An intelligent method for color calibration of dental photographs and an automatic color selection system for restorations
By constructing a color difference conversion model and a deep learning model for tooth segmentation, the color difference problem in the process of color transfer of restorations was solved, and the accurate calibration of tooth color was achieved, improving treatment efficiency and patient satisfaction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HOSPITAL OF STOMATOLOGY SUN YAT SEN UNIV
- Filing Date
- 2025-12-17
- Publication Date
- 2026-05-26
AI Technical Summary
In anterior aesthetic restoration treatment, the color transmission process of the restoration in the existing technology is affected by factors such as ambient lighting, differences in camera parameters, and inconsistencies in the color gamut of display devices, resulting in color distortion. This leads to a mismatch between the restoration and the natural tooth color, affecting treatment efficiency and patient satisfaction.
By receiving oral images containing a shade guide and teeth, identifying the shade guide and target teeth, constructing a color difference conversion model, performing color correction on the teeth based on the model, and using image processing algorithms and deep learning models for tooth segmentation to eliminate color differences caused by environmental factors, tooth color calibration is achieved.
It improves the accuracy and reliability of tooth color in photographs, reduces the phenomenon of mismatch between restorations and natural tooth color, and enhances treatment efficiency and patient satisfaction.
Smart Images
Figure CN122089618A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of medical image processing technology, and in particular to an intelligent method for color calibration of dental photographs and an automatic color selection system for restorations. Background Technology
[0002] In anterior aesthetic restorations, achieving a high degree of color matching between the restoration and natural teeth is one of the core clinical goals. The key lies in accurately and completely transmitting the color information of the teeth within the mouth to the dental technician during the fabrication process. Currently, color transmission in clinical practice mainly relies on a combination of visual color comparison by the dentist and digital photography. The dentist determines the tooth color by comparing it to a standard shade guide and records the image of the teeth against the shade guide using intraoral photographs. These photographs are then provided to the restorative technician as the basis for color matching in the fabrication of the restoration.
[0003] However, this traditional method has significant flaws in the entire color transmission chain, often leading to inaccurate final restoration colors. First, the color matching process relies on the dentist's subjective visual judgment, which is easily affected by ambient lighting, visual fatigue, and individual differences in color perception, resulting in poor repeatability. Second, during the photographic recording stage, differences in camera parameter settings (such as white balance and exposure) and lighting conditions directly cause the colors in the captured images to deviate significantly from the true colors. Third, at the technician's end, inconsistent color gamuts and calibration states of display devices cause further discrepancies between the colors seen by the technician and the dentist's intended colors. These factors create multiple uncontrollable distortion points in the color information transmission chain from clinical observation to image recording, ultimately forcing technicians to rely solely on experience to infer colors. This results in restorations that do not match the natural tooth color, requiring multiple reworks and severely impacting treatment efficiency and patient satisfaction. Summary of the Invention
[0004] To overcome the defects of environmental factors causing color deviation in photographic images described in the prior art, this invention provides an intelligent method for color calibration of dental photographs and an automatic color selection system for restorations.
[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: An intelligent method for color calibration of dental photographs includes the following steps: Receive an oral cavity image containing a shade guide and teeth; identify the shade guide and target teeth in the oral cavity image; Obtain the distorted color value of the colorimeter; construct a color difference conversion model based on the distorted color value and the pre-stored standard color value of the colorimeter; The target teeth in the oral cavity image are color corrected based on the color difference conversion model to obtain the corrected image.
[0006] As a preferred embodiment, the step of identifying the shade guide and target tooth in the oral cavity image includes: locating the shade guide region using an image processing algorithm, and segmenting the target tooth region using a tooth segmentation model.
[0007] As a preferred embodiment, the step of segmenting the region of the target tooth using a tooth segmentation model includes: The oral cavity image is processed using a semantic segmentation model to obtain a pixel-level mask for the tooth region; Based on the pixel-level mask of the tooth region, a single tooth instance is separated using an instance segmentation algorithm; The target tooth region is determined from the isolated individual tooth instances.
[0008] As a preferred approach, after determining the target tooth region, the region color statistics are also calculated, and pixels with color deviation greater than a first threshold are removed.
[0009] As a preferred embodiment, the color value is the LAB value, and the step of constructing a color difference conversion model based on the distorted color value and the pre-stored standard color values of the colorimeter includes: Compare the pre-stored standard LAB values of the colorimeter with the LAB values of the colorimeter in the oral cavity image; A color transformation function is fitted based on the comparison results. The color transformation function is used to map the LAB values of the colorimeter in the oral image to the color space of the standard LAB values of the colorimeter.
