Intelligent toothbrush camera real-time plaque analysis method and device

Abstract

This specification provides a method and apparatus for real-time plaque analysis using a smart toothbrush camera. The method includes: acquiring raw RGB image sequences of the tooth surface in a user's mouth using a miniature camera integrated into the handle of the smart toothbrush, generating a standardized RGB image sequence; performing precise edge segmentation on the tooth regions in each frame image to generate a set of numbered masks for each tooth, obtaining a set of plaque pixel clusters for each tooth region; retaining plaque clusters that persist for more than a predetermined number of frames and have smooth motion trajectories as real plaque regions; calculating the plaque coverage and plaque distribution density of each tooth; mapping the plaque coverage and distribution density of each tooth to a standardized three-dimensional oral cavity model, assigning color gradients according to the coverage range, and superimposing dot density to represent the degree of plaque aggregation, generating a three-dimensional visualized plaque distribution map; determining the risk level of each tooth, and generating a personalized suggestion report containing specific cleaning operation instructions.

Description

Technical Field

[0001] This document relates to the field of artificial intelligence technology, and in particular to a method and device for real-time plaque analysis using a smart toothbrush camera. Background Technology

[0002] While existing smart toothbrushes have basic cleaning reminders or usage time monitoring functions, they lack the ability to identify oral plaque in real time and accurately. They usually rely on users to manually take photos and upload them or analyze them offline, resulting in plaque detection delays, strong subjectivity, and the inability to provide dynamic feedback during brushing. Furthermore, most systems do not integrate miniature cameras and edge computing capabilities, making it difficult to achieve high-precision real-time plaque image acquisition and intelligent analysis under low power consumption conditions, which seriously affects the timeliness and effectiveness of oral health interventions. Summary of the Invention

[0003] The purpose of this invention is to provide a method and device for real-time plaque analysis using a smart toothbrush camera, aiming to solve the aforementioned problems in the prior art.

[0004] This invention provides a method for real-time plaque analysis using a smart toothbrush camera, comprising: A miniature camera integrated into the handle of a smart toothbrush is used to acquire raw RGB image sequences of the surface of teeth in the user's mouth at a predetermined frequency, and simultaneously record the device's posture sensor data. The raw RGB image sequences are then denoised, light compensated, and distortion corrected to generate a standardized RGB image sequence. Based on a preset 3D oral anatomy template, the tooth region in each frame of the image is precisely segmented at the edges to generate a set of numbered masks for each tooth. For each tooth mask region in each frame of the image, the plaque sensitivity index of each pixel is calculated. Pixel sets with plaque sensitivity indices greater than the threshold are selected and cluster analysis is performed to remove artifacts, thus obtaining a set of plaque pixel clusters for each tooth region. Based on attitude sensor data, rigid transformation compensation is performed on adjacent frame images. Cross-frame centroid matching is performed on plaque pixel clusters in the same numbered tooth region. The Hungarian algorithm is used to establish the tracking trajectory. Only plaque clusters that persist for more than a predetermined number of frames and have smooth motion trajectories are retained as real plaque regions. Based on the actual plaque area and the tooth mask, the plaque coverage and plaque distribution density of each tooth are calculated; the plaque coverage and distribution density of each tooth are mapped to a standardized three-dimensional oral model, a color gradient is assigned according to the coverage range, and a dot matrix density is superimposed to represent the degree of plaque aggregation, generating a three-dimensional visualized plaque distribution map. Based on preset dental plaque assessment standards, the risk level of each tooth is determined, high-risk areas with plaque coverage exceeding a preset area threshold and distribution density exceeding a predetermined density threshold are identified, a personalized suggestion report containing specific cleaning operation instructions is generated, and the personalized suggestion report and three-dimensional plaque distribution map are pushed to the user terminal and stored in the local historical database.

