An intelligent clothing design method based on human body data

Abstract

The present invention discloses an intelligent clothing design method based on human body data, specifically relating to the technical field of clothing design. By collecting the posture data of users in daily environments and using a deep learning bone key point recognition algorithm to evaluate the degree of daily behavior misalignment of users, the correction parameters of the clothing pattern are determined based on the degree of daily behavior misalignment. For the corrected clothing pattern, the balance degree of clothing stress distribution is evaluated by analyzing the parameters of the clothing mechanical model. At the same time, the epidermal sensitivity data of users is obtained by using a skin conductance sensor, and the analytic hierarchy process is used to evaluate the clothing comfort. Finally, the balance degree of clothing stress distribution and the clothing comfort are analyzed to evaluate the adaptability degree of the clothing pattern, so as to realize the intelligent matching of clothing design and human body characteristics and improve the comfort and functionality of clothing wearing.

Description

Technical Field

[0001] The present invention relates to the technical field of clothing design, and more specifically, to an intelligent clothing design method based on human body data. Background Art

[0002] In the prior art, clothing design is mostly based on static or simple dynamic posture measurements, which are difficult to accurately reflect the posture misalignment in users' daily behaviors. Due to habits such as sitting for long periods, lifting heavy objects, or using mobile phones, there may be mild or moderate squeezing and stretching in some parts of the human body that are inconsistent with the standard posture, resulting in differences between the body circumference or curve of the human body and the original data, and it is difficult to meet the needs of personalized clothing design. Especially in office, travel, and daily life scenarios, users' body postures change at any time due to different postures, and traditional measurement methods are difficult to dynamically capture these subtle changes, making it difficult to balance the fit and comfort of clothing.

[0003] To solve the above problems, a technical solution is provided now. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent clothing design method based on human body data to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An intelligent clothing design method based on human body data, comprising the following steps:

[0007] By monitoring the multi-dimensional posture data of the user in the daily environment, a deep learning skeleton key point recognition algorithm is used to evaluate the degree of daily behavior misalignment of the user;

[0008] Based on the degree of daily behavior misalignment of the user, the correction parameters of the clothing pattern are determined;

[0009] For the clothing pattern determined according to the correction parameters: by analyzing the parameters of the clothing mechanics model, the balance degree of the stress distribution of the clothing is evaluated; by obtaining the user's epidermal sensitivity data through a skin conductance sensor, the comfort of the clothing is evaluated using the analytic hierarchy process;

[0010] Based on the balance degree of the stress distribution of the clothing and the comfort of the clothing, the adaptability degree of the clothing pattern is analyzed.

[0011] In a preferred embodiment, by monitoring the multi-dimensional posture data of the user in the daily environment, a deep learning skeleton key point recognition algorithm is used to evaluate the degree of daily behavior misalignment of the user, specifically:

[0012] Image acquisition devices and sensor devices are arranged in the user's daily scenarios to collect multi-angle posture data in real time;

[0013] Preprocess the multi-angle pose data collected;

[0014] Use a convolutional neural network to achieve precise positioning of skeletal key points;

[0015] Compare the positioning data of the skeletal key points with the standard body posture model to quantify the user's behavior deviation.

[0016] In a preferred embodiment, comparing the positioning data of the skeletal key points with the standard body posture model to quantify the user's behavior deviation specifically includes:

[0017] Compare the overall deviation index with a preset standard threshold to determine the deviation between the user's actual posture and the standard body posture:

[0018] When the overall deviation index is greater than or equal to the preset standard threshold, it is determined that the deviation between the user's actual posture and the standard body posture exceeds the acceptable range;

[0019] When the overall deviation index is less than the preset standard threshold, it is determined that the deviation between the user's actual posture and the standard body posture is within the acceptable range.

[0020] In a preferred embodiment, determine the correction parameters of the clothing pattern based on the degree of the user's daily behavior misalignment, specifically:

[0021] Construct a clothing mechanics model based on finite element analysis, and define structural nodes, material properties, and boundary conditions:

[0022] Use the domain decomposition algorithm to divide the human body into different blocks;

[0023] Calculate the correction coefficient of each human body block using a linear regression equation based on the regional data;

[0024] Integrate the correction coefficients of different human body blocks to generate clothing pattern correction parameters adapted to the user's body posture.

