Clothing Data Analysis Methods and Systems Based on Smart Wearables
Through dynamic modeling and simulation analysis, the optimal wearing position of smart wearable devices is recommended, solving the problems of device detachment and structural damage in existing technologies, and achieving stable signal acquisition and user comfort under different postures.
Patent Information
- Application Number
- CN202611282107.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-24
- Publication Date
- 2026-11-17
AI Technical Summary
The wearing position of existing smart wearable devices is usually determined by user experience, without taking into account changes in the user's body shape in different scenarios. This leads to device detachment, structural damage, and unstable signal acquisition, failing to balance structural durability, functional effectiveness, and user comfort.
By acquiring data on user clothing and equipment, a dynamic body model is established to simulate typical movement sequences, calculate the force and deformation distribution of clothing at different locations, analyze structural damage risk, functional effectiveness, and user comfort, and recommend the optimal wearing position.
It achieves precise adaptation under different postures, balances structural safety and functional efficiency, avoids stress concentration, and improves the durability of the equipment and user comfort.
Smart Images

Figure CN123115518A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of smart wearables, specifically to a method and system for analyzing clothing data based on smart wearables. Background Technology
[0002] With the development of the Internet of Things and flexible electronics technology, smart wearable devices (such as physiological monitoring patches, motion sensors, and smart fabrics) have been widely used in military training, sports rehabilitation, medical monitoring, and daily health management. However, the current wearing position of smart wearable devices is usually determined by the user's experience or general recommendations in the product manual (e.g., heart rate monitors are usually worn on the left chest, and gait sensors are worn on the outside of the thigh).
[0003] Traditional solutions often provide a one-size-fits-all, fixed wearing position, failing to consider the changes in a user's body shape under different circumstances (such as limb deformation during movement, bending, and twisting). Smart devices are typically attached to or embedded in clothing. Existing technologies do not adequately consider the dynamic stress and deformation distribution generated by clothing when users perform typical movements (such as running, squatting, and stretching). This can lead to: devices detaching or shifting due to excessive stretching of the clothing, affecting the stability of signal acquisition (decreased functional effectiveness); localized stress concentration at the connection between the device and the clothing, accelerating structural damage to the clothing or device (structural fatigue failure); and existing technologies often determine the wearing position only from a single dimension (such as the optimal physiological signal acquisition position), without systematically balancing the conflicting relationships between the structural durability of the clothing / device (damage risk), the effectiveness of device function (such as sensor alignment), and the user's long-term wearing comfort. For example, a theoretically optimal signal position may lead to rapid device damage or severe skin discomfort due to repeated folding, stretching, or friction of the clothing at that location. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this application proposes a method and system for clothing data analysis based on smart wearables.
[0005] To achieve the above objectives, this application provides the following technical solution: The data analysis method and system for clothing based on smart wearables includes the following specific steps: S1: Acquire the target user's clothing data, the target smart wearable device's device data, and the user's dynamic body model in a specific attendance scenario; S2: Based on the dynamic body model, simulate the user's typical action sequence in the specific attendance scenario, and calculate the dynamic force and deformation distribution of the clothing at various positions on the body under the typical action sequence; S3: Define multiple candidate wearing positions on the garment, and analyze the garment structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position. S4: Based on the clothing structure damage risk value, the functional effectiveness index, and the user comfort index corresponding to each candidate wearing position, calculate and output the recommended optimal wearing position of the target smart wearable device for the target user in the specific attendance scenario.
[0006] Preferably, the steps of acquiring the target user's clothing data, the target smart wearable device's device data, and the user's dynamic body model in a specific attendance scenario include the following specific steps: S11. Using a 3D scanner, acquire high-precision 3D point cloud data of the clothing selected by the target user in a natural hanging state, generate a surface mesh model, and assign material properties and identify weak points in the mesh model. S12. Obtain the three-dimensional geometric model of the equipment, including curvature, contact area and protrusion features. Obtain the physical property data of the equipment, including total mass, center of mass position, and static preload distribution generated by the external fixing method of the equipment (such as magnetic buckle, elastic clip or lock). The preload can be obtained by measuring on a standard flat plate material through a pressure mapping sensor and converted into a force vector field. Obtain the dynamic excitation spectrum characterizing the operating state of the equipment itself. The operation of the equipment is not limited to active vibration (such as motor, speaker, optical image stabilization component), but also includes the inertial reaction caused by human movement. Collect the working frequency, amplitude and direction of the internal moving parts of the equipment, as well as the inertial force time sequence template caused by the equipment's own mass and rotational inertia at different acceleration levels. All dynamic excitation spectra are normalized into a set of time-related multi-channel excitation signals as input for subsequent operating state and micro-displacement simulation. S13. Using sensing devices, the three-dimensional shape and skeletal dimensions of the user's body are collected while maintaining a standard anatomical posture. A static digital human body model with standard skeletal binding and a 1:1 reproduction of the user's body size is generated. Combined with bioelectrical impedance analysis, the soft tissue thickness and stiffness coefficients of various body regions (such as chest, waist, and thighs) are obtained. Non-uniform elastic support characteristics are assigned to the surface shell units of the digital human body model, and a skin deformation resistance model is established. A specific attendance scene is a collection of ordered typical movements. The user must first select a scene template, call a pre-set general action library, and adapt it to the user's actual movement characteristics. The user uploads their own... In everyday videos within a scene, a pose estimation algorithm is used to extract skeletal motion sequences, resulting in time-stamped joint angle temporal data covering the main activity range of the scene—a personalized typical action sequence. This acquired joint angle temporal data is then used to drive a static digital human body model after skeletal binding in real time. A dual quaternion skinning algorithm is employed to allow the skin surface to naturally bulge, stretch, and slide during body movement. The instantaneous spatial position, velocity, acceleration, and local strain tensor caused by skin stretching at each vertex of the body surface are updated in real time. The resulting dynamic body model outputs the geometry of the entire body surface, three-dimensional forces (including reaction forces applied by clothing and equipment), and deformation field.
