Skin and hair analysis method and system based on mobile terminal screen mirroring interconnection detector

Abstract

The present invention relates to the technical field of skin and hair analysis, and in particular, to a skin and hair analysis method and system based on a mobile terminal screen mirroring interconnection detector. The method comprises the following steps: acquiring, on the basis of a camera at a detection end of a detector, a skin-hair multispectral image set; performing multi-image spectral quantitative analysis on the skin-hair multispectral image set to construct a skin-hair pigment distribution map; performing contour detail feature analysis and detailed image reconstruction on the skin-hair multispectral image set to construct a detailed reconstructed skin-hair image; performing refined visual recognition of hair follicles on the detailed reconstructed skin-hair image to obtain spatial position coordinates of hair follicles; and performing regional hair feature change trend analysis and feature region clustering division on the skin pigment distribution map and a hair follicle density distribution trend model to obtain a plurality of change trend feature regions. The present invention achieves fast and efficient skin and hair analysis and can visualize analysis results on a mobile terminal and a detector display device.

Description

A skin and hair analysis method and system based on mobile terminal projection interconnection detector TECHNICAL FIELD

[0001] The present application relates to the technical field of skin and hair analysis, and particularly relates to a skin and hair analysis method and system based on mobile terminal projection interconnection detector. BACKGROUND

[0002] Personal skin and hair condition is one of the important indicators for measuring the health condition of human body. With the increasing demand for beauty and health, accurate assessment and monitoring of skin and hair condition becomes more and more important. Traditional skin and hair analysis methods usually rely on professional detection instruments, which not only need to be operated by professionals, but also have problems such as long detection time and inability to monitor in real time. Traditional detection instruments cannot be connected to user terminal devices, and cannot present skin and hair analysis results to users in real time and quickly. At the same time, as the most popular smart terminal device at present, mobile phones are reshaping the medical and health field with their powerful computing power and rich functions, bringing new opportunities for personal health detection. Therefore, it is urgent to explore a skin and hair analysis method based on mobile smart terminal projection interconnection detector, which uses mobile phones, tablets and other smart mobile terminals as data analysis carriers, and realizes rapid, efficient and non-invasive detection of skin and hair condition through projection interconnection detection equipment. SUMMARY

[0003] To solve the above technical problems, the present application provides a skin and hair analysis method and system based on mobile terminal projection interconnection detector to solve at least one of the above technical problems.

[0004] To achieve the above purpose, the present application provides a skin and hair analysis method based on mobile terminal projection interconnection detector, which is realized based on a mobile terminal and a detector based on real-time communication, wherein the detector includes a detector probe end and a detector display device, and the mobile terminal is used to provide data analysis computing power for the detector, including the following steps:

[0005] Step S1: obtaining a skin multispectral image set based on a detector probe end camera; performing multi-image spectral quantitative analysis on the skin multispectral image set to construct a skin pigment distribution atlas;

[0006] Step S2: performing contour detail feature analysis on the skin multispectral image set, and performing detail image reconstruction to construct a detail reconstructed skin and hair image;

[0007] Step S3: performing hair follicle fine visual recognition on the detail reconstructed skin and hair image to obtain hair follicle spatial position coordinates; performing hair follicle density distribution analysis according to the hair follicle spatial position coordinates to construct a hair follicle density distribution trend model;

[0008] Step S4: analyzing the regional hair feature change trend of the skin pigment distribution atlas and the hair follicle density distribution trend model, and performing feature region clustering division, so as to obtain a plurality of change trend feature regions;

[0009] Step S5: performing multi-scene growth time sequence evolution simulation on the plurality of change trend feature regions, and performing growth trend change prediction, to construct a hair region growth trend evolution curve;

[0010] Step S6: performing hair growth condition evaluation on the hair region growth trend evolution curve, so as to obtain a growth condition evaluation result; and performing visual projection on a detector display device based on the growth condition evaluation result, to present a hair analysis result.

[0011] The present application obtains more information about skin and hair by obtaining a skin multi-spectrum image set, provides a data basis for subsequent analysis, performs multi-image spectral quantitative analysis, constructs a skin pigment distribution atlas to understand the pigment distribution of hair, provides a basis for subsequent analysis and recognition, highlights the detail features of skin and hair through contour detail feature analysis and detail image reconstruction, improves the understanding and display of hair structure, constructs a detail reconstructed skin and hair image to observe and analyze the microstructure of hair, accurately obtains the spatial position coordinates and hair follicle density information of hair follicles through fine visual recognition and hair follicle density distribution analysis, constructs a hair follicle density distribution trend model to understand the distribution and density change trend of hair follicles on the skin surface, performs regional hair feature change trend analysis and feature region clustering division, identifies and classifies the hair feature change trend of different regions, obtains a plurality of change trend feature regions to deeply understand the hair features and change rules of different regions, simulates the growth process of hair and predicts its change trend through multi-scene growth time sequence evolution simulation and growth trend change prediction, constructs a hair region growth trend evolution curve to understand the growth rules and trends of hair, evaluates the hair growth condition to obtain a growth condition evaluation result, comprehensively evaluates the health condition and growth state of hair, visually projects and presents the hair analysis result to intuitively display the analysis result, improves the communication efficiency and the clarity of the result display.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: obtaining a skin multi-spectrum image set based on a detector probe camera;

[0014] Step S12: performing multi-image spectral quantitative analysis on the skin multi-spectrum image set, and extracting an image spectral feature vector;

[0015] Step S13: performing regional color difference analysis on the image spectral feature vector, to obtain color difference data of different image regions;

[0016] Step S14: color space difference decomposition is performed on the skin multispectral image set according to the color difference data of different image regions, to generate hair pigment region feature data;

[0017] Step S15: pigment distribution fitting is performed on the hair pigment region feature data, to construct a skin pigment distribution atlas.

[0018] The present application obtains rich spectral information by obtaining a skin multispectral image set, provides a data basis for subsequent analysis, performs multispectral quantitative analysis on images, extracts image spectral feature vectors, quantifies spectral information, provides data support for subsequent analysis, performs regional color difference analysis on image spectral feature vectors, obtains color difference data of different image regions, helps to understand the color difference of different regions, performs color space difference decomposition according to the color difference data of different image regions, generates hair pigment region feature data, distinguishes the features of different color spaces, and the hair pigment region feature data provides more specific pigment distribution information. Pigment distribution fitting is performed on the hair pigment region feature data, to construct a skin pigment distribution atlas, understand the distribution of hair pigments in different regions, and the constructed skin pigment distribution atlas provides a visual reference for subsequent analysis and diagnosis.