[0025] In a preferred embodiment, evaluate the stress distribution balance degree of the clothing by analyzing the clothing mechanics model parameters, specifically:

[0026] Construct a clothing mechanics model based on finite element analysis, and define material properties and boundary conditions;

[0027] Perform numerical simulation to calculate the stress distribution;

[0028] Collect local stress data and statistically analyze the stress of each node;

[0029] Evaluate the stress distribution balance degree of the clothing according to the statistical results.

[0030] In a preferred embodiment, user epidermal sensitivity data is obtained through a skin conductance sensor, and the analytic hierarchy process is used to evaluate the comfort of clothing, specifically as follows:

[0031] Collect skin conductance data in real time according to the skin conductance sensor;

[0032] Perform signal filtering and feature extraction on the conductance data to generate a skin sensitivity index;

[0033] Establish a hierarchical structure of comfort evaluation indicators and determine the evaluation factor weight matrix;

[0034] Calculate the clothing comfort index according to the regional comprehensive weight and the skin sensitivity index.

[0035] In a preferred embodiment, analyze the adaptability of the clothing pattern based on the stress distribution balance degree of the clothing and the comfort of the clothing, specifically as follows:

[0036] Normalize the stress distribution balance index and the clothing comfort index. After assigning preset proportional coefficients to the normalized stress distribution balance index and the clothing comfort index respectively, calculate the fitness score;

[0037] Compare the fitness score with the fitness score threshold: when the fitness score is greater than or equal to the fitness score threshold, it indicates that the adaptability of the clothing pattern is good; when the fitness score is less than the fitness score threshold, it indicates that the adaptability of the clothing pattern is poor.

[0038] The technical effects and advantages of an intelligent clothing design method based on human data according to the present invention:

[0039] 1. By monitoring the multi-dimensional posture data of the user in the daily environment and combining with the deep learning skeleton key point recognition algorithm, the evaluation of the misalignment degree of the user's daily behavior is realized, breaking through the limitation of traditional clothing design relying on static measurement data, being able to capture the body posture deviation caused by the user's non-standard posture, and then adjusting the clothing pattern correction parameters based on the deviation data, effectively improving the fit between the clothing and the human body. By analyzing the clothing mechanical model parameters and evaluating the stress distribution balance degree of the clothing, it is possible to identify and optimize the possible local pressure concentration problems in the pattern, reducing the discomfort during wearing. In addition, combining the user epidermal sensitivity data obtained by the skin conductance sensor and using the analytic hierarchy process to evaluate the clothing comfort, ensuring that the clothing takes into account the user's wearing comfort while meeting the structural stability.

[0040] 2. By comprehensively analyzing the stress distribution balance degree of the clothing and the comfort of the clothing, evaluating the adaptability of the clothing pattern, meeting the wearing needs of users' personalization and dynamic posture changes, and improving the comfort and adaptability of clothing design. Description of the Drawings

[0041] Figure 1 Schematic diagram of an intelligent clothing design method based on human body data according to the present invention. Specific implementation manner

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Embodiment

[0043] Figure 1 An intelligent clothing design method based on human body data according to the present invention is given, which includes the following steps:

[0044] By monitoring the multi-dimensional pose data of the user in the daily environment, a deep learning bone key point recognition algorithm is used to evaluate the degree of misalignment of the user's daily behavior;

[0045] Based on the degree of misalignment of the user's daily behavior, the correction parameters of the clothing pattern are determined;

[0046] For the clothing pattern determined according to the correction parameters: by analyzing the parameters of the clothing mechanical model, the balance degree of the stress distribution of the clothing is evaluated; by using a skin conductance sensor to obtain the user's epidermal sensitivity data, the analytic hierarchy process is used to evaluate the comfort of the clothing;

[0047] Based on the balance degree of the stress distribution of the clothing and the comfort of the clothing, the adaptability degree of the clothing pattern is analyzed.

