A care regimen and generation system for a portable personalized cosmetic instrument

Abstract

The application discloses a nursing scheme and a generating system of a portable personalized cosmetic instrument, and belongs to the technical field of portable cosmetic electronic equipment and intelligent nursing. The system comprises a cloud server and at least one portable cosmetic instrument. The cloud server comprises a skin analysis module, a scheme generating module and a data storage module. The portable cosmetic instrument is integrated with a multi-sensor data acquisition module, a nursing execution module and a communication module. According to the method, multi-dimensional physiological data and environmental data of user skin are collected, skin problems are analyzed by an AI model, personalized nursing schemes are generated by combining user personal information and historical records, and the schemes are issued to the instrument for execution and real-time dynamic adjustment of parameters. The application solves the core defects of the existing portable cosmetic instrument nursing scheme, such as fixation, dependence on subjective judgment and lack of dynamic adjustment, improves the skin problem recognition accuracy, improves the nursing efficiency, and is widely applicable to various portable cosmetic instruments such as radio frequency, introduction and light therapy.

Description

Technical Field

[0001] This invention belongs to the field of portable beauty electronic devices and intelligent care technology, specifically relating to a care plan and generation system for a portable personalized beauty instrument. Background Technology

[0002] With the rapid development of the home beauty market, portable beauty devices have become an increasingly popular daily skincare choice for consumers due to their ease of use, affordability, and privacy. However, existing portable beauty devices still suffer from the following unavoidable core technological shortcomings in practical use: First, most portable beauty devices currently available only offer 3-5 fixed settings, meaning all users, regardless of skin type, severity of problems, or tolerance, use the same parameters. However, different users have vastly different skin conditions. For example, oily skin and dry skin have different tolerances to radiofrequency energy, and mild acne and severe acne require different phototherapy doses. Fixed settings lead to poor treatment results for some users, and may even cause adverse reactions such as redness, stinging, and peeling.

[0003] Secondly, users need to assess their own skin problems and select the corresponding intensity level. However, ordinary users lack professional skin knowledge and are prone to making mistakes. For example, they may misdiagnose redness of sensitive skin as rosacea and use excessively high-intensity light therapy, which may further damage the skin barrier; or they may misdiagnose oiliness caused by dehydration as oily skin and over-cleanse, which may worsen the skin problem.

[0004] Third, once the existing instruments are started, they operate according to preset fixed parameters throughout the entire process, while the skin's temperature, moisture content, and impedance value will change significantly during the treatment. For example, during radiofrequency treatment, the skin temperature will gradually rise, and continuing to use the original power may cause burns; during iontophoresis treatment, the skin's moisture content increases and the impedance value decreases, requiring adjustment of the iontophoresis parameters to improve the absorption effect.

[0005] Fourth, most existing instruments lack data upload and storage functions, making it impossible to accumulate users' historical skin data and care records, continuously optimize care plans based on users' skin changes, or provide users with long-term, systematic skin care guidance.

[0006] In view of the above-mentioned shortcomings of existing technologies, there is currently no effective integrated solution in the industry. Therefore, the development of a portable beauty instrument skin care plan generation method and system that can automatically detect skin condition, generate personalized care plans, and dynamically adjust parameters in real time has become an urgent need in the home beauty industry. Summary of the Invention

[0007] In response to the problems raised in the background art above, the purpose of this invention is to provide a portable personalized beauty device care plan and generation system. This invention solves the core defects of existing portable beauty device care plans that are fixed, rely on subjective judgment, and lack dynamic adjustment. It improves the accuracy of skin problem identification, increases the effectiveness of care, and is widely compatible with various portable beauty devices such as radio frequency, iontophoresis, and phototherapy.

[0008] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows: A method for generating a skincare plan using a portable, personalized beauty device includes the following steps: S1. Through the multi-sensor array of the portable beauty device, multi-dimensional physiological data of the user's skin and current environmental data are collected simultaneously; S2. Upload the standardized collected data to the cloud server, process the data using the trained skin analysis model, and identify the type and severity of the user's skin problems. S3. Combining the user's basic personal information, historical nursing records and skin analysis results, the system matches the optimal combination of nursing parameters to generate a personalized nursing plan. S4. Send the personalized care plan to the corresponding portable beauty instrument via wireless communication, and control the instrument to perform the care operation according to the preset steps and parameters of the plan. S5. During the nursing process, skin feedback data is collected in real time, and nursing parameters are dynamically adjusted based on the feedback data until the nursing process is completed.

