Porcelain Denture Color Adjustment Method and System Based on Image Analysis

Abstract

The present invention relates to the technical field of color correction, specifically a method and system for adjusting the color of porcelain-fused-to-metal dentures based on image analysis, including the following steps: identifying teeth, gums and surrounding tissues in the image, using edge detection technology to determine the regional boundaries, and calibrating the regions according to the gray-scale differences. In the present invention, by using edge detection technology to accurately segment teeth from surrounding tissues, precise identification of the tooth region in the image is achieved, providing a clear starting point for color adjustment. Selecting bands with high reflectivity to optimize the quality of image data and ensure the accuracy of subsequent color analysis. Adjusting hue, brightness and saturation in the CIELAB color space, improving the matching degree between the denture color and natural teeth, enhancing visual coherence. Integrating neural network analysis and time series data to predict long-term color changes, and a real-time feedback mechanism automatically adjusts color parameters according to lighting conditions to continuously optimize the visual effect and ensure the naturalness and consistency of colors in different environments.

Description

Technical Field

[0001] The present invention relates to the technical field of color correction, and particularly to a method and system for adjusting the color of porcelain-fused-to-metal dentures based on image analysis. Background Art

[0002] Color correction technology mainly involves methods and systems for adjusting and optimizing colors in image processing to make the color representation of an image more realistic, more in line with specific visual effects, or meet specific output requirements. This technical field is widely applied in digital photography, film production, the printing industry, and any field that requires precise color reproduction. Color correction can be dynamically adjusted according to light sources, environmental variables, and device characteristics, including modifying color balance, adjusting contrast and brightness, and enhancing or suppressing specific colors, etc. These processes help improve the quality of visual presentation and ensure the consistency and accuracy of color output.

[0003] Among them, the method for adjusting the color of porcelain-fused-to-metal dentures refers to using image analysis technology to precisely adjust the color of porcelain-fused-to-metal dentures to ensure that the color of the dentures matches the natural tone of the patient's other teeth. The main use of this technology is in the dental restoration process, through advanced image capture and analysis software, to precisely control the color effect of porcelain materials during the production process, and finally achieve seamless docking with the patient's own tooth color. This not only improves the aesthetics but also enhances the naturalness of the dentures and overall oral health.

[0004] Existing color correction technologies fail to effectively handle color deviations caused by light changes. Especially in the color matching of dental dentures, this limitation results in inconsistent colors between the dentures and natural teeth under different lighting conditions, affecting aesthetics and patient satisfaction. In addition, these technologies lack the ability to predict and dynamically adjust color changes during the use of dentures and cannot adapt to the possible color changes of natural teeth over time, which reduces the durability and naturalness of their effects in long-term applications and limits their actual application effects in dental aesthetics. Summary of the Invention

[0005] The purpose of the present invention is to solve the drawbacks existing in the prior art and propose a method and system for adjusting the color of porcelain-fused-to-metal dentures based on image analysis.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: A method for adjusting the color of porcelain-fused-to-metal dentures based on image analysis, comprising the following steps:

[0007] S1: Identify the teeth, gums, and surrounding tissues in the image, use edge detection technology to determine the regional boundaries, calibrate the regions according to the gray-scale differences, and divide the image into two parts: teeth and non-teeth to obtain the result of the segmented image;

[0008] S2: Perform band screening on the segmented image results, extract the visible light, near-infrared, and ultraviolet band data of the tooth area image, analyze the multi-band reflectance according to the spectral resolution, and select the bands with high reflectance as the preferred band results;

[0009] S3: In the CIELAB color space, adjust the hue, brightness, and saturation of the preferred band results. By analyzing the multi-color channel data one by one, extract the color features within the target wavelength range, and adjust the ratio between the color channels to obtain the color adjustment parameters;

[0010] S4: Through the color adjustment parameters, use a hybrid neural network to analyze the long-term color change trend, combine time series data, compare the color differences before and after adjustment to optimize the model, and generate the dynamic adjustment results;

[0011] S5: Through a real-time feedback mechanism, automatically adjust the color parameters of the dynamic adjustment results according to changes in lighting and monitoring conditions. By comparing the differences between the target color and the actual output color, adjust the color depth and saturation, continuously monitor the actual performance, and generate the color matching results.

[0012] The segmented image results specifically include tooth images and non-tooth images. The preferred band results include visible light bands, near-infrared bands, and ultraviolet bands. The color adjustment parameters specifically refer to hue adjustment parameters, brightness adjustment parameters, and saturation adjustment parameters. The dynamic adjustment results include long-term color change trends and time series data. The color matching results include color depth and saturation.

