A hair iron with a scalp health detection system and a detection method

Abstract

The application discloses a hair perming device with a scalp health detection system and a detection method. Through an image processing system integrating a camera array and a processor, quantitative detection and objective analysis of key indicators of the scalp (such as hair grease, dandruff, redness, etc.) can be performed before, during and after hair perming, which completely changes the traditional subjective and rough judgment mode relying on naked eyes and experience and provides scientific data basis. Moreover, through real-time monitoring, the sub-health or risk state (such as inflammation, sensitivity, damage, etc.) of the scalp can be identified in time, and when high risk is detected, the system can automatically issue an alarm or intervention, thereby effectively preventing safety problems such as contact dermatitis, hair follicle damage and aggravated hair loss caused by improper hair perming operation.

Description

Technical Field

[0001] This invention relates to the technical field of scalp health detection, specifically to a hair perming device and detection method with a scalp health detection system. Background Technology

[0002] Perming is a widely used hair styling technique that primarily uses chemical agents and physical heat to alter the keratin structure of the hair, thereby achieving a long-lasting curl or style. Traditional perming equipment mainly focuses on technical aspects such as temperature control, even application of chemicals, and curl shaping, with the design goal of optimizing perming results and ease of operation. However, the entire perming process—including the chemical effects of the chemicals, heat stimulation, and mechanical pulling—can potentially impact the scalp microenvironment and hair follicle health.

[0003] Currently available hair perming devices and styling equipment generally lack the ability to provide real-time, objective assessment of the scalp and hair's basic condition. Operators typically rely solely on visual observation and experience to judge scalp health, which is not only highly subjective but also makes it difficult to accurately identify early or subtle pathological changes. For example, if pre-existing issues such as imbalanced sebum secretion, keratin buildup, folliculitis, or redness and sensitivity on the scalp are not identified before perming, forcibly performing a chemical perm may exacerbate scalp sensitivity, induce contact dermatitis, and even lead to follicle damage and hair loss. Summary of the Invention

[0004] This application provides a hair perming device and its detection method with a scalp health detection system, which solves the problem mentioned in the background art that existing hair perming devices cannot identify early or subtle pathological changes in the scalp before perming, resulting in irreversible damage to the scalp and hair follicles that are already in a sub-healthy state.

[0005] To address the aforementioned technical problems, in a first aspect, this application provides a hair perming device with a scalp health detection system, comprising a hair perming device body, which is a straightener or curling iron with a comb structure. The hair perming device body includes a plurality of cameras, a processor, a storage module, and a power supply module. The plurality of cameras are disposed in the comb structure area and are used to capture raw images of different locations on the human scalp. The processor is used to analyze and process the real-time images captured by the cameras and generate a scalp health analysis report. The storage module is used to store the scalp health analysis report. The power supply module is used to supply power to the cameras, the processor, and the storage module.

[0006] In one embodiment, the hair perming device with a scalp health detection system further includes a wireless connection module and an APP that can be installed on a handheld smart device. The wireless connection module is used to connect to the handheld smart device and send a scalp health analysis report generated by the processor so that the user can view it through the APP.

[0007] In one embodiment, the processor includes: The image preprocessing module is used to standardize the raw images captured by cameras from different light sources and output preprocessed images. The multispectral image fusion module performs precise registration and fusion of preprocessed images to generate a fused image; The feature extraction and quantization module is used to identify at least one feature data from the fused image and output a structured multidimensional feature vector. The health status analysis and diagnosis module is used to compare the structured multidimensional feature vector with a preset threshold to generate a scalp health score or risk warning information. The real-time control and feedback module is used to generate adjustment instructions and safety alarms for the operating parameters of the hair perming device based on the health score or risk warning information.

[0008] In one embodiment, the multispectral image fusion module includes: The image input interface unit receives and buffers multiple preprocessed images from the image preprocessing module, and is responsible for the unification and synchronization of data formats to ensure that subsequent units receive time-aligned image frames. The feature point extraction and matching unit extracts stable image feature points for each input multi-channel preprocessed image with different light sources, performs feature point matching between images with different light sources, finds the correspondence of the same scalp area under different imaging conditions, and outputs matching point pairs. The spatial transformation and registration unit, based on the matching point pairs output by the feature point matching unit, transforms all images under different light sources to the same coordinate system and outputs a registered multispectral image. The feature layer fusion unit performs fusion processing on the registered multispectral images to generate a fused image.

[0009] In one embodiment, the feature extraction and quantization module includes: The dandruff detection unit is used to identify dandruff areas in the fused image and calculate the percentage of the dandruff area to the total area of ​​the detection area as a quantitative indicator of dandruff coverage. The sebum analysis unit is used to receive the fused image from the multispectral image fusion module, identify the bright areas caused by sebum fluorescence through an image segmentation algorithm, and calculate the percentage of the area of ​​the bright areas to the total area of ​​the detection area as a quantitative indicator of sebum coverage. The redness sensitivity analysis unit is used to convert the fused image from the RGB color space and calculate the erythema index map of the fused image; based on the erythema index map, it identifies areas with index values ​​exceeding a preset threshold as effective erythema areas, and outputs the area ratio of the effective erythema areas and the average erythema index of the effective erythema areas as quantitative indicators of redness sensitivity. The data integration and standardization unit is used to receive quantitative data from the keratin and dandruff detection unit, sebum analysis unit, and redness sensitivity analysis unit, and to integrate, standardize units, and unify formats to generate structured multidimensional feature vectors. And an output interface unit, used to output the structured multidimensional feature vector to the health status analysis and diagnosis module.

