Intelligent clothing design method and system based on customer preference

Abstract

The present application discloses a method and system for intelligent clothing design based on customer preferences, belonging to the technical field of clothing design, the method comprising: identifying the customer's body type; determining the clothing style to be selected based on a deep learning neural network; determining the clothing color to be selected; obtaining a three-dimensional model of the customer's body; determining the clothing type, color and style of the clothing type that the customer needs to search for, and putting the clothing selected by the customer on the three-dimensional body model for a preview of the effect; after the customer has determined the selected clothing, displaying the fabric material and fabric price information for the customer to choose; after the customer has determined the selected fabric, counting the area of ​​each fabric used, the total purchase price of the fabric and the labor cost, and completing the self-selected clothing design. Compared with the current purchase of finished clothing, the present application allows customers to selectively screen clothing styles, clothing colors, and clothing fabrics to meet the customer's personalized requirements, improve customer experience, and make prices more transparent.

Description

Technical Field

[0001] The present application relates to the technical field of clothing design, and in particular to a method and system for intelligent clothing design based on customer preferences. Background Art

[0002] There are standard sizes of clothing produced by clothing companies, and there are also customized clothing. Customized clothing is tailored according to the customer's body shape. The traditional design method is that the tailor uses a ruler to measure the dimensions of various parts of the body and then designs. The process is boring and the errors are large. Usually, due to errors in certain dimensions, the final designed clothing is unsatisfactory. In addition, after the customer selects the design style and style, he cannot quickly view the design renderings, and the design efficiency is low.

[0003] Existing clothing design is designed by designers based on their subjective consciousness. However, when designers design, they often improve and optimize existing elements. Designers generally draw and display manually, and their three-dimensional images are not intuitive enough, so designers cannot intuitively feel their own design samples. Improving the actual clothing after it is made affects the progress of the design, and the workload is large. At present, there is an urgent need for a clothing design method based on customer preferences to realize customers' independent design and quickly view the design effect. Therefore, the existing technology still has the problem of not being able to realize customers' independent personalized design based on customer preferences in clothing design. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a method, system, device and storage medium for intelligent clothing design based on customer preferences, so as to solve the problem that the prior art in clothing design is still unable to enable customers to perform autonomous personalized design based on customer preferences.

[0005] In order to solve the above technical problems, the embodiment of the present application provides a clothing intelligent design method based on customer preferences, which adopts the following technical solutions:

[0006] An intelligent clothing design method based on customer preferences, comprising:

[0007] Based on the human-computer interaction mode, the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer are obtained, the parameters are prioritized, and the parameters are transferred into a preset body shape recognition model according to the priority order to identify the body shape type corresponding to the customer;

[0008] Based on the big data platform, network tasks are captured to obtain a number of pictures corresponding to hats, tops, pants, and shoes, and the pictures are classified into styles based on the deep learning neural network as the styles to be selected;

[0009] Identify the colors in the pictures corresponding to the hat, coat, pants and shoes, and obtain the top n colors of clothing used the most in the pictures as the colors to be selected based on a bubble sort method, where n is a positive integer;

[0010] Based on the customer's corresponding body type, reverse engineering is performed to obtain the customer's corresponding body three-dimensional model;

[0011] Based on the preset display interface, a search box is provided for the customer to input, the character string input by the customer is split, the type of clothing to be searched, the color and style of the clothing type are determined through the split character string, and the effect of wearing the clothing selected by the customer on the three-dimensional body model is previewed;

[0012] After the customer has determined the selected clothing, the fabric material and fabric price information that can be used to cut the clothing are displayed to the customer on a preset display interface for selection;

[0013] After the customer has decided on the fabric of their choice, the preset price statistics algorithm is used to calculate the area of ​​each fabric used, the total purchase price of the fabric and the labor costs to complete the self-selected clothing design.

[0014] Furthermore, the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer are obtained, and the priority of the parameters is set. The specific setting method is:

[0015] Setting the height and weight as first priority parameters;

[0016] Set gender as the second priority parameter;

[0017] Set shoulder width, chest circumference, and hip width as the third priority parameter.

[0018] Furthermore, the parameters are input into a preset body shape recognition model according to the priority order to identify the body shape type corresponding to the customer. The specific identification method is:

[0019] The first priority parameter is passed into the preset body shape recognition model to obtain the first fuzzy body shape type corresponding to the customer;

[0020] The second priority parameter is passed into a preset body shape recognition model, and based on the first fuzzy body shape type, a second fuzzy body shape type corresponding to the customer is obtained;

[0021] The third priority parameter is passed into the preset body shape recognition model, and based on the second fuzzy body shape type, a third body shape type corresponding to the customer is obtained as the body shape type corresponding to the customer;

[0022] The preset body shape recognition model pre-caches several human body shape types in the form of three-layer cache. During recognition, the first priority parameter is used to determine the initial search range, and the result after the initial search is obtained as the first fuzzy body shape type. The second priority parameter is then used to perform a secondary screening on the result after the initial search, and the secondary screening result is obtained as the second fuzzy body shape type. Finally, the third priority parameter is used to perform a second screening on the secondary screening result, and the second screening result is obtained as the body shape type corresponding to the customer.