[0013] As a further solution of the present invention, the steps for obtaining the segmented image results are specifically as follows:

[0014] S111: Identify teeth, gums, and surrounding tissues, identify the regional boundaries by applying edge detection technology, determine the differentiated regions, and obtain the boundary detection results;

[0015] S112: Use the boundary detection results to compare the gray values inside and outside the boundary, adopt the threshold segmentation method to determine the tooth and non-tooth regions, and comprehensively refer to the gray distribution characteristics to obtain the gray segmentation results;

[0016] S113: Based on the gray segmentation results, classify the tooth and non-tooth regions in the image, and use the formula:

[0017]

[0018] Optimize the regional segmentation effect to obtain the segmented image results ;

[0019] Among them, Represents the average gray value of the tooth region, focusing on the brightness and darkness of the tooth region, Represents the average gray value of the non-tooth region, used to balance the contrast between the tooth and the surrounding tissues, Is the display weight of the tooth region, adjusting the contrast between the tooth region and the non-tooth region in the image.

[0020] As a further solution of the present invention, the specific steps for obtaining the preferred band result are as follows:

[0021] S211: Identify and extract the visible light, near-infrared, and ultraviolet band data of the tooth region from the segmented image result, determine the spectral range of each band, and measure the spectral signal intensity of multiple bands to obtain the band data result;

[0022] S212: Based on the band data result, calculate the average reflectance of multiple bands, compare the statistical data of the multiple-band reflectance, and select the band with the highest average reflectance to generate the reflectance comparison result;

[0023] S213: Using the reflectance comparison result, use the formula:

[0024]

[0025] Select the band with the smallest standard deviation as the preferred band result ;

[0026] Among them, Represents the reflectance of the th band, which directly affects the quality of the selected band, Is the weight of the th band, which is inversely proportional to the stability of the band reflectance.

[0027] As a further solution of the present invention, the specific steps for obtaining the color adjustment parameters are as follows:

[0028] S311: From the preferred band result, analyze multiple color channels through the CIELAB color space, calculate the hue, brightness, and saturation data of multiple channels, and comprehensively identify the basic color characteristics of each channel to obtain the color characteristic result;

[0029] S312: Based on the color characteristic result, analyze the color data of each color channel within the target wavelength range, extract key color characteristics, evaluate the adjustment requirements of multiple channels, and generate the color characteristic evaluation result;

[0030] S313: Using the color characteristic evaluation result, use the formula:

[0031]

[0032] Calculate the new ratio between color channels, perform color equalization, and obtain color adjustment parameters ;

[0033] Among them, represents the original parameter of the th color channel, referring to the initial color value of the channel, is the adjustment parameter obtained based on color analysis, which determines the intensity and direction of the adjustment, is the sum of all adjustment parameters, which is used to normalize multi-channel parameters.

[0034] As a further solution of the present invention, the steps for obtaining the dynamic adjustment result are specifically as follows:

[0035] S411: Starting from the color adjustment parameters, configure a neural network model, input long-term color data, and set the network to analyze the color change trend, extract preliminary color trend features from the data, and obtain a color trend feature result;

[0036] S412: Using the color trend feature result, combined with time series analysis technology, evaluate the color change, adjust and optimize the parameters of the neural network model, and generate a model adjustment result;

[0037] S413: Based on the optimized model in the model adjustment result, use the formula:

[0038]

[0039] Calculate and compare the color differences before and after adjustment, and generate a dynamic adjustment result;

[0040] Among them, is the color difference after adjustment, indicating the color change amount after model optimization, is the color difference calculated by the previous model, , The adjustment coefficients obtained through data analysis regulate the sensitivity and adjustment range of the color difference, and match the differential change rate and range.

[0041] As a further solution of the present invention, the steps for obtaining the color matching result are specifically as follows:

[0042] S511: Configure a real-time feedback mechanism, monitor the ambient light and the status of the color display device, collect data on the ambient light and monitoring conditions in real time, and automatically adjust the color parameters of the dynamic adjustment result according to the data to generate an environmental monitoring data result;

[0043] S512: Using the environmental monitoring data results, compare the difference between the target color and the actual output color, calculate the deviation of color depth and saturation, adjust the color parameters to match the current monitoring conditions, and generate a color adjustment result;

[0044] S513: Adopt the color adjustment result and use the formula:

[0045]

[0046] Adjust the color matching to minimize the color difference between the target and the actual output, and generate a color matching result ;

[0047] where, represents the target color value, which is a preset or desired color parameter, represents the actually displayed color value, which is affected by environmental factors and device performance, is an adjustment coefficient, which is calculated based on the color difference monitored in real time and is used to dynamically adjust the color parameters.