[0007] Preferably, based on the dynamic body model, simulating the user's typical action sequence in the specific attendance scenario, and calculating the dynamic force and deformation distribution of the clothing at various body positions under the typical action sequence includes the following specific steps: S21. From the typical action sequence of the dynamic body model built in step S13, extract the first posture as the standard anatomical standing posture. The clothing digital twin (mesh model with weak position map) generated in step S11 is virtually worn on the static body model through automatic alignment and penetration processing algorithm. The wearing process simulates the spatial relationship between clothing and body, and obtains a coupled initial state in the initial frame where clothing and body fit perfectly and there is no penetration. The initial stress of each vertex on the clothing model is set to zero. S22. Using the joint angle timing data obtained in step S13, the body's skeletal movement is driven, and the mesh on the body surface is displaced and deformed accordingly. Under the combined action of body drive, gravity, its own inertial force and contact force, the clothing model generates a continuous dynamic response. The simulation uses a nonlinear finite element method cloth solver to simulate the mechanical state of the clothing from the initial posture through a series of actions until the end of the sequence. S23. For each time step, the contact pressure vector of the inner surface of the garment covered by the integral calculation is obtained to obtain the resultant force curve and the maximum pressure envelope surface that change with time. The tensile force of the garment itself on the edge of the region is extracted. The average in-plane elongation rate, the direction of the maximum principal strain, and the local area change rate of the garment fabric in the region are statistically analyzed. For the marked weak positions, the equivalent stress-time curve of these positions in the whole action sequence, the time of peak stress occurrence and the corresponding body posture are output. At the same time, the stress concentration factor is calculated, that is, the ratio of the actual peak stress at the weak position to the average stress of the surrounding area. By extraction, a complete dynamic force and deformation distribution map covering all positions of the body is obtained, which is recorded as the baseline garment mechanical heat map. The heat map binds the mechanical load of the garment caused by dynamic movement with the body position analysis. A set of candidate positions is preset. For the local area of the garment surface corresponding to each candidate position, the following are extracted from the baseline mechanical heat map: the force component time sequence, the deformation gradient time sequence, the relative displacement field with the surrounding area, and the stress time sequence of all weak positions in the region in the whole action sequence.
[0008] Preferably, the step of defining multiple candidate wearing positions on the garment and analyzing the garment structural damage risk, functional effectiveness index, and user comfort index corresponding to each candidate wearing position includes the following specific steps: Defining multiple candidate wearing positions on the garment, and performing the following analysis for each candidate wearing position: S31: Combining the device data, simulating the operating state and / or micro-displacement of the target smart wearable device at the candidate wearing position and / or relative to the garment; S32: Analyzing the operating state and / or micro-displacement simulated in step S31, and comparing it with the dynamic stress and deformation distribution of the garment in the corresponding candidate wearing position area obtained in step S2, to calculate the garment structural damage risk caused by the candidate wearing position; wherein, the garment structural damage risk is related to the weak position overload index, alternating stress fatigue damage accumulation index, micro-displacement friction and wear index, and stress concentration deterioration degree of the weak position of the garment; S33: Evaluating the functional effectiveness and user comfort of the target smart wearable device at the candidate wearing position. Specifically, S31 includes: extracting the analytical geometric contour of the device's bottom surface, positioning the bottom surface on the outer surface area of the garment corresponding to the current candidate wearing position, making initial contact between the device's bottom surface and the garment's outer surface, and simulating the attachment method based on the actual fixing type of the device: establishing a magnetic attraction field that decays with distance between the device's bottom surface and the magnetic accessories on the inside of the garment, so that the garment is clamped in the middle; after the initial attachment is completed, the device maintains a stable fit with the garment in the absence of external force, and recording the device's initial pose matrix.
[0009] The specific content of S32 includes: the equipment simultaneously bears two types of dynamic loads, including human motion-induced excitation and internal self-excitation of the equipment. The two types of loads are synchronously superimposed in time, forming a resultant force and resultant torque acting on the center of mass of the equipment. The instantaneous acceleration of the clothing surface point at the candidate wearing position. Since the equipment has a certain mass and rotational inertia, when the clothing surface accelerates, the equipment will generate an inertial tendency to move relative to the clothing surface. Based on the equipment mass, center of mass position and rotational inertia of the three main axes, the inertial force and inertial torque generated by the equipment due to translational acceleration and angular acceleration are calculated. If the target equipment contains moving parts, then according to the dynamic excitation spectrum collected in step S12, periodic or transient disturbance forces corresponding to the working state of the moving parts are applied at the center of mass of the equipment. These disturbance forces have specific frequencies, amplitudes and directions.
[0010] Treating the equipment as a rigid body and the local area of the clothing as a deformable body, a coupled mechanical model of equipment-clothing-body is constructed. At each time step, the following equations are solved simultaneously: Equipment rigid body motion equation: Under the combined action of external forces, contact forces, frictional forces, and adhesion constraints, the translational displacement and angular displacement about the center of mass of the equipment are calculated; Clothing local deformation equation: Under the action of contact pressure from the equipment, its own tension, and contact reaction force from the body, the local area of the clothing undergoes in-plane stretching, bending, and out-of-plane wrinkling deformation; The solution process requires real-time monitoring of changes in the contact state between the equipment and the clothing, including adhesion, sliding, and separation. Feature extraction is performed on the time-series displacement data of the equipment relative to the clothing to generate a micro-displacement feature vector of the equipment attachment, which includes the maximum tangential slip amplitude, normal runout amplitude, peel angle amplitude, proportion of attachment failure duration, and contact pressure redistribution spectrum.