[0023] Furthermore, the network task crawling based on the big data platform is performed to obtain several pictures corresponding to the hat, top, pants and shoes respectively, and the specific implementation method is as follows:

[0024] Before performing network task crawling based on the big data platform and obtaining a number of pictures corresponding to hats, tops, pants, and shoes, respectively, a hat picture collection, a top picture collection, a pants picture collection, and a shoe picture collection are pre-built;

[0025] When performing network task crawling based on the big data platform and obtaining several pictures corresponding to hats, tops, pants and shoes respectively, hats, tops, pants and shoes are used as search terms respectively, and the cache paths corresponding to the hat picture set, top picture set, pants picture set and shoe picture set are used as cache addresses, and the retrieved pictures are stored in the corresponding cache addresses.

[0026] Furthermore, the style classification of the plurality of pictures based on the deep learning neural network is specifically implemented as follows:

[0027] Based on a neural network, feature values ​​of the pictures in the hat picture set, the top picture set, the pants picture set, and the shoe picture set are respectively acquired;

[0028] The feature values ​​are classified in advance according to the value range, and the pictures corresponding to the feature values ​​in the same value range are put into a preset array, and the arrays are named differently. The external data is constructed as a collection type, and the internal data is constructed as json type data composed of several arrays, where each array corresponds to a style.

[0029] Furthermore, when constructing the hat picture set, top picture set, pants picture set and shoe picture set, the hat picture set, top picture set, pants picture set and shoe picture set are distinctively named using binary values ​​"00", "01", "10" and "11" respectively.

[0030] Furthermore, the feature values ​​are classified in advance according to the value range, the pictures corresponding to the feature values ​​in the same value range are put into a preset array, and the arrays are distinguished by naming, specifically in the following manner:

[0031] Obtaining the maximum and minimum values ​​of the eigenvalues, and determining the initial interval corresponding to the eigenvalues;

[0032] Divide the initial interval into 8 range intervals according to the values, set a corresponding array for each range interval, and name them one by one using binary values ​​"000", "001", "010", "011", "100", "101", "110" and "111";

[0033] After deep learning, the eigenvalue corresponding to the current image is obtained, the corresponding eigenvalue range is determined, and the current image is placed in the corresponding array.

[0034] Furthermore, the method of obtaining the top n colors most used in the clothing in the plurality of pictures as the colors to be selected based on the bubble sorting method also includes:

[0035] Based on gender, the colors to be selected are divided into two different categories: male colors and female colors;

[0036] When customers choose clothing colors, they are directly provided with the color scheme corresponding to their gender for selection.

[0037] Furthermore, based on the preset display interface, a search box is provided for the customer to input, the character string input by the customer is split, and the clothing type, color and style of the clothing type to be searched are determined by the split character string. The specific determination method is:

[0038] The search box provided for the customer to input is limited to the customer to first input a five-digit binary value, the five-digit binary value is split into a two-digit binary value and a three-digit binary value, the clothing type searched by the customer is determined based on the two-digit binary value, and the style classification of the clothing type searched by the customer is determined based on the three-digit binary value, wherein the clothing type includes: hats, tops, pants and shoes;

[0039] After determining the style category of the clothing type searched by the customer based on the three-digit binary value, all styles under the same style category are displayed on the preset display interface in a list display manner for the customer to select;

[0040] After the customer has selected the style, the numbers of the colors to be selected are provided for the customer to select and color the clothes of the selected style, and the coloring result is used as the final selected clothes.

[0041] In order to solve the above technical problems, the embodiment of the present application also provides a clothing intelligent design system based on customer preferences, which adopts the following technical solutions:

[0042] An intelligent clothing design system based on customer preferences, comprising:

[0043] A body shape recognition module is used to obtain the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer based on a human-computer interaction mode, set priorities for the parameters, and transfer the parameters to a preset body shape recognition model according to the priority order to identify the body shape type corresponding to the customer;

[0044] The style acquisition and classification module is used to crawl network tasks based on the big data platform, obtain several pictures corresponding to hats, tops, pants and shoes respectively, and classify the pictures into styles based on the deep learning neural network as the styles to be selected;

[0045] A module for determining colors to be selected is used to identify the colors in the pictures corresponding to the hat, coat, pants and shoes, and obtain the top n colors used the most in the clothes in the pictures based on a bubble sort method as the colors to be selected, where n is a positive integer;

[0046] A body 3D model building module is used to perform reverse engineering based on the body type corresponding to the customer to obtain the body 3D model corresponding to the customer;

[0047] The clothing selection module is used to provide a search box for customers to input based on a preset display interface, split the string input by the customer, determine the type of clothing to be searched, the color and style of the clothing type through the split string, and preview the effect of the clothing selected by the customer on the three-dimensional body model;

[0048] The fabric cutting selection module is used to display the fabric material and fabric price information that can be used to cut the garment to the customer for selection after the customer has determined the selected garment.