[0048] A color adjustment system for porcelain-fused-to-metal dentures based on image analysis, the color adjustment system for porcelain-fused-to-metal dentures based on image analysis is used to execute the above-mentioned color adjustment method for porcelain-fused-to-metal dentures based on image analysis, and the system includes:

[0049] The image segmentation module, based on the input porcelain-fused-to-metal denture image, identifies the teeth, gums and surrounding tissues, uses edge detection to divide the region boundaries, completes region calibration according to the gray-scale difference, divides the image into tooth and non-tooth regions, and outputs the segmented image result;

[0050] The band screening module, based on the segmented image result, screens the visible light, near-infrared and ultraviolet band data of the tooth region, analyzes the multi-band reflectance, selects the band with high reflectance as the processing basis, and outputs the preferred band result;

[0051] The color adjustment module, based on the preferred band result, adjusts the hue, brightness and saturation in the CIELAB color space, analyzes the multi-color channel data, extracts and adjusts the color channel ratio, and outputs the color adjustment parameters;

[0052] The color matching monitoring module, based on the color adjustment parameters, analyzes the long-term color change trend through a hybrid neural network, combines time series data, optimizes the color difference model before and after adjustment, dynamically generates the adjustment result, automatically adjusts the color depth and saturation according to the changes in light and monitoring conditions through a real-time feedback mechanism, continuously monitors the actual performance, and outputs the color matching result.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are:

[0054] In the present invention, by using edge detection technology to accurately segment teeth from surrounding tissues, precise identification of the tooth region in the image is achieved, providing a clear starting point for color adjustment. Selecting bands with high reflectivity optimizes the quality of the image data, ensuring the accuracy of subsequent color analysis. Adjusting hue, brightness, and saturation in the CIELAB color space improves the color matching between the denture and natural teeth, enhancing visual coherence. Integrating neural network analysis and time series data to predict long-term color changes, a real-time feedback mechanism automatically adjusts color parameters according to lighting conditions, continuously optimizing the visual effect and ensuring color naturalness and consistency in different environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a schematic diagram of the workflow of the present invention;

[0056] Figure 2 is a flowchart of the steps for obtaining the segmentation image results of the present invention;

[0057] Figure 3 is a flowchart of the steps for obtaining the preferred band results of the present invention;

[0058] Figure 4 is a flowchart of the steps for obtaining the color adjustment parameters of the present invention;

[0059] Figure 5 is a flowchart of the steps for obtaining the dynamic adjustment results of the present invention;

[0060] Figure 6 is a flowchart of the steps for obtaining the color matching results of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0062] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the accompanying drawings. These are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0063] Embodiment 1

[0064] Please refer to Figure 1 , the present invention provides a technical solution: a method for adjusting the color of porcelain-fused-to-metal dentures based on image analysis, comprising the following steps:

[0065] S1: Identify the teeth, gums and surrounding tissues in the image, use edge detection technology to determine the regional boundaries, calibrate the regions according to the gray-scale differences, divide the image into two parts: teeth and non-teeth, and obtain the segmentation image result;

[0066] S2: Perform band screening on the segmentation image result, extract the visible light, near-infrared and ultraviolet band data of the tooth region image, analyze the multi-band reflectance according to the spectral resolution, and select the band with high reflectance as the preferred band result;

[0067] S3: In the CIELAB color space, adjust the hue, brightness and saturation of the preferred band result. By analyzing the multi-color channel data one by one, extract the color features within the target wavelength range, and adjust the ratio between the color channels to obtain the color adjustment parameters;

[0068] S4: Through the color adjustment parameters, use a hybrid neural network to analyze the long-term color change trend, combine the time series data, compare and optimize the model for the color difference before and after adjustment, and generate a dynamic adjustment result;

[0069] S5: Through a real-time feedback mechanism, automatically adjust the color parameters of the dynamic adjustment result according to the changes in lighting and monitoring conditions. By comparing the target color and the actual output color difference, adjust the color depth and saturation, continuously monitor the actual performance, and generate a color matching result.

[0070] The segmentation image result specifically includes a tooth image and a non-tooth image. The preferred band result includes a visible light band, a near-infrared band, and an ultraviolet band. The color adjustment parameters specifically refer to hue adjustment parameters, brightness adjustment parameters, and saturation adjustment parameters. The dynamic adjustment result includes a long-term color change trend and time series data. The color matching result includes color depth and saturation.