[0010] In one embodiment, the erythema index is calculated as follows: EI=k*[log(R)-log(G)], where R and G are the red and green channel values ​​of the pixel in the RGB color space, respectively, and k is the calibration coefficient.

[0011] In one embodiment, the data integration and standardization unit includes: A multi-channel data receiving and buffering subunit is used to receive raw quantitative data from the keratin and dandruff detection unit, the sebum analysis unit, and the redness sensitivity analysis unit in parallel, and to perform synchronous buffering. The data cleaning and validity verification subunit is connected to the multi-channel data receiving and caching subunit. It is used to perform range verification and logical consistency checks on the raw quantized data, mark or process abnormal data values, and generate verified data with validity status markers. The unit conversion and normalization processing subunit is connected to the data cleaning and validity verification subunit. The unit conversion and normalization processing subunit has an index conversion configuration table pre-stored inside, which is used to convert the units of the verified data with validity status markers into predetermined standardized data according to the index conversion configuration table. The structured feature vector assembly subunit is connected to the unit conversion and normalization processing subunit. It is used to fill the standardized data after unit conversion into the corresponding dimensions according to the predefined vector template, and assemble and generate a structured multidimensional feature vector. And a data packet encapsulation and output buffer subunit, connected to the structured feature vector assembly subunit, for encapsulating the multidimensional feature vector and associated metadata into a standard format data packet and outputting it to the output interface unit.

[0012] In one embodiment, the health status analysis and diagnosis module includes: The feature vector input and parsing unit is used to receive the structured multidimensional feature vector from the feature extraction and quantization module, and parse the various quantization indicators contained therein into independent variables; The multidimensional indicator database and knowledge base unit are used to store the preset thresholds corresponding to various scalp health indicators, the association rules between different indicator combinations and scalp symptoms, and the weight configuration of each indicator in the comprehensive score. The single-indicator state evaluation unit is connected to the feature vector input and parsing unit and the multi-dimensional indicator database and knowledge base unit. It is used to compare the measured values ​​of each quantitative indicator with their corresponding preset thresholds to generate an independent state evaluation result for each indicator. A multi-indicator comprehensive reasoning and diagnosis unit is connected to the feature vector input and parsing unit, the multi-dimensional indicator database and knowledge base unit, and the single indicator state evaluation unit. It is used to generate a comprehensive scalp health score and at least one preliminary diagnostic conclusion based on the quantitative indicators, the independent state evaluation results, the association rules, and the weight configuration through a preset reasoning model. The risk warning and priority determination unit is connected to the multi-indicator comprehensive reasoning and diagnosis unit, and is used to determine the current scalp health risk level and generate corresponding warning information based on the comprehensive health score and / or the preliminary diagnosis conclusion. The system also includes a report generation and formatting unit, which is connected to the single-indicator status assessment unit, the multi-indicator comprehensive reasoning and diagnosis unit, and the risk warning and priority determination unit. This unit integrates the independent status assessment results, the comprehensive health score, the preliminary diagnostic conclusion, and the warning information to generate a standard format scalp health analysis report.

[0013] In one embodiment, the real-time control and feedback module includes: The instruction parsing and strategy mapping unit is used to receive health scores, risk levels and early warning information from the health status analysis and diagnosis module, and generate preliminary control strategy instructions based on the pre-stored health status and control strategy mapping relationship. A multi-parameter collaborative control unit, connected to the instruction parsing and strategy mapping unit, is used to generate a collaborative control signal to adjust at least two independent operating parameters of the hair perming device based on the preliminary control strategy instruction; the independent operating parameters include at least two of heating temperature, chemical action time, and mechanical force. The security alarm and status synchronization unit is connected to the instruction parsing and strategy mapping unit, and is used to generate corresponding audible and visual alarm signals according to the risk level, and broadcast the current security control status to other modules of the system. The control command encapsulation and output interface unit is connected to the multi-parameter collaborative control unit and the safety alarm and status synchronization unit. It is used to encapsulate the collaborative control signal and the alarm signal into drive commands that conform to the communication protocol of the hair perming device actuator, and output them to the corresponding actuators and alarm devices.

[0014] Secondly, this application also provides a scalp health detection method, applied to a hair perming device with a scalp health detection system, the method comprising: By using several cameras placed in different locations, images of different parts of the human scalp are captured to obtain raw images of the scalp area. The processor receives and processes the original image, identifies and analyzes the scalp features in the image, and generates a scalp health analysis report containing quantitative indicators. The scalp health analysis report is stored in the storage module to form a queryable scalp health analysis report.