[0049] The clothing price statistics module is used to determine the area of ​​each fabric used, the total purchase price of the fabric and the labor cost based on the preset price statistics algorithm after the customer has determined the selected fabric, so as to complete the self-selected clothing design.

[0050] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0051] The embodiment of the present application discloses a method and system for intelligent clothing design based on customer preferences, which identifies the customer's body type; determines the style of clothing to be selected based on a deep learning neural network; determines the color of clothing to be selected; obtains a three-dimensional model of the customer's body; determines the type of clothing, color and style of the clothing type that the customer needs to search for, and previews the effect of the clothing selected by the customer on the three-dimensional body model; after the customer has determined the selected clothing, displays the fabric material and fabric price information for the customer to choose; after the customer has determined the selected fabric, counts the area of ​​each fabric used, the total purchase price of the fabric and the labor cost, and completes the self-selected clothing design. Compared with the current purchase of finished clothing, the present application allows customers to selectively screen clothing styles, clothing colors, and clothing fabrics to meet the customer's personalized requirements, improve customer experience, and make prices more transparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 A flowchart of an embodiment of a clothing intelligent design method based on customer preferences in an embodiment of the present application;

[0054] Figure 2 This is an embodiment of the clothing intelligent design system based on customer preferences in the embodiments of the present application;

[0055] Figure 3 This is an execution logic diagram of an embodiment of the clothing intelligent design method based on customer preferences in the embodiments of the present application. DETAILED DESCRIPTION

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0057] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0059] It should be noted that the intelligent clothing design method based on customer preferences provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the intelligent clothing design system based on customer preferences is generally set in the server / terminal device.

[0060] refer to Figure 1 , the figure shows a flow chart of an embodiment of the clothing intelligent design method based on customer preferences of the present application, the clothing intelligent design method based on customer preferences comprises the following steps:

[0061] Step 101, based on the human-computer interaction mode, obtain the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer, set priorities for the parameters, and transfer the parameters to a preset body shape recognition model according to the priority order to identify the body shape type corresponding to the customer.

[0062] In some embodiments of the present application, the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer are obtained, and the parameters are prioritized. The specific setting method is: setting the height and weight as the first priority parameters; setting the gender as the second priority parameter; setting the shoulder width, chest circumference, and hip width as the third priority parameters.

[0063] In some embodiments of the present application, the parameters are passed into a preset body shape recognition model in the order of priority to identify the body shape type corresponding to the customer. The specific identification method is as follows: the first priority parameter is passed into the preset body shape recognition model to obtain the first fuzzy body shape type corresponding to the customer; the second priority parameter is passed into the preset body shape recognition model to obtain the second fuzzy body shape type corresponding to the customer based on the first fuzzy body shape type; the third priority parameter is passed into the preset body shape recognition model to obtain the third body shape type corresponding to the customer based on the second fuzzy body shape type as the body shape type corresponding to the customer; the preset body shape recognition model caches several human body shape types in advance in the form of three-layer cache. During identification, the first priority parameter is first used to determine the preliminary search range, and the result after the preliminary search is obtained as the first fuzzy body shape type. The second priority parameter is then used to perform a secondary screening on the result after the preliminary search, and the secondary screening result is obtained as the second fuzzy body shape type. Finally, the secondary screening result is screened again by the third priority parameter, and the secondary screening result is obtained as the body shape type corresponding to the customer.

[0064] Explanation: By dividing the parameters input by the customer into three priority levels, height and weight are used as the first priority parameters to first obtain the customer's approximate body shape. Then, the customer's gender is obtained as the second priority parameter to distinguish between male and female customers. Finally, the customer's shoulder width, chest circumference, and hip width are obtained as the third priority parameters to determine the customer's body type.

[0065] Step 102, crawling network tasks based on the big data platform, obtaining a number of pictures corresponding to hats, tops, pants and shoes respectively, and classifying the styles of the pictures based on the deep learning neural network as styles to be selected.