[0019] Preferably, step S2 comprises the following steps:

[0020] Step S21: hair root recognition is performed on the skin multispectral image set, to mark the hair root region;

[0021] Step S22: brightness enhancement processing is performed on the hair root region, to obtain a brightness-enhanced hair image;

[0022] Step S23: hair contour analysis is performed on the brightness-enhanced hair image, to extract a hair contour line;

[0023] Step S24: contour expansion processing is performed on the hair contour line, to obtain a hair contour expansion image;

[0024] Step S25: contour detail feature analysis is performed on the hair contour detail expansion image, to generate hair contour detail features;

[0025] Step S26: detail image reconstruction is performed on the brightness-enhanced hair image according to the hair contour detail features, to construct a detail-reconstructed skin hair image.

[0026] The present application can accurately locate and mark the growth starting point of the hair by recognizing the hair root of the skin multi-spectrum image set, providing an accurate basis for subsequent analysis, enhancing the brightness of the hair root area, highlighting the brightness characteristics of the hair root, improving the visualization effect of the hair root area, analyzing the hair contour of the brightness enhanced hair image, extracting the hair contour line, accurately capturing the shape and contour information of the hair, performing contour expansion processing on the hair contour line to obtain a hair contour expansion image, enhancing the display effect of the hair contour, making the hair morphology more clear, performing contour detail feature analysis on the hair contour detail expansion image, highlighting the detail features of the hair contour, improving the understanding and display of the hair structure, reconstructing the detail image of the brightness enhanced hair image according to the hair contour detail features, and constructing a detail reconstructed skin hair image to observe the microstructure and features of the hair in depth.

[0027] Preferably, step S3 is specifically as follows:

[0028] Step S31: performing fine visual recognition on the hair follicle of the detail reconstructed skin hair image, and marking each hair follicle feature point;

[0029] Step S32: performing space position calculation on each hair follicle feature point, so as to obtain a hair follicle space position coordinate;

[0030] Step S33: performing inter-follicle distance calculation according to the hair follicle space position coordinate, so as to obtain a plurality of inter-follicle distances;

[0031] Step S34: performing hair follicle density distribution analysis on each hair follicle feature point based on the plurality of inter-follicle distances, so as to generate hair follicle density distribution data;

[0032] Step S35: performing density trend mapping on the hair follicle density distribution data, so as to construct a hair follicle density distribution trend model.

[0033] The present application can accurately locate and mark the growth starting point of the hair by recognizing the hair root of the skin multi-spectrum image set, providing an accurate basis for subsequent analysis, enhancing the brightness of the hair root area, highlighting the brightness characteristics of the hair root, improving the visualization effect of the hair root area, analyzing the hair contour of the brightness enhanced hair image, extracting the hair contour line, accurately capturing the shape and contour information of the hair, performing contour expansion processing on the hair contour line to obtain a hair contour expansion image, enhancing the display effect of the hair contour, making the hair morphology more clear, performing contour detail feature analysis on the hair contour detail expansion image, highlighting the detail features of the hair contour, improving the understanding and display of the hair structure, reconstructing the detail image of the brightness enhanced hair image according to the hair contour detail features, and constructing a detail reconstructed skin hair image to observe the microstructure and features of the hair in depth.

[0034] Preferably, the specific steps of step S4 are:

[0035] Step S41: image segmentation processing is performed on the detail reconstructed skin hair image to obtain a plurality of image regions;

[0036] Step S42: regional hair feature change trend analysis is performed on the plurality of image regions based on a skin pigment distribution atlas and a hair follicle density distribution trend model to generate feature change trends of different image regions;

[0037] Step S43: similarity analysis is performed on the feature change trends of different image regions to generate regional change trend similarity;

[0038] Step S44: feature region clustering division is performed on the detail reconstructed skin hair image according to the regional change trend similarity, thereby obtaining a plurality of change trend feature regions.

[0039] The present application discloses a method for analyzing skin hair growth and change trend.

[0040] Preferably, the specific steps of step S5 are:

[0041] Step S51: multi-scene growth time sequence evolution simulation is performed on the plurality of change trend feature regions to generate multi-scene time sequence evolution simulation data;

[0042] Step S52: growth time sequence response rate analysis is performed on the multi-scene time sequence evolution simulation data to obtain hair growth time sequence response rate data;

[0043] Step S53: growth trend change prediction is performed on the hair growth time sequence response rate data to generate hair growth trend prediction data;

[0044] Step S54: stage evolution trajectory analysis is performed on the hair growth trend prediction data to construct a hair region growth trend evolution curve.

[0045] The present application simulates the growth process and changes of hair under different scenarios by multi-scene growth time evolution simulation of multiple change trend characteristic regions, analyzes the growth of hair under different conditions, provides data basis for subsequent growth rate analysis, performs growth time response rate analysis on multi-scene time evolution simulation data, evaluates the growth rate and response of hair under different conditions, understands the dynamic changes and response characteristics of hair growth, provides basis for growth trend prediction, performs growth trend change prediction according to hair growth time response rate data, generates hair growth trend prediction data, predicts the growth trend change trend of hair in the future period, provides reference for the formulation of individualized care and treatment scheme, performs stage evolution trajectory analysis on hair growth trend prediction data, constructs hair region growth trend evolution curve, and reveals the stage changes and trends of hair growth.

[0046] Preferably, the specific steps of step S51 are:

[0047] The growth trend of different regions is obtained by performing growth trend deduction on multiple change trend characteristic regions.

[0048] Define multi-scene growth environment parameters.

[0049] Based on the multi-scene growth environment parameters, time-domain growth numerical simulation calculation is performed on the growth trend rules of different regions to obtain regional growth trend parameters of multiple scenes.

[0050] Based on the regional growth trend parameters of multiple scenes, multi-time-point growth evolution simulation is performed to generate multi-scene time evolution simulation data.

[0051] The present application reveals the growth rules and trends of different regions by performing growth trend deduction on multiple change trend characteristic regions, understands the differences of hair growth in different regions, defines multi-scene growth environment parameters to simulate the growth of hair under different conditions, considers the influence of environment on growth trend, simulates the growth situation closer to the actual situation, improves the accuracy of simulation results, performs time-domain growth numerical simulation calculation on the regional growth trend rules based on multi-scene growth environment parameters to obtain regional growth trend parameters of multiple scenes, quantifies the characteristics and trends of hair growth under different scenes, provides data support for subsequent simulation and analysis, performs multi-time-point growth evolution simulation based on the regional growth trend parameters of multiple scenes to generate multi-scene time evolution simulation data, simulates the growth change process of hair under different scenes, and provides basis for growth trend prediction and individualized care scheme.

[0052] Preferably, the specific steps of step S6 are:

[0053] Step S61: identifying the periodic fluctuation amplitude of the hair region growth trend evolution curve, and extracting the periodic fluctuation amplitude of the trend curve;

[0054] Step S62: calculating the trend peak time point of the hair region growth trend evolution curve, and generating a plurality of trend peak time points;

[0055] Step S63: evaluating the hair growth condition based on the plurality of trend peak time points and the trend curve periodic fluctuation amplitude, thereby obtaining the growth condition evaluation result;

[0056] Step S64: visualizing the growth condition evaluation result and the hair region growth trend evolution curve on the detector display device to finely present the hair analysis result.