[0015] The beneficial effects of the aforementioned hair perming device and detection method with a scalp health detection system are as follows: Through an image processing system integrating a camera array and a processor, key scalp indicators (such as hair oil, dandruff, and redness) can be quantitatively detected and objectively analyzed before, during, and after perming. This completely changes the traditional subjective and rough judgment mode that relies on visual observation and experience, providing scientific data support. Furthermore, through real-time monitoring, it can promptly identify sub-healthy or risky states of the scalp (such as inflammation, sensitivity, and damage). When a high risk is detected, the system can automatically issue an alarm or intervene, thereby effectively preventing safety issues such as contact dermatitis, hair follicle damage, and aggravated hair loss caused by improper perming operations. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of a hair perming device with a scalp health detection system shown in an embodiment of this application; Figure 2 for Figure 1 This application embodiment shows a schematic diagram of the processor of a hair perming device with a scalp health detection system; Figure 3 for Figure 1 The flowchart of the scalp health detection method shown in the embodiments of this application is as follows. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0018] It should be noted that when a component is considered to be "connected" to another component, it can be directly connected to the other component or there may be an intermediary component present. Conversely, when a component is said to be "directly" connected to another component, there is no intermediary component.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0020] like Figure 1 As shown, this application provides a hair perming device with a scalp health detection system, including a hair perming device body, which is a straightener or curling iron with a comb structure. The hair perming device body includes a plurality of cameras 1, a processor 2, a storage module 3, and a power supply module 4. The plurality of cameras 1 are arranged in the comb structure area and are used to capture raw images of different positions of the human scalp. The processor 2 is used to analyze and process the real-time images captured by the cameras and generate a scalp health analysis report. The storage module 3 is used to store the scalp health analysis report. The power supply module 4 is used to supply power to the cameras, the processor, and the storage module.

[0021] Multiple miniature high-definition cameras (e.g., five, positioned at different angles on the device) on the hair perming equipment are simultaneously or sequentially aimed at different areas of the scalp. These cameras are equipped with special light sources (such as white light, polarized light, or UV light) to illuminate the scalp and highlight different features (such as oil fluorescing under UV light). Each camera captures a high-resolution raw image of the scalp within its field of view. For example, one camera might capture a clear image of the scalp at the crown, magnified dozens of times, clearly showing hair follicles, hair, and any dandruff and oil.

[0022] In one embodiment, the hair perming device with a scalp health detection system further includes a wireless connection module 5 and an APP that can be installed on a handheld smart device. The wireless connection module 5 is used to connect to the handheld smart device and send a scalp health analysis report generated by the processor so that the user can view it through the APP.

[0023] like Figure 2 As shown, in one embodiment, the processor 2 includes: Image preprocessing module 21 is used to standardize the original images captured by cameras from different light sources and output preprocessed images; Suppose the system has three cameras that simultaneously capture images of the same area above a customer's head using white light, polarized light, and UV light, respectively, resulting in three original images.

[0024] Problems with the original images: Image A (white light): The colors are generally normal, but some areas appear too dark due to hair shadows, and there is slight noise overall; Image B (polarized light): Effectively eliminates the shine and reflection from the scalp, but the image edges are slightly distorted due to the lens, and the brightness is uneven (bright in the center and dark around the edges); Image C (UV light): Sebum fluoresces, but the overall image is dark with low contrast, and is slightly blurry due to slight hand tremors. Directly feeding these three "original images" to the analysis algorithm would lead to inconsistent results. At this point, the image preprocessing module 21 begins its work.

[0025] After the above series of automated processes, the image preprocessing module 21 outputs three new images: Image A': a white light scalp image with realistic colors, low noise, and clear details. Image B': a polarized light scalp image with uniform brightness, no distortion, and suppressed surface reflection. Image C': a UV light fluorescent scalp image with high contrast and clear outlines.

[0026] The multispectral image fusion module 22 performs precise registration and fusion of the preprocessed images to generate a fused image; The multispectral image fusion module 22 includes: The image input interface unit receives and buffers multiple preprocessed images from the image preprocessing module, and is responsible for the unification and synchronization of data formats to ensure that subsequent units receive time-aligned image frames. The image input interface unit simultaneously receives three standard images from the image preprocessing module 21: image A' (white light), image B' (polarized light), and image C' (UV light). It checks the timestamps and numbers of these three images to ensure they belong to the same shooting session and the same scalp area, and then temporarily stores them in memory for synchronous processing.

[0027] The feature point extraction and matching unit extracts stable image feature points for each input multi-channel preprocessed image with different light sources, performs feature point matching between images with different light sources, finds the correspondence of the same scalp area under different imaging conditions, and outputs matching point pairs. The feature point extraction and matching unit searches for stable, unique, and easily identifiable landmarks in images A', B', and C', respectively. What are these landmarks? For example: a uniquely shaped hair follicle opening (visible in both A' and B'); a distinctive intersection of several hairs; a tiny pigmented spot or mole (stable in all three images); a prominent sebaceous gland opening (a concave point in A', and possibly a bright spot in C' due to sebum). The feature point extraction and matching unit calculates a feature descriptor for each landmark, incorporating information about the surrounding texture. The unit then compares the fingerprints of these landmarks across the different images. For example, it finds a hair follicle at coordinates (100, 150) in image A' with fingerprint F_A; a hair follicle at coordinates (102, 148) in image B' with fingerprint F_B highly similar to F_A; and a point at coordinates (99, 152) in image C' with fingerprint F_C also highly similar to F_A. The feature point extraction and matching unit determines that these are the same physical points, and thus generates matching point pairs: [(100,150)_A,(102,148)_B,(99,152)_C]. It will find dozens to hundreds of such matching points.