[0066] In some embodiments of the present application, the network task crawling based on the big data platform is performed to obtain several pictures corresponding to hats, tops, pants and shoes respectively. The specific implementation method is as follows: before the network task crawling based on the big data platform is performed to obtain several pictures corresponding to hats, tops, pants and shoes respectively, a hat picture set, a top picture set, a pants picture set and a shoe picture set are pre-constructed; when the network task crawling based on the big data platform is performed to obtain several pictures corresponding to hats, tops, pants and shoes respectively, hats, tops, pants and shoes are used as search terms respectively, and the cache paths corresponding to the hat picture set, top picture set, pants picture set and shoe picture set are used as cache addresses, and the retrieved pictures are stored in the corresponding cache addresses.

[0067] In some embodiments of the present application, when constructing the hat picture set, top picture set, pants picture set and shoe picture set, the hat picture set, top picture set, pants picture set and shoe picture set are distinctively named using binary values ​​"00", "01", "10" and "11" respectively.

[0068] Explanation: When capturing images, first, construct a hat image set, a top image set, a pants image set and a shoe image set. In order to facilitate distinction, the above sets are named with binary values ​​"00", "01", "10" and "11" respectively. When hat is used as the search term, the image is cached in the set named "00". Similarly, when top is used as the search term, the image is cached in the set named "01". When pants are used as the search term, the image is cached in the set named "10". When shoes are used as the search term, the image is cached in the set named "11".

[0069] In some embodiments of the present application, the style classification of the several pictures is performed based on a deep learning neural network, and the specific implementation method is: based on the neural network, feature values ​​of the pictures in the hat picture set, top picture set, pants picture set and shoe picture set are respectively obtained; the feature values ​​are pre-classified according to the value range, and the pictures corresponding to the feature values ​​in the same value range are put into a preset array, and the arrays are named differently, and the external data is constructed as a collection type, and the internal data is constructed as json type data consisting of several arrays, wherein each array corresponds to a style.

[0070] Explanation: The hat picture set, top picture set, pants picture set and shoe picture set are taken as four test sets for deep learning, and the feature values ​​of the pictures in the test sets are obtained based on the preset neural network learning model respectively. Taking the hat picture set as an example, first, the neural network learning model is started, and the hat picture set is passed into the learning model as the current test set to obtain the feature values ​​corresponding to each picture in the hat picture set.

[0071] In some embodiments of the present application, the eigenvalues ​​are pre-classified according to value ranges, and pictures corresponding to eigenvalues ​​within the same value range are placed in a preset array, and the arrays are named differently, specifically in the following manner: the maximum and minimum values ​​of the eigenvalues ​​are obtained, and an initial interval corresponding to the eigenvalues ​​is determined; the initial interval is divided into 8 range intervals according to numerical values, a corresponding array is set for each range interval, and the arrays are named one by one using binary values ​​"000", "001", "010", "011", "100", "101", "110" and "111"; after deep learning, the eigenvalue corresponding to the current picture is obtained, the corresponding eigenvalue range is determined, and the current picture is placed in its corresponding array.

[0072] Explanation: After obtaining the eigenvalues ​​corresponding to each picture in the hat picture set, construct a eigenvalue set for the eigenvalues, obtain the maximum and minimum eigenvalues ​​in the eigenvalue set, determine the initial interval corresponding to the eigenvalue, divide the interval equally, and divide the eigenvalues ​​in the eigenvalue set into 8 range intervals. Use binary values ​​"000", "001", "010", "011", "100", "101", "110" and "111" as array names for the above range intervals, name them one by one, obtain the pictures corresponding to the eigenvalues ​​in the interval, and store the pictures in the corresponding array. It should be noted that the corresponding relationship between eigenvalues ​​and pictures is a one-to-many relationship, that is, the same eigenvalue may correspond to multiple pictures. Compared with manual selection and classification of styles, style types can be directly divided by eigenvalues, which can quickly classify batches of pictures into styles and avoid time consumption during manual classification.

[0073] Step 103, identifying the colors in the several pictures corresponding to the hat, top, pants and shoes, and obtaining the top n colors most used in the clothing in the several pictures based on a bubble sort method as the colors to be selected, where n is a positive integer.

[0074] In some embodiments of the present application, the bubble sort-based method obtains the top n colors most commonly used in the clothing in the several pictures as the colors to be selected, and also includes: based on gender, dividing the colors to be selected into two different categories: male colors and female colors; when the customer selects the color of the clothing, the color system corresponding to his or her gender is directly provided for his or her selection.

[0075] Explanation: Based on the RGB extraction method, the colors appearing in the pictures are obtained and expressed in RGB form. The number of pictures in which each RGB value appears is determined, and the colors corresponding to the first n RGB values ​​in which the largest number of pictures appear are selected as the colors to be selected. In order to facilitate the customer's independent selection, male and female color systems are set according to the different color choices of men and women for clothing.