[0048] Specifically, by monitoring the multi-dimensional pose data of the user in the daily environment, a deep learning bone key point recognition algorithm is used to evaluate the degree of misalignment of the user's daily behavior, including:

[0049] Image acquisition devices and sensor devices are arranged in the user's daily scenarios to collect multi-angle pose data in real time: in the user's daily activity areas (such as home, office environment), multiple fixed installation points are selected to arrange high-definition cameras and depth sensors to ensure that different angles of the user's actions can be covered. A binocular or multi-camera system is used to ensure that the acquired image data has stereoscopic information. At the same time, an inertial measurement unit is installed to capture information such as the acceleration and angular velocity of the user's actions.

[0050] To ensure data consistency, a synchronous timestamp mechanism is adopted to unify the data collected by each device to the same time reference. The video is continuously collected at a preset frame rate (e.g., 30 frames per second), and the data collected by the inertial measurement unit is recorded at the same time step. All collected data is timestamped , to ensure the temporal consistency during data fusion.

[0051] By deploying sensors from multiple angles and synchronously collecting multi-source data, the user's movement trajectory in real daily scenarios can be comprehensively captured, providing high-quality data input.

[0052] Preprocess the multi-angle attitude data collected: The raw image data and sensor signals collected usually have noise and environmental interference and need to be preprocessed to extract high-quality human feature data.

[0053] Denote the raw image data as , and perform the following processing:

[0054] Use a Gaussian filter to smooth the raw image data, and its expression is:

[0055] ; where represents the smoothed image data; represents the mathematical function of the Gaussian filter, which is used to smooth the raw image data and reduce noise; represents the raw image data. Normalize the smoothed image data to the interval [0,1], denoted as .

[0056] For the raw signal data collected by the inertial measurement unit , use a low-pass filter to remove high-frequency noise, and its expression is: ; where represents the signal data after low-pass filtering; represents the mathematical function of the low-pass filter, which is used to filter out high-frequency noise; represents the raw signal data collected by the inertial measurement unit.

[0057] According to the timestamp , align the normalized image data and the signal data after low-pass filtering in time to generate composite data . The multi-source data fusion uses interpolation algorithms and time series alignment techniques to ensure spatio-temporal consistency between data points. represents the fused and normalized image data and the signal data after low-pass filtering through time synchronization and data interpolation The obtained composite data.

[0058] Using a convolutional neural network to achieve the localization of skeletal key points: The fused multi-source composite data , is input into a pre-trained convolutional neural network model for achieving precise localization of human skeletal key points, specifically:

[0059] An improved convolutional neural network structure is adopted, which can process image data and sensor signal data simultaneously. The input is the fused multi-source composite data . The first layer of the network is a convolutional layer to extract local image features; then, through multiple convolutional and pooling layers, higher-level human body structure features are obtained. Residual modules are used in the network design to prevent gradient disappearance and improve the localization accuracy of skeletal key points.

[0060] Set the output of the improved convolutional neural network structure as a set of key point coordinate sets ; where, represents the total number of human skeletal key points, determined according to the specific human body model; represents the th coordinate vector of the skeletal key point, denoted as:

[0061] ; where, respectively represent the coordinates of the th skeletal key point on the X-axis, Y-axis, and Z-axis.

[0062] Through backpropagation and training with a large amount of labeled data, the convolutional neural network model can achieve precise localization of skeletal key points in multiple poses and output stable skeletal key point coordinate data.

[0063] Fuse image data and sensor signal data to improve the robustness of skeletal key point detection; adopt an improved convolutional neural network structure, combined with residual modules, to solve the positioning error problem of traditional networks under multi-angle and large-amplitude pose changes; the output skeletal key point coordinate data has high-precision three-dimensional positioning ability, providing a basis for quantifying user behavior deviations.