[0011] For the clothing grid area covered by the current candidate position, on the basis of the existing body contact force and clothing internal stress, the surface distribution pressure applied by the bottom of the equipment and the line distribution clamping force of the equipment clamping edge are superimposed. If there is a micro displacement of the equipment, the area of action of the contact pressure will migrate on the clothing surface over time. The movement path of the pressure is tracked according to the time step and loaded onto the clothing unit at the corresponding position. For the local area of the clothing after the force superposition, the stress field and strain field after each time step are obtained. The structural damage risk value of clothing is related to the stress concentration at weak points in the clothing. It is calculated using a weighted combination of the weak point overload index, alternating stress fatigue damage accumulation index, micro-displacement friction and wear index, and the increase in the weak point stress concentration coefficient. Specifically, for each weak point within the current candidate location area and its influence neighborhood, the ratio of its maximum equivalent stress in the entire action sequence to the damage tolerance threshold recorded in step S11 is calculated. The weighted maximum value of all weak point ratios is taken as the weak point overload index. For each weak point, the stress cycle amplitude and mean stress in its equivalent stress-time curve are statistically analyzed using the rainflow counting method. Combined with the SN fatigue curve of the clothing material, the equivalent fatigue damage degree within a single action sequence cycle is calculated using Miner's linear cumulative damage rule. The equivalent fatigue damage degree is the standardized fatigue life. Finally, the alternating stress fatigue damage accumulation index is calculated using all... The maximum value of the equivalent fatigue damage at the weak position is calculated using the tangential slip amplitude and normal pressure time history obtained in step S31. The frictional work density between the bottom surface of the equipment and the surface of the clothing is calculated, which is the density of energy generated by friction over a long period of time. Combined with the abrasion resistance parameter of the fabric, where the abrasion resistance parameter is the wear coefficient, which is the volume wear caused by sliding a unit distance under a unit load, the micro-displacement friction wear index is calculated by weighting. The stress concentration coefficient increase at the weak position is calculated by dividing the baseline stress concentration coefficient obtained in step S23 by the stress concentration coefficient after superimposing the equipment to obtain the stress concentration deterioration degree. The four indicators of weak position overload index, alternating stress fatigue damage accumulation index, micro-displacement friction wear index, and stress concentration deterioration degree are normalized and merged into a comprehensive clothing structure damage risk value through a preset weight matrix. A structured damage risk assessment report is generated for the current candidate wearing position. S33 specifically includes: evaluating the functional effectiveness and user comfort indicators of the target smart wearable device at the candidate wearing position; calculating the sensor field-of-view occlusion rate; calculating the contact angle between the sensor window and the skin surface, the gap, and the proportion of the field of view obscured by clothing fabric at the current candidate position using the target smart wearable device; simultaneously, using the device pose data from step S31, checking the angle offset between the sensor window normal and the skin surface normal, as well as the average distance between the window and the skin, step by step; comparing these data with corresponding data threshold ranges; defining valid sampling as when all data are within the threshold range; where valid sampling is the sampling process where the accuracy of sensor sampling data meets the requirements; and defining the effective sampling time ratio as the time required for normal sensor operation. The ratio of the time interval to the field of view occlusion rate is calculated as 1 - the effective sampling time ratio. Using the contact pressure distribution map from step S31, the maximum pressure peak in the action sequence is identified. The peak value is divided by the pain pressure threshold of each part of the body to obtain the probability of exceeding the pressure threshold. Simultaneously, the shear deformation caused to the skin by the device edge is assessed. The sum of the shear deformation during wear is divided by the corresponding shear deformation safety value to obtain the local pressure discomfort index. Combining the device's bottom surface area and material thermal resistance, as well as the basic sweating rate data of the body area, the standardized and weighted summation yields the heat and moisture accumulation index of the device coverage area. After normalization, the probability of exceeding the pressure threshold, the local pressure discomfort index, and the heat and moisture accumulation index of the device coverage area are weighted and combined to form a comprehensive user comfort score, ranging from 0 to 1, with higher values indicating greater comfort.
[0012] Preferably, calculating and outputting the recommended optimal wearing position of the target smart wearable device for the target user in the specific attendance scenario includes the following specific steps: For all candidate wearing positions, an evaluation vector is generated based on the risk of clothing structural damage, functional effectiveness index, and user comfort index corresponding to the wearing position. All feature values in the vector are processed by range normalization. Weights are set according to the wearing scenario, and a weighted comprehensive evaluation score is calculated. The comprehensive evaluation scores are sorted, and the position with the highest evaluation score is selected as the best wearing position for recommendation. Among them, the risk of clothing structural damage has a negative weight, while the functional effectiveness index and user comfort index have positive weights.
[0013] The smart wearable-based clothing data analysis system is implemented based on the aforementioned smart wearable-based clothing data analysis method, and specifically includes: The data acquisition module is used to acquire clothing data of the target user, device data of the target smart wearable device, and a dynamic body model of the user in a specific attendance scenario; The simulation module, based on the dynamic body model, simulates the user's typical action sequence in the specific attendance scenario, and calculates the dynamic force and deformation distribution of the clothing at various positions on the body under the typical action sequence; The indicator analysis module is used to define multiple candidate wearing positions on the garment, and analyze the garment structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position. The wearable recommendation module calculates and outputs the optimal wearable position recommended by the target smart wearable device for the target user in the specific attendance scenario, based on the clothing structure damage risk value, the functional effectiveness index, and the user comfort index corresponding to each candidate wearable position.
[0014] An electronic device includes: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the above-described clothing data analysis method based on smart wearables by calling the computer program stored in the memory.
[0015] A computer-readable storage medium is characterized by storing instructions that, when executed on a computer, cause the computer to perform the aforementioned clothing data analysis method based on smart wearables.
[0016] Compared with the prior art, the beneficial effects of this application are: This application fundamentally solves the pain points of traditional fitting solutions that ignore individual differences and rely solely on fixed settings or manual fine-tuning. Its core breakthrough lies in: through dynamic modeling, simulating the mechanical interaction between the user and the clothing in different postures to achieve precise fitting; multi-objective decision-making that takes into account both structural safety and functional efficiency, avoiding sacrificing comfort or causing stress concentration in pursuit of a single indicator. This solution elevates fitting from static estimation to a new level of dynamic simulation and intelligent decision-making. Attached Figure Description
[0017] Figure 1 This is a schematic diagram of the overall process of the clothing data analysis method based on smart wearables in this application; Figure 2 This is a schematic diagram of step S3 of this application; Figure 3 This is a schematic diagram illustrating the coupling mechanics and damage risk assessment in this application; Figure 4 This is a schematic diagram of the overall framework of the clothing data analysis system based on smart wearables in this application. Detailed Implementation
[0018] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some preferred embodiments of this application, and not all embodiments.