[0028] The spatial transformation and registration unit, based on the matching point pairs output by the feature point matching unit, transforms all images under different light sources to the same coordinate system and outputs a registered multispectral image. Upon receiving the matching point pairs, the spatial transformation and registration unit analyzes the coordinate differences between these point pairs and calculates a precise mathematical transformation model. Then, it uses this model to "reproject" images B' and C'. This precisely "moves" the point with coordinates (102, 148) in image B' to a position that completely overlaps with (100, 150) in image A'. This transformation is performed on every pixel in the image, generating three new images: image A'', image B'', and image C''. Now, the pixel coordinates of the same scalp point are completely consistent across these three images. These three images are the registered multispectral images.

[0029] The feature layer fusion unit performs fusion processing on the registered multispectral images to generate a fused image.

[0030] The feature layer fusion unit uses intelligent algorithms (such as weighted fusion and multi-scale fusion) to integrate the advantages of the three images. For example, it extracts from image A'' (white light): true color information (used to determine redness), the overall shape and density of the hair. From image B'' (polarized light): details of the scalp surface after removing reflections (making the stratum corneum texture and fine dandruff clearer). From image C'' (UV light): the fluorescence distribution map of sebum (bright areas represent oil). Generating the fused image: The feature layer fusion unit does not simply overlay the three images, but assigns appropriate weights to different regions and features based on the importance and clarity of each piece of information, ultimately generating a completely new image. This "fused image" may not look like any of the original images, but it contains all the key information: it has the color foundation of A''. In oily areas, it overlays the fluorescence highlights of C''. In all areas, the surface texture is as clear and glare-free as B''. The slight redness under the skin that was not obvious in a single image (masked by the reflection of oil in A'') may now become more noticeable because the reflection has been eliminated in B''.

[0031] The feature extraction and quantization module 23 is used to identify at least one feature data from the fused image and output a structured multidimensional feature vector; The feature extraction and quantization module 23 includes: The dandruff detection unit is used to identify dandruff areas in the fused image and calculate the percentage of the dandruff area to the total area of ​​the detection area as a quantitative indicator of dandruff coverage. The dandruff detection unit searches for white or grayish-white flaky / granular areas in the fused image. It uses an image segmentation algorithm to accurately delineate all dandruff areas. Assuming the detection area has a total of 1 million pixels, and the algorithm delineates 80,000 dandruff pixels, then the dandruff coverage rate = (80,000 / 1,000,000) * 100% = 8.0%.

[0032] The sebum analysis unit is used to receive the fused image from the multispectral image fusion module, identify the bright areas caused by sebum fluorescence through an image segmentation algorithm, and calculate the percentage of the area of ​​the bright areas to the total area of ​​the detection area as a quantitative indicator of sebum coverage. The sebum analysis unit focuses on analyzing the parts of the fused image highlighted by UV fluorescence information. It identifies areas that are abnormally bright due to sebum. Assuming it identifies 250,000 highlight (oil) pixels, then the sebum coverage = (250,000 / 1,000,000) * 100% = 25.0%.

[0033] The redness sensitivity analysis unit is used to convert the fused image from the RGB color space and calculate the erythema index map of the fused image; based on the erythema index map, it identifies areas with index values ​​exceeding a preset threshold as effective erythema areas, outputs the area ratio of the effective erythema areas and the average erythema index of the effective erythema areas as quantitative indicators of redness sensitivity; the erythema index is calculated as: EI=k*[log(R)-log(G)], where R and G are the red and green channel values ​​of the pixel in the RGB color space, respectively, and k is a calibration coefficient.

[0034] The redness sensitivity analysis unit calculates the redness index (EI) for each pixel of the fused image using the RGB color space and the formula EI = k * [log(R) - log(G)]. Hemoglobin absorbs green light (G) and reflects red light (R). The larger the value of log(R) - log(G), the more prominent the redness at that point (the more hemoglobin). k is a calibration coefficient used to standardize the results. This ultimately generates an "erythema index map" of the same size as the fused image, where each pixel value represents the redness (EI value) of that point. The redness sensitivity analysis unit sets a threshold (e.g., EI > 2.5). In the erythema index map, all pixels with EI values ​​exceeding 2.5 are marked as "effective erythema". Assuming there are 120,000 effective erythema pixels, and the average EI value of these pixels is 3.8, then the redness area percentage = (120,000 / 1,000,000) * 100% = 12.0%, and the average erythema index = 3.8. Thus, the three units have produced three raw quantized data points in parallel.