[0071] Please refer to Figure 2 , the specific steps for obtaining the segmentation image result are as follows:

[0072] S111: Identify the teeth, gums and surrounding tissues, identify the regional boundaries by applying edge detection technology, determine the differentiated regions, and obtain the boundary detection result;

[0073] Edge detection technology identifies the boundaries of teeth, gums, and surrounding tissues. By adjusting the detection parameters to match the differential image contrast, it ensures the capture of boundaries. Through image preprocessing steps, including noise removal and grayscale conversion, the image quality is optimized to improve the accuracy of edge detection. This process operates on the image, and the processed image is then processed through an edge detection algorithm. The boundary detection results include the delineation of tooth and non-tooth regions.

[0074] S112: Using the boundary detection results, compare the grayscale values inside and outside the boundary, and adopt the threshold segmentation method to determine the tooth and non-tooth regions. With reference to the grayscale distribution characteristics, obtain the grayscale segmentation result.

[0075] The acquisition of the boundary detection results is based on the data generated in the grayscale segmentation result. The data is used in the threshold segmentation method, classified according to the grayscale value difference between the tooth and non-tooth regions, and the segmentation threshold is determined through statistical analysis of the grayscale values. Adjust the grayscale segmentation parameters to match different types of tooth images. During the segmentation process, call the image processing library to process the grayscale values to effectively distinguish the teeth from the surrounding tissues. The grayscale segmentation result determines the tooth region and the non-tooth region, laying a foundation for the next image processing step.

[0076] S113: According to the grayscale segmentation result, classify the tooth and non-tooth regions in the image, and adopt the formula:

[0077]

[0078] Optimize the region segmentation effect to obtain the segmented image result ;

[0079] Among them, represents the average grayscale value of the tooth region, which focuses on the brightness and darkness of the tooth region, represents the average grayscale value of the non-tooth region, which is used to balance the contrast between the teeth and the surrounding tissues, is the display weight of the tooth region, which adjusts the contrast between the tooth region and the non-tooth region in the image.

[0080] Formula:

[0081]

[0082] The benefit of the formula is that by weighting the grayscale values of the tooth region and the non-tooth region, the contrast between regions is enhanced, the accuracy and visual effect of image segmentation are improved, details can be better displayed at the boundary between the tooth region and the non-tooth region, which is beneficial for subsequent tooth lesion analysis and diagnosis.

[0083] Detailed explanation of the formula and the formula calculation derivation process:

[0084] Set the average gray value of the tooth region to 180, and the average gray value of the non - tooth region is 80. The weight parameter is set to 0.5. The calculation process is as follows:

[0085] 1. Calculate the weighted square of the tooth region: ;

[0086] 2. Calculate the weighted square root of the non - tooth region: ;

[0087] 3. Apply the formula to calculate the gray value: ;

[0088] The results show that: through weight processing, the boundary between the tooth region and the non - tooth region becomes more obvious, and the image segmentation result is 10802.98, which indicates that the visual effect of the tooth region in the image has been enhanced, helping to diagnose and evaluate the dental health status.

[0089] Please refer to Figure 3 , and the specific steps for obtaining the preferred band results are as follows:

[0090] S211: Identify and extract the visible light, near - infrared, and ultraviolet band data of the tooth region from the segmented image results, determine the spectral range of each band, and measure the spectral signal intensity of multiple bands to obtain the band data results;

[0091] In the segmented image results, filter the visible light, near - infrared, and ultraviolet bands, determine the range of each band through wavelength identification technology, and thus measure the spectral signal intensity of multiple bands. This step is based on the technical parameters of the spectral analysis instrument. By adjusting the equipment, optimizing the scanning speed and resolution, ensure the accuracy and repeatability of the data. The data acquisition strictly depends on the band identification algorithm and spectral analysis to obtain the band data results.

[0092] S212: Based on the band data results, calculate the average reflectance of multiple bands, compare the statistical data of the multi - band reflectance, and filter the band with the highest average reflectance to generate the reflectance comparison result;

[0093] Use the obtained band data to calculate the average reflectance of multiple bands for comparison, and filter the band with the highest reflectance as the target band. In this process, through statistical processing of multiple scan data, ensure the accuracy of the results. Compare the statistical data of the multi - band reflectance, and select the band with high reflectance as the research object to provide data support for spectral analysis and material research.