[0005] This invention provides a smart toothbrush camera real-time plaque analysis device, comprising: The acquisition module is used to acquire raw RGB image sequences of the tooth surface in the user's mouth at a predetermined frequency through a miniature camera integrated in the smart toothbrush handle, and simultaneously record the device posture sensor data. The raw RGB image sequences are then denoised, light compensated, and distortion corrected to generate a standardized RGB image sequence. The segmentation calculation module is used to perform precise edge segmentation of the tooth region in each frame of the image based on a preset oral anatomy 3D template, generate a set of numbered masks for each tooth, calculate the plaque sensitivity index of each pixel for each tooth mask region in each frame of the image, filter the set of pixels with plaque sensitivity index greater than the threshold, and perform cluster analysis to remove artifacts, thereby obtaining the set of plaque pixel clusters for each tooth region. The tracking module is used to perform rigid transformation compensation on adjacent frame images based on attitude sensor data, perform cross-frame centroid matching on plaque pixel clusters in the same numbered tooth region, establish the tracking trajectory using the Hungarian algorithm, and retain only plaque clusters that have been present for more than a predetermined number of frames and have smooth motion trajectories as real plaque regions. The calculation module is used to calculate the plaque coverage and plaque distribution density of each tooth based on the real plaque area and the tooth mask; map the plaque coverage and distribution density of each tooth to a standardized three-dimensional oral model, assign color gradients according to the coverage range, and superimpose lattice density to represent the degree of plaque aggregation, thereby generating a three-dimensional visualized plaque distribution map. The assessment module is used to determine the risk level of each tooth based on preset dental plaque assessment standards, identify high-risk areas where the plaque coverage is greater than a preset area threshold and the distribution density is greater than a predetermined density threshold, generate a personalized suggestion report containing specific cleaning operation instructions, and push the personalized suggestion report and three-dimensional plaque distribution map to the user terminal and store them in the local historical database.

[0006] This invention also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the above-described real-time plaque analysis method for a smart toothbrush camera.

[0007] This invention also provides a computer-readable storage medium storing an information transmission implementation program, which, when executed by a processor, implements the steps of the above-described real-time plaque analysis method for a smart toothbrush camera.

[0008] By integrating a miniature high-resolution camera and an embedded AI processing unit into the handle of a smart toothbrush, real-time optical acquisition and localized deep learning analysis of plaque on the tooth surface can be achieved during the user's brushing process. Combined with dynamic illumination compensation and adaptive segmentation algorithms for tooth surface areas, a plaque distribution map and cleaning efficiency score can be output instantly. The user can be guided to focus on cleaning key areas through vibration or APP prompts. This achieves real-time plaque management with a closed loop of "brushing-detection-feedback" for the first time, breaking through the timeliness bottleneck of traditional offline analysis mode. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart of the real-time plaque analysis method using a smart toothbrush camera according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the real-time plaque analysis device for a smart toothbrush camera according to an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0011] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0012] Method Implementation Examples According to embodiments of the present invention, a method for real-time plaque analysis using a smart toothbrush camera is provided. Figure 1 This is a flowchart of the real-time plaque analysis method using a smart toothbrush camera according to an embodiment of the present invention, as follows: Figure 1As shown, the real-time plaque analysis method for a smart toothbrush camera according to an embodiment of the present invention specifically includes: Step S101: Using a miniature camera integrated into the smart toothbrush handle, the original RGB image sequence of the user's teeth surface is acquired at a predetermined frequency, and the device posture sensor data is recorded simultaneously. The original RGB image sequence is then denoised, illumination compensated, and distortion corrected to generate a standardized RGB image sequence. The simultaneous recording of the device posture sensor data specifically includes: synchronously recording the device posture sensor data, wherein the posture sensor data includes: six-axis fusion data of the accelerometer and gyroscope, and motion compensation of the image sequence is achieved through a rigid transformation matrix, with a compensation accuracy ≤1.5 pixels.