[0035] The data integration and standardization unit is used to receive quantitative data from the keratin and dandruff detection unit, sebum analysis unit, and redness sensitivity analysis unit, and to integrate, standardize units, and unify formats to generate structured multidimensional feature vectors. The data integration and standardization unit includes: A multi-channel data receiving and buffering subunit is used to receive raw quantitative data from the keratin and dandruff detection unit, the sebum analysis unit, and the redness sensitivity analysis unit in parallel, and to perform synchronous buffering. Simultaneously receive three data points: [Dandruff: 8.0%], [Sebum: 25.0%], [Redness area: 12.0%, Average EI: 3.8], and temporarily cache them, waiting to be processed together.

[0036] The data cleaning and validity verification subunit is connected to the multi-channel data receiving and caching subunit. It is used to perform range verification and logical consistency checks on the raw quantized data, mark or process abnormal data values, and generate verified data with validity status markers. Check if the data is within a reasonable range; if so, label each data entry and generate validated data.

[0037] The unit conversion and normalization processing subunit is connected to the data cleaning and validity verification subunit. The unit conversion and normalization processing subunit has an index conversion configuration table pre-stored inside, which is used to convert the units of the verified data with validity status markers into predetermined standardized data according to the index conversion configuration table. Query the indicator conversion configuration table within the unit conversion and normalization processing sub-unit. The table specifies that the target unit for the three indicators—dandruff coverage rate, sebum coverage rate, and redness area percentage—is "%," and the original data is already in percentage form, so no conversion is needed. The target unit for the average erythema index is "EI," which also does not require conversion. The data units are confirmed and standardized, and the standardized data is output as follows: [Dandruff coverage rate: 8.0%], [Sebum coverage rate: 25.0%], [Redness area percentage: 12.0%], [Average erythema index: 3.8EI].

[0038] The structured feature vector assembly subunit is connected to the unit conversion and normalization processing subunit. It is used to fill the standardized data after unit conversion into the corresponding dimensions according to the predefined vector template, and assemble and generate a structured multidimensional feature vector. The structured feature vector assembly subunit has a fixed template, for example: [Dimension 1: Dandruff coverage, Dimension 2: Sebum coverage, Dimension 3: Redness area percentage, Dimension 4: Average erythema index]. Standardized data is sequentially filled into the template as follows: [8.0, 25.0, 12.0, 3.8]. Metadata is then attached to this array, such as timestamp, detection location (e.g., crown area), and user ID.

[0039] And a data packet encapsulation and output buffer subunit, connected to the structured feature vector assembly subunit, for encapsulating the multidimensional feature vector and associated metadata into a standard format data packet and outputting it to the output interface unit.

[0040] The assembled feature vector [8.0, 25.0, 12.0, 3.8] and its metadata are packaged into a standard format agreed upon by the downstream modules. This data packet is then sent out through the output interface unit.

[0041] And an output interface unit, used to output the structured multidimensional feature vector to the health status analysis and diagnosis module 24.

[0042] The health status analysis and diagnosis module 24 is used to compare the structured multidimensional feature vector with a preset threshold to generate a scalp health score or risk warning information. The health status analysis and diagnosis module 24 includes: The feature vector input and parsing unit is used to receive the structured multidimensional feature vector from the feature extraction and quantization module, and parse the various quantization indicators contained therein into independent variables; The feature vector input and parsing unit receives data packets and parses out the feature vector [8.0, 25.0, 12.0, 3.8] and the corresponding index labels ["dandruff coverage (%)", "sebum coverage (%)", "erythema area percentage (%)", "average erythema index (EI)"]. The data is then transformed into internal independent variables: dandruff=8.0, sebum=25.0, erythema_area=12.0, erythema_index=3.8.

[0043] The multidimensional indicator database and knowledge base unit are used to store the preset thresholds corresponding to various scalp health indicators, the association rules between different indicator combinations and scalp symptoms, and the weight configuration of each indicator in the comprehensive score. The preset thresholds for storing multidimensional indicator database and knowledge base units are as follows: Dandruff coverage: <5% normal, 5%-15% mild, >15% moderate, >30% severe. Sebum coverage: <20% dry, 20%-35% neutral, >35% oily. Average erythema index (EI): <2.0 normal, 2.0-3.5 mildly sensitive, >3.5 moderately sensitive, >5.0 severe sensitive / inflammatory.

[0044] Association rules, for example: Rule 1: IF (sebum coverage > 30%) AND (dandruff coverage > 10%) THEEN associated symptom = "seborrheic tendency". Rule 2: IF (average erythema index > 3.5) AND (redness area > 5%) THEEN risk level increase = "potential for active inflammation".

[0045] Weighting (for overall scoring): Inflammation (erythema index) has the highest weight, set at 0.4. Sebum and dandruff have the next highest weights, each at 0.3. Other indicators have a weight of 0.2.

[0046] The single-indicator state evaluation unit is connected to the feature vector input and parsing unit and the multi-dimensional indicator database and knowledge base unit. It is used to compare the measured values ​​of each quantitative indicator with their corresponding preset thresholds to generate an independent state evaluation result for each indicator. The single-indicator status assessment unit compares the measured values ​​with preset thresholds in the multi-dimensional indicator database and knowledge base unit: dandruff=8.0% → Comparison threshold: 5%-15% indicates mild abnormality, so the status is mildly abnormal. sebum=25.0% → Comparison threshold: 20%-35% indicates neutrality, so the status is normal. erythema_index=3.8 → Comparison threshold: >3.5 indicates moderate sensitivity, so the status is moderately abnormal. erythema_area=12.0% → Used as a reference for severity.