[0009] Further specifying, the multi-dimensional physiological data in S1 includes at least three of the following: skin moisture content, sebum secretion, pH value, texture roughness, surface temperature, and redness value; the environmental data includes at least one of the following: ambient temperature, relative humidity, and ultraviolet intensity; during data collection, the sensor is kept in vertical contact with the skin, and the collection time is controlled within 2-5 seconds.

[0010] Further specified, the skin analysis model in S2 is a multi-class convolutional neural network model. The input of the skin analysis model is a standardized feature vector, and the output is the probability value and severity score of 1-5 for 8 skin problems, including dryness, oiliness, sensitivity, pigmentation, fine lines, acne, dullness and sagging. The severity scoring rules for levels 1-5 are as follows: Level 1 is no obvious corresponding skin symptoms, Level 2 is mild symptoms, Level 3 is moderate symptoms, Level 4 is severe symptoms, and Level 5 is extremely severe symptoms; the training dataset of the skin analysis model contains at least 500,000 user skin data points labeled by skin color, age, and gender.

[0011] Further defining the personalized care plan in S3, it includes a combination of care modes, single-step care duration, energy output intensity, division of treatment sites, care interval cycle, and subsequent home care recommendations; different skin problems correspond to different care mode priorities, and different degrees of severity of the same skin problem correspond to different energy intensity coefficients and duration coefficients.

[0012] Further specifying, the real-time skin feedback data collected in S5 includes skin surface temperature, impedance value, and redness change value; When the skin temperature exceeds the preset threshold of 42℃, the energy intensity will be automatically reduced by 30% or the treatment will be paused for 30 seconds; when the skin resistance value is lower than the preset threshold, the treatment ingredient delivery time will be extended by 1-2 minutes.

[0013] A portable personalized beauty device skincare plan generation system includes a cloud server and at least one portable beauty device; the cloud server includes a skin analysis module, a plan generation module, and a data storage module; the input end of the skin analysis module is electrically connected to the output end of the data storage module, and the input end of the plan generation module is electrically connected to the output end of the skin analysis module. The portable beauty instrument includes a multi-sensor data acquisition module, an instrument control module, a communication module, a nursing execution module, and an interaction module; the output end of the multi-sensor data acquisition module is electrically connected to the input end of the instrument control module, the input end of the nursing execution module is electrically connected to the output end of the instrument control module, and the communication module is bidirectionally connected to the instrument control module and the cloud server.

[0014] Furthermore, the multi-sensor data acquisition module includes a capacitive moisture sensor, a semiconductor oil sensor, a planar pH sensor, an infrared temperature sensor, and an AC impedance sensor. All sensors are integrated in a ring at the front end of the portable beauty instrument's treatment head, and the acquisition range covers a circular skin area.

[0015] Furthermore, the skin analysis module has a built-in multi-class convolutional neural network inference engine and a skin care knowledge base. The skin care knowledge base stores nursing parameter thresholds, contraindications, and treatment plan templates corresponding to different skin problems. The data storage module is used to encrypt and store the user's personal identity information, historical skin detection data, nursing execution records, and treatment plan version information.

[0016] Further specifying, the nursing execution module includes a radio frequency energy unit, an iontophoresis unit, a semiconductor cooling unit, and an LED phototherapy unit. Each unit is electrically connected to the instrument control module through an independent drive circuit, and its output parameters can be adjusted separately. The instrument control module adopts a low-power microcontroller that supports dual-mode Bluetooth and Wi-Fi wireless communication.

[0017] Furthermore, the interactive module includes a small OLED touch screen and a voice prompt unit; the touch screen is used to display skin analysis results, details of the care plan, and real-time execution progress, and supports users to manually fine-tune care parameters; the voice prompt unit is used for operation guidance and abnormal situation warnings during the care process; the cloud server also supports communication with a mobile APP, and users can view the complete care report and obtain personalized skin care suggestions through the APP.

[0018] The beneficial effects of this invention are: This invention collects real-time skin data from users using multiple sensors and combines this data with AI analysis to generate personalized skincare plans, completely solving the problem of fixed settings on beauty devices. Clinical trials show that it improves the effectiveness of skincare for combination skin conditions and shortens the treatment cycle.

