBQC) station to detect one or more quality problems (e.g., deformity) in the appliance. Alternatively, appliances marked for quality inspection may be sent to an IBQC station. For example, digital files of the appliances may be input into machine learning models, numerical simulations, rule engines, and / or other modules to determine whether there is a higher risk that any of these appliances will be defective. Machine learning models, numerical simulations, rule engines, and / or other modules may identify a subset of appliances to be inspected using the IBQC system. Optionally, the IBQC system and methods may classify inspected appliances as deformed, potentially deformed, or not deformed, and may also provide recommendations including their analysis results (e.g., further inspection required, remanufacturing required, approval, etc.).

[0036] Now refer to the attached diagram, Figure 1A A perspective view of one embodiment of an IBQC system 100 according to an embodiment of the present disclosure is shown, which performs automatic defect detection of transparent three-dimensional (3D) objects. Figure 1B An embodiment according to this disclosure is shown. Figure 1A A side view of the IBQC system.

[0037] refer to Figures 1A to 1BThe IBQC system 100 may include a workbench 112, wherein a transparent platform 103 is disposed within a cavity of the workbench 112. In an embodiment, the transparent platform 103 is a fixed platform (e.g., without actuators for moving, rotating, etc.). Furthermore, a vertical bracket supporting a top-view camera 101 and one or more displays 110 may be mounted to the workbench 112.

[0038] In some embodiments, the transparent platform 103 (e.g., a glass plate or plexiglass plate) is located at the center of the workbench 112. Alternatively, the transparent platform 103 may not be located at the center. As shown, the transparent platform 103 may have a square shape, and side-view cameras may be positioned on each side of the square shape of the transparent platform 103. Alternatively, the transparent platform 103 may have other shapes, such as circular, elliptical, rectangular, etc. In one embodiment, the transparent 3D object 114 is positioned on the transparent platform 103 during quality control operations concerning the transparent 3D object 114.

[0039] In an embodiment, the transparent platform 103 may include feature patterns marked on or within it. The feature patterns may be geometric patterns capable of determining the position and / or orientation of the transparent 3D object 114 relative to the top-view camera 101 and / or the side-view camera 102 and / or other components of the IBQC system 100 (e.g., relative to the coordinate system of the IBQC system 100). The feature patterns may be circular patterns of points and / or other shapes arranged around the center of the transparent platform and configured to surround the transparent 3D object. In an embodiment, the transparent 3D object should be placed within the area defined by the feature patterns such that the feature patterns surround the 3D object 114. In some embodiments, the number of points and / or other shapes in the feature patterns and the diameter of the circles may be selected such that at least four points and / or other shapes are always within the field of view of the top camera, and the 3D object 114 rarely obscures points or other shapes in the side-view image. In an embodiment, the 3D object 114 may be positioned on the platform 104 surrounded by the feature patterns while an image of the transparent 3D object is captured and subsequently processed by processing logic.

[0040] As shown, a top-view camera 101 can be positioned above a transparent platform 103 and has an imaging axis that is substantially perpendicular to the transparent platform 103 (e.g., perpendicular to the plane defined by the transparent platform). In an embodiment, the top-view camera 101 is located approximately directly above the center of the transparent platform 103, and the optical axis of the top-view camera 101 is located at the center of a feature pattern on the transparent platform 103. In some embodiments, the top-view camera 101 is a high-definition camera. The top-view camera 101 can be a two-dimensional camera or a 3D camera (e.g., a pair of cameras that generate stereoscopic image pairs, a camera that projects a structured light pattern onto a 3D object 114 and an associated structured light projector, etc.). The top-view camera 101 can be configured to acquire a top-view image of the 3D object 114 using specific illumination settings so that the 3D object 114 is visible in the top-view image. In one embodiment, the top-view camera 101 has a fixed position. In some embodiments, the top-view camera 101 is configured to capture a wide field of view of the transparent 3D object in a first configuration. In some embodiments, the top-view camera 101 is configured to capture a narrow field of view of the transparent 3D object in a second configuration. In some embodiments, the top-view camera 101 includes a spectral filter configured to allow light of wavelengths (or wavelength ranges) output by the first illumination system 106 while blocking light outside of those wavelengths (or wavelength ranges). This allows for the filtering out of external visible light, preventing it from reaching the top-view camera 101. For example, the top-view camera 101 may include a spectral filter that transmits infrared or near-infrared wavelengths (e.g., wavelengths from 820 to 930 nm) while blocking other wavelengths. In embodiments, the top-view camera 101 has flexible positioning and includes a wide range of adjustment capabilities in both spatial and angular coordinates (e.g., adjustable in up to five to six degrees of freedom) to ensure proper camera setup. For example, in embodiments, the top-view camera 101 may be mounted to a bracket that allows movement in three orthogonal directions and tilting about two axes. Once adjusted, the top-view camera 101 can be held in a fixed position until further adjustments are required.

[0041] Multiple side-view cameras 102 may also have fixed positions around the transparent platform 103. As shown, each of the side-view cameras 102 may be positioned above the plane of the transparent platform 103 and may have a fixed angle relative to the transparent platform (e.g., relative to the plane of the transparent platform). In an embodiment, the side-view cameras 102 are positioned around the center of a feature pattern on the transparent platform 103, wherein the optical axis of the side-view camera 102 is tilted at an angle and points to the center of the feature pattern on the transparent platform 103. In an embodiment, each of the side-view cameras 102 has the same angle relative to the plane defined by the transparent platform 103. The side-view cameras 102 may be two-dimensional cameras or 3D cameras. In one embodiment, the side-view camera is a high-resolution camera and / or a high-speed camera (e.g., capable of capturing an image per millisecond). The side-view cameras 102 may have fixed positions and may respectively acquire one or more images of different regions or viewpoints of the 3D object 114. In some embodiments, the side-view cameras 102 each include a spectral filter configured to transmit light of a certain wavelength output by the second illumination system 124. This allows for the filtering out of external visible light, preventing it from reaching the side-view camera 102. For example, the side-view camera 102 may include a spectral filter that transmits light of infrared or near-infrared wavelengths (e.g., 820 to 930 nm) while blocking other wavelengths.

[0042] In embodiments, the side-view camera 102 has flexible positioning and includes a wide range of adjustment capabilities in both spatial and angular coordinates (e.g., adjustable in up to five to six degrees of freedom) to ensure proper camera setup. For example, in embodiments, the side-view camera 102 can be mounted to a bracket that allows movement in three orthogonal directions and tilting about two axes. Once adjusted, the side-view camera 102 can be held in a fixed position until further adjustment is required. In some embodiments, each of the side-view cameras 102 includes a lens with a focal plane, an image sensor with an image plane, and a tilt shift adapter that angles the focal plane relative to the image plane. In embodiments, the tilt shift adapter ensures increased depth of field for the side-view image.

[0043] In this embodiment, the specific focal length of the lens, the size of the camera sensor, and the size of the light can be selected based on the size of the transparent 3D object 114 being inspected, system design constraints, system footprint, space required for interaction with other devices (e.g., a robotic arm), and the conditions for obtaining high-quality images of the transparent object as described in the application. For example, for a square inspection area with a side length of 100 mm, in order to ensure that the upper projection of the system does not exceed 1 m with a camera sensor size of 10 × 10 mm, a lens with a focal length of 25-35 mm is needed for the side camera and a lens with a focal length of 25-75 mm is needed for the top camera. A top light 106 with a size of 200 × 200 mm is required, and the size of the matte white board 126 of the second illumination system should be 200 × 120 mm.

[0044] The IBQC system 100 includes multiple light sources, which may include one or more first light sources as part of a first lighting system 106 and one or more second light sources as part of a second lighting system 124. In some embodiments, the axis of the first light source is substantially perpendicular to the transparent platform. In embodiments, both the first lighting system 106 and the second lighting system 124 may be positioned below the transparent platform 104. Each lighting system may include one or more light sources (also referred to as light-emitting elements). Each light source may include at least one of an incandescent bulb, a fluorescent bulb, a light-emitting diode (LED), a neon light, etc. In one embodiment, each light source may emit light of a specific wavelength or spectrum. For example, in one embodiment, each light source may emit infrared or near-infrared light (e.g., light with a wavelength range of 820-930 nm). In embodiments, the first lighting system 106 and / or the second lighting system 124 are configured to provide a spatial brightness distribution that makes defects in the transparent 3D object 114 visible in the image. The first lighting system 106 and the second lighting system 124 may be configured to achieve appropriate contrast of build lines in the image and ensure no glare due to external light.

[0045] The first illumination system 106 may be a top-down illumination system for providing uniform illumination of the transparent 3D object 114 during image capture by the top-down camera 101. In some embodiments, the top-down camera 101 is a variable lens aperture imaging system having at least two lens aperture settings (e.g., lens F-value settings). In one embodiment, the variable lens aperture imaging system has a first lens aperture value setting of approximately 6.0 (e.g., approximately 5.6-8) and a second lens aperture value setting of approximately 2.0 (e.g., approximately 1.8-2.5). The first lens aperture can be used to acquire an image for reading laser markings in the 3D object 114. The second lens aperture can be used for imaging for defect detection. In an embodiment, the lens aperture can be automatically changed by using a motorized lens. This allows for remote control of the lens.

[0046] The second lighting system 124 may be a side-view lighting system comprising a plurality of lighting sections that provide uniform illumination of the transparent 3D object 114 during image capture by the side-view camera 102. Each lighting section of the second lighting system 124 may include one or more light sources 125, a matte white panel 126, and a matte black panel 128. The matte white panel 126 is configured to guide light output from the one or more light sources 125 through the transparent 3D object and toward a corresponding side-view camera 102 pointing toward the lighting assembly 124. The matte black panel 128 is disposed around the matte white panel 126 and configured to absorb light output from the one or more light sources 125. The lighting system 124 may be configured, positioned, and oriented to provide uniform illumination to the side-view camera 102.

[0047] In one embodiment, each illumination segment is configured such that a side-view camera 102 pointing towards that segment provides uniform illumination to the transparent 3D object 114. In some embodiments, all illumination segments simultaneously illuminate the transparent 3D object, and the side-view cameras 102 capture images simultaneously (or nearly simultaneously), resulting in a short inspection cycle. Alternatively, the illumination segments may illuminate the transparent 3D object 114 sequentially. For example, a first illumination segment may provide illumination to the transparent 3D object during image capture by a first side-view camera 102 facing and pointing towards the first illumination segment. Subsequently, a second illumination segment may provide illumination to the transparent 3D object during image capture by a second side-view camera 102 facing and pointing towards the second illumination segment. Subsequently, a third illumination segment may provide illumination to the transparent 3D object during image capture by a third side-view camera 102 facing and pointing towards the third illumination segment. Subsequently, a fourth illumination segment may provide illumination to the transparent 3D object during image capture by a fourth side-view camera 102 facing and pointing towards the fourth illumination segment. If additional lighting sections and associated side-view cameras are included, each of these lighting sections and the associated side-view cameras can be operated sequentially as described above.

[0048] In one embodiment, the IBQC system 100 includes four side-view cameras and four lighting sections. In other embodiments, the IBQC system 100 includes fewer or more than four side-view cameras and associated lighting sections (e.g., three, five, six, seven, eight, etc.). In some embodiments, the IBQC system 100 can rapidly and continuously cycle through multiple side-view cameras 102 and associated lighting sections. Figure 3A more detailed view of the second lighting system 124 and its various lighting sections is shown. Regardless of whether the side-view camera 102 captures images simultaneously or rapidly and sequentially, the inspection time span can be reduced from approximately 10 seconds (e.g., for a system where a single side-view camera captures images from multiple locations around the transparent 3D object 114) to approximately 4 seconds.

[0049] The IBQC system 100 may include one or more controllers 104 and / or computing devices 105. The controllers 104 and / or computing devices 105 may be configured to control and manage various components of the IBQC system (e.g., top-view camera 101, side-view camera 102, first lighting system 106, second lighting system 124, etc.). The controllers 104 and / or computing devices 105 may process image captures from cameras 101, 102, may preprocess images before use (e.g., before processing images to identify the ID of the transparent 3D object and / or detect defects in the transparent 3D object 114), and may process images to identify the ID of the transparent 3D object and / or detect defects in the transparent 3D object 114. Additionally, the controllers 104 and / or computing devices 105 may be integrated with external systems (e.g., to retrieve digital files and / or information about the transparent 3D object 114), store defect detection results for the transparent 3D object, etc.

[0050] In one embodiment, controller 104 and / or computing device 105 may interact with a conveyor that can transport a transparent 3D object to and / or remove a 3D object from IBQC system 100. In another embodiment, controller 104 and / or computing device 105 may interact with one or more robotic arms to pick up a transparent 3D object 114 from a conveyor, place the transparent 3D object 114 on a transparent platform 103, retrieve the transparent 3D object 114 from the transparent platform 103, and / or place the transparent 3D object 114 on a conveyor. The robotic arm may be part of an IBQC (defect detection) system or a related system. For example, computing device 105 may determine the position and / or orientation of the transparent 3D object 114 on the transparent platform 103 and may provide coordinates to the robotic arm used to pick up the transparent 3D object 114 based on the determined position and / or orientation.

[0051] The controller 104 can control the timing of activation of each of the cameras 101, 102 and / or lighting systems 106, 124, control the intensity of generated light, etc. In an embodiment, the controller 104 provides automatic lighting control. For example, the controller 104 can send instructions to the top-view camera 101 and the first lighting system 106 to cause the top-view camera 101 to capture one or more top-view images of a transparent 3D object set on the transparent platform 103 and illuminated by the first lighting system 106. Additionally, the controller 104 can send instructions to the side-view camera 102 and the second lighting system 124 to cause the side-view camera 102 to capture side-view images of multiple different areas or perspectives of the transparent 3D object 114 set on the transparent platform 103 and illuminated by the second lighting system 124. In an embodiment, the side-view camera 102 can be instructed to generate images simultaneously or sequentially. The controller 104 can cause the top-view camera 101 and / or the side-view camera 102 to capture images of the 3D object 114. In an embodiment, the controller 104 can receive images and, according to... Figures 6 to 10 The method shown provides the image to the computing device 105 for processing. The captured image can be sent to the computing device 105, and the image inspection module on the computing device 105 can analyze the image of the transparent 3D object 114 to determine whether there are any defects in the transparent 3D object 114.

[0052] In one embodiment, a first image of the 3D object 114 may be generated by a top-view camera 101 and may include a representation of an identifier (ID) (such as a part number) printed on or formed within the 3D object 114 (e.g., as a laser mark). The first image may include a sequence of symbols displayed in different areas or from different perspectives of the 3D object 114. In one embodiment, the first image of the 3D object 114 may be generated by the top-view camera 101 while a first light source 106 is activated to provide a wide light field. The laser mark may provide identification for a specific transparent 3D object (e.g., a specific dental instrument) and may correspond to a specific digital model of the 3D object. The computing device 105 may perform optical character recognition (OCR) on the sequence of symbols to determine the ID. In other embodiments, a technician may manually enter the ID associated with the 3D object at an inspection station using an interface (such as a user interface (UI) output in display 110). In other embodiments, the ID may be obtained based on the known order and / or location of the 3D object in an object sorting system. For example, a robotic object sorting system may retrieve a 3D object and place it at a specific location in a staging area. The robotic arm can then retrieve the 3D object from the temporary storage area and determine the ID associated with the 3D object based on its position in the temporary storage area. The determined ID can then be sent to the computing device 105.

[0053] The computing device 105 can associate an image of the 3D object 114 with a determined ID. In one embodiment, the processing logic can determine a digital file associated with the ID. The digital file may include one or more characteristics associated with the 3D object 114. In one embodiment, a first characteristic may include geometry associated with at least one surface of the 3D object 114. In another embodiment, a second characteristic may include the composition of the 3D object 114.

[0054] In one embodiment, when the first illumination system 106 is configured to provide a narrow light field, the top-view camera 101 generates an image. This image can then be used to determine the outline of a 3D object. In one embodiment, the side-view camera 102 can generate multiple side-view images. Each of the multiple side-view images can depict a different viewpoint or region of the 3D object 114. In one embodiment, one or more segments of the second illumination system 124 are activated while generating the side-view images. In some embodiments, during the generation of one or more side-view images, the camera lens aperture is contracted (at least partially closed) to achieve a greater depth of field. This facilitates obtaining a sharp image in a side projection at a large angle to the vertical direction.

[0055] The quality of transparent 3D objects (e.g., orthodontic appliances, retainers, etc.) can be inspected by analyzing images (e.g., 2D images) captured by the top-view camera 101 and the side-view camera 102.

[0056] In an embodiment, the computing device 105 preprocesses the image to make its format more suitable for defect detection. Image preprocessing may include, for example, resizing, cropping, rotating, flipping, affine transformation, noise reduction and / or filtering (e.g., Gaussian filtering to smooth the image and reduce noise, median filtering to remove noise, bilateral filtering to smooth the image while preserving edges), color and / or intensity adjustment (e.g., grayscale conversion, histogram equalization to improve contrast, gamma correction, normalization, etc.), edge detection and / or feature extraction (e.g., preset edge detection, applying the Sobel operator, applying the Laplacian operator, etc.), image thresholding and / or segmentation (e.g., global thresholding, adaptive thresholding, applying the watershed algorithm, etc.) and / or morphological operations (e.g., dilation, erosion, etc.). In an embodiment, the computing device 105 may adjust the image resolution and / or scaling to achieve optimal inspection quality and speed. In embodiments, computing device 105 may use conventional image processing algorithms and / or artificial intelligence (AI) models (e.g., machine learning models) for segmentation, the AI ​​models being trained to perform semantic segmentation or instance segmentation to identify transparent 3D objects in the image. In embodiments, the image may then be cropped to remove portions of the image where transparent 3D objects are not shown. Computing device 105 may use one or more image processing algorithms and / or trained artificial intelligence (AI) models to process the image to identify defects. In some embodiments, computing device 105 uses one or more AI models trained on various defect images to process the image. In some embodiments, computing device 105 uses one or more defect detection services (e.g., those that can identify defects related to trimming and / or deformation) to process the image.