[0013] Step S102: Based on a preset 3D oral anatomy template, the tooth regions in each frame of the image are precisely segmented to generate a set of numbered masks for each tooth. For each tooth mask region in each frame of the image, the plaque sensitivity index of each pixel is calculated. Pixel sets with plaque sensitivity indices greater than a threshold are selected and cluster analysis is performed to remove artifacts, obtaining a set of plaque pixel clusters for each tooth region. Specifically, calculating the plaque sensitivity index of each pixel and selecting pixel sets with plaque sensitivity indices greater than a threshold includes: Calculate the plaque sensitivity index (BSI) for each pixel as R / (G+B), and filter the set of pixels with a plaque sensitivity index greater than 1.2.

[0014] Step S103: Based on the attitude sensor data, perform rigid transformation compensation on adjacent frame images, perform cross-frame centroid matching on plaque pixel clusters within the same numbered tooth region, and establish a tracking trajectory using the Hungarian algorithm. Only plaque clusters that persist for more than a predetermined number of frames and have smooth motion trajectories are retained as real plaque regions. Specifically, establishing the tracking trajectory using the Hungarian algorithm includes: The Hungarian algorithm is used to establish the tracking trajectory, where the matching cost function of the Hungarian algorithm is: Cost(i,j) = α·d_p + β·d_a + γ·d_tFormula 1; Where d_p is the Euclidean distance between the centroids, d_a is the rate of change of area, d_t is the time interval, and α=0.5, β=0.3, γ=0.2 are preset weighting coefficients.

[0015] Only plaque clusters that persist for more than a predetermined number of frames and have smooth motion trajectories are retained as real plaque regions. Specifically, these include: Only plaque clusters that persist for more than 3 frames and require a centroid displacement of ≤2 pixels between adjacent frames and an area change rate of ≤25% are retained as real plaque areas to exclude instantaneous reflections and bubble artifacts.

[0016] Step S104: Calculate the plaque coverage and plaque distribution density of each tooth based on the real plaque area and the tooth mask; map the plaque coverage and distribution density of each tooth to a standardized three-dimensional oral model, assign a color gradient according to the coverage range, and superimpose a dot matrix density to represent the degree of plaque aggregation, thereby generating a three-dimensional visualized plaque distribution map. Step S105: Based on preset dental plaque assessment standards, the risk level of each tooth is determined, high-risk areas with plaque coverage exceeding a preset area threshold and plaque distribution density exceeding a predetermined density threshold are identified. A personalized suggestion report containing specific cleaning operation instructions is generated, and the personalized suggestion report and a three-dimensional plaque distribution map are pushed to the user terminal and stored in the local historical database. The plaque distribution density is the number of actual plaque clusters per square millimeter of tooth surface, calculated as: density = Σ(number of plaque clusters) / tooth mask area (mm²), where the tooth mask area is converted using a pixel-to-millimeter calibration coefficient, which is obtained by a calibration plate under standard lighting. The high-risk areas with plaque coverage exceeding the preset area threshold and distribution density exceeding the predetermined density threshold specifically include: High-risk areas are those with plaque coverage >40% and a distribution density >5 clusters / mm², located at the gingival margin, interproximal space, or occlusal surface of posterior teeth.

[0017] This invention integrates a miniature high-resolution camera and an embedded AI processing unit within the handle of a smart toothbrush, enabling real-time optical acquisition and localized deep learning analysis of plaque on the tooth surface during brushing. Combined with dynamic illumination compensation and adaptive segmentation algorithms for tooth surface areas, it can instantly output plaque distribution maps and cleaning efficiency scores. Furthermore, it guides users to focus on cleaning key areas through vibration or app prompts, thus achieving, for the first time, real-time plaque management through a closed-loop "brushing-detection-feedback" process, breaking through the timeliness bottleneck of traditional offline analysis modes.

[0018] The technical solutions described above in the embodiments of the present invention will be explained in detail below.