[0047] A multi-indicator comprehensive reasoning and diagnosis unit is connected to the feature vector input and parsing unit, the multi-dimensional indicator database and knowledge base unit, and the single indicator state evaluation unit. It is used to generate a comprehensive scalp health score and at least one preliminary diagnostic conclusion based on the quantitative indicators, the independent state evaluation results, the association rules, and the weight configuration through a preset reasoning model. According to rule 1: Sebum (25.0% < 30%) and dandruff (8.0% < 10%) do not trigger "seborrheic tendency". According to rule 2: Erythema index (3.8 > 3.5) and reddened area (12.0% > 5%) trigger "potential for active inflammation".

[0048] Generate a comprehensive health score: using a weighted deduction method (assuming a maximum score of 100): Mild dandruff abnormality: deduct (8-5) / (15-5)*100*0.3=9 points; Normal sebum: deduct 0 points; Moderate erythema abnormality: deduct (3.8-3.5) / (5.0-3.5)*100*0.4=8 points; Therefore, the comprehensive score = 100-9-0-8=83 points. Preliminary diagnostic conclusion: Main problem: "Moderate scalp sensitivity was detected, accompanied by a local inflammatory reaction." Secondary findings: "Mild dandruff is present on the scalp, and sebum secretion is within the neutral range."

[0049] The risk warning and priority determination unit is connected to the multi-indicator comprehensive reasoning and diagnosis unit, and is used to determine the current scalp health risk level and generate corresponding warning information based on the comprehensive health score and / or the preliminary diagnosis conclusion. Based on the diagnostic conclusion (moderate sensitivity / inflammation) and the comprehensive score (83 points, which is "good" but not "excellent").

[0050] Risk level: Determined as "moderate risk". Extreme caution is required when performing hair perming procedures.

[0051] The generated warning message is: The scalp is currently in a moderately sensitive state with signs of inflammation. Proceeding with standard perming procedures may worsen the irritation and increase the risk of contact dermatitis. Protective measures or adjustments to the treatment plan are recommended.

[0052] The system also includes a report generation and formatting unit, which is connected to the single-indicator status assessment unit, the multi-indicator comprehensive reasoning and diagnosis unit, and the risk warning and priority determination unit. This unit integrates the independent status assessment results, the comprehensive health score, the preliminary diagnostic conclusion, and the warning information to generate a standard format scalp health analysis report.

[0053] The real-time control and feedback module 25 is used to generate adjustment instructions and safety alarms for the working parameters of the hair perming device based on the health score or risk warning information.

[0054] The real-time control and feedback module 25 includes: The instruction parsing and strategy mapping unit is used to receive health scores, risk levels and early warning information from the health status analysis and diagnosis module, and generate preliminary control strategy instructions based on the pre-stored health status and control strategy mapping relationship. The instruction parsing and strategy mapping unit has a preset "health status-control strategy mapping table". When the diagnosis conclusion is "moderate risk" & "inflammation", a matching strategy is performed: for example, the goal is to minimize stimulation of the inflamed area while completing the modeling.

[0055] Specific instructions: Global action: Switch the concentration of the perming agent to the "mild" formula setting.

[0056] Zoned temperature control: For the detected reddish areas (coordinates obtained from the image analysis module), the heating temperature is reduced by 30%; the temperature of non-reddish areas remains unchanged.

[0057] Time adjustment: The overall heating time is shortened by 15%.

[0058] Mechanical protection: The mechanical tension applied to the reddened area by the curling iron is reduced by 50% when the curling iron is wrapped.

[0059] Alarm level: Triggers a yellow audible and visual alarm to alert the operator.

[0060] The instruction parsing and strategy mapping unit generates a detailed "Preliminary Control Strategy Instruction" which includes all the above-mentioned action requirements.

[0061] A multi-parameter collaborative control unit, connected to the instruction parsing and strategy mapping unit, is used to generate a collaborative control signal to adjust at least two independent operating parameters of the hair perming device based on the preliminary control strategy instruction; the independent operating parameters include at least two of heating temperature, chemical action time, and mechanical force. The multi-parameter collaborative control unit translates the initial control strategy commands into precise numerical signals that the equipment can understand. Assuming the original planned parameters are: temperature 180°C, time 15 minutes, and standard tension, then for the temperature controller: signal A is generated. The global command is: switch to the mild reagent pump. And the target temperature of the heating element within the coordinate region [X1, Y1, X2, Y2] is set to 180°C * (1 - 30%) = 126°C.

[0062] For the timer: Generate signal B. Set the total heating time to 15 minutes * (1 - 15%) = 12 minutes and 45 seconds.

[0063] For the curling iron motor: Signal C is generated. When the robotic arm moves to the coordinate region [X1,Y1,X2,Y2], the maximum output torque of the motor is limited to 50% of the standard value.