[0057] In some embodiments, a feature pattern on the transparent platform 103 can be used to determine the rotation angle and / or spatial position of a 3D object in each image. Knowing the positions of four or more points / shapes, the computing device 105 and / or controller 104 can determine the position and orientation of the transparent 3D object in the image based on calibration information of the camera relative to the feature pattern. Furthermore, some points / shapes may exist outside and / or on the circle, marked with additional points / shapes of the same or smaller size and / or shape. This allows for precise determination of the rotation angle. Circular points are objects that can be detected quickly and reliably on the image.

[0058] In some embodiments, the feature pattern on the transparent platform 103 is used to determine the blur metric of one or more regions of the captured image and to calibrate the IBQC system 100.

[0059] Once the computing device 105 receives an image of the 3D object 114, it can process the image to determine whether the 3D object 114 depicted in the image includes any defects. In one embodiment, the computing device 105 may include one or more trained artificial intelligence (AI) models (e.g., artificial neural networks, deep neural networks, machine learning models, etc.) trained to identify defects in an image of the 3D object. The AI ​​model may output a confidence value and an indication of whether a defect is identified (and / or the type of defect identified). In some embodiments, the computing device 105 includes additional logic or modules for performing defect detection, such as a comparison between a digital 3D model of the transparent 3D object and an image of the transparent 3D object. In embodiments, the AI ​​model and / or the additional logic / module may identify differences between planned and actual cut lines of the transparent 3D object (e.g., an orthodontic appliance), and / or may identify deformations in the transparent 3D object.

[0060] In this embodiment, defect detection results can be output to display 110. Additional information about the transparent 3D object can also be displayed on the display. In this embodiment, the IBQC system 100 interacts with the manufacturing enterprise system by sending information about defects, retrieving 3D models of the inspected transparent 3D object (e.g., dental appliances), etc. The IBQC system 100 can also manage interactions with one or more conveyors and / or one or more robotic arms.

[0061] In some embodiments, a calibration procedure can be performed before the initial operation of the IBQC system 100 and during maintenance. The calibration procedure can be performed to precisely position the camera such that the image scaling and angular orientation allow for accurate linear distance measurements, which can be used for deformation estimation. In embodiments, the calibration process is entirely software-controlled and uses parameters of the feature pattern captured in the image stored in a configuration file, known parameters of the feature pattern, and / or geometric parameters of the top and side imaging systems (such as camera tilt angle, distance from the rear principal point of the lens to the image plane, distance from the center of the feature pattern to the front principal point of the lens, etc.) to estimate the mechanical adjustments required for normal camera operation. Based on the calibration procedure, the user can perform mechanical adjustments according to the determined adjustments to properly calibrate the IBQC system 100.

[0062] In some embodiments, the top-view camera 101 and / or the side-view camera 102 are replaceable and can be detached and replaced with different cameras having different characteristics (e.g., different angles of field of view, depth of focus, resolution, etc.). Furthermore, the distance between the camera and the transparent platform 103 and / or the angle of the camera relative to the transparent platform 103 can be adjusted. If one or more cameras are replaced and / or adjusted, a calibration procedure can be run to recalibrate the ICQB system 100.

[0063] In some embodiments, IBQC system 100 includes improvements to the IBQC lighting system of U.S. Patent No. 11,189,021, issued November 30, 2021, the contents of which are incorporated herein by reference in their entirety. In some embodiments, IBQC system 100 includes improvements to the IBQC lighting system of U.S. Patent No. 10,783,629, issued September 22, 2020, the contents of which are incorporated herein by reference in their entirety.

[0064] Figure 2A It is an optical image of an image-based quality control system (e.g., IBQC system 100) according to an embodiment of the present disclosure. Figure 2A The optical components of the IBQC system 100 and their relative positions are shown. A top-view camera 101 and a side-view camera 102 are shown. The top-view camera 101 includes an image sensor 205, a lens 204, and a spectral filter 203. As shown, the lens 204 is positioned between the image sensor 205 and the spectral filter 203. However, in an alternative embodiment, the spectral filter 203 may be positioned between the image sensor 205 and the lens 204. The side-view camera 102 includes an image sensor 211, a lens 210, and a spectral filter 209. As shown, the lens 210 is positioned between the image sensor 211 and the spectral filter 209. However, in an alternative embodiment, the spectral filter 209 may be positioned between the image sensor 211 and the lens 210.

[0065] The side-view camera 102 has a focal plane 252 (also referred to as the lens plane) defined by a lens 210, with the imaging axis 225 perpendicular to the focal plane 252. Furthermore, the side-view camera 102 has an image plane 250 defined by the surface of an image sensor 211. A tilt-shift adapter (not shown) causes the focal plane 252 to have an angle θ relative to the image plane 250. This causes the image plane 250 to tilt at angle θ to increase the depth of field of the tilted image plane. In an embodiment, the angle θ at which the image plane 205 is tilted relative to the focal plane 252 is a function of the tilt angle α of the focal plane 252 relative to the plane defined by the transparent platform 103. Because the side-view camera 102 (e.g., the focal plane 252) is tilted relative to the transparent platform 103 (and therefore relative to a transparent 3D object placed on the transparent platform 103), this results in the captured image being focused at the center of the image while blurred at the top and bottom. However, by tilting the image plane 250 relative to the focal plane 252 by an angle θ of a suitable amount (which is a function of angle α), the image captured by the side-view camera 102 has a greater depth of field, and the top, center, and bottom of the image remain in focus.

[0066] In this embodiment, the closer the angle α is to 90 degrees, the higher the defect detection accuracy for some types of defects, but the lower the defect detection accuracy for other types of defects. Therefore, in this embodiment, the angle α is selected to optimize the accuracy for each type of defect that can be detected by the IQCB system 100. In some embodiments, the angle α is approximately 15-50 degrees (e.g., 15, 20, 25, 30, 35, 40, 45, or 50 degrees). In one embodiment, for an angle α of 25 degrees, the tilt shift adapter provides an angle θ of 10.5 degrees. In other embodiments, for an angle α of 15 degrees, angle θ is 17.7 degrees; for an angle α of 20 degrees, angle θ is 13.2 degrees; for an angle α of 30 degrees, angle θ is 8.4 degrees; for an angle α of 35 degrees, angle θ is 6.8 degrees; for an angle α of 40 degrees, angle θ is 5.7 degrees; for an angle α of 45 degrees, angle θ is 4.6 degrees; and for an angle α of 50 degrees, angle θ is 4.0 degrees. Therefore, angle θ ranges from approximately 0 to 35 degrees, depending on the angle α. The aperture values ​​of camera lenses 204 and 210 are selected to ensure a suitable depth of field for high-quality imaging.

[0067] The transparent 3D object to be examined is positioned approximately at the center of the transparent platform 103, such that the transparent 3D object is surrounded by a feature pattern of the transparent platform 103. In an embodiment, the imaging axis 220 of the top-view camera 101 is located at the center of the transparent platform 103, which includes the feature pattern discussed above. Similarly, the imaging axis 225 of the side-view camera 102 is located at the center of the transparent platform 103. In an embodiment, the image scaling ratios of the top-view camera 101 and the side-view camera 102 are selected such that the feature pattern on the transparent platform 103 occupies approximately 90% of the image size (e.g., region) of the images captured by the top-view camera 101 and the side-view camera 102.

[0068] In one embodiment, a first lighting system 106 (also referred to as a bottom lighting system or a top-view lighting system) is positioned below the transparent platform 103, and the center of the first lighting system 106 is aligned with the center of the transparent platform 103. In one embodiment, the first lighting system 106 is a planar backlight. The first lighting system 106 is located at a certain distance below the transparent platform 103.

[0069] The second illumination system 124 (e.g., a side-view imaging system) includes a white matte diffuser 126 illuminated by one or more light sources 125 that may be located above an additional diffuser 207, which ensures a uniform brightness distribution on the white matte diffuser 126. The size of the white matte diffuser 126 is designed to cover the background of the image captured by the side-view camera 102. For example, the size of the white matte diffuser 126 may be approximately equal to the field of view and / or angular field of view of the side-view camera 102 at its distance from the white matte diffuser 126. In an embodiment, the angle and distance at which the side-view camera 102 is positioned relative to the transparent plate 103 are such that the image captured by the side-view camera 102 approximately captures the complete feature pattern of the transparent plate 103. The size of the white matte diffuser 126 and the angle and distance at which it is positioned relative to the transparent plate 103 are such that the field of view of the side-view camera 102 is approximately equal to the size of the white matte diffuser 126. In some embodiments, the white matte diffuser 126 includes multiple (e.g., four) segments. Alternatively, each segment may have a separate diffuser plate. Each segment may be located on the side of the transparent platform opposite to the side-view camera of that segment facing the light diffuser plate 126. For example, the side-view camera may be on the left side of the transparent platform and above the plane of the transparent platform, while the associated light diffuser plate 126 may be on the right side of the transparent platform and below the plane of the transparent platform. In one embodiment, each segment or plate has a plane in which the imaging axis of the side-view camera facing the segment or light diffuser plate is perpendicular to the plane of the segment or light diffuser plate. In some embodiments, the light reflection characteristics of the matte diffuser plate 126 are as follows: reflectivity 0.85–0.95, gloss (60°) <5 GU (very matte), and bidirectional reflectance distribution function with respect to the uniformity of an ideal Lambertian body ±10%; in some embodiments, for the additional light diffuser plate 207, the light transmission characteristics are as follows: surface finish - polished / high gloss, total transmittance 70–95%, haze ≥90%, and gloss (60°) >70 GU.

[0070] The positions, dimensions, and shapes of the first illumination system 106 and the second illumination system 124 are fixed to provide a suitable brightness distribution on the transparent platform 103 where the transparent 3D object to be inspected will be placed. This distribution ensures that defects in the transparent 3D object are visible in the image, providing sufficient contrast for reliable detection.

[0071] The first illumination system 106 forms a background in the image captured by the top-view camera 101 and illuminates a transparent 3D object placed on the transparent platform 103. Similarly, each illumination component of the second illumination system 124 forms a background in the image captured by the corresponding side-view camera 102 in the side-view camera 102 and illuminates a transparent 3D object placed on the transparent platform 103. The image of the transparent 3D object is generated by light emitted from either the backlight (first illumination system 106) or the sidelight (second illumination system 124), refracted by the transparent 3D object, and captured by appropriate camera lenses 204 and 210. The IBQC system 100 can be configured such that the light refracted by the transparent 3D object makes the surface of the transparent 3D object visible in the captured image. The distance from the illumination systems 106 and 124 and the shape of the luminescent surfaces of the illumination systems 106 and 124 significantly affect the contrast of the transparent 3D object in the image. These factors are selected such that the background (e.g., the first illumination system 106 or the white matte diffuser 126) occupies the entire image area (e.g., field of view) of the respective top-view camera 101 or side-view camera 102, and the angular magnitude of the light ensures proper illumination of the transparent 3D object to achieve the required contrast for defect detection. In an embodiment, the first illumination system 106 includes a planar backlight with a uniform brightness distribution on its surface. Each diffuser may be positioned close to the light source and configured to diffuse light from that source.

[0072] Figure 2B This is a diagram of a side-view camera 102 according to an embodiment of the present disclosure. As shown, the side-view camera includes a lens 210 having a focal plane 250 and an image sensor 211 having an image plane 252. Due to a tilt shift adapter 260, the image plane 252 and the focal plane 250 are at an angle θ relative to each other.

[0073] Figure 3 A side lighting system 300 according to an embodiment of the present disclosure is shown, which can correspond to Figure 1AThe second lighting system 124 is divided into quadrants and includes a separate lighting segment in each quadrant, which is configured to be opposite each side-view camera 102 to provide background and backlighting for images captured by the associated side-view camera 102. In an embodiment, the side lighting system 300 includes a plurality of white matte panels 301 fixed to a black matte panel 302. In an embodiment, the side lighting system 300 may include a separate white matte panel 301 and a black matte panel 302 for each segment. The white matte panel 301 and / or the black matte panel 302 are fixed using brackets 303, 304. Light is emitted from a light source (e.g., an LED) 306 located above a light diffuser 305 (which is also fixed in the bracket 303). In an embodiment, the position of the light source 306 and the distance between the light source 306, the light diffuser 305, and the white panels 301 are selected to ensure uniform illumination of the white panels 301. In one embodiment, direct light from light source 306 is blocked by a black panel (not shown) located above the side lighting system 300. In some embodiments, one or more light curtains may be attached to brackets 303, 304 to block light from adjacent light sources 306 from illuminating adjacent white panels 301. In some embodiments, flexible control of the lighting can be achieved by activating only the light source 306 corresponding to a single side segment of the side lighting system 300.

[0074] Figure 4 This is a ray propagation diagram of an image-based quality control system according to an embodiment of the present disclosure. When the transparent 3D object under inspection is positioned on plane 404 (e.g., which may correspond to transparent platform 103), light rays within solid angle 401 may be deflected by the transparent 3D object. Those rays falling within the cone formed by certain edge rays 403 of the camera lens produce an image of the transparent 3D object on the image plane of the camera. Light rays located outside solid angle 401 and / or falling outside the cone formed by the edge rays 403 of the camera lens are not captured by the camera and do not form part of the captured image. In the example, if there is no surface of the transparent 3D object in the path of the light ray within solid angle 401, the light ray will appear as a white dot or spot on the captured image. However, if there is a surface of the transparent 3D object in the path of the light ray, the surface will deflect the light. If the light (e.g., a ray) is deflected out of the cone formed by a specific edge ray 403 of the camera lens, the light ray will not reach camera sensor 406, resulting in dark dots or spots on camera sensor 406. If only a portion of the light falls within a specific edge ray 403 of the camera lens, an intensity value (e.g., grayscale value) between white and black will be captured for the corresponding pixel.

[0075] In embodiments, light source 407 (e.g., corresponding to first illumination system 106 or second illumination system 124) can be considered a Lambertian light source with a uniform light distribution 402. The brightness of a particular pixel on camera sensor 406 (which is projected onto a transparent 3D object through lens 405) is proportional to the amount of light deflected by the transparent 3D object and falling within a solid angle 401 formed by the edge rays 403 of the lens corresponding to that pixel. The contrast of defects in the transparent 3D object visible in the image depends on the shape, size, and orientation of light source 407, as well as its brightness distribution.

[0076] Figure 5A This is an example top-view image captured by a top-view camera of an image-based quality control system according to an embodiment of the present disclosure. The example top-view image depicts a transparent 3D object (e.g., a polymer orthodontic appliance) 500 surrounded by a feature pattern 505. As shown, the feature pattern includes a plurality of features 510, which, in the illustrated example, are a plurality of points.

[0077] Figures 5B to 5E These are example top-view images captured by different side-view cameras of an image-based quality control system according to embodiments of the present disclosure. The example top-view images depict a transparent 3D object 500 surrounded by a feature pattern 505 from different perspectives and show different portions of the transparent 3D object. As shown, the feature pattern includes a plurality of features 510, which, in the illustrated example, are a plurality of points.

[0078] Figures 6 to 10 This is a flowchart illustrating various methods for performing automated defect detection on transparent 3D objects (e.g., polymer orthodontic appliances or other transparent dental instruments) according to embodiments of the present disclosure. Some operations of the methods may be performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, etc.), software (e.g., instructions running on a processing device to perform hardware simulations), or a combination thereof. The processing logic may be (e.g., Figure 1B The processing logic is executed on one or more processing devices (such as controller 104 and / or computing device 105). In embodiments, the processing logic may be the processing logic of an image inspection module and / or an image control module. Some operations of the method may be performed by an IBQC system (such as...). Figures 1A to 1B The IBQC system 100 is executed.

[0079] For the sake of simplicity, the method is depicted and described as a series of actions. However, the actions according to this disclosure may occur in various orders and / or simultaneously, and may occur together with other actions not presented and described herein. Furthermore, not all of the actions shown may be required to implement the method according to the disclosed subject matter. Additionally, those skilled in the art will understand and recognize that these methods may alternatively be represented by a series of interrelated states via state diagrams or events.

[0080] In some embodiments, a mold of a patient's dental arch can be manufactured based on a digital file, and dental appliances (such as orthodontic appliances) can be formed on that mold. The mold can be manufactured by a computing device (such as...) Figure 11 The processing logic is executed by the processing device (a computing device in the computer). The processing logic may include hardware (e.g., circuits, special-purpose logic, programmable logic, microcode, etc.), software (e.g., instructions executed by the processing device), firmware, or a combination thereof. For example, one or more operations may be performed by the processing device executing a computer-aided design (CAD) program or module.

[0081] To create molds for dental appliances, the shape of the patient's dental arch during the treatment phase is determined based on the treatment plan. In an orthodontic example, a treatment plan can be generated based on an intraoral scan of the dental arch to be modeled. An intraoral scan of the patient's dental arch can be performed to generate a three-dimensional (3D) virtual model of the patient's dental arch (mold). For example, a full scan of the patient's mandibular and / or maxillary dental arch can be performed to generate its 3D virtual model. Intraoral scanning can be performed by creating multiple overlapping intraoral images from different scanning stations and then stitching the intraoral images together to provide a composite 3D virtual model. In other applications, virtual 3D models can also be generated based on scans of the object to be modeled or based on the use of computer-aided drawing techniques (e.g., for designing virtual 3D molds). Alternatively, an initial negative mold can be generated from the actual object to be modeled (e.g., dental impressions, etc.). The negative mold can then be scanned to determine the shape of the positive mold to be produced.

[0082] Once a virtual 3D model of the patient's dental arch is generated, the dentist can determine the desired treatment outcome, including the final position and orientation of the patient's teeth. The processing logic can then determine multiple treatment phases to advance the teeth from their initial position and orientation to their target final position and orientation. By calculating the progression of tooth movement throughout the entire orthodontic treatment, from the initial tooth placement and orientation to the final corrected tooth placement and orientation, the shape of the final virtual 3D model and each intermediate virtual 3D model can be determined. For each treatment phase, a separate virtual 3D model of the patient's dental arch can be generated for that phase. The shape of each virtual 3D model will differ. The original virtual 3D model, the final virtual 3D model, and each intermediate virtual 3D model are unique and customized for the patient.