[0019] Step 1: After the smart toothbrush is turned on, the camera captures a sequence of high-definition images of the surface of the teeth inside the user's mouth. Specifically: The smart toothbrush's built-in camera captures a sequence of raw RGB images (containing multiple consecutive frames, each frame showing the complete dental arch area); the camera continuously acquires images at a frequency of 30 frames per second, and the built-in image sensor converts the light signal into digital pixel data. Simultaneously, the main control unit records the timestamp of each frame and data from the device's posture sensor (for subsequent motion compensation); the output is a sequence of raw RGB images, each frame with a resolution of at least 1920×1080 pixels, containing the complete upper and lower jaw teeth area and surrounding soft tissue background. Step 2: Perform noise reduction and illumination compensation preprocessing on the original RGB image sequence.

[0020] The system takes as input the original RGB image sequence and the current light intensity value collected by the device's built-in ambient light sensor. A Gaussian filter is applied to each frame of the image to eliminate sensor noise. The global brightness gain of each frame of the image is dynamically adjusted based on the ambient light sensor data, and histogram equalization is used to enhance the contrast between the tooth surface and plaque. Subsequently, radial distortion correction is performed on the image based on pre-calibrated camera distortion parameters. The output is a normalized RGB image sequence after denoising and illumination compensation. The brightness distribution in the tooth area of ​​the image is uniform, background interference is suppressed, and pixel values ​​are comparable across frames.

[0021] Step 3: Extract the tooth surface region from the standardized RGB image sequence and segment the tooth boundaries.

[0022] The system takes a standardized RGB image sequence and a pre-stored 3D template of oral anatomy (including a mapping of standard dental arch morphology and tooth numbers) as input. It then uses color thresholding and edge detection algorithms to initially locate the tooth surface region. Based on the constraints of the 3D template, it uses template matching and an active contour model to accurately fit the edge contour of each tooth. Each tooth in each frame of the image is numbered and labeled to generate a corresponding tooth region mask. The system outputs a set of precise boundary masks for each tooth in each frame of the image, with each mask corresponding to a uniquely numbered tooth region. Background and non-tooth surface regions are masked.

[0023] Step 4: Extract the spectral feature vector of the plaque staining area based on the tooth surface region mask.

[0024] Input a standardized RGB image sequence and a set of tooth region masks for each frame. For each numbered tooth region in each frame, extract the RGB three-channel values ​​of all pixels within its mask. Calculate the plaque sensitivity index (BSI = R / (G+B)) for each pixel in the red-green-blue space, and filter out the set of pixels with BSI values ​​higher than a preset threshold (1.2). Perform cluster analysis on these pixels to remove artifacts (such as water stains and reflections) and retain continuous regions that conform to the spectral distribution characteristics of plaque. Output a set of plaque pixel clusters corresponding to each tooth region in each frame, where each cluster contains pixel coordinates, mean BSI, area, and color variance.

[0025] Step 5: Perform cross-frame spatiotemporal correlation and motion stabilization tracking on the bacterial plaque pixel clusters in consecutive frames.

[0026] The system takes as input a set of plaque pixel clusters in consecutive frames and synchronously recorded device attitude sensor data (accelerometer and gyroscope). Based on the device attitude data, it performs rigid transformation compensation on each frame to eliminate image shift caused by user hand tremors. For plaque pixel clusters within the same numbered tooth region in adjacent frames, it calculates the centroid distance and area change rate. It uses the Hungarian algorithm for cross-frame matching to establish a stable tracking trajectory. Plaque clusters that persist for more than 3 frames and have smooth motion trajectories are marked as "real plaque regions," while the rest are marked as noise or transient artifacts and removed. The output is a list of stable real plaque regions on each tooth, with each region including its coordinates, area, mean BSI, number of consecutive frames, and timestamp in image space.

[0027] Step 6: Based on the actual plaque area data, calculate the plaque coverage and distribution density of each tooth.

[0028] Input a list of real plaque regions after stable tracking and a precise boundary mask for each tooth. For each tooth, calculate the total pixel area of ​​all real plaque regions within its mask. Divide this area by the total area of ​​the tooth mask to obtain the plaque coverage rate (%). Simultaneously, count the number of plaque clusters per square millimeter and calculate the plaque distribution density (clusters / mm²). Generate an independent plaque quantification index for each tooth according to its number. Output the plaque coverage rate and plaque distribution density values ​​for each tooth, for a total of 16 sets (8 teeth each for the upper and lower jaws).