[0057] The present application extracts the periodic fluctuation of the trend curve by identifying the periodic fluctuation amplitude of the hair region growth trend evolution curve, identifies the periodic change in the hair growth process, provides an important reference for subsequent growth condition evaluation, calculates the trend peak time point of the hair region growth trend evolution curve, determines the peak time point in the curve, i.e. the peak period of the growth trend, identifies the peak period in the hair growth process, provides key information for growth condition evaluation and prediction, evaluates the hair growth condition in combination with the plurality of trend peak time points and the trend curve periodic fluctuation amplitude, comprehensively understands the state and trend of hair growth, provides a basis for formulating a personalized care plan and predicting the growth condition, projects the growth condition evaluation result and the hair region growth trend evolution curve onto the detector display device to realize visual display, intuitively displays the hair analysis result, provides a user-friendly data display and interpretation method, and facilitates user understanding and application.

[0058] In the present specification, a skin and hair analysis system based on a mobile terminal projection interconnected detector is provided, which comprises a mobile terminal and a detector, and the mobile terminal and the detector perform real-time data transmission through wired or wireless network, the mobile terminal is used to execute the algorithm program of the skin and hair analysis method based on the mobile terminal projection interconnected detector as described above, the detector comprises a detector detection end and a detector display device, the detector detection end is used to acquire multispectral image data of skin and hair, and the detector display device is used to synchronize the screen of the mobile terminal and display network push information; the skin and hair analysis system based on the mobile terminal projection interconnected detector specifically comprises:

[0059] A pigment distribution module is used to acquire a skin multispectral image set based on a detector detection end camera, perform multispectral quantitative analysis on the skin multispectral image set, and construct a skin pigment distribution atlas;

[0060] an image reconstruction module, configured to perform contour detail feature analysis on the skin multi-spectrum image set and to perform detail image reconstruction to construct a detail reconstructed skin hair image;

[0061] a density distribution module, configured to perform follicle fine visual recognition on the detail reconstructed skin hair image to obtain follicle spatial position coordinates, and to perform follicle density distribution analysis according to the follicle spatial position coordinates to construct a follicle density distribution trend model;

[0062] a feature change trend module, configured to perform regional hair feature change trend analysis on the skin pigment distribution atlas and the follicle density distribution trend model, and to perform feature region clustering division to obtain a plurality of change trend feature regions;

[0063] a growth trend evolution module, configured to perform multi-scene growth time sequence evolution simulation on the plurality of change trend feature regions and to perform growth trend change prediction to construct a hair region growth trend evolution curve;

[0064] a condition assessment module, configured to perform hair growth condition assessment on the hair region growth trend evolution curve to obtain a growth condition assessment result, and to perform visual screen projection based on the growth condition assessment result to present a hair analysis result.

[0065] The present application can accurately capture the pigment distribution of skin hair by performing multi-image spectral quantitative analysis on the multi-spectrum image set, construct a skin pigment distribution atlas to provide basic data for subsequent analysis, perform contour detail feature analysis on the multi-spectrum image set and detail image reconstruction to reveal the detail features of skin hair, provide clearer and more accurate skin hair images, and provide a more reliable data basis for subsequent analysis, obtain follicle spatial position coordinates and a density distribution trend model through follicle fine visual recognition and density distribution analysis, can deeply understand the distribution of follicles, and provide more detailed data support for subsequent analysis, realize regional hair feature change trend analysis and regional clustering division by analyzing the skin pigment distribution atlas and the follicle density distribution trend model, find the feature change law of different regions, provide a basis for subsequent growth trend evolution, construct a hair region growth trend evolution curve through multi-scene growth time sequence evolution simulation and growth trend change prediction, can deeply understand the hair growth condition, provide prediction of future growth trend, and provide strong support for condition assessment, perform hair growth condition assessment on the hair region growth trend evolution curve to obtain a growth condition assessment result, can provide personalized care suggestions according to the assessment result, present the hair analysis result through visual screen projection to help users understand their own hair health condition and further improve the care scheme. BRIEF DESCRIPTION OF DRAWINGS

[0066] Fig. 1 is a step flowchart of a skin hair analysis method based on a mobile terminal screen projection interconnection detector according to the present application.

[0067] Fig. 2 is a detailed implementation step flow diagram of step S1;

[0068] Fig. 3 is a detailed implementation step flow diagram of step S2;

[0069] Fig. 4 is a detailed implementation step flow diagram of step S3;

[0070] Fig. 5 is a structural schematic diagram of a mobile terminal projection screen interconnection detector. DETAILED DESCRIPTION

[0071] It should be understood that the specific embodiments described herein are intended to be illustrative only and not limiting of the present application.

[0072] The present application provides a skin hair analysis method and system based on a mobile terminal projection screen interconnection detector. The execution subject of the skin hair analysis method and system based on the mobile terminal projection screen interconnection detector includes but is not limited to mechanical equipment, a data processing platform, a cloud server node, and a network upload device, which can be regarded as a general computing node of the present application. The data processing platform includes but is not limited to an audio image management system, an information management system, and a cloud data management system.

[0073] Referring to Figs. 1 to 4, the present application provides a skin hair analysis method based on a mobile terminal projection screen interconnection detector, which is realized based on real-time communication between a mobile terminal and a detector. The detector includes a detector probe end and a detector display device. The mobile terminal is used to provide data analysis computing power for the detector. The method includes the following steps:

[0074] Step S1: Obtain a skin multispectral image set based on a detector probe end camera. Perform multispectral quantitative analysis on the skin multispectral image set to construct a skin pigment distribution map.

[0075] Step S2: Perform contour detail feature analysis on the skin multispectral image set and reconstruct detail images to construct a detail reconstructed skin hair image.

[0076] Step S3: Perform follicle fine visual recognition on the detail reconstructed skin hair image to obtain follicle spatial position coordinates. Perform follicle density distribution analysis based on the follicle spatial position coordinates to construct a follicle density distribution trend model.

[0077] Step S4: Perform regional hair feature change trend analysis on the skin pigment distribution map and the follicle density distribution trend model, and perform feature region clustering and division to obtain a plurality of change trend feature regions.

[0078] Step S5: multi-scene growth time evolution simulation is performed on the multiple change trend feature regions, growth trend changes are predicted, and a hair region growth trend evolution curve is constructed;

[0079] Step S6: hair growth condition assessment is performed on the hair region growth trend evolution curve, so that a growth condition assessment result is obtained; and a mobile terminal screen projection interconnection detector is used to visually project a display device, so that a hair analysis result is presented.