[0064] The multi-parameter collaborative control unit outputs a set of collaborative control signals (A, B, C) to ensure that temperature, time, and intensity are adjusted synchronously according to the strategy, rather than being changed individually.

[0065] The security alarm and status synchronization unit is connected to the instruction parsing and strategy mapping unit, and is used to generate corresponding audible and visual alarm signals according to the risk level, and broadcast the current security control status to other modules of the system. The safety alarm and status synchronization unit activates the device's yellow warning light to flash and emits a medium-frequency beeping sound when the risk level is "medium (yellow)".

[0066] The control command encapsulation and output interface unit is connected to the multi-parameter collaborative control unit and the safety alarm and status synchronization unit. It is used to encapsulate the collaborative control signal and the alarm signal into drive commands that conform to the communication protocol of the hair perming device actuator, and output them to the corresponding actuators and alarm devices.

[0067] Control signals A, B, and C are converted into specific communication protocols (such as PWM waves, CAN bus messages, and specific serial port commands) that can be recognized by the temperature controller, timer, and motor driver, respectively. These drive commands are then sent via physical circuitry to: the heating element and temperature sensor inside the curling iron, and the control valve of the chemical pump. The servo motor of the hair curler's robotic arm, as well as the LED indicator lights and buzzer on the machine body.

[0068] like Figure 3 As shown, in a second aspect, this application also provides a scalp health detection method, applied to a hair perming device having a scalp health detection system, the method comprising: S1. By using several cameras placed at different locations, images are taken of different parts of the human scalp to obtain the original images of the scalp area; S2. The processor receives and processes the original image, identifies and analyzes the scalp features in the image, and generates a scalp health analysis report containing quantitative indicators. S3. Store the scalp health analysis report in the storage module to form a queryable scalp health analysis report.

[0069] The options described in the above system embodiments are also applicable to this embodiment, and will not be detailed here. The remaining contents of this embodiment can be found in the above system embodiments, and will not be repeated in this embodiment.

[0070] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0071] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A hair perming device with a scalp health detection system, comprising a hair perming device body, wherein the hair perming device body is a straightener or curling iron with a comb-like structure, characterized in that: The hair perming device includes several cameras, a processor, a storage module, and a power supply module. The cameras are located in the comb structure area and are used to capture raw images of different locations on the human scalp. The processor is used to analyze and process the real-time images captured by the cameras and generate a scalp health analysis report. The storage module is used to store the scalp health analysis report. The power supply module is used to supply power to the cameras, processor, and storage module.

2. The hair perming device with a scalp health detection system according to claim 1, characterized in that: The hair perming device with a scalp health detection system also includes a wireless connection module and an APP that can be installed on a handheld smart device. The wireless connection module is used to connect to the handheld smart device and send a scalp health analysis report generated by the processor so that the user can view it through the APP.

3. The hair perming device with a scalp health detection system according to claim 2, characterized in that: The processor includes: The image preprocessing module is used to standardize the raw images captured by cameras from different light sources and output preprocessed images. The multispectral image fusion module performs precise registration and fusion of preprocessed images to generate a fused image; The feature extraction and quantization module is used to identify at least one feature data from the fused image and output a structured multidimensional feature vector. The health status analysis and diagnosis module is used to compare the structured multidimensional feature vector with a preset threshold to generate a scalp health score or risk warning information. The real-time control and feedback module is used to generate adjustment instructions and safety alarms for the operating parameters of the hair perming device based on the health score or risk warning information.

4. The hair perming device with a scalp health detection system according to claim 3, characterized in that: The multispectral image fusion module includes: The image input interface unit receives and buffers multiple preprocessed images from the image preprocessing module, and is responsible for the unification and synchronization of data formats to ensure that subsequent units receive time-aligned image frames. The feature point extraction and matching unit extracts stable image feature points for each input multi-channel preprocessed image with different light sources, performs feature point matching between images with different light sources, finds the correspondence of the same scalp area under different imaging conditions, and outputs matching point pairs. The spatial transformation and registration unit, based on the matching point pairs output by the feature point matching unit, transforms all images under different light sources to the same coordinate system and outputs a registered multispectral image. The feature layer fusion unit performs fusion processing on the registered multispectral images to generate a fused image.

5. The hair perming device with a scalp health detection system according to claim 4, characterized in that: The feature extraction and quantization module includes: The dandruff detection unit is used to identify dandruff areas in the fused image and calculate the percentage of the dandruff area to the total area of ​​the detection area as a quantitative indicator of dandruff coverage. The sebum analysis unit is used to receive the fused image from the multispectral image fusion module, identify the bright areas caused by sebum fluorescence through an image segmentation algorithm, and calculate the percentage of the area of ​​the bright areas to the total area of ​​the detection area as a quantitative indicator of sebum coverage. The redness sensitivity analysis unit is used to convert the fused image from the RGB color space and calculate the erythema index map of the fused image; based on the erythema index map, it identifies areas with index values ​​exceeding a preset threshold as effective erythema areas, and outputs the area ratio of the effective erythema areas and the average erythema index of the effective erythema areas as quantitative indicators of redness sensitivity. The data integration and standardization unit is used to receive quantitative data from the keratin and dandruff detection unit, sebum analysis unit, and redness sensitivity analysis unit, and to integrate, standardize units, and unify formats to generate structured multidimensional feature vectors. And an output interface unit, used to output the structured multidimensional feature vector to the health status analysis and diagnosis module.