[0029] Step 7: Map the plaque quantification index to the oral cavity 3D model to generate a visualized plaque distribution map.

[0030] Input the plaque coverage and distribution density of each tooth, and a pre-stored standardized 3D oral model (including tooth number and spatial location mapping). Based on the tooth number, map the plaque coverage and distribution density values ​​of each tooth to the corresponding tooth surface in the 3D model. Assign different color gradients (green→yellow→orange→red) according to the coverage range (0–20%, 20–40%, 40–60%, >60%). Overlay the distribution density information to represent the degree of plaque aggregation with lattice density. Generate a 3D plaque distribution map containing color coding and heatmap. Output a 3D visualized plaque distribution map, in which the severity of plaque for each tooth is represented by both color and density, and interactive rotation and zoom are supported.

[0031] Step 8: Generate a personalized cleaning recommendation report based on the plaque distribution map and preset dental assessment criteria.

[0032] The system inputs a 3D visualized plaque distribution map and preset dental plaque assessment criteria (including plaque coverage thresholds and cleaning priority rules for each tooth area). Based on these criteria, it classifies the plaque coverage of each tooth into low / medium / high risk levels. It identifies high-risk areas with plaque coverage >40% and a density >5 clusters / mm². It analyzes frequently missed areas based on the user's past usage records. Based on the location of high-risk areas, it generates targeted cleaning suggestions (e.g., "Severe plaque buildup on the left lower molar; it is recommended to use dental floss and interdental brushes for focused cleaning"). It then generates a total score and improvement suggestions. Finally, it outputs a personalized cleaning suggestion report, including the risk level of each tooth, the location of high-risk areas, improvement operation guidelines, the total score (0–100 points), and the recommended usage time.

[0033] Step 9: Simultaneously push the cleaning recommendation report and plaque distribution map to the user's mobile app and save them to the local historical database.

[0034] Input personalized cleaning suggestion report, 3D visualized plaque distribution map, device unique identifier, and current timestamp; encapsulate the report text and map image into a structured JSON data packet; send the data packet to the user's bound mobile app via Bluetooth connection; simultaneously write the data packet to the historical analysis record database in the smart toothbrush's local non-volatile storage and archive it by time; the mobile app interface displays real-time cleaning suggestions and plaque distribution map, and a new complete analysis record containing time, device ID, report content, and map is added to the local database for user traceability and long-term trend analysis.

[0035] Device Example 1 According to an embodiment of the present invention, a real-time plaque analysis device using a smart toothbrush camera is provided. Figure 2 This is a schematic diagram of the real-time plaque analysis device for a smart toothbrush camera according to an embodiment of the present invention, as shown below. Figure 2 As shown, the real-time plaque analysis device for a smart toothbrush camera according to an embodiment of the present invention specifically includes: The acquisition module 20 is used to acquire the original RGB image sequence of the tooth surface in the user's mouth at a predetermined frequency through a miniature camera integrated in the smart toothbrush handle, and simultaneously record the device posture sensor data, and perform noise reduction, illumination compensation and distortion correction on the original RGB image sequence to generate a standardized RGB image sequence. The segmentation calculation module 22 is used to perform precise edge segmentation of the tooth region in each frame image based on a preset oral anatomy 3D template, generate a set of numbered masks for each tooth, calculate the plaque sensitivity index of each pixel for each tooth mask region in each frame image, filter the set of pixels with plaque sensitivity index greater than the threshold, and perform cluster analysis to remove artifacts, thereby obtaining a set of plaque pixel clusters for each tooth region. Tracking module 24 is used to perform rigid transformation compensation on adjacent frame images based on attitude sensor data, perform cross-frame centroid matching on plaque pixel clusters in the same numbered tooth region, establish tracking trajectory using Hungarian algorithm, and retain only plaque clusters that have been present for more than a predetermined number of frames and have smooth motion trajectory as real plaque regions. The calculation module 26 is used to calculate the plaque coverage and plaque distribution density of each tooth based on the real plaque area and the tooth mask; map the plaque coverage and distribution density of each tooth to a standardized three-dimensional oral model, assign a color gradient according to the coverage range, and superimpose the dot density to represent the degree of plaque aggregation, thereby generating a three-dimensional visualized plaque distribution map. The assessment module 28 is used to determine the risk level of each tooth based on the preset dental plaque assessment standards, identify high-risk areas where the plaque coverage is greater than a preset area threshold and the distribution density is greater than a predetermined density threshold, generate a personalized suggestion report containing specific cleaning operation instructions, and push the personalized suggestion report and the three-dimensional plaque distribution map to the user terminal and store them in the local historical database.