[0019] Automated skin detection lowers the barrier to entry, requiring no professional skin knowledge from the user. The instrument automatically completes skin detection and problem identification with high accuracy, avoiding improper care caused by subjective judgment errors. Ordinary users can also obtain a professional-grade skin care experience.

[0020] Real-time dynamic adjustments maximize nursing safety. During the nursing process, changes in skin condition are monitored in real time, and nursing parameters are automatically adjusted. When abnormalities occur, energy is reduced or nursing is paused in a timely manner, significantly reducing the incidence of adverse reactions.

[0021] This invention establishes a full-cycle skin profile for continuous skincare management. All user skin data and care records are stored on a cloud server, and care plans are continuously optimized through big data analysis, providing users with long-term, systematic skincare guidance and enhancing user engagement. The system architecture and methodology of this invention are flexibly adaptable to different types of portable beauty devices, such as radiofrequency, iontophoresis, phototherapy, and cleansing devices. Only the corresponding care execution modules and parameter templates need to be replaced; there is no need to redesign the entire system. Attached Figure Description

[0022] The present invention can be further illustrated by the non-limiting embodiments given in the accompanying drawings; Figure 1 This is a flowchart illustrating the steps of an embodiment of a portable personalized beauty instrument's care plan and generation system according to the present invention. Figure 2 This is a system module diagram illustrating a portable personalized beauty instrument care plan and generation system embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments. The technical solutions in the embodiments of the present invention 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 the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. The technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention. It should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0024] like Figure 1 As shown, the present invention provides a portable personalized beauty device and its care plan generation system, comprising the following steps: S1. Through the multi-sensor array of the portable beauty device, multi-dimensional physiological data of the user's skin and current environmental data are collected simultaneously; S2. Upload the standardized collected data to the cloud server, process the data using the trained skin analysis model, and identify the type and severity of the user's skin problems. S3. Combining the user's basic personal information, historical nursing records and skin analysis results, the system matches the optimal combination of nursing parameters to generate a personalized nursing plan. S4. Send the personalized care plan to the corresponding portable beauty instrument via wireless communication, and control the instrument to perform the care operation according to the preset steps and parameters of the plan. S5. During the nursing process, skin feedback data is collected in real time, and nursing parameters are dynamically adjusted based on the feedback data until the nursing process is completed.

[0025] The specific implementation process is as follows: First, the portable beauty device's image acquisition component completely captures the original surface image of the user's skin to be treated, ensuring that the image covers the entire area to be treated with light. Then, the original image undergoes standardized preprocessing such as noise reduction and contrast enhancement to eliminate interference factors such as ambient light and skin reflection, accurately extracting core features such as skin texture, color, and blemishes, thereby dividing the skin into at least two sub-regions with distinct features. The device's built-in light parameter knowledge base can intelligently match exclusive combinations of light wavelength, intensity, and duration parameters based on the type of skin problem and the severity of blemishes in each sub-region. Then, a spatial coordinate matching algorithm establishes a one-to-one mapping relationship between each skin sub-region and the device's independent light source, finally driving the corresponding light source module to independently output light according to the matched parameters. This method, based on image recognition and regional light control principles, abandons the traditional uniform irradiation mode across the entire area, achieving differentiated light treatment for different skin regions. It effectively solves the problems of insufficient treatment or excessive light damage caused by single-parameter irradiation, significantly improving the accuracy and adaptability of light-based beauty treatments.

[0026] In the practical application of this embodiment, the multi-dimensional physiological data in S1 includes at least three of the following: skin moisture content, sebum secretion, pH value, texture roughness, surface temperature, and redness value; the environmental data includes at least one of the following: ambient temperature, relative humidity, and ultraviolet intensity; during data collection, the sensor is kept in vertical contact with the skin, and the collection time is controlled within 2-5 seconds.