[0064] Compare the localization data of skeletal key points with the standard body posture model to quantify user behavior deviations: Use the localization data of the detected skeletal key points to compare with the preset standard body posture model to quantify the deviations of users in daily behaviors, specifically:

[0065] Preset a standard body posture model, defined as a set of reference skeletal key point coordinate sets ; where, represents the coordinate vector of the th skeletal key point in the standard body posture model, denoted as ; where respectively represent the reference coordinates of the -th skeletal key point in the standard body posture model on the X-axis, Y-axis, and Z-axis.

[0066] The standard body posture model data can be obtained through ergonomic research and corrected based on user group statistical data.

[0067] For the -th skeletal key point, define its deviation as the Euclidean distance between the detected coordinates (i.e., ) and the reference coordinate , and its calculation formula is:

[0068] ; where represents the Euclidean distance between the detected -th skeletal key point and the -th skeletal key point in the standard body posture model.

[0069] The deviation degree between the user's actual posture and the standard body posture is reflected by calculating the straight-line distance in three-dimensional space.

[0070] To comprehensively consider the deviations of each skeletal key point, the overall deviation index is defined, and its calculation formula is: ; where represents the overall deviation index; represents the weight coefficient of the -th skeletal key point, which reflects the importance of the -th skeletal key point in clothing adaptation design and is determined by expert experience or statistical data.

[0071] The larger the overall deviation index, the more obvious the difference between the user's actual posture and the standard body posture. A large deviation from the standard body posture may be due to long-term bad sitting postures, exercise habits, or working environment influences, indicating that the user has a relatively serious posture misalignment problem in daily behaviors. A high overall deviation index indicates that personalized corrections are needed in clothing design to better match the user's actual body posture and improve wearing comfort and functionality.

[0072] Compare the overall deviation index with a preset standard threshold to determine the deviation between the user's actual posture and the standard body posture:

[0073] When the overall deviation index is greater than or equal to the preset standard threshold, it is determined that the deviation between the user's actual posture and the standard body posture exceeds the acceptable range; at this time, the user can be prompted to improve daily behavior habits first, such as adjusting the sitting or standing posture, to provide more accurate data support for subsequent clothing design;

[0074] When the overall deviation index is less than the preset standard threshold, it is determined that the deviation between the user's actual posture and the standard body posture is within an acceptable range; at this time, the difference between the user's daily body posture and the ideal standard is limited and will not have a significant impact on the comfort or functionality during clothing wearing.

[0075] The setting of the preset standard threshold is a numerical range obtained through statistical analysis based on a large number of ergonomic studies, historical sample data, and clinical health standards.

[0076] Specifically, based on the degree of daily behavior misalignment of the user, correction parameters for the clothing pattern are determined, including:

[0077] Using the region decomposition algorithm, the human body is divided into different blocks: in the analysis of the user's body posture, when the deviation between the user's actual posture and the standard posture exceeds the acceptable range, it usually cannot indicate which specific part most needs intervention. It is necessary to divide the human body into different blocks through the region decomposition algorithm, such as the shoulder and neck area, the chest and back area, the waist and abdomen area, and the limbs area, etc.

[0078] Introduce indicating the total number of human body blocks. For example, the shoulder and neck area can be regarded as the first block, the chest and back area as the second block, the waist and abdomen area as the third block, the limbs area as the fourth block, etc. Use to identify the block number, that is .

[0079] When distributing the overall deviation index to each block, weight distribution is adopted, specifically:

[0080] ; among them, represents the deviation allocation value corresponding to the th human body block; represents the weight coefficient of the th human body block, which is pre-calibrated by ergonomic experts based on experimental data, and the sum of the weight coefficients of each human body block is 1.

[0081] Traditional methods often only perform overall correction and ignore the different sensitivities of each part of the human body to the clothing fit. By using the weight coefficients of the human body blocks, when distributing the deviation, a higher proportion can be given to key areas such as the shoulder and neck or the waist and abdomen, so as to more accurately obtain the deviation allocation value of each block.

[0082] Based on the regional data, a linear regression equation is used to calculate the correction coefficient of each human body block: after obtaining the decomposed deviation allocation value, it is necessary to combine the body shape data (such as girth, length, radian, etc.) of the user in a specific block for regression analysis to find a more accurate correction amount.