[0083] Therefore, multiple distinct virtual 3D models can be generated for a single patient. The first virtual 3D model can be a unique model of the patient's current dental arch and / or teeth, and the final virtual 3D model can be a model of the patient's dental arch and / or teeth after orthodontic treatment of one or more teeth and / or jaws. Multiple intermediate virtual 3D models can be modeled, each progressively different from the previous virtual 3D models.

[0084] Each virtual 3D model of a patient's dental arch can be used to generate a unique, customized physical mold of the arch for a specific treatment phase. The shape of the mold can be at least partially based on the shape of the virtual 3D model for that treatment phase. The virtual 3D model can be represented in a file such as a computer-aided drafting (CAD) file or a 3D printable file such as a stereolithography (STL) file. The virtual 3D model of the mold can be sent to a third party (e.g., a clinician's office, laboratory, manufacturing plant, or other entity). The virtual 3D model (e.g., it can be in the form of a digital file) can include instructions that control a manufacturing system or apparatus to produce a mold with a specified geometry.

[0085] A clinician's office, laboratory, manufacturing plant, or other entity can receive a virtual 3D model of the mold, which has been created as described above. The entity can then input the digital model into a 3D printer. The 3D printer then uses the digital model to manufacture the mold. 3D printing includes any layer-based additive manufacturing process. 3D printing can be achieved using additive processes, where continuous layers of material are formed in a prescribed shape. 3D printing can be performed using extrusion deposition, granular material bonding, lamination, photopolymerization, continuous liquid interface production (CLIP), or other techniques. Subtractive processes, such as milling, can also be used to achieve 3D printing.

[0086] In some cases, stereolithography (SLA), also known as optical fabrication solid-state imaging, is used to create SLA molds. In SLA, a mold is created by sequentially printing thin layers of a photocurable material (e.g., a polymeric resin) onto one layer of the other. A platform is placed in a bath of liquid photopolymer or resin, just below the surface of the bath. A light source (e.g., an ultraviolet laser) tracks a pattern on the platform, curing the photopolymer as the light source points to it, to form the first layer of the mold. The platform is lowered incrementally, and the light source tracks a new pattern on the platform to form another layer of the mold at each increment. This process is repeated until the mold is completely fabricated. Once all the layers of the mold have been formed, the mold can be cleaned and cured.

[0087] Materials such as polyesters, copolyesters, polycarbonates, polycarbonates, thermopolymerized polyurethanes, polypropylene, polyethylene, polypropylene and polyethylene copolymers, acrylic acid, cyclic block copolymers, polyetheretherketones, polyamides, polyethylene terephthalate, polybutylene terephthalate, polyetherimide, polyethersulfone, polypropylene terephthalate, styrene block copolymers (SBCs), silicone rubber, elastomer alloys, thermopolymerized elastomers (TPEs), thermopolymerized vulcanized rubber (TPV) elastomers, polyurethane elastomers, block copolymer elastomers, polyolefin blend elastomers, thermopolymerized copolyester elastomers, thermopolymerized polyamide elastomers, or combinations thereof, can be used to directly form molds. Materials used to manufacture molds can be provided in uncured form (e.g., as liquids, resins, powders, etc.) and can be cured (e.g., by photopolymerization, light curing, gas curing, laser curing, crosslinking, etc.). The properties of the material before curing may differ from the properties of the material after curing.

[0088] Orthodontic appliances can be formed from individual molds and, when applied to a patient's teeth, provide force to move the teeth as prescribed in the treatment plan. Each appliance is uniquely shaped and tailored to the specific patient and stage of treatment. In the example, the appliance can be pressure-formed or thermoformed on a mold. Each mold can be used to manufacture the appliance, which will apply force to the patient's teeth at a specific stage of orthodontic treatment. Each appliance has a tooth-receiving cavity that accommodates the teeth and flexibly repositions them according to the specific stage of treatment.

[0089] In one embodiment, a sheet of material is pressed or thermoformed onto a mold. This sheet can be, for example, a plastic sheet (e.g., an elastic thermoplastic, a polymer sheet, etc.). To thermoform the shell onto the mold, the sheet can be heated to a temperature at which it becomes pliable. Pressure can be applied simultaneously to the sheet to form the now-pliable sheet around the mold. Once cooled, the sheet will have a shape conforming to the mold. In one embodiment, a release agent (e.g., a non-stick material) is applied to the mold prior to shell formation. This facilitates subsequent removal of the shell from the mold.

[0090] Additional information can be added to the orthodontic appliance. This additional information can be any information related to the appliance. Examples of such additional information include part number identifiers, patient name, patient identifier, case number, sequence identifier (e.g., indicating which appliance in the treatment sequence a particular liner belongs to), manufacturing date, clinician name, logo, etc. For example, after the appliance is thermoformed, it can be laser-marked with a part number identifier (e.g., serial number, barcode, etc.). In some embodiments, the system can be configured to read (e.g., optically, magnetically, etc.) identifiers of the mold (barcode, serial number, electronic tag, etc.) to determine a part number identifier associated with the appliance formed on the mold. After determining the part number identifier, the system can then mark the appliance with a unique part number identifier. The component number identifier can be computer-readable and can be associated with a specific patient, a specific stage in a treatment sequence, whether the orthodontic appliance is an upper or lower shell, a digital model representing the mold from which the orthodontic appliance is manufactured, and / or a digital file including a virtually generated digital model of the orthodontic appliance or an approximate characteristic thereof (e.g., generated by approximating the outer surface of the orthodontic appliance based on a digital model of manipulating the mold, expanding or scaling the projection of the mold in different planes, etc.). In some embodiments, the virtually generated digital model of the orthodontic appliance or its approximate characteristics can be compared with characteristics of the manufactured orthodontic appliance (e.g., the shape of the orthodontic appliance) determined from an image of the manufactured orthodontic appliance for image-based quality control.

[0091] After the orthodontic appliance is formed on a mold for the treatment phase, it is then trimmed along the incision lines (also called trimming lines), and the appliance can be removed from the mold. Processing logic determines the incision lines of the appliance. The incision lines can be determined based on a virtual 3D model of the dental arch at the specific treatment phase, a virtual 3D model of the appliance to be formed on the dental arch, or a combination of virtual 3D models of the dental arch and the appliance. The location and shape of the incision lines are important for the appliance's function (e.g., its ability to apply desired forces to the patient's teeth) and its fit and comfort. For shells such as orthodontic appliances, retainers, and splints, shell trimming plays a role in the shell's effectiveness for its intended purpose (e.g., aligning, retaining, or positioning one or more of the patient's teeth) and its fit on the patient's dental arch. For example, if the shell is trimmed too much, it may lose rigidity, and its ability to apply forces to the patient's teeth may be impaired.

[0092] On the other hand, if the housing is under-trimmed, some parts of the housing may impact the patient's gums and cause discomfort, swelling, and / or other dental problems. Additionally, if the housing is under-trimmed in one location, the housing may be too rigid at that location. In some embodiments, the incision line can be a straight line passing through the aligner at, below, or above the gingival line. In some embodiments, the incision line can be a gingival incision line, which represents the interface between the aligner and the patient's gums. In such embodiments, the incision line controls the distance between the edge of the aligner and the patient's gingival line or gingival surface.

[0093] Each patient has a unique dental arch with unique gingiva. Therefore, the shape and position of the incision line can be unique and customized for each patient and each stage of treatment. For example, the incision line is customized to run along the gingival line (also known as the gum line). In some embodiments, the incision line may be off the gingival line in some areas and on the gingival line in others. For example, in some cases, it may be desirable for the incision line to be off the gingival line (e.g., not touching the gum), where, in the interproximal region between teeth, the shell will touch the teeth and be on the gingival line (e.g., touching the gum). Therefore, it is important to trim the shell along the predetermined incision line.

[0094] In some embodiments, a dental appliance may have multiple cutting lines. A first cutting line, or main cutting line, can control the distance between the edge of the housing and the patient's gingival line. Additional cutting lines can be used to cut grooves, holes, or other shapes in the housing. For example, additional cutting lines can be used to remove occlusal surfaces of the housing, additional surfaces of the housing, or portions of the housing that, after removal, allow for the formation of hooks that can be used with an elastomer.

[0095] Once the cutting lines are determined, the orthodontic appliance can be cut along the cutting lines (or multiple cutting lines) using markings and / or elements imprinted in the appliance. In some embodiments, the appliance can be manually cut by a technician using scissors, a drill, a cutting wheel, a scalpel, or any other cutting tool. In another embodiment, the appliance is cut along the cutting lines by a computer-controlled trimming machine (such as a CNC machine or a laser trimming machine). The computer-controlled trimming machine may include a camera capable of identifying the cutting lines in the appliance. The computer-controlled trimming machine can use images from the camera to determine the location of the cutting lines based on markings in the appliance, and can control the angle and position of the trimming machine's cutting tools to trim the appliance along the cutting lines using the identified markings.

[0096] Additionally or alternatively, the orthodontic appliance may include coordinate system reference markers that can be used to orient the coordinate system of the trimming machine to the predetermined coordinate system of the orthodontic appliance. The trimming machine may receive a digital file containing trimming instructions (e.g., indicating the position and angle of the trimming machine's laser or cutting tool to trim the orthodontic appliance along the cutting line). By aligning the coordinate system of the trimming machine with the orthodontic appliance, the accuracy of computer-controlled trimming of the orthodontic appliance at the cutting line can be improved. The coordinate system reference markers may include markers sufficient to identify the origin and the x, y, and z axes.

[0097] In some embodiments, transparent 3D objects can be manufactured directly from digital files using additive manufacturing techniques (also referred to herein as "3D printing"), rather than thermoforming them on a 3D printing mold. For example, orthodontic appliances can be directly 3D printed. To manufacture 3D objects, the shape of the object can be determined and designed using computer-aided engineering (CAE) or computer-aided design (CAD) programs. In some cases, stereolithography (SLA) can be used to manufacture 3D printed objects from digital files containing or representing a 3D model of the transparent 3D object.

[0098] In some embodiments, other additive manufacturing techniques may be used to produce transparent 3D objects. Other additive manufacturing techniques may include: (1) material spraying, wherein material is sprayed onto a build platform using a continuous or on-demand dripping (DOD) method; (2) binder spraying, wherein alternating layers of build material (e.g., powder-based material) and binder material (e.g., liquid binder) are deposited through a printhead; (3) fused deposition modeling (FDM), wherein material is extracted through a nozzle, heated, and deposited layer by layer; (4) powder bed delivery, including but not limited to direct metal laser sintering (DMLS), electron beam melting (EBM), selective thermal sintering (SHS), selective laser melting (SLM), and selective laser sintering (SLS); (5) sheet lamination, including but not limited to layered solid fabrication (LOM) and ultrasonic additive manufacturing (UAM); and (6) directional energy deposition, including but not limited to laser engineered mesh forming, directional light fabrication, direct metal deposition, and 3D laser cladding.

[0099] For both directly 3D printed transparent 3D objects (e.g., dental appliances) and thermoformed transparent 3D objects on 3D printing molds (e.g., dental appliances), a variety of different types of defects can occur, such as delamination defects from the 3D printing process (e.g., surface defects, internal defects, interface defects, etc.), cutting line defects, gingival line defects, deformation defects (e.g., warping of dental appliances, such as warping caused by removing the dental appliance from the mold with excessive force), and so on. Any one or more of these types of defects can be detected by the IBQC system described in the embodiments herein.

[0100] Figure 6 A flowchart of a method 600 for detecting manufacturing defects in a transparent 3D object according to one embodiment is shown. One or more operations of method 600 can be performed by the processing logic of a computing device. It should be noted that method 600 can be performed for multiple unique transparent 3D objects. In one embodiment, method 600 can be performed for each unique orthodontic appliance for each stage of a patient's orthodontic treatment plan. The appliance can be a 3D-printed appliance or can be formed by thermoforming a plastic sheet on a 3D-printed mold of the dental arch.

[0101] At frame 602, a first light source can be used to provide initial illumination of a transparent 3D object (e.g., an orthodontic appliance or other transparent dental instrument). This initial illumination can be provided by an imaging system (e.g., by...). Figures 1A to 1B The imaging system 106 is provided based on instructions from the processing logic. In one embodiment, the first illumination may be provided by a light-emitting element from a top-view illumination system positioned below a transparent platform on which the transparent 3D object is placed during imaging.

[0102] At block 604, a top-view imaging device (e.g., a top-view camera) can be used to generate one or more images of a transparent 3D object based on instructions from processing logic. In one embodiment, the one or more images include a first image and a second image. The first image may include a representation of IDs (e.g., a sequence of symbols) for different regions or viewpoints of the transparent 3D object. In one embodiment, the first image is generated under specific lighting conditions that enhance the sharpness and / or contrast of laser markings (e.g., where the ID is represented as a laser mark). In some embodiments, at block 606, processing logic processes the first image to determine the ID of the transparent 3D object. This may include performing optical content recognition (OCR) on the first image to determine the ID using a sequence of symbols. The ID may be associated with the transparent 3D object depicted in the first image. The processing logic may determine a digital file associated with the ID. The digital file may include one or more characteristics associated with at least one surface of the transparent 3D object.

[0103] In some embodiments, a first image and / or a second image captured by a top-view camera are processed to identify one or more defects. In some embodiments, a second image is captured using different camera settings of the top-view camera and / or different illumination settings of the first illumination system. The first image and / or the second image may be processed to determine whether the transparent 3D object has any defects. For example, the first image and / or the second image may be processed to identify obvious defects (e.g., deformation) in the transparent 3D object.

[0104] At box 610, the processing logic uses multiple second light sources to provide a second illumination of the transparent 3D object. In an embodiment, the multiple second light sources may be a second lighting system (such as a side-view lighting system, e.g., Figures 1A to 1B The second lighting system 124) is a component of this system. At box 612, multiple side-view cameras capture one or more images of the transparent 3D object, thereby generating multiple side-view images of the transparent 3D object that depict multiple different regions of the transparent 3D object. In some embodiments, the multiple cameras capture side-view images of the transparent 3D object simultaneously. Alternatively, the multiple cameras may capture side-view images sequentially.

[0105] Each of the multiple side-view images can depict a different region or perspective of a transparent 3D object (e.g., different sides of a dental instrument) and can be captured by different side-view cameras. Both the transparent platform supporting the transparent 3D object and the side-view cameras can have fixed positions during the image capture process.

[0106] At box 614, processing logic processes multiple side-view images to identify any defects in the transparent 3D object. In some embodiments, the images are processed by one or more AI models. In some embodiments, rule-based logic is used to process the images. The AI ​​model can be trained to receive an image of the transparent 3D object as input and provide an indication of one or more defect types, the location of the defects, and / or the probability of the defects as output. In one embodiment, the AI ​​model outputs a classification result indicating whether a defect is detected. In one embodiment, the AI ​​model can identify multiple different defect types and, for each defect type, indicate whether a defect of that type is identified in the image. For example, the output may include a vector with multiple elements, where each element may include a value representing the probability that the image contains a defect of a specific defect type. In some embodiments, the output is the probability that the transparent 3D object has a defect. In another example, the output may indicate a probability of 90% for internal volume defects, 15% for surface defects, 2% for interface defects, 5% for line thickness defects, and 8% for delamination defects. Defect probabilities above a threshold (e.g., 80%, 90%, etc.) can be classified as defects. Defect probabilities below a threshold can be classified as no defects. In some embodiments, the AI ​​model outputs bounding boxes around identified defects. In some embodiments, the AI ​​model performs semantic segmentation or instance segmentation and outputs segmentation information identifying defects.

[0107] In one embodiment, in addition to identifying the presence of a defect, the AI ​​model also outputs coordinates associated with the identified defect. These coordinates can be the x and y pixel positions in the input image. In some embodiments, processing logic can then use the defect coordinates to label the image. Furthermore, in an embodiment, the processing logic can indicate the type of defect identified at the identified coordinates. This allows users to quickly review and verify positive defect classification results.

[0108] In one embodiment, the AI ​​model may output a confidence metric for each defect probability it outputs. The confidence metric indicates the level of confidence associated with the output defect probability. A low confidence metric may indicate that the detail and / or contrast in the image are insufficient to accurately determine the presence of a defect. In some embodiments, the confidence metric output by the AI ​​model is compared to a confidence threshold.

[0109] In some embodiments, the processing logic determines the characteristics of a transparent 3D object from the captured image and compares the determined characteristics with the planned characteristics of the transparent 3D object specified in the treatment plan or with additional data associated with the transparent 3D object (e.g., as indicated in the 3D model of the transparent 3D object from the treatment plan). If the difference between the determined characteristics and the planned characteristics exceeds a threshold amount, one or more defect types (e.g., cut line defects and / or deformation defects) can be identified.

[0110] At box 616, the processing logic can output defect detection results. These results can be output to a display of the IBQC system, stored in a data storage device, and / or transmitted to a remote computing device. The remote computing device can receive the defect detection results and can store and / or output the results to a display. In one embodiment, the AI ​​model and / or other defect detection algorithm can output a defect level for each defect or group of defects identified in one or more input images. The defect level can rate the defect based on its severity and / or the likelihood that it will cause problems in the future. The defect level can be based on the density or number of defects identified in the image, as well as the size or dimensions of the defects and / or the type of defect.

[0111] If a transparent 3D object has defects, the severity of the defects can be compared to a defect threshold. If the combined severity of the detected defects exceeds the severity threshold, the transparent 3D object is marked as failing quality control and identified as defective. In one embodiment, the defective transparent 3D object can be repaired to remove the manufacturing defects. In another embodiment, the transparent 3D object can be prevented from further use in the manufacturing process or delivered to the user. In yet another embodiment, the transparent 3D object can be scrapped and a replacement manufactured.

[0112] If a transparent 3D object has no defects, or if the combined severity of detected defects is below a severity threshold, the transparent 3D object passes quality control and is identified as acceptable.