[0076] Step 104, based on the body type corresponding to the customer, reverse engineering is performed to obtain a body three-dimensional model corresponding to the customer.

[0077] Explanation: The reverse engineering directly uses the height, weight, shoulder width, chest circumference, and hip width input by the customer as reverse engineering parameters to obtain a simple three-dimensional model of the body, eliminating the accuracy requirements for equipment and parameters during precise modeling.

[0078] Step 105, based on a preset display interface, provides a search box for the customer to input, splits the character string input by the customer, determines the type of clothing to be searched, the color and style of the clothing type through the split character string, and previews the effect of wearing the clothing selected by the customer on the three-dimensional body model.

[0079] In some embodiments of the present application, a search box is provided for customers to input based on a preset display interface, and the character string input by the customer is split. The type of clothing that the customer needs to search for, the color and style of the clothing type are determined through the split character string. The specific determination method is as follows: the search box provided for customers to input limits the customer to enter a five-digit binary value first, and the five-digit binary value is split into a two-digit binary value and a three-digit binary value. The type of clothing searched by the customer is determined based on the two-digit binary value, and the style classification of the clothing type searched by the customer is determined based on the three-digit binary value, wherein the clothing types include: hats, tops, pants and shoes; after determining the style classification of the clothing type searched by the customer based on the three-digit binary value, all styles under the same style classification are displayed on the preset display interface based on a list display method for the customer to choose; after the customer selects the style, the number of each color to be selected is provided, so that the customer can select and color the clothes of the selected style, and the coloring result is used as the final selected clothing.

[0080] Explanation: First, the customer is restricted to input a five-digit binary value, such as "00001". The first two digits of the binary value are used to distinguish and name the hat image set, top image set, pants image set and shoe image set in step 102, indicating that the set names "00", "01", "10" and "11" corresponding to the clothing types correspond. The last three digits correspond to the style classification names corresponding to different clothing types in step 102. By splitting the five-digit binary value, the clothing type retrieved by the customer and the style classification of the clothing type are determined. Then, all styles corresponding to the style classification are displayed in a preset interface for the customer to choose. After the style selection is completed, the number of the color to be selected in step 103 is provided for the customer to preview the coloring of the clothing after selecting the style.

[0081] Step 106, after the customer has confirmed the selected clothing, the fabric material and fabric price information that can be used to cut the clothing are displayed to the customer on a preset display interface for selection.

[0082] In some embodiments of the present application, after the customer has determined the selected clothing, the fabric material and fabric price information that can be used to cut the clothing are displayed to the customer on a preset display interface for selection. The specific implementation method is: the material information and price information of several types of processed fabrics are pre-stored in a preset database. After the customer's final selection of clothing is determined, the fabric data is displayed in a list form on a preset human-computer interaction display interface for the customer to screen clothing fabrics according to personal preferences.

[0083] Compared with the current purchase of ready-made clothing, customers can selectively select clothing styles, colors, and fabrics, which can better meet their personalized requirements and improve customer experience.

[0084] Step 107, after the customer has determined the selected fabric, the area of ​​each fabric used, the total purchase price of the fabric and the labor cost are counted based on a preset price statistical algorithm to complete the self-selected clothing design.

[0085] In some embodiments of the present application, after the customer has determined the selected fabric, the area of ​​each fabric used, the total purchase price of the fabric and the labor cost are counted based on a preset price statistical algorithm. The specific implementation method is: a preset team of tailors estimates the required fabric area based on the clothing style selected by the customer, determines the total price required to purchase the required fabric area through the fabric price information in step 106, adds the labor production cost, determines the final price of the clothing, and completes the self-selected clothing design.

[0086] Compared with the current purchase of ready-made clothing, customers can clearly understand the price of fabrics and labor costs, and clothing prices are more transparent.

[0087] The intelligent clothing design method based on customer preferences described in the embodiments of the present application can identify the customer's body type; determine the style of clothing to be selected based on a deep learning neural network; determine the color of clothing to be selected; obtain a three-dimensional model of the customer's body; determine the type of clothing that the customer needs to search for, the color and style of the clothing type, and preview the effect of the clothing selected by the customer on the three-dimensional body model; after the customer has determined the selected clothing, display the fabric material and fabric price information for the customer to choose; after the customer has determined the selected fabric, count the area of ​​each fabric used, the total purchase price of the fabric and the labor cost, and complete the self-selected clothing design. Compared with the current purchase of finished clothing, the present application allows customers to selectively screen clothing styles, clothing colors, and clothing fabrics to meet the customer's personalized requirements, improve customer experience, and make prices more transparent.