[0010] As a preferred embodiment, the color transformation function is constructed based on prefitted color transformation parameters, and its expression is as follows: = +
[0011] in,[ L , a , b [ represents the original LAB value of the target tooth,] L’ , a’ , b’ [ represents the LAB value of the target tooth after correction.] α, β, γ The prefitted color transformation parameters, where α, β, and γ are the color transformation parameters for the L, A, and B channels, respectively.
[0012] As a preferred embodiment, the fitting step for the color transformation parameters includes: The luminance L channel and chrominance AB channel were fitted separately. The luminance L channel was transformed by a one-dimensional mapping function, and the color transformation parameters of the L channel were calculated based on the weighted least squares method. After aligning the chroma A and B channels with the luminance L channel, a nonlinear function with a constraint term is used to correct the chroma. The color transformation parameters of the A and B channels are solved with the goal of minimizing the color difference between the mapped chroma AB value and the standard AB value of the colorimeter.
[0013] This invention also proposes an automatic color selection system for restorations, the system comprising: Image input module: used to receive oral images containing a shade guide and teeth; Color chart database module: Used to store pre-stored standard color values for color charts; Color correction module: Performs color correction on oral images containing shade guides and teeth based on an intelligent dental photograph color calibration method; Color selection module: Based on the corrected oral image of the teeth, similarity calculation is performed in the color matching database module to obtain at least one restoration color selection result.
[0014] As a preferred embodiment, the color correction module includes an image extraction module and a color conversion module. The image extraction module includes a tooth segmentation model, which segments the target tooth region. The color conversion module includes a color difference conversion model, which corrects the color of the target tooth in the oral cavity image.
[0015] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an intelligent dental photograph color calibration method as described in the present invention.
[0016] Compared with the prior art, the beneficial effects of the technical solution of the present invention are as follows: This invention captures the color difference effect caused by the environment by comparing a standard color chart and the color chart in the image taken under the same environment. It then corrects the color difference introduced by environmental factors in the tooth area of the image, thereby solving the color difference caused by environmental factors during shooting. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the intelligent dental photograph color calibration method in Example 1; Figure 2 Example image of an oral cavity including a colorimeter and teeth, as described in Example 1; Figure 3 For Example 1 Figure 2 Example image after color correction; Figure 4 This is a schematic diagram of the highlight region division in Example 1; Figure 5 This is an architecture diagram of an automatic color selection system for restorations according to Example 2. Detailed Implementation
[0018] The accompanying drawings are for illustrative purposes only and should not be construed as limiting the scope of this patent. To better illustrate this embodiment, some parts in the accompanying drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions; It will be understood by those skilled in the art that certain well-known structures and their descriptions may be omitted in the accompanying drawings.
[0019] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0020] Example 1 This embodiment proposes an intelligent method for color calibration of dental photographs, such as... Figure 1 The diagram shown is a flowchart of an intelligent dental photograph color calibration method according to this embodiment.
[0021] An intelligent method for color calibration of dental photographs includes: S1. Receive an oral cavity image containing a shade guide and teeth; identify the shade guide and target teeth in the oral cavity image; S2. Obtain the distorted color value of the colorimeter; construct a color difference conversion model based on the distorted color value and the pre-stored standard color value of the colorimeter; S3. Based on the color difference conversion model, perform color correction on the target teeth in the oral cavity image to obtain the corrected image.
[0022] In this embodiment, by receiving an original image containing a standard reference shade guide and the target tooth, the regions of both are identified and separated. Then, by utilizing the difference between the distorted color of the shade guide in the image and its known standard color, a color difference conversion model for that specific shooting scenario is constructed. Finally, this model is applied to correct the color of the target tooth. This application can effectively correct systematic color shifts caused by differences in ambient lighting, camera model, and parameter settings, restoring the tooth color and improving the color accuracy and reliability of dental photographs in applications such as remote diagnosis and digital color matching.
[0023] like Figure 2 The image shown is an example image of an oral cavity containing a colorimeter and teeth.
[0024] like Figure 3 As shown, this is for Figure 2 Example image after color correction.
[0025] It can be seen that, Figure 2 Due to overall lighting issues, the overall color is dark and yellowish, relying on Figure 2Misjudgment can occur during clinical diagnosis; after correction, its original color is restored.