[0113] Figure 7 A flowchart of a method 700 for determining an ID associated with a transparent 3D object according to one embodiment is shown. At block 702, a first illumination system can be used to provide initial illumination to the transparent 3D object. At block 704, a top-view image of the transparent 3D printed object can be generated using a top-view camera. At block 706, OCR is performed on the image. Processing logic can process the image to identify the location of a sequence of symbols in the image. The sequence of symbols can contain letters, numbers, special symbols, punctuation marks, etc. The processing logic can then perform OCR on the sequence of symbols. OCR can be performed using any known method, such as matrix matching, feature extraction, or a combination thereof. In one embodiment, the processing logic can apply one or more transformation operations to the image to more clearly process the sequence of symbols. Transformation operations can include changing the resolution of the image, binarizing the image using specific parameters, correcting distortion, glare, or blur, performing noise reduction operations, etc. These operations can be applied to the entire image or a portion of the image containing the identified sequence of symbols.

[0114] In one embodiment, the image generated by the imaging system may not fully display the symbol sequence. For example, the image may depict a specific region or viewpoint of a transparent 3D object containing half of the symbol sequence. In one embodiment, the processing logic may generate at least one additional image of the location of the symbol sequence. If the newly generated image includes the complete symbol sequence, the processing logic can process the image according to the procedure described above. If the newly generated image depicts the other half of the symbol sequence, the processing logic can perform a stitching operation to generate a single image containing that symbol sequence. The processing logic can then perform OCR according to the procedure described above.

[0115] At box 708, an ID associated with a transparent 3D object can be determined based on the OCR result. After performing OCR according to box 706, the processing logic can produce a first result containing a computer-readable version of the symbol sequence identified in the image. The processing logic can compare the first result with one or more transparent 3D object IDs (e.g., dental appliance IDs) to determine whether the first result corresponds to a known transparent 3D object ID. If it is determined that the first result does not correspond to a known transparent 3D object ID, the first result is rejected. In one embodiment, a second OCR operation can be performed on the image to generate a second result. In another embodiment, an OCR operation can be performed on an image different from the image that generated the first result to generate a second result. If it is determined that the first result or the second result corresponds to a known transparent 3D object identifier, the processing logic determines that the first result or the second result is an ID associated with that transparent 3D object.

[0116] In another embodiment, a technician can manually enter the ID associated with a transparent 3D object at the IBQC system using an interface. In yet another embodiment, the sorting system can sort a series of transparent 3D objects in a known order. The processing logic can obtain the order of the transparent 3D objects from the sorting system to understand which transparent 3D objects are currently being processed and in what order they arrived at the imaging system.

[0117] Figure 8 A flowchart of a method 800 for detecting obvious defects in a transparent 3D object according to one embodiment is shown. In one embodiment, obvious defects on a dental arch mold used to form a dental appliance (e.g., an orthodontic appliance) or on a directly 3D-printed dental appliance may include arch misalignment, deformation, bending (compression or stretching), cutline deviation, adhesion, improper attachment trimming, missing attachments, burrs, eversion, dynamic ridge problems, material breakage, hooks that are too short, etc. At box 802, an ID associated with the transparent 3D object is determined. A method for detecting obvious defects in a transparent 3D object can be used... Figure 7 The method described in the document describes any ID determination method to determine the ID.

[0118] At box 804, a digital file associated with the transparent 3D object can be identified. This digital file can be identified from a group of digital files. The digital file can be associated with the 3D object based on an ID. Each digital file in the group can include a digital model of the 3D object (e.g., a virtual 3D model). In one embodiment, each digital file can include a digital model of a mold used to manufacture orthodontic appliances or other dental instruments. Each digital file can be for a unique, customized 3D object. In one embodiment, each digital file can be for a specific mold customized for a particular patient at a specific stage of their treatment plan.

[0119] In one embodiment, the transparent 3D object may be a directly manufactured (e.g., 3D printed) dental appliance. In one embodiment, the digital file associated with the ID may include a digital model of the first dental appliance. In some embodiments, the digital file is generated by processing logic or received from other sources. The digital model of the first dental appliance can be dynamically generated by manipulating a digital model of the dental arch representing the state of the patient's dentition during treatment. The digital model of the first dental appliance can be generated by enlarging the digital model of the dental arch into an enlarged digital model (e.g., by scaling or expanding the surface of the digital model). Furthermore, generating the digital model of the first dental appliance may include projecting cleavage lines onto the enlarged digital model, virtually cutting the enlarged digital model along the cleavage lines to create a cut enlarged digital model, and selecting the outer surface of the cut enlarged digital model. In one embodiment, the digital model of the first dental appliance includes the outer surface of the first dental appliance but does not necessarily have thickness and / or exclude the inner surface of the first orthodontic appliance, although in other embodiments it may include thickness or an inner surface.

[0120] In one embodiment, the digital file may include a virtual 3D model of a mold used to manufacture the first dental appliance. In one embodiment, the digital file may include multiple files associated with the first dental appliance, wherein the multiple files include a first digital file having a digital model of the mold and a second digital file having a digital model of the first dental appliance. Alternatively, a single digital file may include both a digital model of the mold and a digital model of the first dental appliance.

[0121] At block 806, the processing logic may determine the geometry or shape associated with at least one surface of a transparent 3D object based on a digital file. In one embodiment, the processing logic determines a first silhouette of the first 3D object from a first digital file. In one embodiment, the first silhouette is included in the digital file of the first 3D object. In one embodiment, the first silhouette is based on the projection of a digital model of the first 3D object onto a plane defined by an image of the first 3D object (e.g., an image generated by a top-down or side-view camera). In one embodiment, the first silhouette is based on manipulation of the digital model of the first 3D object. For example, in some embodiments, the first silhouette may be based on the projection of the digital model of the 3D object onto a plane defined by an image of the 3D object. In this case, the projection of the 3D object may be scaled or otherwise adjusted to approximate the projection of the 3D object from a particular viewpoint (e.g., the viewpoint of a top-down or side-view camera). In another embodiment, the first silhouette may be based on manipulation of the digital model, wherein the manipulation causes the outer surface of the digital model to have an approximate shape of the 3D object, and is also based on the projection of the outer surface of the digital model onto a surface defined by an image of the 3D object. In some embodiments, the first silhouette may be determined from the approximate outer surface of the 3D object. In some embodiments, the first contour may include a first shape of the projection of the outer surface of the first 3D object onto a plane defined by an image of the 3D object.

[0122] At box 808, the processing logic can perform defect detection using the determined geometry of the transparent 3D object and the captured image. In one embodiment, the processing logic determines a second contour of the transparent 3D object from at least one image of the transparent 3D object. The image of the transparent 3D object may define a plane. The second contour may include the outline of a second shape of the transparent 3D object projected onto the plane defined by the image. The second contour may be determined directly from one or more images (e.g., top view, side view, etc.). In one embodiment, the contour of the second shape is drawn from the image to form the second contour (e.g., the contour is identified based on edge detection performed on the image).

[0123] The processing logic can compare a first contour with a second contour. Based on this comparison, the processing logic can identify one or more differences between the first and second contours. In some embodiments, the processing logic can identify one or more differences by identifying one or more regions where the first shape of the first contour does not match the second shape of the second contour. The processing logic can also determine differences in regions (e.g., at least one of the thickness or area of ​​one or more regions).

[0124] In some embodiments, the processing logic inputs an image of a transparent 3D object and an associated projection of a digital 3D model of that transparent 3D object onto the image plane of the image into a trained AI model. The trained AI model can then output any defects perceived based on the difference between the transparent 3D object in the image and its projection onto the image plane.

[0125] In one embodiment, the processing logic may generate a difference index between the first contour and the second contour based on the comparison results and / or the output of the AI ​​model. The difference index may include a numerical representation of the difference between the expected geometry or shape of the transparent 3D object and the actual geometry or shape of the transparent 3D object (e.g., between the first contour and the second contour).

[0126] The processing logic can determine whether the difference index exceeds a difference threshold. The difference threshold can be any suitable configurable amount (e.g., a difference greater than 3 mm, 5 mm, 10 mm, or an area greater than 100 square millimeters, etc.). If the difference index exceeds the difference threshold, the processing logic can classify the 3D object as having a significant defect. In one embodiment, a 3D object classified as having a significant defect can be further classified as deformed. If it is determined that the difference index does not exceed the difference threshold, the processing logic can determine that the shape of the 3D object does not have a significant defect. In one embodiment, a transparent 3D object with a significant defect can be repaired to remove the severe defect. In another embodiment, a transparent 3D object with a significant defect can be scrapped, and a replacement transparent 3D object can be manufactured before using or delivering the transparent 3D object.

[0127] Figure 9 A flowchart of a method 900 for processing images captured by an IBQC system to detect defects is shown according to one embodiment. In one embodiment, method 900 may be performed for each unique dental appliance associated with a treatment plan (e.g., an orthodontic appliance for an orthodontic treatment plan).

[0128] At box 902, the processing logic acquires an image of the 3D object. For example, the image may be one of multiple images generated by the IBQC system 100. In one embodiment, the image may be generated by a top-view camera or a side-view camera. Each of the multiple images may depict a different region or viewpoint of the transparent 3D object.

[0129] At box 904, edge detection (or other differential processing) is performed on the image to determine the boundaries of transparent 3D objects in the image. Edge detection may include applying automatic image processing functions, such as edge detection algorithms. One example edge detection operation or algorithm that can be used is multi-scale combined grouping. Other examples of edge detection algorithms that can be used include the Canny edge detector, the Deriche edge detector, first- and second-order differential edge detectors (e.g., second-order Gaussian derivative kernels), the Sobel operator, the Prewitt operator, the Roberts crossover operator, and so on. Instead of edge detection or in addition to edge detection, segmentation operations (e.g., tooth segmentation operations) may also be performed on the image. In one embodiment, a segmentation operation may be applied to segment the transparent 3D object into individual objects, thereby highlighting different regions or viewpoints of the transparent 3D object depicted in the image. In one embodiment, a combination of multiple edge detection algorithms and / or segmentation algorithms is used for edge detection.

[0130] At box 906, a set of points on the boundary can be selected. In one embodiment, the portion of the image within the boundary may include a higher contrast than another portion of the image within the boundary. In this embodiment, a set of points on the boundary can be selected toward the portion of the image with higher contrast.

[0131] At box 908, the region of interest is defined using a set of points and / or boundaries. In one embodiment, the processing logic may generate one or more shapes corresponding to the set of points and / or boundaries.

[0132] In one embodiment, the processing logic may further define a region of interest. The processing logic may identify portions of the image within a geometric shape as regions of interest. In one embodiment, the processing logic may determine that the region of interest includes at least a minimum height and / or width.

[0133] At box 916, the image can be cropped to exclude areas of the image outside the region of interest.

[0134] At box 918, an AI model (e.g., an artificial neural network) and / or other algorithms can be used to process the cropped image (or uncropped image) to identify manufacturing defects on transparent 3D objects. For example, one or more image processing operations can be performed on the cropped image (or uncropped image), and the results of the image processing operations can be compared with a defined set of rules or other image data.

[0135] In one embodiment, a machine learning model (or a defined set of rules) can determine whether a specific type of layering defect exists (e.g., by identifying multiple lines present in the region of interest), whether a deformation defect exists in the region of interest, whether a cutting line defect exists in the region of interest, and / or whether other defect types exist. In the example, multiple lines may originate from a manufacturing process (e.g., SLA) used to create transparent 3D objects. The defined set of rules may include an acceptable threshold for the number of lines that should exist in regions of similar size to the region of interest. Processing logic can determine whether the number of lines in the region of interest is within the acceptable threshold. If the processing logic determines that the number of lines is not within the acceptable threshold, the machine learning model can indicate that the region of interest contains a defect. Additionally, the processing logic can determine the severity of the defect and the likelihood that the defect will cause significant deformation. If the processing logic determines that the defect is within the acceptable threshold, the processing logic can indicate that the region of interest does not contain a specific type of defect.

[0136] The AI ​​model can be trained on a training dataset to identify each of the defect types discussed above, which includes annotated images of transparent 3D objects without defects and annotated images of transparent 3D objects including defects of these types. Furthermore, the AI ​​model can be trained to identify other types of defects (e.g., layering defects). For example, the AI ​​model can determine whether debris, air bubbles (voids), or cavities (pits) are present in the region of interest. If debris or cavities are present, the AI ​​model can indicate the region of interest as containing layering defects (e.g., surface defects or interface defects), and can optionally indicate the type of layering defect and / or the coordinates of defects detected on the image (e.g., layering defects, cutting line defects, deformation, etc.). Additionally, the AI ​​model can determine the severity of the defect and the likelihood that the defect will cause problems (e.g., hindering the comfortable fit of dental appliances to the patient).

[0137] The AI ​​model can generate output that will be processed by processing logic. In one embodiment, the output of the AI ​​model may include the probability that the image contains a defect. In one embodiment, for each type of defect the AI ​​model has been trained to detect, the output may include the probability that the image contains a defect of that type. In another embodiment, the output of the AI ​​model may include a defect level, which indicates the severity of the defect identified in the image. In yet another embodiment, the output of the AI ​​model may include an identifier within the image that marks the location of the defect. The output may also include a highlighting of the location of the defect in the image.

[0138] AI models can consist of a single layer of linear or nonlinear operations (e.g., a support vector machine (SVM) or a single-layer neural network), or they can be deep neural networks consisting of multiple layers of nonlinear operations. Examples of deep networks and neural networks include recurrent neural networks and / or convolutional neural networks with one or more hidden layers. Some neural networks consist of interconnected nodes, where each node receives input from the previous node, performs one or more operations, and sends the resulting output to one or more other connected nodes for further processing.

[0139] Convolutional neural networks (CNNs) comprise architectures that can provide efficient image recognition. A CNN can include several convolutional layers and subsampling layers that apply filters to different parts of an image of text to detect certain features (e.g., defects). In other words, a CNN includes convolution operations that multiply each image segment element-wise with a filter (e.g., a matrix) and sum the results at the corresponding positions in the output image.

[0140] Recurrent neural networks (RNNs) can propagate data forward and backward from later processing stages to earlier ones. RNNs include the ability to process sequences of information and store information about previous computations in the context of hidden layers. Therefore, RNNs may possess "memory."

[0141] Artificial neural networks generally include feature representation components with classifier or regression layers that map features to a desired output space. For example, convolutional neural networks (CNNs) have multiple layers of convolutional filters. Pooling is performed, and non-linearity can be addressed at lower layers. Multiple perceptrons are typically attached on top of these lower layers, mapping the top-level features extracted by the convolutional layers to a decision (e.g., a classification output). Deep learning is a class of machine learning algorithms that use a cascade of multiple layers of non-linear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. Deep neural networks can learn in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. Deep neural networks consist of a hierarchical structure of layers, where different layers learn representations corresponding to different levels of abstraction. In deep learning, each level learns to transform its input data into a slightly more abstract and complex representation. For example, in image recognition applications, the raw input can be a pixel matrix; the first representation layer can abstract the pixels and encode the edges; the second layer can consist of and encode the arrangement of edges; the third layer can encode higher-level shapes (e.g., teeth, lips, gums, etc.); and the fourth layer can identify whether an image contains a face or defines bounding boxes around teeth in an image. It's worth noting that deep learning processes can learn on their own which features are best placed at which level. The "depth" in "deep learning" refers to the number of layers through which it transforms data. More precisely, deep learning systems have a considerable credit allocation path (CAP) depth. A CAP is a chain of transformations from input to output. CAP describes the underlying causal relationship between the input and output. For feedforward neural networks, the CAP depth can be the network depth and can be the number of hidden layers plus one. For recurrent neural networks where signals can propagate through layers more than once, the CAP depth can be infinite.

[0142] A machine learning model for identifying defects from images of transparent 3D objects can be trained using a training dataset. Training the neural network can be done in a supervised learning manner, involving feeding the network a training dataset consisting of labeled inputs, observing its output, defining the error (by measuring the difference between the output and the labeled value), and using techniques such as deep gradient descent and backpropagation to adjust the weights of the network across all its layers and nodes to minimize the error. In many applications, this process is repeated among many labeled inputs in the training dataset to produce a network that can produce correct outputs when presented with inputs different from those present in the training dataset. This generalization is achieved in high-dimensional settings (such as large images) when sufficiently large and diverse training datasets are available. The training dataset can include many images of transparent 3D objects. Each image can include a label or target for that image. The label or target can indicate whether the image contains a defect, the type of defect, the location of one or more defects, the severity of the defect, and / or other information.

[0143] In one embodiment, the training of the machine learning model is ongoing. Therefore, as new images are generated, the machine learning model can be applied to identify defects in these images. In some cases, a part may be based on the output of the machine learning model, but that part may ultimately fail due to undetected defects. This information can be added to processed images, and these images can be fed back into the machine learning model during updated learning processes to further teach the model and reduce future false negatives.

[0144] At block 920, processing logic determines whether the transparent 3D object includes defects, and / or whether the transparent 3D object includes one or more defects that would adversely affect the intended use of the transparent 3D object. The processing logic may evaluate the AI ​​model output for the image and all other images in a plurality of images according to the methods described above. In one embodiment, the output generated by the AI ​​model may include a defect level for each defect identified in each of the plurality of images. The processing logic may compare the AI ​​model output (e.g., defect level) with a defect threshold. In one embodiment, the processing logic may compare the output for each image with a defect threshold. In another embodiment, the processing logic may generate an overall combined defect level for the plurality of images and compare the overall combined defect level with a defect threshold. If the output is higher than the defect threshold, the processing logic may determine that a defect has been identified in the image associated with the transparent 3D object, and method 900 may proceed to block 922. If the output is lower than the defect threshold, the processing logic may determine that no manufacturing defect has been identified in the transparent 3D object, and method 900 may terminate.

[0145] At box 922, if the processing logic determines that the transparent 3D object contains a defect, the transparent 3D object can be discarded. In one embodiment, the transparent 3D object can be repaired to remove the defect. In another embodiment, the transparent 3D object can be scrapped and a replacement transparent 3D object can be manufactured.