[0094] S213: Using the reflectance comparison result, adopt the formula:

[0095]

[0096] Select the band with the smallest standard deviation as the preferred band result ;

[0097] Among them, represents the reflectance of the th band, which directly affects the quality of the selected band, is the weight of the th band, which is inversely proportional to the stability of the band reflectance.

[0098] Formula:

[0099]

[0100] The advantage of the formula is that by the method of weighted average of the reflectance of each band, the most representative band can be preferably selected. This method is especially suitable for spectral selection in materials science and helps to enhance the accuracy and efficiency of materials analysis.

[0101] Detailed explanation of the formula and the derivation process of formula calculation:

[0102] Set the reflectances of three bands , and their standard deviations are respectively , then the weight . The calculation formula is:

[0103]

[0104] The result shows that the reflectance of the selected preferred band is 0.302, which represents that the band has the optimal average reflection characteristics and is applicable to scientific research and practical applications. This value can be directly used to guide the calibration of the spectrometer and the acquisition of spectral data.

[0105] Please refer to Figure 4 , and the specific steps for obtaining the color adjustment parameters are as follows:

[0106] S311: From the preferred band result, analyze multiple color channels through the CIELAB color space, calculate the hue, brightness and saturation data of multiple channels, comprehensively identify the basic color characteristics of each channel, and obtain the color characteristic result;

[0107] When working in the CIELAB color space, the preferred band results are imported through color analysis, and the hue, brightness, and saturation at different wavelengths are measured and calculated. Based on the measurement data, the initial characteristics of multiple color channels are identified, which involves a large number of numerical operations, including wavelength identification, spectral analysis of colors, and data normalization. The data processing is carried out through algorithms. The algorithm first calculates the average hue and saturation of the color channels according to the spectral data, and then adjusts the color balance according to color theory to ensure that the output color results can truly reflect the true color characteristics of the object being measured. The obtained color characteristic results serve as the basis for color adjustment and simulation.

[0108] S312: Based on the color characteristic results, analyze the color data of each color channel within the target wavelength range, extract key color characteristics, evaluate the adjustment requirements of multiple channels, and generate color characteristic evaluation results;

[0109] Based on the color characteristic results, iteratively analyze the color characteristics within the target wavelength range. Through comparative analysis between color channels, determine the adjustment requirements of each channel. This analysis process includes the deconstruction and feature extraction of the color data of each channel, and uses statistical and image processing techniques to analyze the data, including using principal component analysis and clustering algorithms to identify and separate different color characteristics. These algorithms can effectively extract data from color information for color adjustment. The obtained color characteristic evaluation results will directly affect the adjustment strategy.

[0110] S313: Adopt the color characteristic evaluation results and use the formula:

[0111]

[0112] Calculate the new ratio between color channels to perform color balance and obtain color adjustment parameters ;

[0113] Among them, represents the original parameter of the th color channel, referring to the initial color value of the channel, is the adjustment parameter obtained based on color analysis, which determines the intensity and direction of the adjustment, is the sum of all adjustment parameters, which is used to normalize the multi-channel parameters.

[0114] Formula:

[0115]

[0116] The advantage of the formula is that it improves the flexibility and sensitivity of the influence of multiple color channels on the output during the adjustment process through the method of exponential weighting, allowing for the adjustment of color balance, thus being closer to the true perception of color by the human eye.

[0117] Detailed Explanation of Formulas and Derivation Process of Formula Calculations:

[0118] Set as the initial parameter of the i-th color channel, taken from the results of color characteristic evaluation, as the adjustment coefficient obtained by the channel based on color data analysis, as the sum of all adjustment coefficients. There are three channels, with initial parameters , adjustment coefficient , calculate:

[0119]

[0120] Calculate the new weight of each channel:

[0121]

[0122]

[0123]

[0124] Then the adjusted color parameters are:

[0125]

[0126] The results show that the adjusted color configuration is more balanced. Referring to the perception sensitivity of the human eye to different color channels, the visual effect of the image is effectively improved.

[0127] Please refer to Figure 5 , the specific steps for obtaining the dynamic adjustment results are as follows:

[0128] S411: Starting from the color adjustment parameters, configure the neural network model, input long-term color data, and set the network to analyze the color change trend, extract preliminary color trend features from the data, and obtain the color trend feature results;

[0129] In the process of configuring the neural network model, the key is the setting of the input layer. This requires using the data obtained from the color adjustment parameters as the input. The data is obtained through long-term color monitoring, including RGB values and CIELAB values. The collection of data involves color sensors and time series analysis methods. By continuously monitoring the color changes in different time periods, the collected data will be preprocessed, such as noise removal and data normalization, to match the data input requirements of the neural network, ensuring the quality and accuracy of the data, providing a reliable basis for subsequent trend analysis. The training of the neural network model involves multi-layer perceptrons, and the weights are adjusted cyclically through forward and backward propagation algorithms to predict the color change trend in future time periods.