[0088] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0089] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0090] Further references Figure 2 , as a response to the above Figure 1 The present application provides an embodiment of a clothing intelligent design system based on customer preferences, and the system embodiment is similar to Figure 1 Corresponding to the method embodiment shown, the system can be specifically applied to various electronic devices.

[0091] like Figure 2 As shown, the clothing intelligent design system 2 based on customer preferences described in this embodiment includes: a body shape recognition module 201, a style acquisition classification module 202, a to-be-selected color determination module 203, a body three-dimensional model construction module 204, a clothing matching selection module 205, a cutting fabric selection module 206 and a clothing price statistics module 207. Among them:

[0092] The body shape recognition module 201 is used to obtain the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer based on the human-computer interaction mode, set the priority of the parameters, and transfer the parameters to the preset body shape recognition model according to the priority order to identify the body shape type corresponding to the customer;

[0093] The style acquisition and classification module 202 is used to perform network task crawling based on the big data platform, respectively acquire a number of pictures corresponding to hats, tops, pants and shoes, and classify the pictures into styles based on a deep learning neural network as styles to be selected;

[0094] The module 203 for determining the color to be selected is used to identify the colors in the pictures corresponding to the hat, coat, pants and shoes, and obtain the top n colors of the clothes in the pictures based on the bubble sort method as the colors to be selected, where n is a positive integer;

[0095] A body three-dimensional model building module 204 is used to perform reverse engineering based on the body type corresponding to the customer to obtain the body three-dimensional model corresponding to the customer;

[0096] The clothing selection module 205 is used to provide a search box for the customer to input based on a preset display interface, split the character string input by the customer, determine the type of clothing to be searched, the color and style of the clothing type through the split character string, and preview the effect of the clothing selected by the customer on the three-dimensional body model;

[0097] The fabric cutting selection module 206 is used to display the fabric material and fabric price information that can be used to cut the garment to the customer for selection after the customer has determined the selected garment.

[0098] The clothing price statistics module 207 is used to determine the area of ​​each fabric used, the total purchase price of the fabric and the labor cost based on a preset price statistics algorithm after the customer has determined the selected fabric, so as to complete the self-selected clothing design.

[0099] The intelligent clothing design system based on customer preferences described in the embodiment of the present application identifies the customer's body type; determines the style of clothing to be selected based on a deep learning neural network; determines the color of clothing to be selected; obtains a three-dimensional model of the customer's body; determines the type of clothing, color and style of the clothing type that the customer needs to search for, and previews the effect of the clothing selected by the customer on the three-dimensional body model; after the customer has determined the selected clothing, displays the fabric material and fabric price information for the customer to choose; after the customer has determined the selected fabric, counts the area of ​​each fabric used, the total purchase price of the fabric and the labor cost to complete the self-selected clothing design. Compared with the current purchase of finished clothing, the present application allows customers to selectively screen clothing styles, clothing colors, and clothing fabrics to meet the customer's personalized requirements, improve customer experience, and make prices more transparent.

[0100] Continue to refer Figure 3 , Figure 3The following is an execution logic diagram of an embodiment of the clothing intelligent design method based on customer preferences in the embodiment of the present application, including:

[0101] Based on the human-computer interaction mode, the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer are obtained, and the height and weight are set as the first priority parameters; the gender is set as the second priority parameter; the shoulder width, chest circumference, and hip width are set as the third priority parameters, and the body type corresponding to the customer is obtained by passing the priority parameters into the preset body recognition model step by step, and a hat picture set, a top picture set, a trouser picture set, and a shoe picture set are pre-built; network task crawling is performed based on the big data platform, and several pictures corresponding to hats, tops, trousers, and shoes are respectively obtained, and hats, tops, trousers, and shoes are respectively used as search terms, and the cache paths corresponding to the hat picture set, top picture set, trousers picture set, and shoe picture set are used as cache addresses, and the retrieved pictures are stored in the corresponding cache addresses, and the hat picture set, top picture set, trousers picture set, and shoe picture set are distinguished and named, and the hat picture set, top picture set, trousers picture set, and shoe picture set are respectively identified based on the neural network. Acquire characteristic values ​​of pictures in a picture collection; classify the characteristic values ​​in advance according to value ranges, put pictures corresponding to characteristic values ​​within the same value range into a preset array, and distinguish and name the arrays, construct json type data with external data as a collection type and internal data consisting of several arrays, wherein each array corresponds to a style, obtain the maximum value and the minimum value of the characteristic value, and determine the initial interval corresponding to the characteristic value; divide the initial interval into 8 range intervals according to the numerical value, set a corresponding array for each range interval, obtain the characteristic value corresponding to the current picture, judge the corresponding characteristic value range, put the current picture into its corresponding array as a style to be selected, identify the colors in several pictures corresponding to the hat, top, pants and shoes, and obtain the top n colors of clothing used most in the several pictures based on the bubble sort method as the colors to be selected, wherein n is a positive integer, and perform reverse engineering based on the body type corresponding to the customer to obtain the corresponding three-dimensional body model of the customer. Based on a preset display interface, a search box is provided for customers to input, and the character string input by the customer is split. The type of clothing that the customer needs to search for, the color and style of the clothing type are determined through the split character string, and the effect of the clothing selected by the customer is previewed on the three-dimensional model of the body. After the customer has determined the selected clothing, the fabric material and fabric price information that can be used to cut the clothing are displayed to the customer on the preset display interface for selection. After the customer has determined the selected fabric, based on the preset price statistical algorithm, the area of ​​each fabric used, the total purchase price of the fabric and the labor cost are counted to complete the self-selected clothing design.