[0080] The present application obtains more information about skin and hair by obtaining a skin multi-spectrum image set, provides a data basis for subsequent analysis, performs multi-image spectral quantitative analysis, constructs a skin pigment distribution atlas to understand the pigment distribution of hair, provides a basis for subsequent analysis and recognition, highlights the detail features of skin and hair through contour detail feature analysis and detail image reconstruction, improves the understanding and display of hair structure, constructs a detail reconstructed skin and hair image to observe and analyze the microstructure of hair, accurately obtains hair follicle spatial position coordinates and hair follicle density information through hair follicle fine visual recognition and hair follicle density distribution analysis, constructs a hair follicle density distribution trend model to understand the distribution of hair follicles on the surface of the skin and the density change trend, performs regional hair feature change trend analysis and feature region clustering division, identifies and classifies the hair feature change trends of different regions, obtains multiple change trend feature regions to deeply understand the hair features and change rules of different regions, simulates the growth process of hair and predicts its change trend through multi-scene growth time evolution simulation and growth trend change prediction, constructs a hair region growth trend evolution curve to understand the growth rules and trends of hair, assesses the growth condition of hair, obtains a growth condition assessment result, comprehensively evaluates the health condition and growth state of hair, visually projects a display device to present a hair analysis result to directly display the analysis result, improves the efficiency of communication and the clarity of result display.

[0081] In the embodiment of the present application, referring to FIG. 1, it is a step flowchart of a skin and hair analysis method based on a mobile terminal screen projection interconnection detector, in the present example, the steps of the skin and hair analysis method based on the mobile terminal screen projection interconnection detector include:

[0082] Step S1: obtaining a skin multi-spectrum image set based on a detector probe camera; performing multi-image spectral quantitative analysis on the skin multi-spectrum image set to construct a skin pigment distribution atlas;

[0083] In this embodiment, the multispectral imaging camera on the detection instrument probe is used. By controlling the light source and filter, multi-channel spectral images covering the visible light to near-infrared band can be obtained, ensuring good imaging conditions such as uniform illumination and clear imaging focus, to obtain high-quality multispectral image data. Images at multiple angles and positions are collected to obtain comprehensive skin and hair information. The obtained multispectral image set is preprocessed, including image registration, denoising, and optical correction, to ensure data quality. For each spectral channel image, statistical features such as average reflectance and variance within the region of interest (ROI) are extracted. The ROI features of different spectral channels are combined into a high-dimensional feature vector to describe the spectral features of the region. Clustering algorithms such as K-means are used to classify all ROI features, dividing them into different pigment distribution regions. The classification results are mapped back to the original image space to construct a visualization map reflecting the skin and hair pigment distribution.

[0084] Step S2: Profile detail feature analysis is performed on the skin multispectral image set, and a detail image is reconstructed to construct a detail reconstructed skin and hair image.

[0085] In this embodiment, the extracted profile features are analyzed, including calculating geometric features such as profile line length, curvature, and inflection point number. Texture analysis methods such as gray level co-occurrence matrix and Gabor filtering are used to extract texture features of the profile. The geometric features and texture features are combined into a high-dimensional feature vector to describe the detail information of the profile. Machine learning or deep learning models such as super-resolution convolutional networks are used to learn the mapping relationship between profile details and original images from the multispectral images. The learned model is applied to the original multispectral images to generate reconstructed images containing more detailed information. Post-processing such as sharpening and denoising is performed on the reconstructed images to further improve image quality. The detail reconstructed image is fused with the original multispectral image to obtain the final skin and hair image containing more detailed information. Manual or automatic quality evaluation is performed on the fusion result to ensure that the image detail information is effectively preserved and enhanced.

[0086] Step S3: Fine visual recognition of hair follicles is performed on the detail reconstructed skin and hair image to obtain the spatial position coordinates of the hair follicles. Based on the spatial position coordinates of the hair follicles, the hair follicle density distribution is analyzed, and a hair follicle density distribution trend model is constructed.

[0087] In this embodiment, the skin and hair image after detail reconstruction is processed by using an image segmentation algorithm (such as semantic segmentation or instance segmentation), the hair follicle region is segmented from the background, and further feature extraction and classification are performed on the segmented hair follicle region, such as identifying different types of hair follicles by using a deep learning model. Through these visual recognition technologies, the spatial coordinate position of each hair follicle in the image is accurately located. According to the obtained hair follicle coordinate position information, the distribution density of hair follicles in different regions is calculated, and methods such as kernel density estimation and Voronoi diagram are used to map the density value to a two-dimensional plane to construct a hair follicle density distribution map. The density distribution map is analyzed to identify high-density and low-density regions, which provides a basis for subsequent growth trend analysis. The hair follicle density distribution maps obtained at different time points are compared and analyzed to identify the trend of density value changes. Time series analysis or machine learning methods are used to construct a mathematical model describing the evolution law of hair follicle density distribution, which predicts the changes of hair follicle density in a future period of time, providing quantitative basis for growth trend analysis.

[0088] Step S4: Perform regional hair feature change trend analysis on the skin pigment distribution map and hair follicle density distribution trend model, and perform feature region clustering and division to obtain a plurality of change trend feature regions.

[0089] In this embodiment, time series analysis is performed on the hair features (pigment distribution, density, etc.) in different regions to identify the change trend of the hair features in different regions, such as increase, decrease, fluctuation, etc., and the change trend is quantitatively described. According to the characteristics of the change trend, the entire skin and hair region is preliminarily divided based on the change rate and amplitude of the hair features. Based on the preliminary regional division result as input features, unsupervised learning is performed by using clustering algorithms (such as K-means, hierarchical clustering, etc.). Through clustering analysis, regions with similar hair feature change trends are classified into the same category to form more refined feature region division. According to the silhouette coefficient or other indicators of the clustering result, the optimal number of clusters is determined to obtain the final feature region division scheme. Through the analysis and clustering of the foregoing steps, a plurality of different change trend feature regions are obtained, and the change trend of the hair features in each region is relatively consistent. For each feature region, the change of the hair features is further described, such as the increase or decrease of the average pigment concentration and the change of the average density. These change trend feature regions provide a more detailed and accurate data basis for subsequent growth trend analysis.

[0090] Step S5: Perform multi-scene growth time evolution simulation on a plurality of change trend feature regions, and perform growth trend change prediction to construct a hair region growth trend evolution curve.

[0091] In this embodiment, a corresponding growth time evolution model is constructed for each region. Different external environmental conditions or intervention measures are set as input parameters according to actual conditions, such as nutritional status, hormone level, disease factors, etc. Through numerical simulation, the growth time evolution process of each feature region under different scenarios is predicted, including changes in indicators such as hair density and pigment content. The evolution results under different scenarios are compared and analyzed to understand the degree of influence of various factors on hair growth. The growth trend of each feature region is predicted and analyzed, focusing on the change trend of hair growth indicators (density, pigment content, etc.) in each region. The growth trend change prediction model is established using time series analysis, grey model, neural network, etc. The sensitivity of the prediction results is analyzed to understand the degree of influence of key factors on the growth trend. The growth trend change prediction results of each feature region are integrated and drawn into a growth trend evolution curve. The curve reflects the growth trend changes of different feature regions in time series, directly showing the differences between regions. Combined with actual conditions, the inflection point and fluctuation trend of the curve are analyzed in depth to reveal the underlying physiological mechanisms. The growth trend evolution curve provides a visual data basis for subsequent comprehensive analysis and decision support.