[0026] As a preferred embodiment, the step of identifying the shade guide and target tooth in the oral cavity image includes: locating the shade guide region using an image processing algorithm, and segmenting the target tooth region using a tooth segmentation model.
[0027] In this embodiment, by combining traditional image processing algorithms with a deep learning-based tooth segmentation model, automated and high-precision localization and segmentation of the shade guide and tooth regions are achieved.
[0028] As a preferred embodiment, the step of segmenting the region of the target tooth using a tooth segmentation model includes: The oral cavity image is processed using a semantic segmentation model to obtain a pixel-level mask for the tooth region; Based on the pixel-level mask of the tooth region, a single tooth instance is separated using an instance segmentation algorithm; The target tooth region is determined from the isolated individual tooth instances.
[0029] In this embodiment, the tooth segmentation and target region optimization are refined. First, a two-level segmentation strategy, combining semantic segmentation and instance segmentation, accurately separates individual tooth instances, facilitating calibration for specific teeth. This two-level segmentation strategy improves the segmentation boundary accuracy in cases of closely spaced teeth and prevents color contamination between different tooth instances.
[0030] As an example, this invention uses a deep learning semantic segmentation model (SimpleUNet) to perform pixel-level tooth region segmentation on oral images: the input is preprocessed by RGB conversion and 256×256 normalization, the network adopts an encoder-bottleneck-decoder structure (including skip connections and transposed convolution upsampling) to output a single-channel tooth mask, and then combines the watershed algorithm to perform instance segmentation to obtain an independent pixel-level mask for each tooth.
[0031] As a preferred approach, after determining the target tooth region, the region color statistics are also calculated, and pixels with color deviation greater than a first threshold are removed.
[0032] In this embodiment, by calculating color statistics and removing abnormal pixels, such as highlights, the interference of local extreme values on the overall color representation is eliminated.
[0033] like Figure 4 The diagram shown is a schematic representation of the highlight region division.
[0034] As a preferred embodiment, the color value is the LAB value, and the step of constructing a color difference conversion model based on the distorted color value and the pre-stored standard color values of the colorimeter includes: Compare the pre-stored standard LAB values of the colorimeter with the LAB values of the colorimeter in the oral cavity image; A color transformation function is fitted based on the comparison results. The color transformation function is used to map the LAB values of the colorimeter in the oral image to the color space of the standard LAB values of the colorimeter.
[0035] In an optional embodiment, the color transformation function is constructed based on prefitted color transformation parameters, and its expression is as follows: = +
[0036] in,[ L , a , b [ represents the original LAB value of the target tooth,] L’ , a’ , b’ [ represents the LAB value of the target tooth after correction.] α, β, γ The prefitted color transformation parameters, where α, β, and γ are the color transformation parameters for the L, A, and B channels, respectively.
[0037] Further, optionally, the fitting step for the color transformation parameters includes: The luminance L channel and chrominance AB channel were fitted separately. The luminance L channel was transformed by a one-dimensional mapping function, and the color transformation parameters of the L channel were calculated based on the weighted least squares method. After aligning the chroma A and B channels with the luminance L channel, a nonlinear function with a constraint term is used to correct the chroma. The color transformation parameters of the A and B channels are solved with the goal of minimizing the color difference between the mapped chroma AB value and the standard AB value of the colorimeter.
[0038] In this embodiment, the luminance (L) and chrominance (A, B) channels are modeled separately: a constrained one-dimensional mapping is used for the L channel to ensure the smoothness and rationality of luminance correction and prevent over-adjustment; for the AB channels, a nonlinear function with a constraint term is used for fitting based on luminance correction to capture and correct color distortion relationships.
[0039] Example 2 This embodiment proposes an automatic color selection system for restorations, such as... Figure 5 The diagram shown is an architecture diagram of an automatic color selection system for prostheses. The system includes: Image input module: used to receive oral images containing a shade guide and teeth; Color chart database module: Used to store pre-stored standard color values for color charts; Color correction module: Performs color correction on oral images containing shade guides and teeth based on an intelligent dental photograph color calibration method; Color selection module: Based on the corrected oral image of the teeth, similarity calculation is performed in the color matching database module to obtain at least one restoration color selection result.
[0040] In one optional embodiment, if the color difference similarity of the standard color values of the color chart is greater than a second threshold, the recommendation priority is ranked based on the color difference similarity; otherwise, the recommendation priority is ranked based on the L channel of the LAB value in the standard color of the color chart.
[0041] As an example, the second threshold is 1.8.