[0130] S412: Utilize the results of color trend features, combine with time series analysis techniques, evaluate color changes, adjust and optimize the parameters of the neural network model, and generate model adjustment results;

[0131] Based on the results of color trend features, refine color changes through time series analysis techniques, which involve statistical methods including autoregressive models and moving average models. The model can identify and predict seasonal and trend components in color changes. By long-term collecting color data, including daily RGB color values, calculate the color changes at different time points, and then optimize the parameters of the model, including adjusting the number of neurons in the hidden layer and the learning rate. These parameter adjustments are optimized based on the results of cross-validation to reduce overfitting and improve prediction accuracy. The generated color change model will be used for color adjustment and application, and can dynamically adjust color output according to actual changes.

[0132] S413: Based on the optimized model in the model adjustment results, use the formula:

[0133]

[0134] Calculate and compare the color differences before and after adjustment, and generate dynamic adjustment results;

[0135] Among them, is the color difference after adjustment, representing the amount of color change after model optimization, is the color difference calculated by the previous model, 、 The adjustment coefficients obtained through data analysis regulate the sensitivity and adjustment range of color differences, and match the differential change rate and amplitude.

[0136] Formula:

[0137]

[0138] The advantage of the formula is that by taking the logarithm of the color differences before and after, it smooths the excessive changes in color adjustment, reduces the mutations during the adjustment process, and makes the color changes more delicate and natural. By dynamically adjusting to reflect more accurate color changes, the model can make more reasonable predictions based on actual monitoring data.

[0139] Detailed explanation of the formula and the derivation process of formula calculation:

[0140] Set at a target monitoring point, (color change at the previous time point), (adjustment coefficient), (smoothing parameter):

[0141]

[0142] The results show that after referring to the color changes in the previous state, the color difference to be adjusted in the current state is approximately 1.8792. This value will be used in the actual color adjustment to provide a smooth transition of color change output and match the user's visual perception.

[0143] Please refer to Figure 6 , and the steps for obtaining the color matching result are specifically as follows:

[0144] S511: Configure a real-time feedback mechanism to monitor the status of ambient light and color display devices, collect data on ambient light and monitoring conditions in real time, automatically adjust the color parameters of the dynamic adjustment result according to the data, and generate the ambient monitoring data result;

[0145] Monitor the status of ambient light and color display devices, collect light data and device performance data, feedback real-time information to the central processing unit through environmental sensors and interfaces, adjust the output parameters of the display device according to the algorithm of the processing unit to match the environmental conditions, update the color display parameters in real time to match the environmental changes, ensure the matching of the displayed color with the ambient light, enhance the continuity and comfort of the user's visual experience, and generate the ambient monitoring data result.

[0146] S512: Utilize the ambient monitoring data result to compare the difference between the target color and the actual output color, calculate the deviation of color depth and saturation, and adjust the color parameters to match the current monitoring conditions to generate the color adjustment result;

[0147] Utilize the ambient monitoring data to analyze the difference between the target color and the actual output color, calculate the deviation of color depth and saturation through algorithms, adjust the color parameters according to color theory, perform dynamic adjustment of colors, optimize the color display effect to match the current monitoring conditions, and provide real-time feedback on the adjustment effect. Refine the color adjustment strategy in a data-driven manner to ensure a natural transition of the display effect and generate the color adjustment result.

[0148] S513: Adopt the color adjustment result and use the formula:

[0149]

[0150] Adjust the color matching to minimize the color difference between the target and the actual output and generate the color matching result ;

[0151] Among them, represents the target color value, which is a preset or desired color parameter, represents the actually displayed color value, which is affected by environmental factors and device performance, is the adjustment coefficient, which is calculated based on the color difference monitored in real time and is used to dynamically adjust the color parameters.

[0152] Formula:

[0153]

[0154] The benefit of the formula is that through real-time monitoring and feedback mechanism, the color output of the display device can be continuously adjusted to achieve the color display effect closest to the user's expectation.