[0092] Step S6: Perform hair growth condition assessment on the hair region growth trend evolution curve to obtain growth condition assessment results; and perform visual projection based on the growth condition assessment results to present the hair analysis results.

[0093] In this embodiment, the hair growth conditions of each feature region are comprehensively evaluated. For each curve, key indicators such as change trend, fluctuation frequency, and peak characteristics are analyzed, and actual physiological knowledge is interpreted to determine whether the hair growth of each region is in a good, normal, or abnormal state. The abnormal state is further divided into mild, moderate, and severe abnormalities, etc. Evaluation standards and quantitative indicators for hair growth conditions are established to ensure that the evaluation results are objective, fair, and comparable. The evaluation results of each feature region are integrated to obtain a global hair growth condition assessment result. The hair growth condition assessment result is presented in an intuitive and visual form, using heat maps, color coding, data charts, etc. to directly mark the growth conditions of each region on the human skin map. Different colors or markers are used according to different levels of evaluation results to clearly display the hair growth conditions of each region. An interactive interface is set up to allow users to zoom in, rotate, and observe from different angles to obtain more analysis details. The visual results are output in the form of projection.

[0094] In this embodiment, referring to FIG. 2, a detailed implementation step flowchart of step S1 is shown. In this embodiment, the detailed implementation steps of step S1 include:

[0095] Step S11: Obtain a skin multi-spectral image set based on the detection end camera of the detector;

[0096] Step S12: Perform multi-image spectral quantitative analysis on the skin multi-spectral image set to extract an image spectral feature vector;

[0097] Step S13: Perform regional color difference analysis on the image spectral feature vector to obtain color difference data of different image regions;

[0098] Step S14: Perform color space difference decomposition on the skin multi-spectral image set according to the color difference data of different image regions to generate hair pigment region feature data;

[0099] Step S15: Perform pigment distribution fitting on the hair pigment region feature data to construct a skin pigment distribution atlas.

[0100] In this embodiment, the multi-spectral imaging camera on the detection end of the detector is used to image the skin and hair of the sample to be measured. During the imaging process, it is necessary to ensure stable lighting conditions and avoid interference from environmental light sources. A multi-spectral image set containing different waveband spectral information is obtained to provide a data basis for subsequent spectral analysis. The multi-spectral image set is processed to extract the multi-waveband spectral information corresponding to each image pixel point. For each pixel point, a feature vector containing different spectral waveband values is constructed for subsequent spectral analysis. After this step, detailed image spectral feature data is obtained, which lays the foundation for regional color difference analysis. The spectral feature vectors are used to calculate the color difference values between different pixel points. The entire image is divided into several regions of interest, and the average color difference within each region is calculated to obtain color difference data between different image regions, which provides a reference basis for subsequent pigment distribution analysis. The regional color difference data is used to perform difference decomposition on the color space of the multi-spectral image to determine the key color channels reflecting the hair pigment distribution, and regional features are extracted for these channels to generate feature data describing the hair pigment distribution in different image regions, which provides a basis for subsequent pigment distribution analysis. Mathematical models such as Gaussian mixture models are used to perform fitting analysis on the hair pigment region feature data. Through the fitting process, the distribution characteristics of the hair pigment in different regions are determined, including peak position, distribution width, etc. The fitting results are visualized to construct an atlas describing the pigment distribution of the entire hair sample, which provides an intuitive reference for subsequent pigment analysis.

[0101] In this embodiment, referring to FIG. 3, it is a detailed implementation step flow diagram of step S2. The detailed implementation steps of step S2 in this embodiment include:

[0102] Step S21: Perform hair root recognition on the skin multi-spectral image set to mark the hair root region;

[0103] Step S22: Perform brightness enhancement processing on the hair root region to obtain a brightness-enhanced hair image;

[0104] Step S23: Perform hair contour analysis on the brightness-enhanced hair image to extract a hair contour line;

[0105] Step S24: Perform contour expansion processing on the hair contour line to obtain a hair contour expansion image;

[0106] Step S25: Perform contour detail feature analysis on the hair contour detail expansion image to generate a hair contour detail feature;

[0107] Step S26: Perform detail image reconstruction on the brightness-enhanced hair image according to the hair contour detail feature to construct a detail-reconstructed skin and hair image.

[0108] In this embodiment, image processing techniques such as edge detection, region growing, etc. are used to analyze and process the multispectral image to determine the position and range of the hair root in the image, mark and segment the hair root region, and provide targeted area positioning for subsequent brightness enhancement and contour analysis. For the identified hair root region, local brightness adjustment and enhancement processing is performed, and image processing algorithms such as histogram equalization, adaptive filtering, etc. are used to improve the overall brightness of the hair region. Through brightness enhancement, hair details are highlighted to provide a better image basis for subsequent contour analysis. Edge detection, shape analysis, etc. are used to process the brightness-enhanced hair image to identify and extract the contour line of the hair in the image, obtaining contour information describing the shape and details of the hair. On the basis of the extracted hair contour line, pixel-level contour expansion processing is performed, and morphological operations such as dilation and erosion are used to moderately expand the width and range of the contour line. The expanded contour line better covers the overall shape and detail information of the hair. For the hair contour expansion image, shape, texture, etc. features of the contour line are extracted, and digital image processing techniques such as texture analysis, shape descriptor, etc. are used to quantitatively describe the detail features of the hair contour. These detail features reflect the microscopic structure of the hair and provide a basis for subsequent image reconstruction. The extracted hair contour detail features are used to reconstruct the brightness-enhanced hair image, and the detail feature information is integrated into the original image to generate a reconstructed image with more details. On the basis of preserving the brightness information of the original image, the microscopic details of the hair are restored and strengthened.

[0109] In this embodiment, referring to FIG. 4, a detailed implementation step flowchart of step S3 is provided. In this embodiment, the detailed implementation step of step S3 includes:

[0110] Step S31: Perform hair follicle fine visual recognition on the detail-reconstructed skin and hair image to mark each hair follicle feature point;

[0111] Step S32: calculating the spatial position of each follicle feature point to obtain the spatial position coordinates of the follicle;

[0112] Step S33: calculating the inter-follicle distance according to the spatial position coordinates of the follicle to obtain a plurality of inter-follicle distances;

[0113] Step S34: analyzing the follicle density distribution of each follicle feature point based on the plurality of inter-follicle distances to generate follicle density distribution data;

[0114] Step S35: mapping the density trend of the follicle density distribution data to construct a follicle density distribution trend model.