[0102] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0103] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A clothing intelligent design method based on customer preferences, characterized in that: The steps include: Step 101, based on the human-computer interaction mode, obtaining the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer, setting priorities for the parameters, and transferring the parameters to a preset body shape recognition model in order of priority to identify the body shape type corresponding to the customer; the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer are obtained, and the priorities for the parameters are set, and the specific setting method is: setting the height and weight as the first priority parameters; setting the gender as the second priority parameter; setting the shoulder width, chest circumference, and hip width as the third priority parameters; Step 102, crawling network tasks based on the big data platform, obtaining a number of pictures corresponding to hats, tops, pants and shoes, respectively, and classifying the pictures into styles based on a deep learning neural network as styles to be selected; Step 103, identifying the colors in the pictures corresponding to the hat, coat, pants and shoes, and obtaining the top n colors of the clothes in the pictures based on a bubble sort method as the colors to be selected, where n is a positive integer; Step 104, based on the body type corresponding to the customer, reverse engineering is performed to obtain a body three-dimensional model corresponding to the customer; the reverse engineering directly uses the height, weight, shoulder width, chest circumference, and hip width input by the customer as reverse engineering parameters to obtain a simple body three-dimensional model, eliminating the accuracy requirements for equipment and parameters during accurate modeling; Step 105, based on the preset display interface, a search box is provided for the customer to input, the character string input by the customer is split, the clothing type to be searched, the color and style of the clothing type are determined through the split character string, and the effect of wearing the clothing selected by the customer on the three-dimensional body model is previewed; Step 106, after the customer has confirmed the selected clothing, the fabric material and fabric price information that can be used to cut the clothing are displayed to the customer on a preset display interface for selection; Step 107, after the customer has determined the selected fabric, the area of ​​each fabric used, the total purchase price of the fabric and the labor cost are counted based on a preset price statistical algorithm to complete the self-selected clothing design.

2. The intelligent clothing design method based on customer preferences according to claim 1 is characterized in that: According to the priority order, the parameters are passed into a preset body shape recognition model to identify the body shape type corresponding to the customer. The specific identification method is: The first priority parameter is passed into the preset body shape recognition model to obtain the first fuzzy body shape type corresponding to the customer; The second priority parameter is passed into a preset body shape recognition model, and based on the first fuzzy body shape type, a second fuzzy body shape type corresponding to the customer is obtained; The third priority parameter is passed into the preset body shape recognition model, and based on the second fuzzy body shape type, a third body shape type corresponding to the customer is obtained as the body shape type corresponding to the customer; The preset body shape recognition model pre-caches several human body shape types in the form of three-layer cache. During recognition, the first priority parameter is used to determine the initial search range, and the result after the initial search is obtained as the first fuzzy body shape type. The second priority parameter is then used to perform a secondary screening on the result after the initial search, and the secondary screening result is obtained as the second fuzzy body shape type. Finally, the third priority parameter is used to perform a second screening on the secondary screening result, and the second screening result is obtained as the body shape type corresponding to the customer.

3. The clothing intelligent design method based on customer preference according to claim 2 is characterized in that: The network task crawling based on the big data platform is performed to obtain several pictures corresponding to hats, tops, pants and shoes respectively. The specific implementation method is as follows: Before performing network task crawling based on the big data platform and obtaining a number of pictures corresponding to hats, tops, pants, and shoes, respectively, a hat picture collection, a top picture collection, a pants picture collection, and a shoe picture collection are pre-built; When performing network task crawling based on the big data platform and obtaining several pictures corresponding to hats, tops, pants and shoes respectively, hats, tops, pants and shoes are used as search terms respectively, and the cache paths corresponding to the hat picture set, top picture set, pants picture set and shoe picture set are used as cache addresses, and the retrieved pictures are stored in the corresponding cache addresses.