[0155] Detailed explanation of the formula and the derivation process of formula calculation:

[0156] Set the target color to 255, 255, 255 (white), and the actual output color is 250, 250, 245. When the actual output turns yellow due to light change, set the adjustment coefficient to 0.8. The calculation process is as follows:

[0157]

[0158] Since the color value cannot exceed 255, it is adjusted to 255. Therefore, the white output remains unchanged. Verifying the effectiveness and practicality of the adjustment algorithm, the results show that by dynamically adjusting the color depth and saturation, the expected and actual colors can be matched, optimizing the visual output quality and ensuring the accuracy and matching of color display.

[0159] The porcelain-fused-to-metal denture color adjustment system based on image analysis is used to execute the above-mentioned porcelain-fused-to-metal denture color adjustment method based on image analysis. The system includes:

[0160] The image segmentation module, based on the input porcelain-fused-to-metal denture image, identifies teeth, gums and surrounding tissues, uses edge detection to divide the region boundaries, completes region calibration according to gray-scale differences, distinguishes the image into tooth and non-tooth regions, and outputs the segmented image result;

[0161] The band screening module, based on the segmented image result, screens the visible light, near-infrared and ultraviolet band data of the tooth region, analyzes the multi-band reflectance, selects the band with high reflectance as the processing basis, and outputs the preferred band result;

[0162] The color adjustment module, based on the preferred band result, adjusts the hue, brightness and saturation in the CIELAB color space, analyzes the multi-color channel data, extracts and adjusts the color channel ratio, and outputs the color adjustment parameters;

[0163] Based on color adjustment parameters, the color matching monitoring module analyzes the long-term color change trend through a hybrid neural network, combines time series data, optimizes the color difference model before and after adjustment, dynamically generates adjustment results, automatically adjusts the color depth and saturation according to changes in light and monitoring conditions through a real-time feedback mechanism, continuously monitors the actual performance, and outputs the color matching result.

[0164] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for adjusting the color of porcelain dentures based on image analysis, characterized in that, It includes the following steps: Identify the teeth, gums, and surrounding tissues in the image, use edge detection technology to determine the regional boundaries, calibrate the regions according to the gray-scale differences, divide the image into two parts: teeth and non-teeth, and obtain the segmented image result; Perform band screening on the segmented image result, extract the visible light, near-infrared, and ultraviolet band data of the tooth region image, analyze the multi-band reflectance according to the spectral resolution, and select the bands with high reflectance as the preferred band results; In the CIELAB color space, adjust the hue, brightness, and saturation of the preferred band results, extract the color features within the target wavelength range by analyzing the multi-color channel data one by one, and adjust the ratio between the color channels to obtain the color adjustment parameters; The specific steps for obtaining the color adjustment parameters are as follows: From the preferred band results, analyze the multi-color channels through the CIELAB color space, calculate the hue, brightness, and saturation data of the multi-channels, comprehensively identify the basic color features of each channel, and obtain the color feature results; Based on the color feature results, analyze the color data of each color channel within the target wavelength range, extract the key color characteristics, evaluate the adjustment requirements of the multi-channels, and generate the color characteristic evaluation results; Adopt the color characteristic evaluation results and use the formula: Calculate the new ratio between color channels, perform color equalization, and obtain color adjustment parameters ; Among them, represents the original parameter of the color channel, referring to the initial color value of the channel, is the adjustment parameter obtained based on color analysis, determining the intensity and direction of the adjustment, is the sum of all adjustment parameters, used to normalize multi-channel parameters; Through the color adjustment parameters, the hybrid neural network analyzes the long-term color change trend, combines the time series data, compares the color differences before and after adjustment to optimize the model, and generates the dynamic adjustment results; The specific steps for obtaining the dynamic adjustment results are as follows: Starting from the color adjustment parameters, configure the neural network model, input the long-term color data, and set the network to analyze the color change trend, extract the preliminary color trend features from the data, and obtain the color trend feature results; Utilize the color trend feature results, combine with the time series analysis technology, evaluate the color changes, adjust and optimize the parameters of the neural network model, and generate the model adjustment results; Based on the optimized model in the model adjustment results, use the formula: Calculate and compare the color differences before and after adjustment to generate the dynamic adjustment results; Among them, is the adjusted color difference, representing the amount of color change after model optimization, is the color difference calculated by the previous model, and is the adjustment coefficient obtained through data analysis, which regulates the sensitivity and adjustment range of the color difference, and matches the differential change rate and range; Through the real-time feedback mechanism, automatically adjust the color parameters of the dynamic adjustment results according to the changes in lighting and monitoring conditions, adjust the color depth and saturation by comparing the target color and the actual output color differences, continuously monitor the actual performance, and generate the color matching results.