[0083] Introduce Represents the set of actual observed block physical sign characteristic values. For example, the shoulder and neck area includes neck circumference, shoulder width, etc.

[0084] Define a regression equation for mapping the deviation allocation values and physical sign characteristic values corresponding to different human body blocks to the correction coefficient , and its expression is:

[0085] ; where Represents the correction coefficient for the th human body block, which is used to guide the correction amount of the clothing pattern of this human body block in terms of size, position, and angle; Is the intercept term of the regression equation; Is the deviation allocation value Corresponding regression coefficient; Is the physical sign characteristic value Corresponding regression coefficient.

[0086] Use a large amount of user wearing and measurement data to fit the regression equation to obtain the optimal . In actual application, input the deviation allocation value and the physical sign characteristic value into the regression equation, and the correction coefficient can be obtained.

[0087] When the block division is more complex, the regression dimension can be increased. For example, more feature quantities can be introduced or polynomial regression can be adopted.

[0088] Through multivariable regression, integrating the user's physical sign characteristic values and deviation allocation values improves the correction accuracy and meets the differentiated requirements for different human body blocks.

[0089] Integrate the correction coefficients of different human body blocks to generate clothing pattern correction parameters adapted to the user's body posture: After obtaining the correction coefficient of each human body block, it is necessary to integrate the correction coefficients of each block into the adjustment information of the overall clothing pattern.

[0090] Define a set of correction coefficients , where Represents the total number of human body blocks.

[0091] To achieve the unification of the correction coefficients of each block, calculate the correction parameters, and its calculation formula is: ; where Represents the correction parameter of the clothing pattern, which can be subdivided into, for example, how many millimeters the shoulder and neck area is lifted, how many centimeters the waist is narrowed, how many millimeters the trouser leg is widened, etc.

[0092] Specifically, by analyzing the parameters of the clothing mechanical model, evaluate the stress distribution balance degree of the clothing, including:

[0093] Construct a clothing mechanical model based on finite element analysis and define material properties and boundary conditions: In clothing CAD software, according to the pattern outline and the human body surface, divide the clothing pattern into finite element meshes. Define the set composed of all nodes of the finite element mesh as The set of all nodes in the finite element mesh formed after discretizing the clothing pattern, where each node corresponds to the position coordinates of the fabric or seam. The node set The number of nodes in is defined as , representing the total number of nodes in the finite element mesh, and each node has three-dimensional or two-dimensional position coordinates.

[0094] The clothing fabric has mechanical properties such as elasticity and strength, which are measured through experiments on a fabric tensile testing machine and denoted as , representing the material properties of the clothing fabric, including elastic modulus, Poisson's ratio, friction coefficient, etc.

[0095] Define the area where the clothing contacts or is fixed to the human body in the wearing scenario as the boundary condition, and the boundary condition set is denoted as , corresponding to the degree-of-freedom restrictions of the nodes when subjected to human constraints. If it is necessary to consider external forces generated by gravity, tensile force, or posture movements, then define the external force , representing the external force applied to each node of the clothing mechanical model.

[0096] The clothing mechanical model integrates actual fabric data, human constraints, and external force characteristics, and can reflect the stress distribution and deformation of the clothing in the real wearing state, rather than being limited to the ideal static scenario.

[0097] Perform numerical simulation to calculate the stress distribution: In finite element analysis, establish an elastic equilibrium equation according to the defined mesh and material properties. Denote the stress as , the strain as , and the material property as the elastic matrix , then the expression of the elastic equilibrium equation is: ; where, Represents the stress vector inside the clothing material; Represents the elastic matrix of the clothing material; Represents the strain vector inside the clothing material.

[0098] Introduce the displacement vector , representing the displacement of each node in the finite element model, including translational and rotational components, and its value at node is denoted as , representing the displacement of node . According to the boundary condition and the external force , the overall mesh is assembled and iterated using a finite element solver to obtain the displacements of each node. Then, the strains and stresses within each element are calculated using the displacements of each node to form a clothing stress distribution matrix.