6. The hair perming device with a scalp health detection system according to claim 5, characterized in that: The erythema index is calculated as follows: EI=k*[log(R)-log(G)], where R and G are the red and green channel values ​​of the pixel in the RGB color space, respectively, and k is the calibration coefficient.

7. The hair perming device with a scalp health detection system according to claim 6, characterized in that, The data integration and standardization unit includes: A multi-channel data receiving and buffering subunit is used to receive raw quantitative data from the keratin and dandruff detection unit, the sebum analysis unit, and the redness sensitivity analysis unit in parallel, and to perform synchronous buffering. The data cleaning and validity verification subunit is connected to the multi-channel data receiving and caching subunit. It is used to perform range verification and logical consistency checks on the raw quantized data, mark or process abnormal data values, and generate verified data with validity status markers. The unit conversion and normalization processing subunit is connected to the data cleaning and validity verification subunit. The unit conversion and normalization processing subunit has an index conversion configuration table pre-stored inside, which is used to convert the units of the verified data with validity status markers into predetermined standardized data according to the index conversion configuration table. The structured feature vector assembly subunit is connected to the unit conversion and normalization processing subunit. It is used to fill the standardized data after unit conversion into the corresponding dimensions according to the predefined vector template, and assemble and generate a structured multidimensional feature vector. And a data packet encapsulation and output buffer subunit, connected to the structured feature vector assembly subunit, for encapsulating the multidimensional feature vector and associated metadata into a standard format data packet and outputting it to the output interface unit.

8. The hair perming device with a scalp health detection system according to claim 6, characterized in that, The health status analysis and diagnosis module includes: The feature vector input and parsing unit is used to receive the structured multidimensional feature vector from the feature extraction and quantization module, and parse the various quantization indicators contained therein into independent variables; The multidimensional indicator database and knowledge base unit are used to store the preset thresholds corresponding to various scalp health indicators, the association rules between different indicator combinations and scalp symptoms, and the weight configuration of each indicator in the comprehensive score. The single-indicator state evaluation unit is connected to the feature vector input and parsing unit and the multi-dimensional indicator database and knowledge base unit. It is used to compare the measured values ​​of each quantitative indicator with their corresponding preset thresholds to generate an independent state evaluation result for each indicator. A multi-indicator comprehensive reasoning and diagnosis unit is connected to the feature vector input and parsing unit, the multi-dimensional indicator database and knowledge base unit, and the single indicator state evaluation unit. It is used to generate a comprehensive scalp health score and at least one preliminary diagnostic conclusion based on the quantitative indicators, the independent state evaluation results, the association rules, and the weight configuration through a preset reasoning model. The risk warning and priority determination unit is connected to the multi-indicator comprehensive reasoning and diagnosis unit, and is used to determine the current scalp health risk level and generate corresponding warning information based on the comprehensive health score and / or the preliminary diagnosis conclusion. The system also includes a report generation and formatting unit, which is connected to the single-indicator status assessment unit, the multi-indicator comprehensive reasoning and diagnosis unit, and the risk warning and priority determination unit. This unit integrates the independent status assessment results, the comprehensive health score, the preliminary diagnostic conclusion, and the warning information to generate a standard format scalp health analysis report.

9. The hair perming device with a scalp health detection system according to claim 8, characterized in that, The real-time control and feedback module includes: The instruction parsing and strategy mapping unit is used to receive health scores, risk levels and early warning information from the health status analysis and diagnosis module, and generate preliminary control strategy instructions based on the pre-stored health status and control strategy mapping relationship. A multi-parameter collaborative control unit, connected to the instruction parsing and strategy mapping unit, is used to generate a collaborative control signal to adjust at least two independent operating parameters of the hair perming device based on the preliminary control strategy instruction; the independent operating parameters include at least two of heating temperature, chemical action time, and mechanical force. The security alarm and status synchronization unit is connected to the instruction parsing and strategy mapping unit, and is used to generate corresponding audible and visual alarm signals according to the risk level, and broadcast the current security control status to other modules of the system. The control command encapsulation and output interface unit is connected to the multi-parameter collaborative control unit and the safety alarm and status synchronization unit. It is used to encapsulate the collaborative control signal and the alarm signal into drive commands that conform to the communication protocol of the hair perming device actuator, and output them to the corresponding actuators and alarm devices.

10. A method for detecting scalp health, characterized in that, A method applied to a hair perming apparatus with a scalp health detection system as described in any one of claims 1 to 9, the method comprising: By using several cameras placed in different locations, images of different parts of the human scalp are captured to obtain raw images of the scalp area. The processor receives and processes the original image, identifies and analyzes the scalp features in the image, and generates a scalp health analysis report containing quantitative indicators. The scalp health analysis report is stored in the storage module to form a queryable scalp health analysis report.

Images