[0027] The specific implementation process is as follows: After preprocessing the original image, the standardized image is input into a pre-trained and converged U-Net deep learning semantic segmentation model. This model is trained on a massive labeled skin dataset and has pixel-level feature classification capabilities. The model can identify image content pixel by pixel, accurately distinguishing different skin feature categories such as normal skin, acne, pigmentation, rosacea, and wrinkles, and outputting full-pixel feature classification results. Subsequently, through a connected component aggregation algorithm, consecutive pixels with the same skin features in the image are integrated into independent skin sub-regions, while accurately labeling the boundary coordinates, coverage area, and location information of each sub-region. Its core principle is to rely on the high-precision feature extraction capabilities of deep learning semantic segmentation to achieve refined, pixel-level partitioning of skin problems, avoiding problems such as blurred partitions, overlapping partitions, and missing regions caused by manual partitioning. This effectively ensures the accuracy of subsequent illumination parameter matching and light source mapping, providing a reliable data foundation for differentiated phototherapy based on different regions.

[0028] In the practical application of this embodiment, the skin analysis model in S2 is a multi-class convolutional neural network model. The input of the skin analysis model is a standardized feature vector, and the output is the probability value and severity score of 1-5 for 8 skin problems, including dryness, oiliness, sensitivity, pigmentation, fine lines, acne, dullness, and sagging. The severity scoring rules for levels 1-5 are as follows: Level 1 is no obvious corresponding skin symptoms, Level 2 is mild symptoms, Level 3 is moderate symptoms, Level 4 is severe symptoms, and Level 5 is extremely severe symptoms; the training dataset of the skin analysis model contains at least 500,000 user skin data points labeled by skin color, age, and gender.

[0029] In practice, the equipment uses a pre-set standardized set of light parameters, including five adjustable parameters: wavelength, intensity, duration, pulse frequency, and duty cycle. These parameters can be flexibly combined and matched according to treatment needs. The system pre-establishes binding rules between features and parameters, configuring dedicated basic light parameter sets for different skin problems such as acne, pigmentation, and fine lines, adapting to the treatment mechanisms of different blemishes. Simultaneously, for the same type of skin problem, gradient parameter correction coefficients are set according to the severity of the blemish; mild blemishes use low-dose, gentle parameters, while severe blemishes use enhanced treatment parameters. The principle is to combine the phototherapy mechanisms of different skin problems to achieve differentiated and precise matching of parameters, avoiding the defects of fixed parameters and poor adaptability. Through graded and categorized parameter matching, the treatment effect of severe skin problems can be guaranteed while protecting areas with minor blemishes and normal skin, significantly improving the effectiveness and safety of phototherapy.

[0030] In the practical application of this embodiment, the personalized care plan in S3 includes a combination of care modes, single-step care duration, energy output intensity, division of treatment sites, care interval cycle, and subsequent home care recommendations; different skin problems correspond to different care mode priorities, and different degrees of severity of the same skin problem correspond to different energy intensity coefficients and duration coefficients.

[0031] In practice, the system pre-inputs the physical layout coordinates and effective irradiation range of all independent light source modules of the beauty device, constructing a device light source coordinate database. After completing the skin sub-region division and coordinate calibration, a coordinate transformation algorithm is used to accurately convert the position and boundary parameters of the skin sub-region in the image pixel coordinate system into the device physical coordinate system parameters, achieving spatial alignment between the image region and the device light source. Then, an optimal coverage matching strategy is employed to select the minimum number of light source modules that can completely cover a single skin sub-region, completing a one-to-one or many-to-one precise mapping and binding. Its core principle is to rely on dual coordinate system transformation and intelligent matching algorithms to eliminate the deviation between the image space and the device physical space, achieving precise correspondence between the light source and the treatment area. This effectively solves problems such as misalignment of traditional light sources, incomplete area coverage, and light redundancy, ensuring that each skin sub-region can receive precise and complete targeted light therapy.

[0032] In the practical application of this embodiment, the real-time skin feedback data collected in S5 includes skin surface temperature, impedance value, and redness change value; When the skin temperature exceeds the preset threshold of 42℃, the energy intensity will be automatically reduced by 30% or the treatment will be paused for 30 seconds; when the skin resistance value is lower than the preset threshold, the treatment ingredient delivery time will be extended by 1-2 minutes.

[0033] In practice, the device is pre-set to a fixed time interval, continuously re-acquiring real-time skin images during phototherapy to dynamically monitor changes in skin condition in each sub-region. The system compares the initial skin condition with the real-time data in real time. If skin blemishes are detected to have faded or the condition has improved, the system automatically lowers the light parameters for the corresponding area to avoid overtreatment. If abnormal light reactions such as skin redness, abnormally high body temperature, or mild edema are detected, a safety control mechanism is immediately triggered, rapidly reducing the light source intensity in the corresponding area or pausing local irradiation, and simultaneously issuing an audible and visual warning. Its principle is to construct a closed-loop control logic of real-time monitoring, condition comparison, and dynamic parameter adjustment, breaking the rigid mode of traditional fixed-parameter irradiation throughout the treatment. It can adapt to the dynamic changes in skin condition during treatment in real time, promptly avoiding safety risks such as overheating and skin burns, while dynamically optimizing treatment parameters to balance treatment effectiveness and safety.