[0019] Example 1 Please see Figure 1 One embodiment provided in this application is as follows: Figure 1 As shown, the clothing data analysis method based on smart wearables includes the following specific steps: S1: Acquire the target user's clothing data, the target smart wearable device's device data, and the user's dynamic body model in a specific attendance scenario; S2: Based on a dynamic body model, simulate the typical action sequence of a user in a specific attendance scenario, and calculate the dynamic force and deformation distribution of clothing at various positions on the body under the typical action sequence; S3: Define multiple candidate wearing positions on the clothing, and analyze the clothing structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position. S4: Based on the clothing structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position, calculate and output the recommended best wearing position of the target smart wearable device for the target user in a specific attendance scenario.
[0020] In this embodiment, acquiring the target user's clothing data, the target smart wearable device's device data, and the user's dynamic body model in a specific attendance scenario includes the following specific steps: S11. Using a 3D scanner, acquire high-precision 3D point cloud data of the garment selected by the target user in its natural hanging state, generate a surface mesh model, and restore the original fabric patch splicing information of the garment. Define material properties by region. Based on the fabric patch composition, divide the model into multiple material regions and assign a set of mechanical parameters to each region, including density, elastic modulus, Poisson's ratio, elongation at break, and coefficient of friction. On the mesh model, mark all structural weak points using a combination of automatic recognition and other methods. The definition of weak points is based on garment construction rules and the historical use of the garment. Data includes seams (especially chain stitches in high-tension areas), zipper and button seams, edges of openwork structures such as mesh or lace, warp-weak areas of the fabric itself, and locations of existing micro-damage (such as pilling or snagging). Each weak location is assigned an initial damage tolerance threshold, which is set as the minimum value of the force applied to the weak location in history that caused the damaged area to exceed 10%. This forms a weak map attached to the digital model of the garment. This map is the basis for subsequent calculation of damage risk values, and the mesh model is assigned material properties and weak locations are identified. S12. Obtain the three-dimensional geometric model of the equipment, the bottom contact relationship between the equipment and the clothing, including curvature, contact area and protrusion features, and obtain the physical property data of the equipment, including total mass, center of mass position, and static preload distribution generated by the external fixing method of the equipment (such as magnetic buckle, elastic clip or lock). The preload can be obtained by measuring on a standard flat material through a pressure mapping sensor and converted into a force vector field. Obtain the dynamic excitation spectrum characterizing the operating state of the equipment itself. The operation of the equipment is not limited to active vibration (such as motor, speaker, optical image stabilization components), but also includes the inertial reaction caused by human movement. Collect the working frequency, amplitude and direction of the internal moving parts of the equipment, as well as the inertial force time sequence template caused by the equipment's own mass and rotational inertia at different acceleration levels. All dynamic excitation spectra are normalized into a set of time-related multi-channel excitation signals as input for subsequent operating state and micro-displacement simulation. S13. Using sensing devices, the three-dimensional shape and skeletal dimensions of the user's body are collected while maintaining a standard anatomical posture. A static digital human body model with standard skeletal binding and a 1:1 reproduction of the user's body size is generated. Combined with bioelectrical impedance analysis or body composition data input by the user, the soft tissue thickness and stiffness coefficient of various body regions (such as chest, waist, and thighs) are obtained. Non-uniform elastic support characteristics are given to the surface shell units of the digital human body model, and a skin deformation resistance model is established. This determines the extent of indentation and displacement of the skin and underlying soft tissue when pressure is applied by clothing and equipment, directly affecting the realism of the force and deformation distribution calculation. A specific attendance scenario is a collection of a series of ordered typical movements. The user must first select a scenario template (such as jogging, business commuting, or rehabilitation training), call a pre-set general movement library, and match the user's actual movement characteristics. For adaptation, users upload daily videos of themselves in the scene, and the pose estimation algorithm extracts the skeletal motion sequence to obtain time-stamped joint angle temporal data covering the main activity range of the scene, i.e., personalized typical action sequences. The acquired joint angle temporal data drives the static digital human body model after skeletal binding in real time. A dual quaternion skinning algorithm is used so that the skin surface can naturally bulge, stretch and slide during body movement. The instantaneous spatial position, velocity, acceleration and local strain tensor caused by skin stretching at each vertex of the body surface are updated in real time. It is also necessary to superimpose the dynamic micro-movement of soft tissue relative to the skeleton caused by inertia (such as the shaking of the biceps when the arm swings). This is achieved by establishing a soft tissue layer model based on mass-spring. The generated dynamic body model outputs the geometry of the entire body surface, three-dimensional forces (including the reaction forces applied by clothing and equipment) and deformation field.