[0115] In this embodiment, the computer vision technology, such as the target detection algorithm of deep learning, is used to analyze the generated detail reconstruction image, identify the position and features of each follicle in the image, and mark the feature points of each follicle on the image to accurately locate each independent follicle in the image. According to the marked follicle feature points, the two-dimensional or three-dimensional spatial coordinates of each follicle in the image coordinate system are calculated. The image processing technology, such as feature point matching and projection transformation, is used to map the follicle feature points to the actual spatial coordinate system to obtain the accurate position information of each follicle on the skin surface. The Euclidean distance between any two follicles is calculated using the obtained follicle spatial coordinates. The distance data between each follicle is obtained by traversing all the follicle feature point pairs in the image. A set of distance data describing the distribution regularity of the follicles in the image is output. The density of the follicles within a certain range around each follicle feature point is calculated. The kernel density estimation method is used to estimate the follicle distribution density in the local area according to the distance data to obtain the density distribution data of each follicle position, which reflects the distribution characteristics of the follicles in the image. The local density distribution data is mapped to the density distribution trend of the entire image. An interpolation and smoothing technique is used to construct a continuous follicle density distribution trend model to describe the overall trend of the follicle distribution in the entire skin area.

[0116] In this embodiment, step S4 includes the following steps:

[0117] Step S41: performing image segmentation processing on the detail reconstruction skin hair image to obtain a plurality of image regions;

[0118] Step S42: performing regional hair feature change trend analysis on the plurality of image regions based on the skin pigment distribution map and the follicle density distribution trend model to generate feature change trends of different image regions;

[0119] Step S43: performing change trend similarity analysis on the feature change trends of different image regions to generate regional change trend similarity;

[0120] Step S44: According to the similarity of the change trend, the detail reconstructed skin hair image is clustered and divided into feature regions, thereby obtaining a plurality of change trend feature regions.

[0121] In this embodiment, the generated detail reconstructed image is divided into a plurality of independent regions by using image segmentation techniques such as region growing, image edge detection and the like. Each region should have relatively independent visual features, reflecting the skin hair conditions of different parts, and a set of representative image regions is obtained. The skin pigment distribution atlas is combined with the constructed hair follicle density distribution trend model to analyze the hair feature changes in each image region. For each region, the change trend of the hair pigment content, hair follicle density and the like is evaluated, and the overall change of the hair features in the region is described to obtain a set of trend data representing the change rules of the hair features of different image regions. The change trend similarity between any two regions is calculated by using the feature change trend data of each region. Statistical methods such as correlation coefficient, cosine similarity and the like are used to quantitatively describe the similarity degree of the feature change trends between different regions, and the regions with similar feature change trends in the image are found. According to the similarity matrix between regions, a clustering algorithm such as spectral clustering, hierarchical clustering and the like is used to cluster the image regions. The goal of clustering is to classify regions with similar feature change trends into the same class, form a plurality of change trend feature regions, and output a set of representative feature regions, reflecting the change trends of the hair features of different parts in the entire skin image.

[0122] In this embodiment, the specific steps of step S5 are as follows:

[0123] Step S51: Multi-scene growth time evolution simulation is performed on the plurality of change trend feature regions to generate multi-scene time evolution simulation data;

[0124] Step S52: Growth time response rate analysis is performed on the multi-scene time evolution simulation data to obtain hair growth time response rate data;

[0125] Step S53: Growth trend change prediction is performed on the hair growth time response rate data to generate hair growth trend prediction data;

[0126] Step S54: Stage evolution trajectory analysis is performed on the hair growth trend prediction data to construct a hair region growth trend evolution curve.

[0127] In this embodiment, a hair growth time sequence evolution model is constructed for each region. The model should consider the variation law of the hair features (such as color, density) in the region over time, and be able to simulate the growth trend changes under different environmental conditions. Through parameter adjustment of the model, hair growth time sequence evolution data under various environmental conditions (such as nutrient level, hormone level, etc.) is generated. The response rate of the hair growth features (such as color, density) in each feature region to environmental changes is calculated. The response rate is evaluated by analyzing the change trend of the time series data, such as calculating the slope, derivative, etc. of the feature change, to quantitatively describe the sensitivity of the hair growth in different regions to environmental changes. Combined with existing environmental change prediction information, the change trend of the hair growth features in each feature region in the future is predicted. Time series analysis, machine learning, etc. are used to generate prediction data of the future growth trend according to the historical change pattern and external environment prediction. The evolution trajectory of the hair growth features in each feature region over time is analyzed. Curve fitting, etc. is used to describe the change law of the hair growth condition at different time stages, and a curve representing the evolution of the hair growth trend in different feature regions is obtained.

[0128] In this embodiment, the specific steps of step S51 are:

[0129] The growth trend of the multiple change trend feature regions is deduced, thereby obtaining the growth trend law of different regions;

[0130] The multi-scene growth environment parameters are defined;

[0131] Based on the multi-scene growth environment parameters, time-domain growth numerical simulation calculation is performed on the growth trend law of different regions to obtain the region growth trend parameters of multiple scenes;

[0132] Based on the region growth trend parameters of multiple scenes, multi-time-point growth evolution simulation is performed to generate multi-scene time sequence evolution simulation data.

[0133] In this embodiment, the change law of the hair growth characteristics (such as color, density, etc.) of each region over time is analyzed, time series analysis, curve fitting and other methods are used to describe the change trend of the hair growth characteristics in each region, and the change law of the hair growth characteristics in different regions is obtained. According to the actual application situation, a set of representative growth environment parameters such as nutrient level, hormone level, temperature and humidity are defined, and multiple value ranges are set for each parameter to cover different environmental conditions. A series of growth environment scenarios are constructed, and the time-domain growth numerical simulation calculation is performed according to the growth trend of each region and the set multi-scene growth environment parameters. For each characteristic region and each growth environment scenario, the change trend of the hair growth characteristics in a future period of time is calculated, and the differential equation, difference equation and other numerical simulation methods are used to simulate the dynamic changes in the growth process. The growth trend parameters of different regions under different environmental conditions are obtained, and the multi-time-point growth evolution simulation is performed according to the growth trend parameters of each region under different environmental scenarios. Through the adjustment of the model parameters, the simulation data sequence of the hair growth characteristics over time under different environmental conditions is generated, and a set of representative hair growth time series evolution data is obtained, which provides a basis for subsequent response rate analysis and situation prediction.

[0134] In this embodiment, the specific steps of step S6 are:

[0135] Step S61: Perform cycle fluctuation amplitude identification on the hair region growth trend evolution curve to extract the trend curve cycle fluctuation amplitude.

[0136] Step S62: Perform trend peak time point calculation on the hair region growth trend evolution curve to generate a plurality of trend peak time points.

[0137] Step S63: Perform hair growth condition evaluation on the plurality of trend peak time points and the trend curve cycle fluctuation amplitude to obtain the growth condition evaluation result.