[0042] In one optional embodiment, the color correction module includes an image extraction module and a color conversion module. The image extraction module includes a tooth segmentation model, which is used to segment the target tooth region. The color conversion module includes a color difference conversion model, which is used to perform color correction on the target teeth in the oral cavity image.
[0043] Example 3 This embodiment proposes a computer device, including a memory and a processor. The memory stores computer-readable instructions, wherein when the computer-readable instructions are executed by the processor, the processor performs the steps of the intelligent dental photograph color calibration method proposed in Embodiment 1.
[0044] By way of example, the processor may be a central processing unit (CPU), a microprocessor unit (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.
[0045] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
[0046] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.
Claims
1. An intelligent method for color calibration of dental photographs, characterized in that, Includes the following steps: Receive an oral cavity image containing a shade guide and teeth; identify the shade guide and target teeth in the oral cavity image; Obtain the distorted color value of the colorimeter; A color difference conversion model is constructed based on the distorted color values and the pre-stored standard color values of the colorimeter. The target teeth in the oral cavity image are color corrected based on the color difference conversion model to obtain the corrected image.
2. The intelligent dental photograph color calibration method according to claim 1, characterized in that, The step of identifying the shade guide and target tooth in the oral cavity image includes: locating the region of the shade guide using an image processing algorithm, and segmenting the region of the target tooth using a tooth segmentation model.
3. The intelligent dental photograph color calibration method according to claim 2, characterized in that, The step of segmenting the target tooth region using a tooth segmentation model includes: The oral cavity image is processed using a semantic segmentation model to obtain a pixel-level mask for the tooth region; Based on the pixel-level mask of the tooth region, a single tooth instance is separated using an instance segmentation algorithm; The target tooth region is determined from the isolated individual tooth instances.
4. The intelligent dental photograph color calibration method according to claim 3, characterized in that, After determining the target tooth region, the region color statistics are also calculated, and pixels with color deviation greater than the first threshold are removed.
5. The intelligent dental photograph color calibration method according to claim 1, characterized in that, The color value is the LAB value, and the step of constructing a color difference conversion model based on the distorted color value and the pre-stored standard color values of the colorimeter includes: Compare the pre-stored standard LAB values of the colorimeter with the LAB values of the colorimeter in the oral cavity image; A color transformation function is fitted based on the comparison results. The color transformation function is used to map the LAB values of the colorimeter in the oral image to the color space of the standard LAB values of the colorimeter.
6. The intelligent dental photograph color calibration method according to claim 5, characterized in that, The color transformation function is constructed based on prefitted color transformation parameters, and its expression is as follows: = + in,[ L , a , b [ represents the original LAB value of the target tooth,] L’ , a’ , b’ [ represents the LAB value of the target tooth after correction.] α、β、 γ The prefitted color transformation parameters, where α, β, and γ are the color transformation parameters for the L, A, and B channels, respectively.
7. The intelligent dental photograph color calibration method according to claim 6, characterized in that, The fitting steps for the color transformation parameters include: The luminance L channel and chrominance AB channel were fitted separately. The luminance L channel was transformed by a one-dimensional mapping function, and the color transformation parameters of the L channel were calculated based on the weighted least squares method. After aligning the chroma A and B channels with the luminance L channel, a nonlinear function with a constraint term is used to correct the chroma. The color transformation parameters of the A and B channels are solved with the goal of minimizing the color difference between the mapped chroma AB value and the standard AB value of the colorimeter.
8. An automatic color selection system for restorations, characterized in that, The system includes: Image input module: used to receive oral images containing a shade guide and teeth; Color chart database module: Used to store pre-stored standard color values for color charts; Color correction module: Based on the intelligent dental photograph color calibration method according to any one of claims 1 to 7, perform color correction on oral images containing a shade guide and teeth; Color selection module: Based on the corrected oral image of the teeth, the color difference similarity is calculated in the color matching database module, and at least one restoration color number selection result is obtained.
9. The automatic color selection system for restorations according to claim 8, characterized in that, If the color difference similarity of the standard color values in the color chart is greater than the second threshold, the recommendation priority is ranked based on the color difference similarity; otherwise, the recommendation priority is ranked based on the L channel of the LAB value in the standard color of the color chart.
10. The automatic color selection system for restorations according to claim 9, characterized in that, The color correction module includes an image extraction module and a color conversion module. The image extraction module includes a tooth segmentation model, which segments the target tooth region. The color conversion module includes a color difference conversion model, which corrects the color of the target teeth in the oral cavity image.