[0021] In this embodiment, based on a dynamic body model, the simulation of a user's typical action sequence in a specific attendance scenario, and the calculation of the dynamic force and deformation distribution of clothing at various body positions under the typical action sequence, include the following specific steps: S21. From the typical action sequence of the dynamic body model built in step S13, extract the first posture as a standard anatomical standing posture. The clothing digital twin (a mesh model with a weak point map) generated in step S11 is virtually dressed onto the static body model using an automatic alignment and penetration processing algorithm. The dressing process simulates the spatial relationship between the clothing and the body. Based on the difference between the clothing pattern and the body size, the fabric accumulation and tension are automatically distributed to form a natural initial wrinkle shape. A penalty function-based contact pair is set between the inner surface of the clothing and the outer surface of the skin; that is, a master-slave contact pair is established between the contact surfaces. The penalty function method is used, by setting... Appropriate penalty stiffness establishes a linear relationship between penetration depth and contact force, thereby approximately simulating the normal impermeability and tangential friction behavior of the contact interface. The friction coefficient is set based on the fabric properties in S11 and the skin properties in S13. Weak edge constraint forces are applied to local areas such as the collar, cuffs, and hem to simulate the initial tightening effect of the garment caused by the elastic band. For closures such as zippers and buttons, virtual connectors are added at their corresponding positions to limit their relative separation and ensure the integrity of the garment structure. After wearing, a coupled initial state is obtained in the initial frame where the garment fits the body perfectly and there is no penetration. The initial stress of each vertex on the garment model is set to zero. S22. Using the joint angle time-series data obtained in step S13, the skeletal movement of the body is driven, and the mesh on the body surface is displaced and deformed accordingly. Under the combined action of body drive, gravity, its own inertial force, and contact force, the clothing model produces a continuous dynamic response. The simulation uses a nonlinear finite element method cloth solver. The calculation process for each time step is as follows: body state update: based on the joint angles at the current time point, calculate the world coordinates, velocity, and acceleration of each vertex on the body surface, and use them as the motion boundary conditions for the clothing simulation; apply body force and inertial force to each finite element or particle of the clothing model. The simulation applies gravity and inertial forces resulting from changes in body surface acceleration; it calculates the constitutive response of materials: based on the material parameters of each region, it calculates the elastic stress and viscous damping force within the garment unit, and corrects for identified weak points using higher-order elements or stress intensity factors; it solves for contact collisions, iteratively detecting multiple collisions between the garment and the body, and between the garment itself, applying repulsive and frictional forces to ensure the garment does not penetrate the body and slides along the body surface; it integrates over time, updating the position and velocity of all vertices of the garment, and recording the in-plane stress, bending stress, shear stress, and corresponding tensile and shear deformation rates of each unit at this time step. The simulation then tracks the entire mechanical state of the garment from its initial posture through a series of actions until the end of the sequence. S23. For each time step, integrate the contact pressure vector of the inner surface of the clothing covered by the region to obtain the resultant force curve and the maximum pressure envelope as a function of time. Extract the tensile force of the clothing itself on the edge of the region. Statistically calculate the average in-plane elongation, maximum principal strain direction, and local area change rate of the clothing fabric within the region. For marked weak points, output the equivalent stress-time curves of these locations throughout the entire action sequence, the time of peak stress occurrence, and the corresponding body posture. Simultaneously calculate the stress concentration factor, which is the ratio of the actual peak stress at the weak point to the average stress of the surrounding area. The stress concentration factor is an important intermediate variable for injury risk assessment. By extracting the resultant force curve covering the whole body... The complete dynamic stress and deformation distribution map at each location is recorded as the baseline clothing mechanics heat map. The heat map binds the mechanical load of clothing caused by dynamic movements to the body position analysis. A set of candidate locations is preset, such as the upper left chest, the outer side of the waistband, the outer side of the upper arm, etc. For the local area of the clothing surface corresponding to each candidate location, the following are extracted from the baseline mechanics heat map: the timing of the stress components (normal pressure, tangential shear force), the timing of the deformation gradient (stretch ratio of two orthogonal directions, in-plane shear angle), the relative displacement field with the surrounding area (used to assess whether the clothing fabric in this area will slip significantly), and the stress timing of all weak points in the area.
[0022] In this embodiment, multiple candidate wearing positions are defined on the clothing. For each candidate wearing position, the analysis of the clothing structure damage risk, functional effectiveness index, and user comfort index includes the following specific steps: Multiple candidate wearing positions are defined on the clothing; for each candidate wearing position, such as... Figure 2 As shown, the following analysis is performed: S31: Combining device data, simulate the operating state and / or micro-displacement of the target smart wearable device at the candidate wearing position; S32: Analyze the operating state and / or micro-displacement simulated in step S31, and compare it with the dynamic stress and deformation distribution of the garment in the corresponding candidate wearing position area obtained in step S2, to calculate the risk of garment structural damage caused by the candidate wearing position; wherein, the risk of garment structural damage is related to the weak position overload index, alternating stress fatigue damage accumulation index, micro-displacement friction and wear index, and stress concentration deterioration degree of the weak position of the garment; S33: Evaluate the functional effectiveness and user comfort of the target smart wearable device at the candidate wearing position; In this embodiment, S31 specifically includes: extracting the analytical geometric contour of the bottom surface of the device, positioning the bottom surface on the outer surface area of the garment corresponding to the current candidate wearing position, so that the bottom surface of the device and the outer surface of the garment make initial contact, and simulating the attachment method according to the actual fixing type of the device: establishing a magnetic attraction field that decays with distance between the bottom surface of the device and the magnetic accessories on the inside of the garment, so that the garment is clamped in the middle; after the initial attachment is completed, the device maintains a stable fit with the garment in the absence of external force, and records the initial pose matrix (position and orientation) of the device as a reference benchmark for subsequent relative displacement calculation.
[0023] In this embodiment, as Figure 3 As shown, the specific content of S32 includes: the device simultaneously bears two types of dynamic loads, including human motion-induced excitation and internal self-excitation of the device. The two types of loads are synchronously superimposed in time, forming a resultant force and resultant torque acting on the center of mass of the device. The instantaneous acceleration of the clothing surface point at the candidate wearing position. Since the device has a certain mass and rotational inertia, when the clothing surface accelerates, the device will generate an inertial tendency to move relative to the clothing surface. Based on the mass of the device, the position of the center of mass and the rotational inertia of the three main axes, the inertial force and inertial torque generated by the device due to translational acceleration and angular acceleration are calculated. If the target device contains moving parts, then according to the dynamic excitation spectrum collected in step S12, a periodic or transient disturbance force corresponding to the working state of the moving parts is applied at the center of mass of the device. These disturbance forces have specific frequencies, amplitudes and directions, such as the single axial harmonic force of a linear motor or the multi-axis micro-motion compensation force of an optical image stabilization component.
[0024] Treating the equipment as a rigid body and the local area of the clothing as a deformable body, a coupled mechanical model of equipment-clothing-body is constructed. At each time step, the following equations are solved simultaneously: Equipment rigid body motion equation: Under the combined action of external forces, contact forces, frictional forces, and adhesion constraints, the translational displacement and angular displacement about the center of mass of the equipment are calculated; Clothing local deformation equation: Under the action of contact pressure from the equipment, its own tension, and contact reaction force from the body, the local area of the clothing undergoes in-plane stretching, bending, and out-of-plane wrinkling deformation; The solution process requires real-time monitoring of changes in the contact state between the equipment and the clothing, including adhesion, sliding, and separation. Feature extraction is performed on the displacement time-series data of the device relative to the clothing to generate a device attachment micro-displacement feature vector, which includes the maximum tangential slip amplitude, normal runout amplitude, peel angle amplitude, attachment failure duration ratio, and contact pressure redistribution map. Among them, the maximum tangential slip amplitude is the maximum cumulative slip distance in a plane parallel to the clothing surface from the geometric center of the device bottom surface to the center point of the clothing attachment area. The normal runout amplitude is the maximum value and root mean square value of the normal gap between the device bottom surface and the clothing surface in the action sequence, representing the impact intensity of the device. The peel angle amplitude is the maximum angle change range of the device bottom surface plane relative to the tangential plane of the clothing surface, representing the degree of prying of the device edge. The attachment failure duration ratio is the proportion of the duration of the instantaneous state when the device partially or globally detaches from the clothing to the total duration of the entire action sequence. The contact pressure redistribution map is the contact pressure distribution sequence of the device bottom surface to the clothing surface at each time step, and the peak pressure points and their migration trajectories are extracted to identify the concentrated pressure areas of the clothing caused by the edges or protrusions of the device.