[0138] Step S64: Visualize and screen the growth condition evaluation result and the hair region growth trend evolution curve to the detector display device to finely present the hair analysis result.

[0139] In this embodiment, according to the hair growth time series evolution simulation data, the growth trend curve of each region is analyzed in time series, the periodic fluctuation characteristics in the growth trend curve are identified by using signal processing methods such as Fourier transform and wavelet transform, the amplitude of each periodic fluctuation is quantified as an important indicator for evaluating the growth condition, the peak time point in the growth trend curve is identified according to the hair growth time series evolution data, the peak time point represents the time node at which the hair growth rate reaches the maximum, and is another important indicator for evaluating the growth condition, the peak time point of the growth trend curve of each region is obtained by using curve extreme value analysis and sliding window method, the obtained indicators are comprehensively obtained, a hair growth condition evaluation model is established, an evaluation algorithm based on fuzzy logic and neural network is designed, the periodic fluctuation amplitude and the peak time point are mapped to the excellent degree of the growth condition, the evaluation result of the hair growth condition of each region is obtained, and a basis is provided for subsequent visual display. The obtained growth condition evaluation result and the hair growth trend evolution curve are presented in an intuitive graphical manner on the detector display device, common visual charts such as column chart and line chart are used to display the growth condition evaluation score and the growth trend curve of different regions, an interactive visual interface is made by using the screen characteristics of the detector display device, and the user can conveniently check and analyze the hair growth state.

[0140] In another aspect, the embodiment also provides a skin and hair analysis system based on a mobile terminal projection interconnection detector. As shown in FIG. 5, the mobile terminal projection interconnection detector is a structural schematic diagram. The mobile terminal includes mobile smart devices such as mobile phones, tablets, notebook computers, and wearable devices. The detector includes a detector probe end and a detector display device. Real-time data transmission is performed between the mobile terminal and the detector through wired or wireless networks. Optionally, real-time data transmission can also be performed between the detector probe end and the detector display device through wired or wireless networks. It should be noted that the network includes Wi-Fi Direct, broadband wide area network, local area network, cellular network, Ethernet, radio frequency, and satellite communication, etc. That is, the mobile terminal and the detector both have the functions of wireless data transmission and Internet access. In some scenarios, the detector can push service information to the user through direct Internet access, or be used to receive push information from the mobile terminal to improve user experience. The mobile terminal is used to execute the algorithm program of any one of the skin and hair analysis methods based on the mobile terminal projection interconnection detector in the foregoing embodiments. In this embodiment, the mobile terminal and the display device are wirelessly connected through Mira cast using Wi-Fi Direct technology. Wi-Fi Direct is used to make point-to-point Wi-Fi connection between devices without connecting a router, so as to realize fast wireless projection. The mobile terminal and the detector are wirelessly connected through Bluetooth or Wi-Fi to realize real-time data transmission and wireless control between the mobile terminal and the detector within a short distance. In one embodiment, the detector can also be used alone to view the surface state of skin and hair. When more computing power is needed to deeply analyze the state data of skin and hair, the detector probe end is used to provide captured various skin and hair multi-spectral image data to the mobile terminal for analysis, and the analysis data and results are transmitted in real time to the detector display device through the wireless network for display. Further, the skin and hair analysis system based on the mobile terminal projection interconnection detector specifically includes:

[0141] A pigment distribution module is configured to acquire a skin multi-spectral image set based on a detector probe end camera; perform multi-image spectral quantitative analysis on the skin multi-spectral image set; and construct a skin pigment distribution atlas.

[0142] An image reconstruction module is configured to perform contour detail feature analysis on the skin multi-spectral image set; perform detail image reconstruction; and construct a detail reconstructed skin and hair image.

[0143] A density distribution module is configured to perform follicle fine visual recognition on the detail reconstructed skin and hair image, so as to obtain follicle spatial position coordinates; perform follicle density distribution analysis based on the follicle spatial position coordinates; and construct a follicle density distribution trend model.

[0144] The feature change trend module is configured to analyze the regional hair feature change trend based on the skin pigment distribution atlas and the hair follicle density distribution trend model, and perform feature region clustering division, so as to obtain a plurality of change trend feature regions.

[0145] The growth trend evolution module is configured to simulate the multi-scene growth time sequence evolution of the plurality of change trend feature regions, and predict the growth trend change, so as to construct a hair region growth trend evolution curve.

[0146] The condition assessment module is configured to assess the hair growth condition based on the hair region growth trend evolution curve, so as to obtain a growth condition assessment result; and perform visual projection based on the growth condition assessment result, and present the hair analysis result.

[0147] The present application can accurately capture the pigment distribution of skin and hair by performing multi-image spectral quantitative analysis on a multi-spectral image set, construct a skin pigment distribution atlas, and provide basic data for subsequent analysis. The present application can also provide clearer and more accurate skin and hair images by performing contour detail feature analysis on the multi-spectral image set and reconstructing the detail images, and provide more reliable data basis for subsequent analysis by performing fine visual recognition and density distribution analysis on the hair follicle, obtaining the spatial position coordinates and density distribution trend model of the hair follicle, and providing more detailed data support for subsequent analysis by in-depth understanding of the distribution of the hair follicle. The present application can also realize regional hair feature change trend analysis and regional clustering division by analyzing the skin pigment distribution atlas and the hair follicle density distribution trend model, discover the feature change law of different regions, provide a basis for subsequent growth trend evolution, construct a hair region growth trend evolution curve by simulating the multi-scene growth time sequence evolution and predicting the growth trend change, in-depth understand the hair growth condition, provide prediction of future growth trend, and provide strong support for condition assessment. The present application can also obtain the growth condition assessment result by assessing the hair growth condition based on the hair region growth trend evolution curve, provide personalized care suggestions according to the assessment result, present the hair analysis result by visual projection, help users understand their own hair health condition, and further improve the care scheme.

[0148] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the present application being defined by the appended claims rather than the above description, and it is intended to encompass all variations falling within the meaning and scope of the equivalent elements of the application file.

[0149] The foregoing merely illustrates the principles of the application and application of its principles. Various modifications and alterations to this implementation will occur to those skilled in the art. The scope of the application should be determined, however, by the following claims rather than by the embodiments shown.

