Intraoral Scanner Reflection Suppression for Accurate 3D Impressions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing intraoral scanners suffer from unwanted reflections caused by hygienic barriers, such as sterilized tips or sheaths, leading to artifacts in captured images and compromising the accuracy of 3D digital impressions.
Innovation Solution
A 3D scanner system with a window or sleeve in the optical path, combined with a parameterized model, such as a neural network, to suppress or ignore reflections from the window and/or sleeve, allowing accurate determination of pattern features and generation of a more precise 3D model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a hygienic barrier (window or sleeve) is introduced in the optical path to ensure hygiene, then hygiene is improved, but unwanted reflections and artifacts are generated
Solution Approach 1:
The patent trains a parameterized model (neural network) to recognize and suppress the harmful reflections generated by the hygienic barrier, converting the previously harmful reflected light into usable information for accurate 3D modeling. The model learns to distinguish between useful pattern features and harmful artifacts, effectively turning the barrier's unwanted reflections into a manageable condition rather than a deal-breaking defect.
Solution Approach 2:
The parameterized model acts as an intermediary between the captured images containing artifacts and the final 3D model generation process. This computational intermediary processes the raw images, suppresses reflections from the hygienic barrier, and outputs cleaned image data suitable for accurate 3D reconstruction, thereby mediating between the conflicting requirements of hygiene and image quality.
2Reliability
If a hygienic barrier is used to protect the scanner, then hygiene is improved, but measurement precision deteriorates due to artifacts
Solution Approach 1:
The trained parameterized model converts the harmful effect of reflections on measurement precision into a beneficial outcome by automatically suppressing these artifacts. The model learns during training to identify reflection patterns and subtract them from the captured images, thereby recovering the true surface information and maintaining high measurement precision even when using a hygienic barrier.
Solution Approach 2:
The patent changes the parameters of the image processing system by introducing a trained neural network model with specific weights and biases optimized for reflection suppression. This parameter change in the computational domain allows the system to automatically adapt and correct for the optical distortions introduced by the hygienic barrier, maintaining measurement precision.
3Reliability
If a hygienic barrier is introduced, then hygiene is improved, but device complexity increases
Solution Approach 1:
The parameterized model serves as a computational intermediary that handles the complexity of reflection suppression without requiring physical modifications to the scanner hardware. By offloading the complex artifact removal task to a trained neural network, the system maintains a relatively simple physical structure while achieving sophisticated image processing capabilities through software intelligence.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively suppresses artifacts, enabling more accurate 3D modeling by ignoring unwanted reflections, thereby improving the precision of digital impressions.
Implementation Method 1
These reflections are often caused by e.g. specular reflection(s) of light projected by the scanner
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The present disclosure relates to a 3D scanner system comprising an intraoral scanner comprising: an elongated housing comprising a distal end for being inserted into an oral cavity, wherein the housing comprises an aperture in a sidewall of the distal end of the housing; and a window arranged in the aperture of the housing; and/or a sleeve mounted on the outside of the elongated housing; the 3D scanner system further comprising one or more processors operatively connected to the intraoral scanner, said processors configured to receive one or more two-dimensional images comprising a plurality of pattern features, wherein the images comprises one or more artifacts arising from reflections from the window and/or sleeve; and provide the two-dimensional image(s) as input to a parameterized model trained to determine the position of at least a subset of the pattern features in the image(s), while suppressing or ignoring the artifacts.