[0025] For the clothing grid area covered by the current candidate position, on the basis of the existing body contact force and clothing internal stress, the surface distribution pressure applied by the bottom of the equipment and the line distribution clamping force of the equipment clamping edge are superimposed. If there is a micro displacement of the equipment, the area of action of the contact pressure will migrate on the clothing surface over time. The movement path of the pressure is tracked according to the time step and loaded onto the clothing unit at the corresponding position. For the local area of the clothing after the force superposition, the stress field and strain field after each time step are obtained. The risk value of clothing structural damage is related to the stress concentration at weak points in the clothing. It is calculated using a weighted combination of the weak point overload index, alternating stress fatigue damage accumulation index, micro-displacement friction and wear index, and the increase in the weak point stress concentration coefficient. Specifically, for each weak point within the current candidate location area and its influence neighborhood, the ratio of its maximum equivalent stress in the entire action sequence to the damage tolerance threshold recorded in step S11 is calculated. The weighted maximum value of all weak point ratios is taken as the weak point overload index. For each weak point, the stress cycle amplitude and mean stress in its equivalent stress-time curve are statistically analyzed using the rainflow counting method. Combined with the SN fatigue curve (stress-life curve) of the clothing material, the equivalent fatigue damage degree within a single action sequence cycle is calculated using Miner's linear cumulative damage rule. The equivalent fatigue damage degree is the standardized fatigue life. Finally, the alternating stress fatigue damage accumulation index is the maximum value of the equivalent fatigue damage degree among all weak points. Using the tangential slip amplitude and normal pressure time history obtained in step S31, the frictional work density between the bottom surface of the equipment and the surface of the clothing is calculated, i.e., the density of energy generated by friction over a given time period. The abrasion resistance parameters of the fabric are calculated, where the abrasion resistance parameter is the wear coefficient, which is the volume wear caused by sliding a unit distance under a unit load. The micro-displacement friction wear index is calculated by weighting. The risk of fabric surface wear, pilling, or even yarn breakage caused by repeated equipment sliding is considered. The stress concentration coefficient increase at weak points is calculated. The baseline stress concentration coefficient obtained in step S23 is divided by the stress concentration coefficient corrected after superimposing the equipment to obtain the stress concentration deterioration degree. The four indicators of weak point overload index, alternating stress fatigue damage accumulation index, micro-displacement friction wear index, and stress concentration deterioration degree are normalized and merged into a comprehensive clothing structure damage risk value through a preset weight matrix. The value ranges from 0 to 1. The higher the value, the greater the damage risk. The weight matrix can be customized for different clothing types (such as precision knitwear, stiff shirts) and different user preferences (such as more concerned about tearing or wear). For the current candidate wearing position, a structured damage risk assessment report is generated, including: comprehensive risk value, specific values of each sub-index, the weakest position with the highest risk and its dominant damage mode (fatigue / overload / wear / tear), and the corresponding key action posture. In this embodiment, S33 specifically includes: evaluating the functional effectiveness and user comfort indicators of the target smart wearable device at the candidate wearing position; calculating the sensor field-of-view occlusion rate; the device includes optical sensors (such as a heart rate photoplethysmometer, blood oxygen probe, and ambient light sensor); calculating the contact angle between the sensor window and the skin surface, the gap, and the proportion of the field of view obscured by clothing fabric at the current candidate position using the target smart wearable device; simultaneously, using the device pose data from step S31, checking the angle offset between the sensor window normal and the skin surface normal, as well as the average distance between the window and the skin, step by step; comparing these data with corresponding data threshold ranges; and defining valid sampling when all data are within the threshold range. Valid sampling refers to the sampling process where the accuracy of the sensor sampling data meets the requirements. The valid sampling time ratio is defined as the proportion of time required for normal sensor operation. Field-of-view occlusion rate = 1 - Effective sampling time ratio; using the contact pressure distribution map in step S31, identify the maximum pressure peak in the action sequence, divide the peak by the pain pressure threshold of each part of the body to obtain the probability of exceeding the pressure threshold, and simultaneously assess the shear deformation caused to the skin by the device edge. Under repeated shearing action, the skin is prone to friction blisters or hyperkeratosis. Divide the sum of shear deformation during wearing by the shear deformation safety value of the corresponding position to obtain the local pressure discomfort index. Device coverage will cause local skin heat dissipation obstruction and sweat accumulation. Combining the device bottom area and material thermal resistance, as well as the basic sweating rate data of the body area, after standardization, weighted summation is obtained to obtain the heat and moisture accumulation index of the device coverage area. After normalization, the probability of exceeding the pressure threshold, the local pressure discomfort index, and the heat and moisture accumulation index of the device coverage area are weighted and combined to obtain the user comfort comprehensive score, with a value range of 0 to 1, and the higher the value, the more comfortable. The weights are obtained by using the user's historical usage data to score.
[0026] In this embodiment, calculating and outputting the recommended optimal wearing position of the target smart wearable device for the target user in a specific attendance scenario includes the following specific steps: For all candidate wearing positions, an evaluation vector is generated based on the risk of clothing structural damage, functional effectiveness index, and user comfort index corresponding to the wearing position. All feature values in the vector are processed through range normalization. Weights are set according to the wearing scenario, and a weighted comprehensive evaluation score is calculated. The comprehensive evaluation scores are sorted, and the position with the highest evaluation score is selected as the best wearing position for recommendation. In this process, the risk of clothing structural damage has a negative weight, while the functional effectiveness index and user comfort index have positive weights. The weights are obtained by obtaining the probability of errors occurring in the scenario due to the risk of clothing structural damage, functional effectiveness index, and user comfort index, and the weights are allocated according to the probability ratio.