[0099] Compared with traditional manual or experience - dependent methods, the introduction of finite element numerical simulation can more intuitively and accurately predict the force conditions of various parts of the clothing. It can not only calculate local stress concentration but also capture overall deformation and the stretching and compression between seams, meeting the requirements of high - precision clothing design.

[0100] Collect local stress data and statistically analyze the stress of each node: In the simulation results, the nodes in key areas such as the cutting edge and the seam line need to be focused on and denoted as , which represents the set of nodes in key areas such as the cutting edge and the seam line in the clothing pattern. It is selected according to the clothing design requirements and corresponds to the key parts that are prone to deformation. Extract the local stress within the range of the node set from the clothing stress distribution matrix to form local stress data , which represents the local stress data of each node in the key area and contains the stress vector of each node.

[0101] If the node , then the corresponding stress vector is denoted as , representing the stress vector corresponding to the - th node. Calculate the mean value of the stress vector to obtain , which represents the arithmetic mean of the stresses of all nodes in the key area.

[0102] Evaluate the balance degree of the stress distribution of the clothing according to the statistical results: Define the stress distribution balance index, and its calculation formula is: ; where represents the stress distribution balance index; represents the number of nodes used for statistics in the node set , that is, the total number of sampled nodes in the key area.

[0103] The larger the stress distribution balance index, the more obvious the imbalance in the local stress distribution within the clothing, indicating that in some key areas, such as the cutting edge and the seam line, the degree of force concentration is relatively large, and the stress borne by the local area is higher than the overall average level. This imbalance may lead to fabric fatigue, local deformation, or even damage, thus affecting the durability and wearing comfort of the clothing.

[0104] Specifically, obtain the user's epidermal sensitivity data through a skin conductance sensor, and use the analytic hierarchy process to evaluate the comfort of the clothing, including:

[0105] Collect skin conductance data in real time according to the skin conductance sensor: During the clothing comfort test, special attention should be paid to the areas where the clothing has frequent contact with the human body and high sensitivity, such as the shoulders, waist, armpits, abdomen, and back. Subsequently, high-precision skin conductance sensors are respectively arranged on the skin surfaces of these areas to collect the conductance signals on the skin surface in real time. Define the set of user's key skin areas as: ; where represents the set of areas where the clothing has frequent contact with the human body and high sensitivity; represents the total number of areas where the clothing has frequent contact with the human body and high sensitivity.

[0106] Place skin conductance sensors in each area to collect skin conductance data in real time, denoted as:

[0107] , ; where represents the skin conductance data of the th area; represents the sampling moment.

[0108] Perform signal filtering and feature extraction on the conductance data to generate skin sensitivity indicators: In order to reduce the interference in the skin conductance data, a Butterworth low-pass filter is used to filter the skin conductance data, and the filtering formula is expressed as: ; where represents the conductance signal of the th area after filtering; represents the filtering function.

[0109] Extract the characteristic conductance value in each area as the skin sensitivity indicator. Adopt the method of combining the mean and peak values of the conductance signal within the statistical window: ; where represents the skin sensitivity indicator of the th area; respectively represent the weight coefficients of the mean and peak values of the conductance signal, which are obtained through experimental fitting; represents the mean value of the conductance signal; represents the peak value of the conductance signal.

[0110] Establish the hierarchical structure of the comfort evaluation index and determine the evaluation factor weight matrix: Use the analytic hierarchy process to establish the comfort evaluation index system:

[0111] Establish the evaluation hierarchical structure according to expert experience and experimental data, denoted as: ; where represents the set of factors for comfort evaluation, including various comfort indicators; represents the total number of comfort evaluation factors, which is determined by experts or experiments.

[0112] Based on expert experience and data support, a pairwise comparison matrix is established, denoted as: , ; where represents the pairwise comparison matrix; represents the importance degree of the -th element in the pairwise comparison matrix relative to the -th evaluation factor; and are the row and column index numbers of the elements in the pairwise comparison matrix, respectively.