[0146] Figure 10 A graphical representation of a machine in an example form of a computing device 1000 is shown, wherein actions can be performed to enable the machine to perform reference... Figures 6 to 9 The method discussed refers to a set of instructions for any one or more methods. In alternative embodiments, the machine may be connected (e.g., networked) to other machines on a local area network (LAN), intranet, extranet, or the Internet. For example, the machine may be networked to a rapid prototyping device, such as a 3D printer or SLA device. In another example, the machine may be networked to an IBQC system, or directly connected to an IBQC system, or as a component of an IBQC system. In one embodiment, computing device 1000 corresponds to... Figures 1A to 1B The computing device 105 is described above. The machine can operate as a server or client machine in a client-server network environment, or as a peer-to-peer (or distributed) network environment. The machine can be a personal computer (PC), tablet computer, set-top box (STB), personal digital assistant (PDA), cellular phone, network device, server, network router, switch, or bridge, or any machine capable of executing a set of instructions (sequentially or otherwise) specifying the actions to be taken by the machine. Furthermore, although only a single machine is shown, the term "machine" should also be considered as any collection of machines (e.g., computers) that individually or jointly execute a set (or more) of instructions to perform any one or more methods discussed herein.

[0147] Example computing device 1000 includes processing device 1002, main memory 1004 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM), etc.), static memory 1006 (e.g., flash memory, static random access memory (SRAM), etc.), and auxiliary memory (e.g., data storage device 1028) that communicate with each other via bus 1008.

[0148] Processing device 1002 represents one or more general-purpose processors, such as microprocessors, central processing units, etc. More specifically, processing device 1002 may be a Complex Instruction Set Computing (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing combinations of instruction sets. Processing device 1002 may also be one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, etc. Processing device 1002 is configured to execute processing logic (instructions 1026) for performing the operations and steps discussed herein.

[0149] The computing device 1000 may also include a network interface device 1022 for communicating with the network 1064. The computing device 1000 may also include a video display unit 1010 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1012 (e.g., a keyboard), a cursor control device 1014 (e.g., a mouse), and a signal generation device 1020 (e.g., a speaker).

[0150] Data storage device 1028 may include machine-readable storage medium (or more specifically, non-transitory computer-readable storage medium) 1024 thereon storing one or more sets of instructions 1026 embodying any one or more of the methods or functions described herein. Non-transitory storage medium refers to storage medium other than a carrier wave. Instructions 1026 may also reside wholly or at least partially in main memory 1004 and / or processing device 1002 during execution by computer device 1000, which also constitute computer-readable storage media.

[0151] Computer-readable storage medium 1024 can also be used for image inspection module 145 and / or image control module 140 as described above herein, image inspection module 145 and / or image control module 140 can perform reference Figures 6 to 9One or more operations of the described methods. The computer-readable storage medium 1024 may also store a software library containing methods that invoke the image inspection module 1045 and / or the image control module 1040, which can perform any of the aforementioned operations. Although the computer-readable storage medium 1024 is shown as a single medium in the example embodiment, the term "computer-readable storage medium" should be considered to include a single medium or multiple media (e.g., a centralized or distributed database and / or associated caches and servers) that store one or more sets of instructions. The term "computer-readable storage medium" should also be considered to include any medium capable of storing or encoding a set of instructions for machine execution and causing the machine to perform any one or more methods of this disclosure. Therefore, the term "computer-readable storage medium" should be considered to include, but is not limited to, solid-state memory, optical and magnetic media, and other non-transitory computer-readable media.

[0152] As discussed above, in some embodiments, Figures 1A to 1B The IBQC system 100 can be used to perform automatic defect detection on dental arch molds used for manufacturing orthodontic appliances, and / or to perform automatic defect detection on directly printed orthodontic appliances.

[0153] Figure 11An exemplary transparent tooth repositioning appliance or orthodontic appliance 1200 is shown, which can be worn by a patient to achieve incremental repositioning of individual teeth 1202 in the jaw. The appliance may include a housing (e.g., a continuous polymer housing or a segmented housing) having tooth-receiving cavities that accommodate and resiliently reposition the teeth. The appliance (also referred to as an apparatus) or portions thereof may be fabricated indirectly using a physical model of a tooth. For example, an apparatus (e.g., a polymer apparatus) may be formed using a physical model of a tooth and sheets of suitable layers of polymeric material. As used herein, “polymeric material” may include any material formed of a polymer. As used herein, “polymer” may refer to a molecule composed of repeating structural units linked by covalent chemical bonds, typically characterized by a substantial number of repeating units (e.g., equal to or greater than 3 repeating units, optionally, equal to or greater than 10 repeating units in some embodiments, and greater than or equal to 30 repeating units in some embodiments) and a high molecular weight (e.g., greater than or equal to 10,000 Da, greater than or equal to 50,000 Da or greater than or equal to 100,000 Da in some embodiments). Polymers are typically polymers of one or more monomer precursors. The term polymer includes homopolymers, or polymers consisting essentially of a single repeating monomer subunit. The term polymer also includes copolymers formed when two or more different types of monomers are linked in the same polymer. Useful polymers include organic or inorganic polymers that can be in an amorphous, semi-amorphous, crystalline, or semi-crystalline state. Polymers can include polyolefins, polyesters, polyacrylates, polymethacrylates, polystyrene, polypropylene, polyethylene, polyethylene terephthalate, polylactic acid, polyurethane, epoxy polymers, polyethers, poly(vinyl chloride), polysiloxanes, polycarbonates, polyamides, polyacrylonitrile, polybutadiene, poly(cycloolefins), and copolymers. The systems and / or methods provided herein are compatible with a range of plastics and / or polymers. Therefore, this list is not exhaustive but exemplary. Plastics can be thermosetting or thermoplastic. The plastic may be thermoplastic.

[0154] Examples of materials applicable to the embodiments disclosed herein include, but are not limited to, those described in the following provisional patent applications filed by Align Technology: U.S. Provisional Application No. 62 / 189,259, filed July 7, 2015, entitled “MULTI-MATERIAL ALIGNERS”; U.S. Provisional Application No. 62 / 189,263, filed July 7, 2015, entitled “DIRECT FABRICATION OF ALIGNERS WITH INTERPROXIMAL FORCE COUPLING”; and U.S. Provisional Application No. 62 / 189,291, filed July 7, 2015, entitled “DIRECT FABRICATION OF ORTHODONTIC APPLIANCES WITH VARIABLE”. PROPERTIES (Direct fabrication of orthodontic appliances with variable properties); U.S. Provisional Application No. 62 / 189,271, filed July 7, 2015, entitled "DIRECT FABRICATION OF ALIGNERS FOR ARCH EXPANSION"; U.S. Provisional Application No. 62 / 189,282, filed July 7, 2015, entitled "DIRECT FABRICATION OF ATTACHMENT TEMPLATESWITH ADHESIVE"; U.S. Provisional Application No. 62 / 189,301, filed July 7, 2015, entitled "DIRECT FABRICATION CROSS-LINKING FOR PALATE EXPANSION ANDOTHER". Applications (Direct fabrication crosslinking for palatal expansion and other applications); U.S. Provisional Application No. 62 / 189,312 filed on July 7, 2015, entitled "Systems, Apparatuses and Methods for Dental Appliances with Integrity Forged Features"; U.S. Provisional Application No. 62 / 189,317 filed on July 7, 2015, entitled "Direct Fabrication of Power Arms";The following U.S. Provisional Applications were filed on July 7, 2015: Serial No. 62 / 189,303, entitled "Systems, Apparatuses and Methods for Drug Delivery from Dental Appliances with Integral Formed Reservoirs"; Serial No. 62 / 189,318, entitled "Dental Appliance Having Ornamental Design"; Serial No. 62 / 189,380, entitled "Dental Material Susing Thermoset Polymers"; and Serial No. 62 / 667,354, entitled "Curable". The following are included: "Composiation for use in a high-performance lithography-based photopolymerization process and method of producing crosslinked polymers therefrom"; U.S. Provisional Application Serial No. 62 / 667,364, filed May 4, 2018, entitled "Polymerizable monomers and method of polymerizing the same"; and any conversion applications thereof (including publications and published patents), including any divisional, continuation, or partial continuation thereof.

[0155] The appliance 1200 can be adapted to all or fewer teeth present in the maxilla or mandible. The appliance can be specifically designed to accommodate the patient's teeth (e.g., the morphology of the tooth-accommodating cavity matches the morphology of the patient's teeth) and can be manufactured based on a positive or negative model of the patient's teeth generated by molding, scanning, etc. Alternatively, the appliance can be a general-purpose appliance configured to accommodate teeth, but not necessarily shaped to match the morphology of the patient's teeth. In some cases, only certain teeth accommodated by the appliance will be repositioned by the appliance, while other teeth can provide a base or anchor area to hold the appliance in place when the appliance applies force to one or more teeth targeted for repositioning. In some cases, at some point during treatment, some, most, or even all of the teeth will be repositioned. The moved teeth can also serve as a base or anchor point to hold the appliance in place above the teeth. However, in some cases, it may be desirable or necessary to provide corresponding receptacles or holes 1206 in the appliance 1200 for various attachments or other anchoring elements 1204 on the tooth 1202, allowing the appliance to apply selected forces on the tooth. Example appliances (including those used in the Invisalign® system) have been described in numerous patents and patent applications of Allianz Technologies, Inc. (including, for example, U.S. Patent Nos. 6,450,807 and 5,975,893) and on the company’s website (which is accessible on the Internet, e.g., see the URL “invisalign.com”). Examples of attachments suitable for use with orthodontic appliances are also described in patents and patent applications assigned by Allianz Technologies, Inc., including, for example, U.S. Patent Nos. 6,309,215 and 6,830,450.

[0156] Figure 12A tooth repositioning system 1210 comprising multiple appliances 1212, 1214, and 1216 is illustrated. Any of the appliances described herein can be designed and / or provided as part of a set of multiple appliances for use in a tooth repositioning system. Each appliance can be configured such that the tooth receiving cavity has a geometry corresponding to an intermediate or final tooth arrangement intended for use with the appliance. By placing a series of incremental position-adjusting appliances on a patient's teeth, it is possible to progressively reposition the patient's teeth from an initial tooth arrangement to a target tooth arrangement. For example, tooth repositioning system 1210 may include: a first appliance 1212 corresponding to the initial tooth arrangement; one or more intermediate appliances 1214 corresponding to one or more intermediate arrangements; and a final appliance 1216 corresponding to the target arrangement. The target tooth arrangement may be the planned final tooth arrangement selected for the patient's teeth at the end of all planned orthodontic treatment. Alternatively, the target alignment can be one of several intermediate alignments used for the patient's teeth during orthodontic treatment, and can include a variety of different treatment scenarios, including but not limited to cases where surgery is recommended, interproximal reduction (IPR) is appropriate, scheduling progress checks, optimal anchor placement, desired palatal expansion, and cases involving restorative dentistry (e.g., inlays, onlays, crowns, bridges, implants, veneers, etc.). Therefore, it should be understood that the target tooth alignment can be the result of any planned alignment of the patient's teeth following one or more incremental repositioning phases. Similarly, the initial tooth alignment can be any initial alignment of the patient's teeth following one or more incremental repositioning phases.

[0157] In some embodiments, appliances 1212, 1214, 1216, or a portion thereof, may be manufactured using indirect manufacturing techniques, such as thermoforming on a male or female mold. Appliances 1212, 1214, 1216, or a portion thereof may be inspected using the methods and systems described above. Indirect manufacturing of orthodontic appliances may involve: manufacturing a male or female mold of a patient's dentition in a targeted arrangement (e.g., through rapid prototyping, milling, etc.), and thermoforming one or more sheets on a mold to produce an appliance housing.

[0158] In an example of indirect manufacturing, the mold for the patient's dental arch can be manufactured from a digital model of the dental arch, and the housing can be formed on the mold (e.g., by thermoforming a polymer sheet on the mold of the dental arch and then trimming the thermoformed polymer sheet). The mold can be manufactured using a rapid prototyping machine (e.g., an SLA 3D printer). After the digital models of appliances 1212, 1214, and 1216 have been processed by the processing logic of the computing device, the rapid prototyping machine can receive the digital model of the dental arch mold and / or the digital models of appliances 1212, 1214, and 1216. The processing logic may include hardware (e.g., circuitry, dedicated logic, programming logic, microcode, etc.), software (e.g., instructions executed by the processing device), firmware, or a combination thereof.

[0159] To create a mold, the shape of the patient's dental arch during the treatment phase is determined based on the treatment plan. In an orthodontic example, a treatment plan can be generated based on an intraoral scan of the dental arch to be modeled. An intraoral scan of the patient's dental arch can be performed to generate a three-dimensional (3D) virtual model (mold) of the patient's dental arch. For example, a full scan of the patient's mandibular and / or maxillary dental arch can be performed to generate its 3D virtual model. Intraoral scanning can be performed by creating multiple overlapping intraoral images from different scanning stations and then stitching the intraoral images together to provide a composite 3D virtual model. In other applications, virtual 3D models can also be generated based on scans of the object to be modeled or based on the use of computer-aided drawing techniques (e.g., for designing virtual 3D molds). Alternatively, an initial negative mold can be generated from the actual object to be modeled (e.g., dental impressions, etc.). The negative mold can then be scanned to determine the shape of the positive mold to be produced.

[0160] Once a virtual 3D model of the patient's dental arch is generated, the dentist can determine the desired treatment outcome, including the final position and orientation of the patient's teeth. The processing logic can then determine multiple treatment phases to advance the teeth from their initial position and orientation to their target final position and orientation. By calculating the progression of tooth movement throughout the entire orthodontic treatment, from the initial tooth placement and orientation to the final corrected tooth placement and orientation, the shape of the final virtual 3D model and each intermediate virtual 3D model can be determined. For each treatment phase, the individual virtual 3D model will differ. The original virtual 3D model, the final virtual 3D model, and each intermediate virtual 3D model are unique and customized for the patient.

[0161] Therefore, multiple distinct virtual 3D models (digital designs) of the dental arch can be generated for a single patient. The first virtual 3D model can be a unique model of the patient's currently presented dental arch and / or teeth, and the final virtual 3D model can be a model of the patient's dental arch and / or teeth after orthodontic treatment of one or more teeth and / or jaws. Multiple intermediate virtual 3D models can be modeled, each progressively different from the previous virtual 3D models.

[0162] Each virtual 3D model of a patient's dental arch can be used to generate a customized physical mold of the arch for a specific treatment phase. The shape of the mold can be at least partially based on the shape of the virtual 3D model for that treatment phase. The virtual 3D model can be represented in a file such as a computer-aided drafting (CAD) file or a 3D printable file such as a stereolithography (STL) file. The virtual 3D model of the mold can be sent to a third party (e.g., a clinician's office, laboratory, manufacturing plant, or other entity). The virtual 3D model may include instructions that control the manufacturing system or apparatus to produce a mold with a specific geometry.

[0163] Clinicians' offices, laboratories, manufacturing plants, or other entities can receive virtual 3D models of a mold (i.e., digital models already created as described above). The entity can then input the digital model into a rapid prototyping machine. The rapid prototyping machine then uses the digital model to manufacture the mold. An example of a rapid prototyping machine is a 3D printer. 3D printing encompasses any layer-based additive manufacturing process. 3D printing can be achieved using additive processes, where continuous layers of material are formed in a prescribed shape. 3D printing can be performed using extrusion deposition, granular material bonding, lamination, photopolymerization, continuous liquid interface production (CLIP), or other techniques. 3D printing can also be achieved using subtractive processes such as milling.

[0164] In some cases, SLA is used to manufacture SLA molds. In SLA, a mold is created by sequentially printing thin layers of a photocurable material (e.g., a polymeric resin) onto one substrate at a time. A platform rests in a bath of liquid photopolymer or resin, just below the surface of the bath. A light source (e.g., an ultraviolet laser) tracks a pattern on the platform, curing the photopolymer as the light source is directed, forming the first layer of the mold. The platform is gradually lowered, and the light source tracks a new pattern on the platform to form another layer of the mold with each descent. This process is repeated until the mold is completely manufactured. Once all the layers of the mold have been formed, the mold can be cleaned and cured.

[0165] Materials such as polyesters, copolyesters, polycarbonates, thermopolymerized polyurethanes, polypropylene, polyethylene, polypropylene and polyethylene copolymers, acrylic acid, cyclic block copolymers, polyetheretherketones, polyamides, polyethylene terephthalate, polybutylene terephthalate, polyetherimide, polyethersulfone, polypropylene terephthalate, styrene block copolymers (SBCs), silicone rubber, elastomer alloys, thermopolymerized elastomers (TPEs), thermopolymerized vulcanized rubber (TPV) elastomers, polyurethane elastomers, block copolymer elastomers, polyolefin blend elastomers, thermopolymerized copolyester elastomers, thermopolymerized polyamide elastomers, or combinations thereof, can be used to directly form molds. Materials used to manufacture molds can be provided in uncured form (e.g., as liquids, resins, powders, etc.) and can be cured (e.g., by photopolymerization, light curing, gas curing, laser curing, crosslinking, etc.). The properties of the material before curing may differ from the properties of the material after curing.

[0166] After the mold is generated, it can be inspected using the systems and / or methods described above. If the mold passes inspection, it can be used to form an appliance (e.g., an orthodontic appliance).

[0167] The appliances can be formed from each mold and, when applied to a patient's teeth, can provide force to move the patient's teeth as specified in the treatment plan. Each appliance is uniquely shaped and tailored to the specific patient and stage of treatment. In the examples, appliances 1212, 1214, and 1216 can be pressure-formed or thermoformed on a mold. Each mold can be used to manufacture the appliance that will apply force to the patient's teeth at a specific stage of orthodontic treatment. Each appliance 1212, 1214, and 1216 has a tooth-receiving cavity that accommodates the tooth and resiliently repositions the tooth according to the specific stage of treatment.

[0168] In one embodiment, a sheet of material is pressed or thermoformed onto a mold. This sheet can be, for example, a polymer sheet (e.g., an elastic thermopolymer, a polymer material sheet, etc.). To thermoform the shell onto the mold, the sheet can be heated to a temperature at which it becomes pliable. Pressure can be applied simultaneously to the sheet to form the now-pliable sheet around the mold. Once cooled, the sheet will have a shape conforming to the mold. In one embodiment, a release agent (e.g., a non-stick material) is applied to the mold before the shell is formed. This facilitates subsequent removal of the shell from the mold.