[0027] The advantages of this embodiment compared to the prior art are: it fundamentally solves the pain point of traditional fitting solutions ignoring individual differences and relying only on fixed gears or manual fine-tuning. Its core breakthrough lies in: through dynamic modeling, simulating the mechanical interaction between the user and the clothing in different postures to achieve accurate fitting; multi-objective decision-making takes into account both structural safety and functional efficiency, avoiding sacrificing comfort or causing stress concentration in pursuit of a single indicator. This solution elevates fitting from static estimation to a new level of dynamic simulation and intelligent decision-making.
[0028] Example 2 like Figure 4 As shown, the clothing data analysis system based on smart wearables is implemented based on the above-mentioned clothing data analysis method based on smart wearables. Specifically, it includes a data acquisition module, which is used to acquire clothing data of the target user, device data of the target smart wearable device, and dynamic body model of the user in a specific attendance scenario. The simulation module, based on the dynamic body model, simulates the user's typical action sequence in the specific attendance scenario, and calculates the dynamic force and deformation distribution of the clothing at various positions on the body under the typical action sequence; The indicator analysis module is used to define multiple candidate wearing positions on the garment, and analyze the garment structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position. The wearable recommendation module calculates and outputs the optimal wearable position recommended by the target smart wearable device for the target user in the specific attendance scenario, based on the clothing structure damage risk value, the functional effectiveness index, and the user comfort index corresponding to each candidate wearable position.
[0029] Example 3 This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor executes the aforementioned method for analyzing clothing data based on smart wearables by calling computer programs stored in memory.
[0030] The electronic device can vary considerably depending on its configuration or performance. It may include one or more processors and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the smart wearable clothing data analysis method provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.
[0031] Example 4 This embodiment proposes a computer-readable storage medium on which an erasable and rewritable computer program is stored. When the computer program runs on the computer device, it causes the computer device to perform the above-mentioned clothing data analysis method based on smart wearables.
[0032] For example, computer-readable storage media can be read-only memory, random access memory, read-only optical disc, magnetic tape, floppy disk, and optical data storage devices.
[0033] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired network and / or wireless network. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., a floppy disk), an optical medium (e.g., a DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.
Claims
1. A smart wear-based clothing data analysis method, characterized by, It includes the following specific steps: S1: Acquire the target user's clothing data, the target smart wearable device's device data, and the user's dynamic body model in a specific attendance scenario; S2: Based on the dynamic body model, simulate the user's typical action sequence in the specific attendance scenario, and calculate the dynamic force and deformation distribution of the clothing at various positions on the body under the typical action sequence; S3: Define multiple candidate wearing positions on the garment, and analyze the garment structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position. S4: Based on the clothing structure damage risk value, the functional effectiveness index, and the user comfort index corresponding to each candidate wearing position, calculate and output the recommended optimal wearing position of the target smart wearable device for the target user in the specific attendance scenario. 2.The smart wear based clothing data analysis method of claim 1, wherein, The process of acquiring the target user's clothing data, the target smart wearable device's device data, and the user's dynamic body model in a specific attendance scenario includes the following specific steps: S11. Using a 3D scanner, acquire high-precision 3D point cloud data of the clothing selected by the target user in a natural hanging state, generate a surface mesh model, and assign material properties and identify weak points in the mesh model. S12. Obtain the three-dimensional geometric model of the equipment, collect the working frequency, amplitude and direction of the moving parts inside the equipment, as well as the time sequence template of the inertial force caused by the equipment's own mass and moment of inertia at different acceleration levels. All dynamic excitation spectra are normalized into a set of time-related multi-channel excitation signals as input for subsequent operation status and micro-displacement simulation. S13. Using sensing devices, collect the three-dimensional shape and skeletal dimensions of the user's body when maintaining a standard anatomical posture, generate a static digital human body model with standard skeletal binding that is 1:1 replicated with the user's body size, obtain the soft tissue thickness and stiffness coefficient of each area of the body, generate a dynamic body model, and output the geometric shape, three-dimensional force and deformation field of the entire surface of the body.
3. The clothing data analysis method based on smart wearables as described in claim 2, characterized in that, Based on the dynamic body model, simulating the user's typical action sequence in the specific attendance scenario, and calculating the dynamic force and deformation distribution of the clothing at various body positions under the typical action sequence includes the following specific steps: S21. From the typical action sequence of the dynamic body model built in step S13, extract the first posture as the standard anatomical standing posture. The clothing digital twin generated in step S11 is virtually worn on the static body model through an automatic alignment and penetration processing algorithm. The wearing process simulates the spatial relationship between the clothing and the body, and obtains a coupled initial state in the initial frame where the clothing and the body fit perfectly and there is no penetration. S22. Using the joint angle timing data obtained in step S13, the body's skeletal movement is driven, and the mesh on the body surface is displaced and deformed accordingly. Under the combined action of body drive, gravity, its own inertial force and contact force, the clothing model generates a continuous dynamic response. The simulation uses a nonlinear finite element method cloth solver to simulate the mechanical state of the clothing from its initial posture through a series of actions until the end of the sequence. S23. For each time step, the contact pressure vector of the inner surface of the garment covered by the integral calculation is obtained to obtain the resultant force curve and the maximum pressure envelope surface that change with time. The tensile force of the garment itself on the edge of the region is extracted. The average in-plane elongation rate, the direction of the maximum principal strain and the local area change rate of the garment fabric in the region are statistically analyzed. For the marked weak positions, the equivalent stress-time curve of these positions in the whole action sequence, the time of occurrence of peak stress and the corresponding body posture are output. At the same time, the stress concentration coefficient is calculated. By extraction, a complete dynamic force and deformation distribution map covering all positions of the body is obtained and recorded as the baseline garment mechanical heat map. The heat map binds the mechanical load of the garment caused by dynamic movement with the body position analysis. A set of candidate positions is preset. For the local area of the garment surface corresponding to each candidate position, the following are extracted from the baseline mechanical heat map: the force component time sequence, the deformation gradient time sequence, the relative displacement field with the surrounding area and the stress time sequence of all weak positions in the region in the whole action sequence.