[0034] A portable personalized beauty device skincare plan generation system includes a cloud server and at least one portable beauty device; the cloud server includes a skin analysis module, a plan generation module, and a data storage module; the input end of the skin analysis module is electrically connected to the output end of the data storage module, and the input end of the plan generation module is electrically connected to the output end of the skin analysis module. The portable beauty instrument includes a multi-sensor data acquisition module, an instrument control module, a communication module, a nursing execution module, and an interaction module; the output end of the multi-sensor data acquisition module is electrically connected to the input end of the instrument control module, the input end of the nursing execution module is electrically connected to the output end of the instrument control module, and the communication module is bidirectionally connected to the instrument control module and the cloud server.

[0035] The cloud server is configured with 2 cores and 4GB of memory, running a Linux operating system. It includes built-in skin analysis, treatment plan generation, and data storage modules. The skin analysis module runs an optimized TensorFlow Lite inference engine, loading a pre-trained multi-class convolutional neural network model; the treatment plan generation module uses a rule engine for parameter matching; and the data storage module uses a MySQL database to encrypt and store user data.

[0036] Specifically, this system consists of a cloud-based or local main control unit that links various functional modules of a portable beauty instrument. These modules are electrically connected and operate collaboratively according to a fixed logic. During operation, the image acquisition module acts as the data input, acquiring raw skin images and transmitting them to the skin region recognition module. After completing skin zoning and feature analysis, the skin region recognition module transmits the data to the parameter determination module to match personalized illumination parameters. The parameter data is then transmitted to the light source control module after the region and light source are bound together by the light source mapping module. Finally, the light source control module drives each independent light source module to output differentiated illumination. Each module performs its specific function, data is transmitted unidirectionally and in an orderly manner, and a closed-loop linkage is established, constructing a fully intelligent control system from image acquisition, intelligent analysis, parameter matching to precise light control. This modular architecture features clear division of labor, low coupling, and strong operational stability, enabling highly efficient fully automated regional differentiated light therapy and significantly reducing reliance on manual operation.

[0037] In the practical application of this embodiment, the multi-sensor data acquisition module includes a capacitive moisture sensor, a semiconductor oil sensor, a planar pH sensor, an infrared temperature sensor, and an AC impedance sensor. All sensors are integrated in a ring at the front end of the portable beauty instrument's care head, and the acquisition range covers a circular skin area.

[0038] In practical implementation, the system is equipped with dual acquisition units: a visible light camera and an infrared camera, simultaneously acquiring dual-modal images. The visible light camera accurately captures surface features such as color, texture, and blemish morphology to identify visible skin problems like pigmentation, acne, and fine lines. The infrared camera captures skin temperature distribution and subcutaneous microcirculation thermal imaging features to identify latent skin conditions such as redness, sensitivity, and inflammation. The system then uses an image registration and fusion algorithm to superimpose and calibrate the two sets of image data, combining the morphological features of visible light and the temperature characteristics of infrared light for comprehensive judgment. The principle behind this is to leverage multimodal image fusion technology to overcome the limitations of single-image recognition in terms of limited features and accuracy, achieving comprehensive identification of both visible and latent skin problems. This effectively improves the accuracy of skin region segmentation and skin problem identification, providing more comprehensive and accurate data support for subsequent personalized lighting parameter configuration.

[0039] In the practical application of this embodiment, the skin analysis module has a built-in multi-class convolutional neural network inference engine and a skin care knowledge base. The skin care knowledge base stores the nursing parameter thresholds, contraindications and solution templates corresponding to different skin problems. The data storage module is used to encrypt and store the user's personal identity information, historical skin detection data, nursing execution records and solution version information.