[0169] Additional information can be added to the appliance. This additional information can be any information related to the appliance. Examples of such additional information include part number identifiers, patient name, patient identifier, case number, sequence identifier (e.g., indicating which appliance in the treatment sequence a particular liner belongs to), manufacturing date, clinician name, logo, etc. For example, after the appliance is thermoformed, the appliance can be laser-marked with a part number identifier (e.g., serial number, barcode, etc.). In some embodiments, the system can be configured to read (e.g., optically, magnetically, etc.) identifiers of the mold (barcode, serial number, electronic tag, etc.) to determine the part number associated with the appliance formed on the mold. After determining the part number identifier, the system can then mark the appliance with the unique part number identifier. The component number identifier can be computer-readable and can be associated with a specific patient, a specific stage in the treatment sequence, whether the orthodontic appliance is an upper or lower shell, a digital model representing the mold from which the orthodontic appliance is manufactured, and / or a digital file including a virtually generated digital model of the orthodontic appliance or its approximate characteristics (e.g., generated by approximating the outer surface of the orthodontic appliance by manipulating the digital model of the mold, expanding or scaling the projection of the mold in different planes, etc.).

[0170] After an appliance is formed on a mold for a specific treatment phase, it is then trimmed along a cutting line (also called a trimming line) and removed from the mold. The processing logic determines the cutting line of the appliance. The cutting line can be determined based on a virtual 3D model of the dental arch at the specific treatment phase, a virtual 3D model of the appliance to be formed on the dental arch, or a combination of virtual 3D models of the dental arch and the appliance. The location and shape of the cutting line are important for the appliance's function (e.g., its ability to apply desired forces to the patient's teeth) and its fit and comfort. For housings such as orthodontic appliances, orthodontic retainers, and orthodontic splints, the trimming of the housing plays a role in the housing's effectiveness for its intended purpose (e.g., aligning, retaining, or positioning one or more of the patient's teeth) and its fit on the patient's dental arch. For example, if the housing is trimmed too much, it may lose rigidity and its ability to apply forces to the patient's teeth may be impaired. When the housing is over-trimmed, it may become thinner in that location and could become a point of damage when the patient removes the housing from their tooth or removes it from the mold. In some embodiments, as one of the corrective measures taken when a potential point of damage is identified in the digital design of the appliance, the cut lines may be modified in the digital design of the appliance.

[0171] On the other hand, if the housing is under-trimmed, some parts of the housing may impact the patient's gums and cause discomfort, swelling, and / or other dental problems. Additionally, if the housing is under-trimmed in one location, the housing may be too rigid at that location. In some embodiments, the cutting line can be a straight line passing through the instrument at, below, or above the gingival line. In some embodiments, the cutting line can be a gingival cutting line, which represents the interface between the instrument and the patient's gums. In such embodiments, the cutting line controls the distance between the edge of the instrument and the patient's gingival line or gingival surface.

[0172] Each patient has a unique dental arch with unique gingiva. Therefore, the shape and position of the incision line can be unique and customized for each patient and each stage of treatment. For example, the incision line is customized to run along the gingival line (also known as the gum line). In some embodiments, the incision line may be off the gingival line in some areas and on the gingival line in others. For example, in some cases, it may be desirable for the incision line to be off the gingival line (e.g., not touching the gum), where, in the interproximal region between teeth, the shell will touch the teeth and be on the gingival line (e.g., touching the gum). Therefore, it is important to trim the shell along the predetermined incision line.

[0173] In some embodiments, dental (e.g., orthodontic) appliances (or portions thereof) described herein may be manufactured using direct manufacturing techniques, such as additive manufacturing (also referred to herein as “3D printing”) or subtractive manufacturing techniques (e.g., milling). In some embodiments, direct manufacturing involves forming an object (e.g., an orthodontic appliance or a portion thereof) without using physical templates (e.g., molds, masks, etc.) to define the object's geometry. Additive manufacturing technologies can be categorized as follows: (1) vat photopolymerization (e.g., stereolithography), in which objects are constructed layer by layer through a liquid photopolymer resin tank; (2) material spraying, in which materials are sprayed onto a build platform using a continuous or on-demand dripping (DOD) method; (3) binder spraying, in which alternating layers of build material (e.g., powder-based materials) and binder material (e.g., liquid binder) are deposited through a printhead; (4) fused deposition modeling (FDM), in which materials are extracted through a nozzle, heated, and deposited layer by layer; (5) powder bed fusion, including but not limited to direct metal laser sintering (DMLS), electron beam melting (EBM), selective thermal sintering (SHS), selective laser melting (SLM), and selective laser sintering (SLS); (6) sheet lamination, including but not limited to layered solid fabrication (LOM) and ultrasonic additive manufacturing (UAM); and (7) directional energy deposition, including but not limited to laser engineered mesh forming, directional light fabrication, direct metal deposition, and 3D laser cladding. For example, stereolithography can be used to directly manufacture one or more of appliances 1212, 1214, and 1216. In some embodiments, stereolithography involves selectively polymerizing a photosensitive resin (e.g., a photopolymer) using light (e.g., ultraviolet light) according to a desired cross-sectional shape. By sequentially polymerizing multiple object cross-sections, an object geometry can be built up layer by layer. As another example, appliances 1212, 1214, and 1216 can be directly manufactured using selective laser sintering. In some embodiments, selective laser sintering involves selectively melting and fusing layers of powder material using a laser beam according to a desired cross-sectional shape to build up the object geometry. As yet another example, appliances 1212, 1214, and 1216 are directly manufactured by fused deposition modeling. In some embodiments, fused deposition modeling involves melting and selectively depositing filaments of a thermoplastic polymer layer by layer to form an object. In yet another example, material spraying can be used to directly manufacture appliances 1212, 1214, and 1216. In some embodiments, material spraying involves spraying or extruding one or more materials onto a construction surface to form a continuous layer of the object geometry.

[0174] In some embodiments, the direct manufacturing methods provided herein build up the geometry of an object in a layer-by-layer manner, wherein successive layers are formed in discrete build-up steps. Alternatively or in combination, direct manufacturing methods that allow for the continuous building up of the object's geometry may be used, referred to herein as "continuous direct manufacturing." Various types of continuous direct manufacturing methods can be used. As an example, in some embodiments, "continuous liquid-phase printing" is used to manufacture appliances 1212, 1214, and 1216, wherein the object is continuously built from a reservoir of photopolymerizable resin by forming a gradient of partially cured resin between the object's build surface and a "dead zone" that inhibits polymerization. In some embodiments, a semi-permeable membrane is used to control the delivery of photopolymerization inhibitors (e.g., oxygen) into the dead zone to form a polymerization gradient. Continuous liquid-phase printing can achieve manufacturing speeds approximately 25 to approximately 100 times faster than other direct manufacturing methods, and approximately 1000 times faster by incorporating a cooling system. Continuous liquid phase printing is described in the following documents: U.S. Patents with publication numbers 2015 / 0097315, 2015 / 0097316 and 2015 / 0102532, the contents of each of which are incorporated herein by reference in their entirety.

[0175] As another example, by constructing a platform that moves continuously during the irradiation phase (e.g., along the vertical or Z-direction), thereby controlling the curing depth of the irradiated photopolymer with the speed of movement, a continuous direct manufacturing method can achieve the continuous construction of the object's geometry. Thus, continuous polymerization of the material on the construction surface can be achieved. Such methods are described in U.S. Patent No. 7,892,474, the disclosure of which is incorporated herein by reference in its entirety.

[0176] In another example, a continuous direct manufacturing method may involve extruding a composite material consisting of a curable liquid material surrounding a solid strand. The composite material can be extruded along a continuous three-dimensional path to form an object. Such methods are described in U.S. Patent Publication No. 2014 / 0061974, the entire disclosure of which is incorporated herein by reference.

[0177] In yet another example, the continuous direct manufacturing method utilizes heliolithography, in which a liquid photopolymer is cured using focused radiation while the build platform is continuously rotated and raised. Thus, the geometry of an object can be continuously built along a helical construction path. Such methods are described in U.S. Patent Publication No. 2014 / 0265034, the disclosure of which is incorporated herein by reference in its entirety.

[0178] The direct manufacturing method described herein is compatible with a wide range of materials, including but not limited to one or more of the following: polyesters, copolyesters, polycarbonates, thermoplastic polyurethanes, polypropylene, polyethylene, copolymers of polypropylene and polyethylene, acrylic acid, cyclic block copolymers, polyetheretherketones, polyamides, polyethylene terephthalate, polybutylene terephthalate, polyetherimide, polyethersulfone, polypropylene terephthalate, styrene block copolymers (SBCs), silicone rubber, elastomer alloys, thermoplastic elastomers (TPEs), thermoplastic vulcanizate (TPV) elastomers, polyurethane elastomers, block copolymer elastomers, polyolefin blend elastomers, thermoplastic copolyester elastomers, thermoplastic polyamide elastomers, thermosetting materials, or combinations thereof. Materials used for direct manufacturing may be provided in uncured form (e.g., as liquids, resins, powders, etc.) and may be cured (e.g., by photopolymerization, photocuring, gas curing, laser curing, crosslinking, etc.) to form orthodontic appliances or portions thereof. The properties of the material before curing may differ from those after curing. Once cured, the materials described in this article exhibit sufficient strength, stiffness, durability, and biocompatibility for use in orthodontic appliances. The properties of the cured material can be selected based on the required characteristics of the corresponding part of the appliance.

[0179] In some embodiments, the relatively rigid portions of the orthodontic appliance may be formed by direct manufacturing using one or more of the following materials: polyester, copolyester, polycarbonate, thermoplastic polyurethane, polypropylene, polyethylene, copolymers of polypropylene and polyethylene, acrylic acid, cyclic block copolymers, polyetheretherketone, polyamide, polyethylene terephthalate, polybutylene terephthalate, polyetherimide, polyethersulfone, and / or polypropylene terephthalate.

[0180] In some embodiments, the relatively elastic portion of the orthodontic appliance may be formed by direct manufacturing using one or more of the following materials: styrene block copolymer (SBC), silicone rubber, elastomer alloy, thermoplastic elastomer (TPE), thermoplastic vulcanized rubber (TPV) elastomer, polyurethane elastomer, block copolymer elastomer, polyolefin blend elastomer, thermoplastic copolyester elastomer, and / or thermoplastic polyamide elastomer.

[0181] Machine parameters can include curing parameters. For digital light processing (DLP) based curing systems, curing parameters can include power, curing time, and / or the grayscale of the entire image. For laser-based curing systems, curing parameters can include power, speed, beam size, beam shape, and / or beam power distribution. For printing systems, curing parameters can include material droplet size, viscosity, and / or curing power. These machine parameters (e.g., certain parameters per 1-x layers and certain parameters after each build) can be periodically monitored and adjusted as part of process control on the manufacturing machine. Process control can be achieved by including sensors on the machine that measure power and other beam parameters per layer or every few seconds and automatically adjust them via feedback loops. For DLP machines, depending on system stability, grayscale, etc., can be measured and calibrated before, during, and / or at the end of each build and / or at predetermined time intervals (e.g., every n builds, once per hour, once per day, once per week, etc.). Additionally, material properties and / or optical properties can be provided to the manufacturing machine, and the machine process control module can use these parameters to adjust machine parameters (e.g., power, time, grayscale, etc.) to compensate for variations in material properties. By implementing process control on manufacturing machines, variations in tool precision and residual stress can be reduced.

[0182] Optionally, the direct manufacturing method described herein allows for the manufacture of appliances comprising multiple materials, referred to herein as "multimaterial direct manufacturing." In some embodiments, the multimaterial direct manufacturing method involves simultaneously forming an object from multiple materials in a single manufacturing step. For example, a multi-tip extrusion apparatus can be used to selectively dispense multiple types of materials from different material supply sources to manufacture an object from multiple different materials. Such a method is described in U.S. Patent No. 6,749,414, the disclosure of which is incorporated herein by reference in its entirety. Alternatively or in combination, the multimaterial direct manufacturing method may involve forming an object from multiple materials in multiple sequential manufacturing steps. For example, a first portion of the object may be formed from a first material according to any direct manufacturing method herein, then a second portion of the object may be formed from a second material according to a method herein, and so on, until the entire object has been formed.

[0183] Direct manufacturing offers various advantages compared to other manufacturing methods. For example, compared to indirect manufacturing, direct manufacturing allows for the production of orthodontic appliances without the use of any molds or templates to shape the appliance, thereby reducing the number of manufacturing steps involved and improving the resolution and accuracy of the final appliance geometry. Additionally, direct manufacturing allows for precise control over the three-dimensional geometry of the appliance, such as its thickness. Complex structures and / or auxiliary components can be integrally formed as a monolith with the appliance housing in a single manufacturing step, rather than being added to the housing in separate manufacturing steps. In some embodiments, direct manufacturing is used to produce appliance geometries that are difficult to create using alternative manufacturing techniques, such as appliances with very small or fine features, complex geometries, undercuts, adjacent structures, and housings with variable thickness and / or internal structures (e.g., increasing strength by reducing weight and material usage). For example, in some embodiments, the direct manufacturing method described herein allows for the manufacture of orthodontic appliances with feature dimensions less than or equal to about 5 µm, or in the range of about 5 µm to about 50 µm, or in the range of about 20 µm to about 50 µm.

[0184] The direct manufacturing techniques described herein can be used to produce appliances with substantially isotropic material properties (e.g., substantially the same or similar strength in all directions). In some embodiments, the direct manufacturing method described herein allows the production of orthodontic appliances with strength variations in all directions not exceeding about 25%, about 20%, about 15%, about 10%, about 5%, about 1%, or about 0.5%. Furthermore, the direct manufacturing method described herein can be used to manufacture orthodontic appliances at a faster rate compared to other manufacturing techniques. In some embodiments, the direct manufacturing method described herein allows the production of orthodontic appliances within time intervals of less than or equal to about 1 hour, about 30 minutes, about 25 minutes, about 20 minutes, about 15 minutes, about 10 minutes, about 5 minutes, about 4 minutes, about 3 minutes, about 2 minutes, about 1 minute, or about 30 seconds. Such manufacturing speeds allow for rapid “chairside” production of custom-made appliances, for example, during regular appointments or checkups.

[0185] In some embodiments, the direct manufacturing methods described herein implement process control over various machine parameters for the direct manufacturing system or apparatus to ensure the manufacture of the resulting appliance with high precision. Such precision is advantageous for ensuring that the desired force system is accurately delivered to the teeth to effectively induce tooth movement. Process control can be implemented to account for process variations caused by a variety of sources, such as material properties, machine parameters, environmental variables, and / or post-processing parameters.

[0186] Material properties can vary depending on the characteristics of the raw materials, their purity, and / or process variables during mixing. In many embodiments, resins or other materials intended for direct manufacturing should be manufactured under strict process control to ensure minimal variation in optical properties, material properties (e.g., viscosity, surface tension), physical properties (e.g., modulus, strength, elongation), and / or thermal properties (e.g., glass transition temperature, heat distortion temperature). Process control for material manufacturing processes can be achieved by screening the physical properties of raw materials during mixing and / or controlling temperature, humidity, and / or other process parameters. By controlling the material manufacturing process, the variability of process parameters can be reduced, resulting in more uniform material properties for each batch. As further discussed herein, residual variability in material properties can be compensated for through on-machine process control.

[0187] Machine parameters can include curing parameters. For digital light processing (DLP) based curing systems, curing parameters can include power, curing time, and / or the grayscale of the entire image. For laser-based curing systems, curing parameters can include power, speed, beam size, beam shape, and / or beam power distribution. For printing systems, curing parameters can include material droplet size, viscosity, and / or curing power. These machine parameters (e.g., certain parameters per 1-x layers and certain parameters after each build) can be periodically monitored and adjusted as part of process control on the manufacturing machine. Process control can be achieved by including sensors on the machine that measure power and other beam parameters per layer or every few seconds and automatically adjust them via feedback loops. For DLP machines, grayscale can be measured and calibrated at the end of each build. Additionally, material properties and / or optical properties can be provided to the manufacturing machine, and the machine process control module can use these parameters to adjust machine parameters (e.g., power, time, grayscale, etc.) to compensate for variations in material properties. By controlling the manufacturing machine, variations in instrument accuracy and residual stress can be reduced.

[0188] In many embodiments, environmental variables (e.g., temperature, humidity, sunlight, or exposure to other energy / curing sources) are kept within a narrow range to reduce variations in appliance thickness and / or other characteristics. Optionally, machine parameters may be adjusted to compensate for environmental variables.

[0189] In many embodiments, post-treatment of the appliance includes cleaning, post-curing, and / or support removal processes. Relevant post-treatment parameters may include the purity of the cleaning agent, cleaning pressure and / or temperature, cleaning time, post-curing energy and / or time, and / or the consistency of the support removal process. These parameters can be measured and adjusted as part of a process control scheme. Furthermore, the physical properties of the appliance can be altered by modifying the post-treatment parameters. Adjusting the post-treatment machine parameters can provide another way to compensate for variations in material properties and / or machine properties.

[0190] Once the appliance (e.g., an orthodontic appliance) is manufactured directly, the system and / or methods described above can be used to test the appliance.

[0191] The configuration of orthodontic appliances described herein can be determined based on the patient's treatment plan (e.g., a treatment plan involving the successive application of multiple appliances to progressively reposition teeth). Computer-based treatment planning and / or appliance manufacturing methods can be used to facilitate the design and manufacture of appliances. For example, one or more appliance components described herein can be digitally designed and manufactured using computer-controlled manufacturing equipment (e.g., computer numerical control (CNC) milling, computer-controlled rapid prototyping (e.g., 3D printing), etc.). The computer-based methods proposed herein can improve the accuracy, flexibility, and convenience of appliance manufacturing.