4. The clothing data analysis method based on smart wearables as described in claim 3, characterized in that, Multiple candidate wearing positions are defined on the garment. For each candidate wearing position, the analysis of the garment structure damage risk, functional effectiveness index, and user comfort index includes the following specific steps: Multiple candidate wearing positions are defined on the garment. For each candidate wearing position, the following analysis is performed: S31: Combining the device data, the operating state and micro-displacement of the target smart wearable device at the candidate wearing position are simulated; S32: The operating state and micro-displacement simulated in step S31 are analyzed, along with the dynamic stress and deformation distribution of the garment in the corresponding candidate wearing position area obtained in step S2, to calculate the garment structure damage risk caused by the candidate wearing position; wherein, the garment structure damage risk is related to the weak point overload index, alternating stress fatigue damage accumulation index, micro-displacement friction and wear index, and stress concentration deterioration degree of the weak points of the garment; S33: The functional effectiveness and user comfort of the target smart wearable device at the candidate wearing position are evaluated.
5. The clothing data analysis method based on smart wearables as described in claim 4, characterized in that, Calculating and outputting the recommended optimal wearing position of the target smart wearable device for the target user in the specific attendance scenario includes the following specific steps: For all candidate wearing positions, an evaluation vector is generated based on the risk of damage to the clothing structure, the functional effectiveness index, and the user comfort index corresponding to the wearing position. All feature values in the vector are processed by range normalization, and a weighted comprehensive evaluation score is calculated. The comprehensive evaluation scores are sorted, and the position with the highest evaluation score is selected as the best wearing position for recommendation.
6. The clothing data analysis method based on smart wearables as described in claim 5, characterized in that, S31 includes the following specific content: extracting the analytical geometric contour of the bottom surface of the device, positioning the bottom surface on the outer surface area of the clothing corresponding to the current candidate wearing position, so that the bottom surface of the device and the outer surface of the clothing make initial contact; The attachment method is simulated based on the actual fixing type of the equipment. A magnetic attraction field with a decreasing distance is established between the bottom surface of the equipment and the magnetic accessories inside the garment, so that the garment is clamped in the middle. After the initial attachment is completed, the device maintains a stable fit with the clothing without external force, and the initial pose matrix of the device is recorded as a reference for subsequent relative displacement calculation.
7. The clothing data analysis method based on smart wearables as described in claim 6, characterized in that, S32 includes the following specific contents: The equipment is simultaneously subjected to two types of dynamic loads: the excitation caused by human movement and the self-excitation of the equipment's internal operation, forming a resultant force and resultant torque acting on the equipment's center of mass; a coupled mechanical model of the equipment-clothing local area-body is constructed, and the rigid body motion equation of the equipment and the local deformation equation of the clothing are solved simultaneously at each time step, and the adhesion, sliding and separation states between the equipment and the clothing are detected in real time; the displacement time series data of the equipment relative to the clothing are feature extracted to generate an adhesion micro-displacement feature vector containing the maximum tangential slip amplitude, normal runout amplitude, peel angle amplitude, adhesion failure duration ratio and contact pressure redistribution spectrum; for the local area of the clothing after force superposition, the corrected stress field and strain field at each time step are obtained; the structural damage risk value of the clothing is calculated by a weighted combination of the weak position overload index, alternating stress fatigue damage accumulation index, micro-displacement friction and wear index and the increase of the stress concentration coefficient at the weak position.
8. The clothing data analysis method based on smart wearables as described in claim 7, characterized in that, S33 includes the following specific content: S33 specifically includes: evaluating the functional effectiveness and user comfort indicators of the target smart wearable device at the candidate wearing position; calculating the sensor field-of-view occlusion rate; calculating the contact angle between the sensor window and the skin surface, the gap, and the proportion of the field of view obscured by clothing fabric at the current candidate position using the target smart wearable device; simultaneously, using the device pose data from step S31, checking the angle offset between the sensor window normal and the skin surface normal, as well as the average distance between the window and the skin, step by step; comparing these data with corresponding data threshold ranges; when all these data are within the threshold range, it is defined as valid sampling, and the field-of-view occlusion rate = 1 - Effective sampling time ratio; using the contact pressure distribution map in step S31, identify the maximum pressure peak in the action sequence, divide the peak by the pain pressure threshold of each part of the body to obtain the probability of exceeding the pressure point limit, and at the same time assess the shear deformation of the skin caused by the edge of the device, divide the sum of the shear deformation during wearing by the shear deformation safety value of the corresponding position to obtain the local pressure discomfort index, combine the device bottom area and material thermal resistance, and the basic sweating rate data of the body area, standardize and weight sum to obtain the heat and moisture accumulation index of the device coverage area, and normalize the pressure point excess probability, local pressure discomfort index and the heat and moisture accumulation index of the device coverage area, and then weight sum them to obtain the overall user comfort score.
9. A clothing data analysis system based on smart wearables, implemented based on the clothing data analysis method based on smart wearables as described in any one of claims 1-8, characterized in that, Specifically, it includes: The data acquisition module is used to acquire clothing data of the target user, device data of the target smart wearable device, and a dynamic body model of the user in a specific attendance scenario; The simulation module, based on the dynamic body model, simulates the user's typical action sequence in the specific attendance scenario, and calculates the dynamic force and deformation distribution of the clothing at various positions on the body under the typical action sequence; The indicator analysis module is used to define multiple candidate wearing positions on the garment, and analyze the garment structure damage risk value, functional effectiveness index and user comfort index corresponding to each candidate wearing position. The wearable recommendation module calculates and outputs the optimal wearable position recommended by the target smart wearable device for the target user in the specific attendance scenario, based on the clothing structure damage risk value, the functional effectiveness index, and the user comfort index corresponding to each candidate wearable position.
10. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; The processor is characterized in that it executes the clothing data analysis method based on smart wearables as described in any one of claims 1-8 by calling a computer program stored in the memory.