[0113] By calculating the maximum eigenvalue and the corresponding eigenvector of the pairwise comparison matrix, the weight matrix of the evaluation factors is obtained: ; where represents the weight vector of the evaluation factors, and the sum of its components is 1; represents the transpose symbol.

[0114] Use the analytic hierarchy process to calculate the comprehensive weights of each sensitive area and sort them: The weights of each area are assigned by the analytic hierarchy process. According to the skin sensitivity index and the weight of the evaluation factors of each area, the comprehensive weight of the area is calculated. The calculation formula is: ; where represents the comprehensive weight of the -th area; represents the normalized value of the sensitivity index of the -th area under the

[0115] Sort the comprehensive weights of all areas from high to low to determine the priority ranking of the user's skin sensitivity.

[0116] According to the comprehensive weight of the area and the skin sensitivity index, calculate the clothing comfort index: Combine the comprehensive weights of all areas and the skin sensitivity index to calculate the clothing comfort index. The calculation formula is: ; where represents the clothing comfort index.

[0117] The larger the clothing comfort index, the more it indicates that the clothing has fully considered ergonomics, fabric characteristics, and user individual differences during the design and production process. It can effectively disperse and balance the pressure and friction generated during wearing, thereby reducing skin irritation and discomfort. A higher comfort index reflects that the clothing has reached a high level in aspects such as pattern design, cutting process, sewing structure, and material selection, ensuring that the clothing can closely fit the body curves and provide good freedom of movement and wearing experience. Overall, the higher this index, the more it can meet the user's requirements for comfort and fit in different activity scenarios, thereby enhancing the satisfaction and health of daily wearing.

[0118] Specifically, analyze the adaptability of the clothing pattern based on the stress distribution balance degree of the clothing and the comfort of the clothing, including:

[0119] Normalize the stress distribution balance index and the clothing comfort index. After assigning preset proportional coefficients to the normalized stress distribution balance index and the clothing comfort index respectively, calculate the adaptability score. For example:

[0120] The calculation formula for the adaptability score is: ; where are the preset proportional coefficients of the stress distribution balance index and the clothing comfort index respectively, satisfying .

[0121] Compare the calculated adaptability score with a preset adaptability score threshold:

[0122] When the adaptability score is greater than or equal to the adaptability score threshold, it indicates that the adaptability of the clothing pattern is good; the clothing shows uniform stress distribution and a high level of comfort during wearing, and can achieve a perfect balance between fit and support, enhancing the wearing experience and freedom of movement;

[0123] When the adaptability score is less than the adaptability score threshold, it indicates that the adaptability of the clothing pattern is poor; at this time, the clothing may have local stress concentration, uneven stress distribution, and discomfort during wearing. Users may feel local tightness or looseness during exercise or long-term wearing, and it is necessary to optimize and adjust the pattern parameters or production process to improve the overall adaptability.

[0124] The adaptability score threshold is a quantitative standard determined comprehensively based on a large amount of user trial-wearing data, engineering experiments, and expert reviews, reflecting the minimum acceptance level of the clothing in terms of structural balance, material performance, and wearing comfort.

[0125] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data and performing software simulations to get a formula that is closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0126] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The 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, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0127] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0128] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0129] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or modules can be in electrical, mechanical, or other forms.

[0130] The modules described as separate components may or may not be physically separated. The components displayed as modules may or may not be physical modules. They can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0131] In addition, in each embodiment of the present application, the functional modules can be integrated into a processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0132] If the above functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0133] As described above, this is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0134] Finally, the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent clothing design method based on human body data, characterized in that, It includes the following steps: By monitoring the multi-dimensional posture data of the user in the daily environment, a deep learning skeleton key point recognition algorithm is used to evaluate the degree of misalignment of the user's daily behavior; Based on the degree of misalignment of the user's daily behavior, the correction parameters of the clothing pattern are determined; Using the domain decomposition algorithm, the human body is divided into different blocks: in the user's body posture analysis, the deviation between the actual posture and the standard posture of the user usually exceeds the acceptable range and cannot indicate which specific part needs intervention the most. It is necessary to divide the human body into different blocks through the domain decomposition algorithm; Introduction represents the total number of human body regions. The shoulder and neck region can be regarded as the first region, the chest and back region as the second region, the waist and abdomen region as the third region, and the limbs region as the fourth region. Use to identify the region number, that is ; For the clothing pattern determined according to the correction parameters: by analyzing the clothing mechanics model parameters, the stress distribution balance degree of the clothing is evaluated; by obtaining the user's epidermal sensitivity data through a skin conductance sensor, the analytic hierarchy process is used to evaluate the comfort of the clothing; Based on the stress distribution balance degree of the clothing and the comfort of the clothing, the fitting degree of the clothing pattern is analyzed.