[0036] The embodiments of the present invention are device embodiments corresponding to the above method embodiments. The specific operation of each module can be understood with reference to the description of the method embodiments, and will not be repeated here.

[0037] Device Example 2 This invention provides an electronic device, such as... Figure 3 As shown, it includes: a memory 30, a processor 32, and a computer program stored in the memory 30 and executable on the processor 32, wherein the computer program, when executed by the processor 32, performs the steps as described in the method embodiment.

[0038] Device Example 3 This invention provides a computer-readable storage medium storing an information transmission implementation program, which, when executed by a processor 32, performs the steps described in the method embodiment.

[0039] The computer-readable storage media described in this embodiment include, but are not limited to, ROM, RAM, disk, or optical disk.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for real-time plaque analysis using a smart toothbrush camera, characterized in that, include: A miniature camera integrated into the handle of a smart toothbrush is used to acquire raw RGB image sequences of the surface of teeth in the user's mouth at a predetermined frequency, and simultaneously record the device's posture sensor data. The raw RGB image sequences are then denoised, light compensated, and distortion corrected to generate a standardized RGB image sequence. Based on a preset 3D oral anatomy template, the tooth region in each frame of the image is precisely segmented at the edges to generate a set of numbered masks for each tooth. For each tooth mask region in each frame of the image, the plaque sensitivity index of each pixel is calculated. Pixel sets with plaque sensitivity indices greater than the threshold are selected and cluster analysis is performed to remove artifacts, thus obtaining a set of plaque pixel clusters for each tooth region. Based on attitude sensor data, rigid transformation compensation is performed on adjacent frame images. Cross-frame centroid matching is performed on plaque pixel clusters in the same numbered tooth region. The Hungarian algorithm is used to establish the tracking trajectory. Only plaque clusters that persist for more than a predetermined number of frames and have smooth motion trajectories are retained as real plaque regions. Based on the actual plaque area and the tooth mask, the plaque coverage and plaque distribution density of each tooth are calculated; the plaque coverage and distribution density of each tooth are mapped to a standardized three-dimensional oral model, a color gradient is assigned according to the coverage range, and a dot matrix density is superimposed to represent the degree of plaque aggregation, generating a three-dimensional visualized plaque distribution map. Based on preset dental plaque assessment standards, the risk level of each tooth is determined, high-risk areas with plaque coverage exceeding a preset area threshold and distribution density exceeding a predetermined density threshold are identified, a personalized suggestion report containing specific cleaning operation instructions is generated, and the personalized suggestion report and three-dimensional plaque distribution map are pushed to the user terminal and stored in the local historical database.

2. The method according to claim 1, characterized in that, Calculating the plaque sensitivity index for each pixel and selecting the set of pixels with a plaque sensitivity index greater than a threshold specifically includes: Calculate the plaque sensitivity index (BSI) for each pixel as R / (G+B), and filter the set of pixels with a plaque sensitivity index greater than 1.

2.