[0040] Specifically, the skin analysis module is composed of an image preprocessing unit, a deep learning inference unit, and a region aggregation unit working together. During operation, the image preprocessing unit first performs denoising, cropping, alignment, and standardization on the original bimodal image, removing invalid interference information and standardizing the image input specifications. Then, the deep learning inference unit calls the built-in optimized U-Net semantic segmentation model, relying on the feature logic accumulated from a massive labeled training dataset to complete pixel-level skin feature classification inference. Finally, the region aggregation unit uses a connected component analysis algorithm to integrate similar pixels and calibrate region parameters to complete the final skin sub-region division. Its principle is to achieve a standardized process of image purification, intelligent inference, and region integration through three-level unit hierarchical processing. This progressive processing logic effectively reduces recognition errors, avoids environmental interference, ensures the refinement and standardization of skin partitioning, and significantly improves the overall system's intelligent recognition capabilities.

[0041] In the practical application of this embodiment, the nursing execution module includes a radio frequency energy unit, an iontophoresis unit, a semiconductor cooling unit, and an LED phototherapy unit. Each unit is electrically connected to the instrument control module through an independent drive circuit, and its output parameters can be adjusted separately. The instrument control module adopts a low-power microcontroller that supports dual-mode wireless communication of Bluetooth and Wi-Fi.

[0042] Specifically, the device's light source modules adopt a matrix or honeycomb arrangement to adapt to the contours of the human face, ensuring that the light adheres closely to the skin surface. Each independent light source module integrates multiple wavelength LED chips, corresponding to blue, green, and red light wavelengths specifically designed for beauty, catering to different beauty needs such as acne treatment, blemish removal, and anti-aging. Each wavelength LED chip is equipped with an independent driving circuit, enabling independent start / stop of a single chip, continuous adjustment of light intensity, and independent control of different light source modules without interference. Its core principle is a multi-wavelength light source integration + single-area independent light control architecture, breaking through the limitations of traditional beauty light sources that output the same parameters across the entire area. It allows for free combination of light wavelengths and adjustment of light intensity according to the treatment needs of different skin sub-areas, precisely matching the treatment mechanisms of various skin problems, achieving refined light-based beauty treatments by zone, wavelength, and intensity, adapting to multiple beauty care scenarios.

[0043] In the practical application of this embodiment, the interaction module includes a small OLED touch screen and a voice prompt unit; the touch screen is used to display skin analysis results, nursing plan details and real-time execution progress, and supports users to manually fine-tune nursing parameters; the voice prompt unit is used for operation guidance and abnormal situation warnings during the nursing process; the cloud server also supports communication with a mobile APP, and users can view the complete nursing report and obtain personalized skin care suggestions through the APP.

[0044] Specifically, the skin condition monitoring module works in conjunction with the image acquisition and recognition unit throughout the entire process, continuously collecting real-time data on changes in skin temperature, color, and texture. This data is then synchronously fed back to the parameter determination module, providing data support for dynamic parameter adjustment and enabling real-time closed-loop optimization of the treatment process. The human-computer interaction module visualizes skin recognition results, customized lighting schemes, and real-time treatment progress through a touch-screen interface. It also allows users to manually fine-tune parameters such as light intensity and treatment duration to suit individual skin tolerance. The voice prompt unit provides operation guidance, abnormal warnings, and announcements. The system can also be linked to a mobile app to store care reports, push skincare suggestions, and track historical data. The principle is to ensure treatment safety through the monitoring module and enhance operational flexibility through the interaction module, allowing the device to combine fully automated intelligent care with manual customization capabilities to meet the personalized needs of different users.

[0045] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for generating a skincare plan using a portable personalized beauty device, characterized in that, Includes the following steps: S1. Through the multi-sensor array of the portable beauty device, multi-dimensional physiological data of the user's skin and current environmental data are collected simultaneously; S2. Upload the standardized collected data to the cloud server, process the data using the trained skin analysis model, and identify the type and severity of the user's skin problems. S3. Combining the user's basic personal information, historical nursing records and skin analysis results, the system matches the optimal combination of nursing parameters to generate a personalized nursing plan. S4. Send the personalized care plan to the corresponding portable beauty instrument via wireless communication, and control the instrument to perform the care operation according to the preset steps and parameters of the plan. S5. During the nursing process, skin feedback data is collected in real time, and nursing parameters are dynamically adjusted based on the feedback data until the nursing process is completed.