2. The intelligent clothing design method based on human body data according to claim 1, wherein By monitoring the multi-dimensional posture data of the user in the daily environment, a deep learning skeleton key point recognition algorithm is used to evaluate the degree of misalignment of the user's daily behavior. Specifically: Image acquisition devices and sensor devices are arranged in the user's daily scene to collect multi-angle posture data in real time; Preprocess the collected multi-angle posture data; Use a convolutional neural network to achieve precise positioning of skeleton key points; Compare the positioning data of the skeleton key points with the standard body posture model to quantify the user's behavior deviation.

3. The intelligent clothing design method based on human body data according to claim 2, wherein, Compare the positioning data of the skeleton key points with the standard body posture model to quantify the user's behavior deviation. Specifically: Compare the overall deviation index with the preset standard threshold to determine the deviation between the user's actual posture and the standard posture: When the overall deviation index is greater than or equal to the preset standard threshold, it is determined that the deviation between the user's actual posture and the standard posture exceeds the acceptable range; When the overall deviation index is less than the preset standard threshold, it is determined that the deviation between the user's actual posture and the standard posture is within the acceptable range.

4. An intelligent clothing design method based on human body data according to claim 3, characterized in that, Based on the degree of misalignment of the user's daily behavior, the correction parameters of the clothing pattern are determined. Specifically: Build a clothing mechanics model based on finite element analysis, and define the structural nodes, material properties and boundary conditions: Use the domain decomposition algorithm to divide the human body into different blocks; Based on the sub-region data, use a linear regression equation to calculate the correction coefficient of each human body block; Integrate the correction coefficients of different human body blocks to generate the clothing pattern correction parameters suitable for the user's body posture.

5. The intelligent clothing design method based on human body data according to claim 4, characterized in that By analyzing the clothing mechanics model parameters, the stress distribution balance degree of the clothing is evaluated. Specifically: Build a clothing mechanics model based on finite element analysis, and define the material properties and boundary conditions; Perform numerical simulation to calculate the stress distribution; Collect local stress data and statistically analyze the stress of each node; Evaluate the stress distribution balance degree of the clothing according to the statistical results.

6. The intelligent clothing design method based on human body data according to claim 5, wherein By obtaining the user's epidermal sensitivity data through a skin conductance sensor, the analytic hierarchy process is used to evaluate the comfort of the clothing. Specifically: Collect skin conductance data in real time according to the skin conductance sensor; Filter the conductance data and extract features to generate skin sensitivity indicators; Establish a comfort evaluation index hierarchy structure and determine the evaluation factor weight matrix; Calculate the clothing comfort index according to the regional comprehensive weight and skin sensitivity index.

7. A method for intelligent clothing design based on human body data according to claim 1, characterized in that Analyze the adaptability of the clothing pattern based on the stress distribution balance degree of the clothing and the comfort of the clothing. Specifically: Normalize the stress distribution balance index and the clothing comfort index. After assigning preset proportional coefficients to the normalized stress distribution balance index and clothing comfort index respectively, calculate the fitness score; Compare the fitness score with the fitness score threshold: when the fitness score is greater than or equal to the fitness score threshold, it indicates that the adaptability of the clothing pattern is good; when the fitness score is less than the fitness score threshold, it indicates that the adaptability of the clothing pattern is poor.

Images