Claims

1. A skin and hair analysis method based on a mobile terminal screen projection interconnection detector, characterized in that, The mobile terminal and the detector based on real-time communication are realized, wherein the detector comprises a detector probe end and a detector display device, and the mobile terminal is used to provide data analysis computing power for the detector, and specifically comprises the following steps: Step S1: obtaining a skin multispectral image set based on a detector probe end camera; performing multi-image spectral quantitative analysis on the skin multispectral image set to construct a skin pigment distribution atlas; Step S2: performing contour detail feature analysis on the skin multispectral image set and performing detail image reconstruction to construct a detail reconstructed skin hair image; Step S3: performing fine visual recognition on the detail reconstructed skin hair image to obtain hair follicle spatial position coordinates; performing hair follicle density distribution analysis according to the hair follicle spatial position coordinates to construct a hair follicle density distribution trend model; Step S4: performing regional hair feature change trend analysis on the skin pigment distribution atlas and the hair follicle density distribution trend model, and performing feature region clustering and division to obtain a plurality of change trend feature regions; Step S5: performing multi-scene growth time sequence evolution simulation on the plurality of change trend feature regions, and performing growth trend change prediction to construct a hair region growth trend evolution curve; Step S6: performing hair growth condition evaluation on the hair region growth trend evolution curve to obtain a growth condition evaluation result; and performing visual projection on the detector display device based on the growth condition evaluation result to present a hair analysis result. 2.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 1, wherein, Step S1 specifically comprises the following steps: Step S11: obtaining a skin multispectral image set based on a detector probe end camera; Step S12: performing multi-image spectral quantitative analysis on the skin multispectral image set to extract image spectral feature vectors; Step S13: performing regional color difference analysis on the image spectral feature vectors to obtain color difference data of different image regions; Step S14: performing color space difference decomposition on the skin multispectral image set according to the color difference data of different image regions to generate skin pigment region feature data; Step S15: performing pigment distribution fitting on the skin pigment region feature data to construct a skin pigment distribution atlas. 3.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 1, wherein, Step S2 specifically comprises the following steps: Step S21: performing hair root recognition on the skin multispectral image set to mark hair root regions; Step S22: performing brightness enhancement processing on the hair root regions to obtain brightness enhanced hair images; Step S23: performing hair contour analysis on the brightness enhanced hair images to extract hair contour lines; Step S24: performing contour expansion processing on the hair contour lines to obtain hair contour expansion images; Step S25: performing contour detail feature analysis on the hair contour detail expansion images to generate hair contour detail features; Step S26: performing detail image reconstruction on the brightness enhanced hair images according to the hair contour detail features to construct a detail reconstructed skin hair image. 4.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 1, wherein, Step S3 specifically comprises the following steps: Step S31: performing fine visual recognition on the detail reconstructed skin hair image to mark each hair follicle feature point; Step S32: performing spatial position calculation on each hair follicle feature point to obtain hair follicle spatial position coordinates; Step S33: Calculate the inter-follicle distance according to the spatial position coordinates of the hair follicles, thereby obtaining a plurality of inter-follicle distances; Step S34: Perform follicle density distribution analysis on each follicle feature point based on the plurality of inter-follicle distances, to generate follicle density distribution data; Step S35: Perform density trend mapping on the follicle density distribution data, to construct a follicle density distribution trend model. 5.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 1, wherein, The specific steps of step S4 are: Step S41: Perform image segmentation processing on the detail-reconstructed skin hair image, to obtain a plurality of image regions; Step S42: Perform regional hair feature change trend analysis on the plurality of image regions based on the skin pigment distribution atlas and the follicle density distribution trend model, to generate feature change trends of different image regions; Step S43: Perform change trend similarity analysis on the feature change trends of different image regions, to generate regional change trend similarity; Step S44: Perform feature region clustering division on the detail-reconstructed skin hair image according to the regional change trend similarity, to thereby obtain a plurality of change trend feature regions. 6.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 1, wherein, The specific steps of step S5 are: Step S51: Perform multi-scenario growth time sequence evolution simulation on the plurality of change trend feature regions, to generate multi-scenario time sequence evolution simulation data; Step S52: Perform growth time sequence response rate analysis on the multi-scenario time sequence evolution simulation data, to obtain hair growth time sequence response rate data; Step S53: Perform growth trend change prediction on the hair growth time sequence response rate data, to generate hair growth trend prediction data; Step S54: Perform stage evolution trajectory analysis on the hair growth trend prediction data, to construct a hair region growth trend evolution curve. 7.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 6, wherein, The specific steps of step S51 are: Perform growth trend deduction on the plurality of change trend feature regions, to thereby obtain growth trend rules of different regions; Define multi-scenario growth environment parameters; Perform time-domain growth numerical simulation calculation on the growth trend rules of different regions based on the multi-scenario growth environment parameters, to obtain regional growth trend parameters of a plurality of scenarios; Perform multi-time-point growth evolution simulation based on the regional growth trend parameters of the plurality of scenarios, to generate multi-scenario time sequence evolution simulation data. 8.The skin and hair analysis method based on the mobile terminal screen projection interconnection detector of claim 1, wherein, The specific steps of step S6 are: Step S61: Perform cycle fluctuation amplitude identification on the hair region growth trend evolution curve, to extract the trend curve cycle fluctuation amplitude; Step S62: Perform trend peak time point calculation on the hair region growth trend evolution curve, to generate a plurality of trend peak time points; Step S63: Perform hair growth condition evaluation on the plurality of trend peak time points and the trend curve cycle fluctuation amplitude, to thereby obtain a growth condition evaluation result; Step S64: Perform visual screen projection of the growth condition evaluation result and the hair region growth trend evolution curve on a detector display device, to finely present the hair analysis result.

9. A skin hair analysis system based on a mobile terminal screen projection interconnection detector, characterized in that, The skin and hair analysis system based on the mobile terminal projection interconnection detector comprises a mobile terminal and a detector, and real-time data transmission is performed between the mobile terminal and the detector through wired or wireless networks. The mobile terminal is used to execute an algorithm program containing the skin and hair analysis method based on the mobile terminal projection interconnection detector as claimed in any one of claims 1-8. The detector comprises a detector detection end and a detector display device. The detector detection end is used to acquire multispectral image data of skin and hair. The detector display device is used to synchronize the screen of the mobile terminal and display network push information. The skin and hair analysis system based on the mobile terminal projection interconnection detector specifically comprises: A pigment distribution module is used to acquire a skin multispectral image set based on a detector detection end camera, perform multispectral quantitative analysis on the skin multispectral image set, and construct a skin pigment distribution atlas. An image reconstruction module is used to perform contour detail feature analysis on the skin multispectral image set, perform detail image reconstruction, and construct a detail reconstructed skin and hair image. A density distribution module is used to perform follicle fine visual recognition on the detail reconstructed skin and hair image, thereby obtaining follicle spatial position coordinates, perform follicle density distribution analysis according to the follicle spatial position coordinates, and construct a follicle density distribution trend model. A feature change trend module is used to perform regional hair feature change trend analysis on the skin pigment distribution atlas and the follicle density distribution trend model, perform feature region clustering and division, and thereby obtain a plurality of change trend feature regions. A growth trend evolution module is used to perform multi-scene growth time series evolution simulation on the plurality of change trend feature regions, perform growth trend change prediction, and construct a hair region growth trend evolution curve. A condition evaluation module is used to perform hair growth condition evaluation on the hair region growth trend evolution curve, thereby obtaining a growth condition evaluation result, perform visual projection based on the growth condition evaluation result, and present a hair analysis result.

Images