[0192] Figure 13A method 1300 for orthodontic treatment using multiple appliances according to an embodiment is illustrated. Method 1300 can be practiced using any of the appliances or sets of appliances described herein. In block 1302, a first orthodontic appliance is applied to the patient's teeth to reposition the teeth from a first dental alignment to a second dental alignment. In block 1304, a second orthodontic appliance is applied to the patient's teeth to reposition the teeth from a second dental alignment to a third dental alignment. Method 1300 can be repeated as needed using any suitable number and combination of sequential appliances to progressively reposition the patient's teeth from an initial alignment to a target alignment. All of these appliances can be manufactured in the same phase, in groups, or in batches (e.g., at the beginning of a treatment phase), or one appliance can be manufactured at a time, and the patient can wear each appliance until the teeth no longer feel pressure from each appliance, or until the maximum amount of tooth movement presented at that given phase has been reached. Several different appliances (e.g., a set of appliances) can be designed and even manufactured before the patient wears any of the multiple appliances. After an appliance has been worn for an appropriate period of time, the patient can replace the current appliance with the next appliance in the series until no appliances remain. The appliances are typically not fixed to the teeth, and the patient can place and change appliances at any time during treatment (e.g., removable appliances). The final appliance, or several appliances in the series, may have one or more geometries selected for overcorrecting tooth alignment. For example, one or more appliances may have geometries that (if fully realized) move individual teeth beyond the alignment already selected as “final.” This overcorrection may be desirable to counteract potential relapse after the repositioning method has ended (e.g., allowing individual teeth to move in the opposite direction to their pre-corrected position). Overcorrection may also be beneficial for accelerating correction (e.g., an appliance with a geometry positioned beyond the desired intermediate or final position may move individual teeth toward that position at a greater rate). In this case, the use of the appliance can be terminated before the teeth reach the position determined by the appliance. Furthermore, overcorrection can be intentionally performed to compensate for any inaccuracies or limitations of the appliance.

[0193] Figure 14 A method 1400 for designing an orthodontic appliance to be manufactured by direct fabrication, according to an embodiment, is shown. Method 1400 can be applied to any embodiment of the orthodontic appliance described herein. Some or all of the blocks in method 1400 can be executed by any suitable data processing system or apparatus (e.g., one or more processors configured with suitable instructions).

[0194] In box 1402, a movement path is determined to move one or more teeth from an initial alignment to a target alignment. The initial alignment can be determined, for example, using wax bite method, direct contact scanning, X-ray imaging, tomography, ultrasound imaging, and other techniques for obtaining information about the position and structure of teeth, jaws, gingiva, and other orthodontic-related tissues, from a mold or scan of the patient's teeth or oral tissues. A digital dataset representing the initial (e.g., pre-treatment) alignment of the patient's teeth and other tissues can be derived from the obtained data. Optionally, the initial digital dataset is processed to segment the tissue components to each other. For example, a data structure digitally representing the individual crowns can be generated. Advantageously, a digital model of the entire tooth can be generated, including the measured or inferred hidden surfaces and root structures, as well as the surrounding bone and soft tissue.

[0195] The target alignment of teeth (e.g., the desired and expected end result of orthodontic treatment) can be received from the clinician in the form of a prescription, calculated based on fundamental orthodontic principles, and / or inferred from the clinical prescription. By specifying the desired final position of the teeth and a digital representation of the teeth themselves, the final position and surface geometry of each tooth can be specified to form a complete model of the tooth alignment at the end of the desired treatment.

[0196] Having both an initial position and a target position for each tooth, a movement path can be defined for the movement of each tooth. In some embodiments, the movement path is configured to move the tooth from its initial position to its desired target position in the fastest manner with the least amount of round trips. The tooth path can optionally be segmented (partitioned), and the segments can be computed such that the movement of each tooth within a segment remains within threshold limits of linear translation and rotational translation. In this way, the endpoints of each path segment can constitute a clinically feasible repositioning, and the set of segment endpoints can constitute a clinically feasible sequence of tooth positions such that moving from one point to the next in the sequence does not result in tooth collision.

[0197] In box 1404, the force system that produces the movement of one or more teeth along the movement path is identified. The force system may include one or more forces and / or one or more torques. Different force systems can result in different types of tooth movement, such as tilting, translation, rotation, extrusion, intrusion, root movement, etc. Biomechanical principles, modeling techniques, force calculation / measurement techniques, etc. (including knowledge and methods commonly used in orthodontics) can be used to determine the appropriate force system to be applied to the teeth to achieve tooth movement. Sources may be considered when determining the force system to be applied, including literature, force systems determined through experimental or virtual modeling, computer-based modeling, clinical experience, minimization of unwanted forces, etc.

[0198] Determining the force system can include constraints on permissible forces, such as permissible directions and magnitudes, and constraints on the desired movement induced by the applied forces. For example, different patients may require different movement strategies when fabricating a palatal expander. For instance, the amount of force required to dissect the palate can depend on the patient's age, as very young patients may not have fully formed sutures. Therefore, in adolescent patients and others with incompletely closed palatal sutures, palatal expansion can be accomplished with lower force magnitudes. Slower palatal movement can also aid bone growth to fill the expanding suture. For other patients, faster expansion may be required, which can be achieved by applying greater forces. The structure and materials of the appliance can be selected based on these requirements; for example, by selecting a palatal expander capable of applying large forces to break the palatal suture and / or induce rapid expansion of the palate. Subsequent appliance stages can be designed to apply varying amounts of force, such as initially applying large forces to break the suture, followed by smaller forces to maintain suture separation or gradually expand the palate and / or dental arch.

[0199] Determining the force system may also involve modeling the patient's facial structures, such as the skeletal structure of the jaw and palate. For example, scan data of the palate and dental arch (such as X-ray strip data or 3D optical scan data) can be used to determine parameters of the skeletal and muscular systems of the patient's oral cavity in order to determine the force sufficient to provide the desired expansion of the palate and / or dental arch. In some embodiments, the thickness and / or density of the palatal suture may be measured or input by a treatment professional. In other embodiments, the treatment professional may select appropriate treatment based on the patient's physiological characteristics. For example, the characteristics of the palate may also be estimated based on factors such as the patient's age; for instance, younger adolescent patients typically require less force to expand the suture than older patients because the suture has not yet fully formed.

[0200] In box 1406, an orthodontic appliance configured to generate a force system is identified. The determination of the orthodontic appliance, its geometry, material composition, and / or properties can be performed using a treatment or force application simulation environment. The simulation environment may include, for example, a computer modeling system, a biomechanical system, or an instrument. Optionally, a digital model of the appliance and / or teeth, such as a finite element model, can be generated. The finite element model can be created using computer program application software available from various vendors. To create a solid geometry model, computer-aided engineering (CAE) or computer-aided design (CAD) programs can be used, such as AutoCAD® software products available from Autodesk, Inc., San Rafael, California. To create and analyze the finite element model, program products from multiple vendors can be used, including the finite element analysis package from ANSYS, Inc., Fort Cannons, Pennsylvania, and the SIMULIA (Abaqus) software product from Dassault Systèmes, Inc., Waltham, Massachusetts.

[0201] Optionally, one or more orthodontic appliances can be selected for testing or force modeling. As described above, the desired tooth movement and the force system required or desired to induce the desired tooth movement can be identified. Using a simulation environment, candidate orthodontic appliances can be analyzed or modeled to determine the actual force system generated by using the candidate appliances. One or more modifications to the candidate appliances can be optionally made, and force modeling can be further analyzed as described, for example, to iteratively determine the appliance design that produces the desired force system.

[0202] In block 1408, instructions are generated for manufacturing an orthodontic appliance comprising an orthodontic appliance. The instructions may be configured to control a manufacturing system or apparatus to produce an orthodontic appliance having the specified orthodontic appliance. In some embodiments, the instructions are configured to manufacture the orthodontic appliance using direct manufacturing methods (e.g., stereolithography, selective laser sintering, fused deposition modeling, 3D printing, continuous direct manufacturing, multimaterial direct manufacturing, etc.) according to the various methods presented herein. In alternative embodiments, the instructions may be configured to manufacture the appliance indirectly, such as by thermoforming.

[0203] Method 1400 may include additional boxes: 1) scanning the patient’s maxillary arch and palate intraorally to generate three-dimensional data of the palate and maxillary arch; 2) determining the three-dimensional shape profile of the appliance to provide the gap and tooth occlusion structure as described herein.

[0204] Although the above boxes illustrate a method 1400 for designing orthodontic appliances according to some embodiments, those skilled in the art will recognize some variations based on the teachings described herein. Some boxes may include sub-boxes. Certain boxes may typically be performed repeatedly as needed. One or more boxes of method 1400 can be performed using any suitable manufacturing system or apparatus, such as the embodiments described herein. Some boxes may be optional, and the order of the boxes may be changed as needed.

[0205] Figure 15 A method 1500 for orthodontic treatment and / or design or manufacture of a digital planning appliance according to an embodiment is shown. Method 1500 can be applied to any treatment procedure described herein and can be performed by any suitable data processing system.

[0206] In box 1510, a digital representation of the patient's teeth is received. The digital representation may include surface topography data of the patient's oral cavity (including teeth, gingival tissue, etc.). The surface topography data can be generated by directly scanning the oral cavity, a physical model (positive or negative) of the oral cavity, or an impression of the oral cavity using a suitable scanning device (e.g., a handheld scanner, a desktop scanner, etc.).

[0207] In box 1502, one or more treatment phases are generated based on the digital representation of the teeth. A treatment phase can be an incremental repositioning phase in the orthodontic treatment process, designed to move one or more of the patient's teeth from an initial tooth arrangement to a target arrangement. For example, a treatment phase can be generated by determining the initial tooth arrangement indicated by the digital representation, determining the target tooth arrangement, and determining the movement path for one or more teeth in the initial arrangement required to achieve the target tooth arrangement. The movement path can be optimized based on minimizing the total distance of movement, preventing collisions between teeth, avoiding more difficult tooth movements, or any other suitable criteria.

[0208] In box 1504, at least one orthodontic appliance is manufactured based on the generated treatment phase. For example, a set of appliances can be manufactured, each appliance shaped according to a tooth alignment specified by one of the treatment phases, such that the appliances can be worn sequentially by the patient to incrementally reposition the teeth from the initial alignment to the target alignment. The appliance set may include one or more orthodontic appliances described herein. The manufacture of the appliances may involve creating a digital model of the appliance to serve as input to a computer-controlled manufacturing system. Direct manufacturing methods, indirect manufacturing methods, or a combination thereof may be used to form the appliances, as needed.

[0209] In some cases, the phased division of various orthodontic arrangements or treatment stages may not be necessary for the design and / or manufacture of the appliance. The design and / or manufacture of orthodontic appliances, and possibly specific orthodontic treatments, may include using a representation of the patient's teeth (e.g., receiving a digital representation of the patient's teeth), and then designing and / or manufacturing the orthodontic appliance based on the representation of the patient's teeth in the arrangement presented by the received representation.

[0210] The foregoing description sets forth numerous specific details, such as examples of specific systems, components, methods, etc., to provide a good understanding of several embodiments of the present disclosure. However, it will be apparent to those skilled in the art that at least some embodiments of the present disclosure can be practiced without these specific details. In other instances, well-known components or methods have not been described in detail or presented in a simple block diagram format to avoid unnecessarily obscuring the present disclosure. Therefore, the specific details set forth are merely exemplary. Specific implementations may differ from these exemplary details and are still considered to be within the scope of this disclosure.

[0211] Throughout this specification, references to "an embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment. Therefore, the phrases "in one embodiment" or "in an embodiment" appearing in various places throughout this specification do not necessarily refer to the same embodiment. Furthermore, the term "or" is intended to indicate an inclusive "or" rather than an exclusive "or." When the terms "about" or "approximately" are used herein, this is intended to indicate that the presented nominal values ​​are exactly within ±10%.

[0212] Although the operations of the methods described herein are shown and described in a specific order, the order of operations for each method can be changed so that some operations can be performed in reverse order, or that some operations can be performed at least partially concurrently with other operations. In another embodiment, the instructions or sub-operations of different operations can be intermittent and / or alternating. In one embodiment, multiple metal bonding operations are performed as a single step.

[0213] The following are some example implementation methods.

[0214] A first example embodiment relates to a defect detection system for a transparent three-dimensional (3D) object. The defect detection system includes a transparent platform configured to support the transparent 3D object. The system includes multiple light sources disposed below the transparent platform to illuminate the transparent 3D object through the platform. The system also includes multiple cameras disposed above the plane of the transparent platform, surrounding the platform and angled relative to it. The cameras are configured to generate multiple images of the transparent 3D object from multiple directions when the object is illuminated by one or more of the light sources, wherein each of the multiple images depicts a different region or viewpoint of the transparent 3D object. The system includes a computing device configured to process the multiple images to determine whether the transparent 3D object includes defects and to output an indication of whether the transparent 3D object includes defects.

[0215] The second example implementation can extend the first example implementation. In the second example implementation, the plurality of images are processed using a trained artificial intelligence (AI) model, wherein the output of the trained AI model includes the probability that the transparent 3D object includes the defect.

[0216] The third example implementation can extend the first example implementation to any of the second example implementations. In the third example implementation, the plurality of images includes a set of images captured simultaneously, each image in the set being captured by a different camera among the plurality of cameras.

[0217] The fourth example implementation can extend any of the first to third example implementations. In the fourth example implementation, the defect detection system further includes an additional light source with its axis substantially perpendicular to the transparent platform. The defect detection system also includes an additional camera disposed above the transparent platform and having an imaging axis substantially perpendicular to the transparent platform, the additional camera being configured to capture additional images during illumination of the transparent 3D object by the additional light source. In the fourth example implementation, the computing device is further configured to process the additional images to a) determine a part number of the transparent 3D object and b) determine whether the transparent 3D object includes a defect.

[0218] The fifth example implementation can extend any of the first to fourth example implementations. In the fifth example implementation, the defect detection system further includes a spectral filter for each of the plurality of cameras, wherein the spectral filter is configured to filter out light outside the wavelength range output by the plurality of light sources.

[0219] The sixth example implementation can extend the fifth example implementation. In the sixth example implementation, the wavelength range corresponds to at least one of infrared light or near-infrared light.

[0220] The seventh example implementation can extend any of the fifth to sixth example implementations. In the seventh example implementation, the wavelength range is 820-930 nm.

[0221] The eighth example implementation can extend any of the first to seventh example implementations. In the eighth example implementation, the plurality of cameras includes a plurality of side-view cameras, wherein each of the plurality of side-view cameras includes a lens having a focal plane, an image sensor having an image plane, and a tilt shift adapter that positions the focal plane at an angle relative to the image plane.

[0222] The ninth example implementation can extend the eighth example implementation. In the ninth example implementation, the plurality of side-view cameras have a first angle relative to the transparent platform, and wherein the tilt shift adapter causes a second angle between the focal plane and the image plane, the second angle being a function of the first angle, wherein the tilt shift adapter increases the depth of field of the side-view cameras.

[0223] The tenth example implementation can extend any of the first to ninth example implementations. In the tenth example implementation, the transparent platform includes a feature pattern arranged in a circle around the center of the transparent platform, wherein the feature pattern is configured to surround the transparent 3D object.

[0224] The eleventh example implementation can extend the tenth example implementation. In the eleventh example implementation, the computing device is further configured to determine at least one of the positions or orientations of the transparent 3D object on the transparent platform based on the identification of the feature patterns in the plurality of images.

[0225] The twelfth example implementation can extend the eleventh example implementation. In the twelfth example implementation, the defect detection system further includes a robotic arm configured to pick up the transparent 3D object using at least one of a determined orientation or a determined position of the transparent 3D object on the transparent platform.

[0226] The thirteenth example implementation can extend any of the tenth to twelfth example implementations. In the thirteenth example implementation, the computing device is further configured to perform configuration based on the detection of the feature patterns in images captured by the plurality of cameras and the geometric characteristics of the defect detection system.

[0227] The fourteenth example implementation can extend any of the tenth to thirteenth example implementations. In the fourteenth example implementation, the plurality of cameras are configured such that the feature pattern surrounds approximately 90% of the area of ​​the plurality of images.

[0228] The fifteenth example implementation can extend any of the first to fourteenth example implementations. In the fifteenth example implementation, the defect detection system is configured such that light refracted by the transparent 3D object makes the surface of the transparent 3D object visible in the plurality of images.

[0229] The sixteenth example implementation can extend any of the first to fifteenth example implementations. In the sixteenth example implementation, the defect detection system further includes one or more matte white boards configured to guide light output by the plurality of light sources through the transparent 3D object and toward the plurality of cameras. The defect detection system also includes one or more matte black boards disposed around the one or more matte white boards and configured to absorb light output by the plurality of light sources.

[0230] The seventeenth example implementation can extend the sixteenth example implementation. In the seventeenth example implementation, the defect detection system further includes a plurality of light curtains disposed between the plurality of light sources and configured to block light between the light sources.

[0231] The eighteenth example implementation can extend any of the first to seventeenth example implementations. In the eighteenth example implementation, the computing device is further configured to determine a digital file associated with the transparent 3D object. The computing device is further configured to determine a first contour associated with the transparent 3D object from the digital file. The computing device is further configured to determine a second contour of the transparent 3D object from the plurality of images. The computing device is further configured to compare the first contour with the second contour. The computing device is further configured to determine a difference index between the first contour and the second contour based on the comparison result. The computing device is further configured to determine whether the difference index exceeds a difference threshold. The computing device is further configured to determine that the transparent 3D printed object includes defects in response to determining that the difference index exceeds the difference threshold.

[0232] The nineteenth example implementation can extend any of the first to eighteenth example implementations. In the nineteenth example implementation, the computing device is further configured to perform the following operations on each of the plurality of images: perform edge detection on the image to determine the boundary of the transparent 3D object in the image; select a set of points on the boundary; use the set of points to determine a region of interest, wherein the region of interest includes a first region in the image depicting the transparent 3D object within the boundary; and crop the image to exclude a second region in the image outside the region of interest, wherein the cropped image is processed by the computing device using a trained artificial intelligence (AI) model to identify defects.