2. The method for generating a care plan for a portable personalized beauty instrument according to claim 1, characterized in that: The multidimensional physiological data in S1 includes at least three of the following: skin moisture content, sebum secretion, pH value, texture roughness, surface temperature, and redness value. The environmental data includes at least one of the following: ambient temperature, relative humidity, and ultraviolet intensity. During data collection, the sensor should be kept in vertical contact with the skin, and the collection time should be controlled within 2-5 seconds.

3. The method for generating a care plan for a portable personalized beauty instrument according to claim 1, characterized in that: The skin analysis model in S2 is a multi-class convolutional neural network model. The input of the skin analysis model is a standardized feature vector, and the output is the probability value and severity score of 1-5 for 8 skin problems, including dryness, oiliness, sensitivity, pigmentation, fine lines, acne, dullness and sagging. The severity scoring rules for levels 1-5 are as follows: Level 1 is no obvious corresponding skin symptoms, Level 2 is mild symptoms, Level 3 is moderate symptoms, Level 4 is severe symptoms, and Level 5 is extremely severe symptoms; the training dataset of the skin analysis model contains at least 500,000 user skin data points labeled by skin color, age, and gender.

4. The method for generating a care plan for a portable personalized beauty instrument according to claim 1, characterized in that: The personalized care plan in S3 includes a combination of care modes, single-step care duration, energy output intensity, division of treatment sites, care interval cycle, and subsequent home care recommendations; different skin problems correspond to different care mode priorities, and different degrees of severity of the same skin problem correspond to different energy intensity coefficients and duration coefficients.

5. The method for generating a care plan for a portable personalized beauty instrument according to claim 1, characterized in that: The real-time skin feedback data collected in S5 includes skin surface temperature, impedance value, and redness change value. When the skin temperature exceeds the preset threshold of 42℃, the energy intensity will be automatically reduced by 30% or the treatment will be paused for 30 seconds; when the skin resistance value is lower than the preset threshold, the treatment ingredient delivery time will be extended by 1-2 minutes.

6. A portable personalized beauty device skincare plan generation system, characterized in that: Includes a cloud server and at least one portable beauty device; The cloud server includes a skin analysis module, a solution generation module, and a data storage module; the input terminal of the skin analysis module is electrically connected to the output terminal of the data storage module, and the input terminal of the solution generation module is electrically connected to the output terminal of the skin analysis module. The portable beauty instrument includes a multi-sensor data acquisition module, an instrument control module, a communication module, a nursing execution module, and an interaction module; the output end of the multi-sensor data acquisition module is electrically connected to the input end of the instrument control module, the input end of the nursing execution module is electrically connected to the output end of the instrument control module, and the communication module is bidirectionally connected to the instrument control module and the cloud server.

7. The portable personalized beauty instrument care plan generation system according to claim 6, characterized in that: The multi-sensor data acquisition module includes a capacitive moisture sensor, a semiconductor oil sensor, a planar pH sensor, an infrared temperature sensor, and an AC impedance sensor. All sensors are integrated in a ring at the front end of the portable beauty instrument's treatment head, and the acquisition range covers a circular skin area.

8. The system for generating a care plan for a portable personalized beauty instrument according to claim 6, characterized in that: The skin analysis module has a built-in multi-class convolutional neural network inference engine and a skin care knowledge base. The skin care knowledge base stores the threshold values ​​of care parameters, contraindications and solution templates corresponding to different skin problems. The data storage module is used to encrypt and store the user's personal identity information, historical skin detection data, care execution records and solution version information.

9. The system for generating a care plan for a portable personalized beauty instrument according to claim 6, characterized in that: The nursing execution module includes a radio frequency energy unit, an iontophoresis unit, a semiconductor cooling unit, and an LED phototherapy unit. Each unit is electrically connected to the instrument control module through an independent drive circuit, and its output parameters can be adjusted separately. The instrument control module uses a low-power microcontroller that supports dual-mode Bluetooth and Wi-Fi wireless communication.

10. The system for generating a care plan for a portable personalized beauty instrument according to claim 6, characterized in that: The interactive module includes a small OLED touch screen and a voice prompt unit; the touch screen is used to display skin analysis results, nursing plan details and real-time execution progress, and supports users to manually fine-tune nursing parameters; the voice prompt unit is used for operation guidance and abnormal situation warnings during the nursing process; the cloud server also supports communication with a mobile APP, and users can view the complete nursing report and obtain personalized skin care suggestions through the APP.

Images