3. The method according to claim 2, characterized in that, The specific steps involved in establishing the tracking trajectory using the Hungarian algorithm are as follows: The Hungarian algorithm is used to establish the tracking trajectory, where the matching cost function of the Hungarian algorithm is: Cost(i,j) = α·d_p + β·d_a + γ·d_tFormula 1; Where d_p is the Euclidean distance between the centroids, d_a is the rate of change of area, d_t is the time interval, and α=0.5, β=0.3, γ=0.2 are preset weighting coefficients.

4. The method according to claim 1, characterized in that, The synchronous recording of device attitude sensor data specifically includes: synchronously recording device attitude sensor data, wherein the attitude sensor data includes: six-axis fusion data of accelerometer and gyroscope, and motion compensation of image sequence is achieved through rigid transformation matrix, with a compensation accuracy of ≤1.5 pixels.

5. The method according to claim 1, characterized in that, Only plaque clusters that persist for more than a predetermined number of frames and have smooth motion trajectories are retained as real plaque regions. Specifically, these include: Only plaque clusters that persist for more than 3 frames and require a centroid displacement of ≤2 pixels between adjacent frames and an area change rate of ≤25% are retained as real plaque areas to exclude instantaneous reflections and bubble artifacts.

6. The method according to claim 1, characterized in that, The plaque distribution density is the number of actual plaque clusters per square millimeter of tooth surface, calculated as: density = Σ(number of plaque clusters) / tooth mask area (mm²), where the tooth mask area is converted by a pixel-to-millimeter calibration coefficient, which is obtained by a calibration plate under standard illumination.

7. The method according to claim 1, characterized in that, High-risk areas identified by both plaque coverage exceeding a preset area threshold and distribution density exceeding a preset density threshold include: High-risk areas are those with plaque coverage >40% and a distribution density >5 clusters / mm², located at the gingival margin, interproximal space, or occlusal surface of posterior teeth.

8. A real-time plaque analysis device for a smart toothbrush camera, installed in a smart toothbrush, characterized in that, include: The acquisition module is used to acquire raw RGB image sequences of the tooth surface in the user's mouth at a predetermined frequency through a miniature camera integrated in the smart toothbrush handle, and simultaneously record the device posture sensor data. The raw RGB image sequences are then denoised, light compensated, and distortion corrected to generate a standardized RGB image sequence. The segmentation calculation module is used to perform precise edge segmentation of the tooth region in each frame of the image based on a preset oral anatomy 3D template, generate a set of numbered masks for each tooth, calculate the plaque sensitivity index of each pixel for each tooth mask region in each frame of the image, filter the set of pixels with plaque sensitivity index greater than the threshold, and perform cluster analysis to remove artifacts, thereby obtaining the set of plaque pixel clusters for each tooth region. The tracking module is used to perform rigid transformation compensation on adjacent frame images based on attitude sensor data, perform cross-frame centroid matching on plaque pixel clusters in the same numbered tooth region, establish the tracking trajectory using the Hungarian algorithm, and retain only plaque clusters that have been present for more than a predetermined number of frames and have smooth motion trajectories as real plaque regions. The calculation module is used to calculate the plaque coverage and plaque distribution density of each tooth based on the real plaque area and the tooth mask; map the plaque coverage and distribution density of each tooth to a standardized three-dimensional oral model, assign color gradients according to the coverage range, and superimpose lattice density to represent the degree of plaque aggregation, thereby generating a three-dimensional visualized plaque distribution map. The assessment module is used to determine the risk level of each tooth based on preset dental plaque assessment standards, identify high-risk areas where the plaque coverage is greater than a preset area threshold and the distribution density is greater than a predetermined density threshold, generate a personalized suggestion report containing specific cleaning operation instructions, and push the personalized suggestion report and three-dimensional plaque distribution map to the user terminal and store them in the local historical database.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the real-time plaque analysis method for a smart toothbrush camera as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an information transmission implementation program, which, when executed by a processor, implements the steps of the real-time plaque analysis method for a smart toothbrush camera as described in any one of claims 1 to 7.

Images