4. The intelligent clothing design method based on customer preference according to claim 3 is characterized in that: When constructing the hat picture set, top picture set, pants picture set and shoe picture set, the hat picture set, top picture set, pants picture set and shoe picture set are distinguished and named using binary values ​​"00", "01", "10" and "11" respectively.

5. The intelligent clothing design method based on customer preference according to claim 3 is characterized in that: The method of classifying the styles of the plurality of pictures based on the deep learning neural network is specifically implemented as follows: Based on a neural network, feature values ​​of the pictures in the hat picture set, the top picture set, the pants picture set, and the shoe picture set are respectively acquired; The feature values ​​are classified in advance according to the value range, and the pictures corresponding to the feature values ​​in the same value range are put into a preset array, and the arrays are named differently. The external data is constructed as a collection type, and the internal data is constructed as json type data composed of several arrays, where each array corresponds to a style.

6. The intelligent clothing design method based on customer preferences according to claim 5 is characterized in that: The feature values ​​are classified in advance according to the value range, the pictures corresponding to the feature values ​​in the same value range are put into a preset array, and the arrays are distinguished by naming, specifically: Obtaining the maximum and minimum values ​​of the eigenvalues, and determining the initial interval corresponding to the eigenvalues; Divide the initial interval into 8 range intervals according to the values, set a corresponding array for each range interval, and name them one by one using binary values ​​"000", "001", "010", "011", "100", "101", "110" and "111"; After deep learning, the eigenvalue corresponding to the current image is obtained, the corresponding eigenvalue range is determined, and the current image is placed in the corresponding array.

7. The intelligent clothing design method based on customer preference according to claim 1, characterized in that: The method based on bubble sorting, obtaining the top n colors most used in the clothing in the plurality of pictures as the colors to be selected, further includes: Based on gender, the colors to be selected are divided into two different categories: male colors and female colors; When customers choose clothing colors, they are directly provided with the color scheme corresponding to their gender for selection.

8. The intelligent clothing design method based on customer preference according to any one of claims 1 to 7, characterized in that: Based on the preset display interface, a search box is provided for the customer to input, the character string input by the customer is split, and the clothing type, color and style of the clothing type to be searched are determined through the split character string. The specific determination method is: The search box provided for the customer to input is limited to the customer to first input a five-digit binary value, the five-digit binary value is split into a two-digit binary value and a three-digit binary value, the clothing type searched by the customer is determined based on the two-digit binary value, and the style classification of the clothing type searched by the customer is determined based on the three-digit binary value, wherein the clothing type includes: hats, tops, pants and shoes; After determining the style category of the clothing type searched by the customer based on the three-digit binary value, all styles under the same style category are displayed on the preset display interface in a list display manner for the customer to select; After the customer has selected the style, the numbers of the colors to be selected are provided for the customer to select and color the clothes of the selected style, and the coloring result is used as the final selected clothes.

9. An intelligent clothing design system based on customer preferences, characterized in that: include: The body shape recognition module is used to obtain the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer based on the human-computer interaction mode, set the priority of the parameters, and transfer the parameters to the preset body shape recognition model according to the priority order to identify the body shape type corresponding to the customer; the height, weight, gender, shoulder width, chest circumference, and hip width parameters input by the customer are obtained, and the priority of the parameters is set, and the specific setting method is: setting the height and weight as the first priority parameters; setting the gender as the second priority parameter; setting the shoulder width, chest circumference, and hip width as the third priority parameters; The style acquisition and classification module is used to crawl network tasks based on the big data platform, obtain several pictures corresponding to hats, tops, pants and shoes respectively, and classify the pictures into styles based on the deep learning neural network as the styles to be selected; A module for determining colors to be selected is used to identify the colors in the pictures corresponding to the hat, coat, pants and shoes, and obtain the top n colors used the most in the clothes in the pictures based on a bubble sort method as the colors to be selected, where n is a positive integer; The body 3D model building module is used to perform reverse engineering based on the body type corresponding to the customer to obtain the body 3D model corresponding to the customer; the reverse engineering directly uses the height, weight, shoulder width, chest circumference, and hip width input by the customer as reverse engineering parameters to obtain a simple body 3D model, eliminating the accuracy requirements for equipment and parameters during accurate modeling; The clothing selection module is used to provide a search box for customers to input based on a preset display interface, split the string input by the customer, determine the type of clothing to be searched, the color and style of the clothing type through the split string, and preview the effect of the clothing selected by the customer on the three-dimensional body model; The fabric cutting selection module is used to display the fabric materials and fabric price information that can be used to cut the garment to the customer for selection on a preset display interface after the customer has determined the selected garment; the clothing price statistics module is used to determine the area of ​​each fabric used, the total purchase price of the fabric and the labor cost based on a preset price statistics algorithm after the customer has determined the selected garment, so as to complete the self-selected clothing design.

Images