[0233] The twentieth example implementation can extend any of the first to the nineteenth example implementations. In the twentieth example implementation, the plurality of light sources are configured to provide a uniform brightness distribution during the generation of the plurality of images.

[0234] The twenty-first example embodiment can extend any of the first to the twenty-twentieth example embodiments. In the twenty-first example embodiment, the defect detection system further includes a plurality of light scattering plates, each of which is positioned close to a light source among the plurality of light sources and configured to scatter light from the light source near the light scattering plate.

[0235] The twenty-second example embodiment relates to a method for performing automated quality control on a transparent three-dimensional (3D) object. The method includes illuminating the transparent 3D object using multiple light sources. The method includes simultaneously generating multiple images of the transparent 3D object using multiple cameras, wherein each of the multiple images depicts a different region or viewpoint of the transparent 3D object. The method includes processing the multiple images by a computing device to identify defects in the transparent 3D object. The method includes determining, based on the results of the processing, whether the transparent 3D object includes one or more defects, without user input.

[0236] The twenty-third example implementation can extend the twenty-second example implementation. In the twenty-third example implementation, the transparent 3D object includes a polymer orthodontic appliance for the patient during the treatment phase of an orthodontic treatment plan.

[0237] The twenty-fourth example implementation can extend any of the twenty-second to twenty-third example implementations. In the twenty-fourth example implementation, the method further includes performing the following operations on each of the plurality of images: performing edge detection on the image to determine the boundary of a transparent 3D printed object in the image; selecting a set of points on the boundary; using the set of points to determine a region of interest, wherein the region of interest includes a first region in the image depicting the transparent 3D object within the boundary; and cropping the image to exclude a second region in the image outside the region of interest, wherein the cropped image is processed by the computing device using an artificial intelligence (AI) model.

[0238] The twenty-fifth example implementation can extend any of the twenty-second to twenty-fourth example implementations. In the twenty-fifth example implementation, an artificial intelligence (AI) model is used to process the plurality of images, wherein the output of the AI ​​model also includes an indication of the severity of the detected defects.

[0239] The twenty-sixth exemplary embodiment can extend any of the twenty-second to twenty-fifth exemplary embodiments. In the twenty-sixth exemplary embodiment, the transparent 3D object is disposed on a transparent platform, wherein the plurality of light sources are disposed below the transparent platform and illuminate the transparent 3D object through the transparent platform, and wherein the plurality of cameras are disposed above the plane of the transparent platform, surrounding the transparent platform and at an angle relative to the transparent platform.

[0240] The twenty-seventh example implementation can extend the twenty-sixth example implementation. In the twenty-seventh example implementation, the method further includes illuminating the transparent 3D object with an additional light source whose axis is substantially perpendicular to the transparent platform. The method further includes capturing additional images of the transparent 3D object using an additional camera, the additional camera being positioned above the transparent platform and having an imaging axis substantially perpendicular to the transparent platform, during the illumination of the transparent 3D object by the additional light source. The method further includes processing the additional images to a) determine a part number of the transparent 3D object and b) determine whether the transparent 3D object includes defects.

[0241] The twenty-eighth example implementation can extend any of the twenty-second to twenty-seventh example implementations. In the twenty-eighth example implementation, a trained artificial intelligence (AI) model is used to process the plurality of images, wherein the output of the trained AI model includes the probability that the transparent 3D object includes defects.

[0242] The twenty-ninth exemplary embodiment can extend any of the twenty-second to twenty-eighth exemplary embodiments. In the twenty-ninth exemplary embodiment, the method further includes filtering the light arriving at the plurality of cameras using a spectral filter of each of the plurality of cameras, wherein the spectral filter is configured to filter out light outside the wavelength range output by the plurality of light sources.

[0243] The thirtieth example embodiment can extend the twenty-ninth example embodiment. In the thirtieth example embodiment, the wavelength range corresponds to at least one of infrared light or near-infrared light.

[0244] The thirty-first example embodiment can be an extension of any of the twenty-ninth to thirtieth example embodiments. In the thirty-first example embodiment, the wavelength range is 820-930 nm.

[0245] The thirty-second example implementation can extend any of the twenty-second to thirty-first example implementations. In the thirty-second example implementation, the plurality of cameras includes a plurality of side-view cameras, wherein each of the plurality of side-view cameras includes a lens having a focal plane, an image sensor having an image plane, and a tilt shift adapter that positions the focal plane at an angle relative to the image plane.

[0246] The thirty-third example implementation can extend the thirty-second example implementation. In the thirty-third example implementation, the plurality of side-view cameras have a first angle relative to the transparent platform supporting the transparent 3D object, and wherein the tilt shift adapter causes a second angle between the focal plane and the image plane, the second angle being a function of the first angle, wherein the tilt shift adapter increases the depth of field of the side-view cameras.

[0247] The thirty-fourth exemplary embodiment can extend any of the twenty-second to thirty-third exemplary embodiments. In the thirty-fourth exemplary embodiment, the transparent 3D object is disposed on a transparent platform, the transparent platform comprising a dotted pattern arranged in a circle around the center of the transparent platform. The method further includes determining at least one of the positions or orientations of the transparent 3D object on the transparent platform based on the identification of the dotted pattern in the plurality of images.

[0248] The thirty-fifth example implementation can extend the thirty-fourth example implementation. In the thirty-fifth example implementation, the method further includes: using a robotic arm to pick up the transparent 3D object based on at least one of a determined orientation or a determined position of the transparent 3D object on the transparent platform.

[0249] The thirty-sixth exemplary embodiment can extend any of the thirty-fourth to thirty-fifth exemplary embodiments. In the thirty-sixth exemplary embodiment, the method further includes configuring the defect detection system based on: the detection of the dotted pattern in the images captured by the plurality of cameras and the geometric characteristics of the defect detection system.

[0250] The thirty-seventh exemplary embodiment can extend any of the thirty-fourth to thirty-sixth exemplary embodiments. In the thirty-seventh exemplary embodiment, the plurality of cameras are configured such that the dot pattern surrounds approximately 90% of the area of ​​the plurality of images.

[0251] The thirty-eighth exemplary embodiment can extend any of the twenty-second to thirty-seventh exemplary embodiments. In the thirty-eighth exemplary embodiment, light refracted by the transparent 3D object makes the surface of the transparent 3D object visible in the plurality of images.

[0252] The thirty-ninth example implementation can extend any of the twenty-second to thirty-eighth example implementations. In the thirty-ninth example implementation, the method further includes determining a digital file associated with the transparent 3D object. The method further includes determining a first contour associated with the transparent 3D object from the digital file. The method further includes determining a second contour of the transparent 3D object from the plurality of images. The method further includes comparing the first contour with the second contour. The method further includes determining a difference index between the first contour and the second contour based on the comparison result. The method further includes determining whether the difference index exceeds a difference threshold. The method further includes determining that the transparent 3D object includes a defect in response to determining that the difference index exceeds the difference threshold.

[0253] The fortieth example implementation can extend any of the twenty-second to thirty-ninth example implementations. In the fortieth example implementation, the method further includes performing edge detection on the images of the plurality of images to determine the boundaries of the transparent 3D object in the images. The method further includes selecting a set of points on the boundaries. The method further includes using the set of points to determine a region of interest, wherein the region of interest includes a first region in the image depicting the transparent 3D object within the boundaries. The method further includes cropping the image to exclude a second region in the image outside the region of interest, wherein the cropped image is processed by the computing device using a trained artificial intelligence (AI) model to identify defects.

[0254] The forty-first example implementation can extend any of the twenty-second to forty example implementations. In the forty-first example implementation, the method further includes creating a transparent object based on a digital file associated with the transparent 3D object before performing automated quality control.

[0255] It should be understood that the above description is intended to be illustrative and not restrictive. Many other embodiments will be apparent to those skilled in the art upon reading and understanding the above description. Therefore, the scope of this disclosure should be determined by reference to the appended claims and the full scope of their equivalents.

Claims

1. A defect detection system for transparent 3D objects, comprising: The transparent platform is configured to support transparent 3D objects; Multiple light sources are positioned below the transparent platform to illuminate the transparent 3D object through the transparent platform; Multiple cameras are arranged above the plane of the transparent platform, surrounding the transparent platform and at an angle relative to the transparent platform. The multiple cameras are configured to generate multiple images of the transparent 3D object from multiple directions when the transparent 3D object is illuminated by one or more of the multiple light sources, wherein each of the multiple images depicts a different region or viewpoint of the transparent 3D object. as well as The computing device is configured as follows: Process the plurality of images to determine whether the transparent 3D object includes defects; and Output an indication of whether the transparent 3D object contains defects.

2. The defect detection system according to claim 1, wherein, The multiple images are processed using a trained artificial intelligence (AI) model, wherein the output of the trained AI model includes the probability that the transparent 3D object contains defects.

3. The defect detection system according to claim 1, wherein, The multiple images include a set of images captured simultaneously, each image in the set being captured by a different camera among the multiple cameras.

4. The defect detection system according to claim 1 further includes: An additional light source has an axis that is substantially perpendicular to the transparent platform; as well as An additional camera, disposed above the transparent platform and having an imaging axis substantially perpendicular to the transparent platform, is configured to capture additional images during the illumination of the transparent 3D object by the additional light source; The computing device is further configured to: The additional image is processed to: a) determine the part number of the transparent 3D object and b) determine whether the transparent 3D object includes defects.

5. The defect detection system according to claim 1, further comprising: Each of the plurality of cameras has a spectral filter, wherein the spectral filter is configured to filter out light outside the wavelength range output by the plurality of light sources.

6. The defect detection system according to claim 5, wherein, The wavelength range corresponds to at least one of infrared or near-infrared light.

7. The defect detection system according to claim 5, wherein, The wavelength range is 820-930nm.

8. The defect detection system according to claim 1, wherein, The plurality of cameras includes a plurality of side-view cameras, wherein each of the plurality of side-view cameras includes a lens having a focal plane, an image sensor having an image plane, and a tilt shift adapter that causes the focal plane to have a certain angle relative to the image plane.

9. The defect detection system according to claim 8, wherein, The plurality of side-view cameras have a first angle relative to the transparent platform, and wherein the tilt shift adapter causes a second angle between the focal plane and the image plane, the second angle being a function of the first angle, wherein the tilt shift adapter increases the depth of field of the side-view cameras.

10. The defect detection system according to claim 1, wherein, The transparent platform includes a feature pattern arranged in a circle around the center of the transparent platform, wherein the feature pattern is configured to surround the transparent 3D object.

11. The defect detection system according to claim 10, wherein, The computing device is also configured to: Based on the identification of the feature patterns in the plurality of images, at least one of the positions or orientations of the transparent 3D object on the transparent platform is determined.

12. The defect detection system according to claim 11, further comprising: A robotic arm is configured to pick up the transparent 3D object using at least one of a determined orientation or a determined position of the transparent 3D object on the transparent platform.

13. The defect detection system according to claim 10, wherein, The computing device is also configured to perform configuration based on the detection of the feature patterns in images captured by the plurality of cameras and the geometric characteristics of the defect detection system.

14. The defect detection system according to claim 10, wherein, The plurality of cameras are configured such that the feature pattern surrounds approximately 90% of the area of ​​the plurality of images.

15. The defect detection system according to claim 1, wherein, The defect detection system is configured such that light refracted by the transparent 3D object makes the surface of the transparent 3D object visible in the plurality of images.

16. The defect detection system according to claim 1, further comprising: One or more matte white boards are configured to guide light output by the plurality of light sources through the transparent 3D object and toward the plurality of cameras; as well as One or more matte blackboards are arranged around the one or more matte whiteboards and configured to absorb light output by the plurality of light sources.

17. The defect detection system according to claim 16, further comprising: Multiple light curtains are disposed between the multiple light sources and configured to block the light between the light sources.

18. The defect detection system according to claim 1, wherein, The computing device is also configured to: Determine the digital file associated with the transparent 3D object; Determine a first contour associated with the transparent 3D object from the digital file; Determine the second outline of the transparent 3D object from the plurality of images; Compare the first contour with the second contour; The difference index between the first contour and the second contour is determined based on the comparison results; Determine whether the difference indicator exceeds the difference threshold; as well as In response to determining that the difference index exceeds the difference threshold, the transparent 3D printed object is determined to contain defects.

19. The defect detection system according to claim 1, wherein, The computing device is also configured to perform the following operations on each of the plurality of images: Edge detection is performed on the image to determine the boundaries of the transparent 3D object in the image; Select the set of points on the boundary; The point set is used to determine a region of interest, wherein the region of interest includes a first region in the image that depicts the transparent 3D object within the boundary; and The image is cropped to exclude a second region in the image outside the region of interest, wherein the cropped image is processed by the computing device using a trained artificial intelligence (AI) model to identify defects.

20. The defect detection system according to claim 1, wherein, The plurality of light sources are configured to provide a uniform brightness distribution during the generation of the plurality of images.

21. The defect detection system according to claim 1, further comprising: A plurality of light scattering plates, each of the plurality of light scattering plates being placed adjacent to a light source among the plurality of light sources and configured to scatter light from the light source adjacent to the light scattering plate.

22. A method for performing automated quality control on transparent 3D objects, comprising: Multiple light sources are used to illuminate the transparent 3D object; Multiple images of the transparent 3D object are generated simultaneously using multiple cameras, wherein each of the multiple images depicts a different region or viewpoint of the transparent 3D object; The plurality of images are processed by a computing device to identify defects in the transparent 3D object; and Without requiring user input, the computing device determines, based on the results of the processing, whether the transparent 3D object includes one or more defects.

23. The method according to claim 22, wherein, The transparent 3D object includes polymer orthodontic appliances used by patients during the treatment phase of an orthodontic treatment plan.

24. The method of claim 22, further comprising performing the following operations on each of the plurality of images: Edge detection is performed on the image to determine the boundaries of the transparent 3D printed object in the image; Select the set of points on the boundary; The region of interest is determined using the aforementioned point set, wherein... The region of interest includes a first region in the image that depicts the transparent 3D object within the boundary; as well as The image is cropped to exclude a second region in the image outside the region of interest, wherein the cropped image is processed by the computing device using an artificial intelligence (AI) model.

25. The method according to claim 22, wherein, The multiple images are processed using an artificial intelligence (AI) model, wherein the output of the AI ​​model also includes an indication of the severity of the detected defects.

26. The method according to claim 22, wherein, The transparent 3D object is placed on a transparent platform, wherein the plurality of light sources are positioned below the transparent platform and illuminate the transparent 3D object through the transparent platform, and wherein the plurality of cameras are positioned above the plane of the transparent platform, surrounding the transparent platform and at an angle relative to the transparent platform.

27. The method of claim 26, further comprising: Illuminate the transparent 3D object using an additional light source whose axis is approximately perpendicular to the transparent platform; During the illumination of the transparent 3D object by the additional light source, an additional image of the transparent 3D object is captured using an additional camera, which is positioned above the transparent platform and has an imaging axis that is substantially perpendicular to the transparent platform; as well as The additional image is processed to: a) determine the part number of the transparent 3D object and b) determine whether the transparent 3D object includes defects.

28. The method according to claim 22, wherein, The multiple images are processed using a trained artificial intelligence (AI) model, wherein the output of the trained AI model includes the probability that the transparent 3D object contains defects.

29. The method of claim 22, further comprising: The light arriving at the plurality of cameras is filtered using a spectral filter of each of the plurality of cameras, wherein the spectral filter is configured to filter out light outside the wavelength range output by the plurality of light sources.

30. The method according to claim 29, wherein, The wavelength range corresponds to at least one of infrared or near-infrared light.

31. The method according to claim 29, wherein, The wavelength range is 820-930nm.

32. The method according to claim 22, wherein, The plurality of cameras includes a plurality of side-view cameras, wherein each of the plurality of side-view cameras includes a lens having a focal plane, an image sensor having an image plane, and a tilt shift adapter that causes the focal plane to have a certain angle relative to the image plane.

33. The method according to claim 32, wherein, The plurality of side-view cameras have a first angle relative to the transparent platform supporting the transparent 3D object, and wherein the tilt shift adapter causes a second angle between the focal plane and the image plane, the second angle being a function of the first angle, wherein the tilt shift adapter increases the depth of field of the side-view cameras.

34. The method according to claim 22, wherein, The transparent 3D object is placed on a transparent platform, the transparent platform comprising a dot pattern arranged in a circle around the center of the transparent platform, and the method further includes: Based on the identification of the dot patterns in the plurality of images, at least one of the positions or orientations of the transparent 3D object on the transparent platform is determined.

35. The method of claim 34, further comprising: Based on at least one of the determined orientation or determined position of the transparent 3D object on the transparent platform, a robotic arm picks up the transparent 3D object.

36. The method of claim 34, further comprising: The defect detection system is configured based on the detection of the dot patterns in the images captured by the plurality of cameras and the geometric characteristics of the defect detection system.

37. The method of claim 34, wherein, The plurality of cameras are configured such that the dot pattern surrounds approximately 90% of the area of ​​the plurality of images.

38. The method according to claim 22, wherein, The light refracted by the transparent 3D object makes the surface of the transparent 3D object visible in the plurality of images.

39. The method of claim 22, further comprising: Determine the digital file associated with the transparent 3D object; Determine a first outline associated with the transparent 3D object from the digital file; Determine the second outline of the transparent 3D object from the plurality of images; Compare the first contour with the second contour; The difference index between the first contour and the second contour is determined based on the comparison results; Determine whether the difference indicator exceeds the difference threshold; as well as In response to determining that the difference index exceeds the difference threshold, the transparent 3D object is determined to have a defect.

40. The method of claim 22, further comprising: Perform edge detection on one of the plurality of images to determine the boundaries of the transparent 3D object in the images; Select the set of points on the boundary; The point set is used to determine a region of interest, wherein the region of interest includes a first region in the image that depicts the transparent 3D object within the boundary; and The image is cropped to exclude a second region in the image outside the region of interest, wherein the cropped image is processed by the computing device using a trained artificial intelligence (AI) model to identify defects.

41. The method of claim 22, further comprising: The transparent object is manufactured based on a digital file associated with it before automatic quality control is performed.

Images