2. The method for adjusting the color of a porcelain-fused-to-metal denture based on image analysis according to claim 1, wherein The segmented image result specifically includes the tooth image and the non-tooth image. The preferred band results include the visible light band, the near-infrared band, and the ultraviolet band. The color adjustment parameters specifically refer to the hue adjustment parameters, the brightness adjustment parameters, and the saturation adjustment parameters. The dynamic adjustment results include the long-term color change trend and the time series data. The color matching results include the color depth and the saturation.

3. The method for adjusting the color of a porcelain fused to metal denture based on image analysis according to claim 2, wherein, The specific steps for obtaining the segmented image result are as follows: Identify the teeth, gums, and surrounding tissues, identify the regional boundaries by applying edge detection technology, determine the differentiated regions, and obtain the boundary detection results; Using the boundary detection results, compare the gray values inside and outside the boundary, and use the threshold segmentation method to determine the tooth and non-tooth regions. With reference to the gray distribution characteristics, obtain the gray segmentation result; According to the gray segmentation result, classify the tooth and non-tooth regions in the image, using the formula: Optimize the regional segmentation effect to obtain the segmented image result ; Among them, represents the average gray value of the tooth region, which focuses on the brightness and darkness of the tooth region. represents the average gray value of the non-tooth region, which is used to balance the contrast between the tooth and the surrounding tissues. is the display weight of the tooth region, which adjusts the contrast between the tooth region and the non-tooth region in the image.

4. The method for adjusting the color of a porcelain-fused-to-metal denture based on image analysis according to claim 3, wherein The specific steps for obtaining the preferred band result are as follows: Identify and extract the visible light, near-infrared, and ultraviolet band data of the tooth region from the segmented image result, determine the spectral range of each band, and measure the spectral signal intensity of multiple bands to obtain the band data result; Based on the band data result, calculate the average reflectance of multiple bands, compare the statistical data of the multi-band reflectance, select the band with the highest average reflectance, and generate the reflectance comparison result; Using the reflectance comparison result, use the formula: Select the band with the smallest standard deviation as the preferred band result ; Among them, represents the reflectance of the th band, which directly affects the quality of the selected band. is the weight of the th band, which is inversely proportional to the stability of the band reflectance.

5. The method for adjusting the color of a porcelain fused to metal denture based on image analysis according to claim 1, wherein The specific steps for obtaining the color matching result are as follows: Configure a real-time feedback mechanism to monitor the ambient light and the status of the color display device, collect the data of the ambient light and monitoring conditions in real time, and automatically adjust the color parameters of the dynamic adjustment result according to the data to generate the ambient monitoring data result; Using the ambient monitoring data result, compare the difference between the target color and the actual output color, calculate the deviation of the color depth and saturation, adjust the color parameters to match the current monitoring conditions, and generate the color adjustment result; Using the color adjustment result, use the formula: Adjust color matching to minimize the color difference between the target and the actual output, and generate a color matching result ; Among them, represents the target color value, which is a preset or expected color parameter, represents the actually displayed color value, which is affected by environmental factors and device performance, is an adjustment coefficient, which is calculated according to the color difference monitored in real time and is used to dynamically adjust the color parameters.

6. A color adjustment system for porcelain fused to metal dentures based on image analysis, characterized in that According to the method for adjusting the color of a porcelain fused to metal denture based on image analysis according to any one of claims 1-5, the system includes: The image segmentation module is based on the input porcelain fused to metal denture image, identifies teeth, gums and surrounding tissues, uses edge detection to divide the regional boundary, completes regional calibration according to the gray difference, divides the image into tooth and non-tooth regions, and outputs the segmented image result; The band screening module is based on the segmented image result, screens the visible light, near-infrared and ultraviolet band data of the tooth region, analyzes the multi-band reflectance, selects the band with high reflectance as the processing basis, and outputs the preferred band result; The color adjustment module is based on the preferred band result, adjusts the hue, brightness and saturation in the CIELAB color space, analyzes the multi-color channel data, extracts and adjusts the color channel ratio, and outputs the color adjustment parameters; The color matching monitoring module is based on the color adjustment parameters, analyzes the long-term color change trend through a hybrid neural network, combines time series data, optimizes the color difference model before and after adjustment, dynamically generates the adjustment result, automatically adjusts the color depth and saturation according to the changes in light and monitoring conditions through a real-time feedback mechanism, continuously monitors the actual performance, and outputs the color matching result.

Images