CN113362133A

CN113362133A - Underwear customization method and device in mutual error correction mode and electronic equipment

Info

Publication number: CN113362133A
Application number: CN202110613214.6A
Authority: CN
Inventors: 周滢滢
Original assignee: Atog Health Technology Beijing Co ltd
Current assignee: Atog Health Technology Beijing Co ltd
Priority date: 2021-06-02
Filing date: 2021-06-02
Publication date: 2021-09-07

Abstract

The invention provides an underwear customization method and device in a mutual error correction mode and electronic equipment, wherein the method comprises the following steps: acquiring qualitative input of a current user for qualitative parameters of each chest and measurement input for measurement parameters of each chest; correcting the qualitative data of each chest qualitative parameter and the measurement input of each chest measurement parameter based on the mutual error correction constraint relation between each chest qualitative parameter and each chest measurement parameter to obtain the chest data of the current user; underwear customization is performed based on the chest data. The method, the device and the electronic equipment provided by the invention can correct the qualitative input and the measurement input obtained by the user through self-measurement, ensure the reliability and the accuracy of the chest data recommended by the underwear model, and provide accurate and reliable underwear customization service for the user.

Description

Underwear customization method and device in mutual error correction mode and electronic equipment

Technical Field

The invention relates to the technical field of computers, in particular to an underwear customizing method and device in a mutual error correction mode and electronic equipment.

Background

Compared with the characteristic that the plate making of the underwear is unified and only the size is distinguished, the customized underwear follows the principle that one person can make one plate, and the underwear which fits the body type of the user is provided for the user.

Currently, underwear customization usually requires a user to go to a store for professional measurement, the user is required to be provided with the store in a city, and the user himself/herself has time to go to the measurement, and requirements for store distribution and user time exist.

In response to this problem, on-line measurement has been carried out. The on-line measuring body has no limitation on the position of the user, and the user only needs to finish the measuring body by himself and input the data obtained by the measuring body into the customized underwear system. However, the volume data required for underwear customization is extremely complex, and the user may not know the parameters to be measured, so that the wrong data is measured and selected, and the reliability of the underwear type recommended according to the data is directly influenced.

Disclosure of Invention

The invention provides an underwear customizing method and device in a mutual error correction mode and electronic equipment, which are used for solving the problem of low underwear type recommendation reliability caused by inaccurate data of a user's own volume.

The invention provides an underwear customization method under a mutual error correction mode, which comprises the following steps:

acquiring qualitative input of a current user for qualitative parameters of each chest and measurement input for measurement parameters of each chest;

correcting errors of the qualitative data of the chest qualitative parameters and the measurement input of the chest measurement parameters based on the mutual error correction constraint relation between the chest qualitative parameters and the chest measurement parameters to obtain the chest data of the current user, wherein the chest data comprises the chest measurement data and the chest qualitative data, and the mutual error correction constraint relation is determined based on the chest data of the historical user;

underwear customization is performed based on the chest data.

The invention provides an underwear customizing device under a mutual error correction mode, which comprises:

the input acquisition unit is used for acquiring qualitative input of a current user for qualitative parameters of each chest and measurement input for measured parameters of each chest;

the input error correction unit is used for correcting the qualitative data of each chest qualitative parameter and the measurement input of each chest measurement parameter based on the mutual error correction constraint relation between each chest qualitative parameter and each chest measurement parameter to obtain the chest data of the current user, wherein the chest data comprises the chest measurement data and the chest qualitative data, and the mutual error correction constraint relation is determined based on the chest data of the historical user;

a recommending unit for customizing underwear based on the chest data.

The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the underwear customization method in any one of the above mutual error correction modes.

The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method for customizing an undergarment in the mutual error correction mode as described in any of the above.

According to the underwear customizing method and device in the mutual error correction mode and the electronic equipment, the qualitative input and the measurement input obtained by the user through self-measurement are corrected by excavating the mutual error correction constraint relation between the qualitative parameters of the breasts and the measurement parameters of the breasts, so that the reliability and the accuracy of the chest data for recommending the underwear format are ensured, the online measurement can be customized, the time cost of purchasing underwear by the user is reduced, the accurate and reliable underwear format recommending service can be provided for the user, and the popularization of online underwear customization is facilitated.

Drawings

In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.

FIG. 1 is a flow chart of the underwear customization method in the mutual error correction mode provided by the invention;

FIG. 2 is a flow chart of an input error correction method provided by the present invention;

FIG. 3 is a schematic diagram of chest measurement parameters provided by the present invention;

FIG. 4 is a second schematic diagram of chest measurement parameters provided by the present invention;

FIG. 5 is a third schematic diagram of chest measurement parameters provided by the present invention;

FIG. 6 is a fourth schematic diagram of chest measurement parameters provided by the present invention;

FIG. 7 is a schematic structural view of an undergarment customizing apparatus in a mutual error correction mode provided by the present invention;

fig. 8 is a schematic structural diagram of an electronic device provided in the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Because the volume data required by underwear customization are extremely complex, the user can measure and select wrong data because the parameters required to be measured are not known when measuring the body online, and the reliability of the underwear type recommended according to the data is directly influenced. To solve this problem, the present invention provides an underwear customization method in a mutual error correction mode. Fig. 1 is a schematic flow chart of an underwear customization method in a mutual error correction mode provided by the present invention, as shown in fig. 1, the method includes:

step 110, obtaining the qualitative input of the current user for each chest qualitative parameter, and the measurement input for each chest measurement parameter.

The current user referred to herein is the user who needs to make underwear style recommendations.

The chest qualitative parameter can qualitatively reflect the chest state, such as any one or more of chest shape, chest prolapse condition, chest expansion condition, axillary flesh, chest softness, fault condition, milk space condition, milk root dimension condition, breast axillary adhesion condition, and breast compression deformation.

The chest measurement parameter may be any one or more of left and right breast inner radius vertical data, left and right breast inner radius close data, left and right breast outer radius vertical data, left and right breast outer radius close data, left and right breast upper radius vertical data, left and right breast upper radius close data, left and right breast lower radius vertical data, left and right breast lower radius close data, left and right breast height data, bi-breast nipple distance data, left and right breast lock center distance, left and right breast shoulder distance, lower chest circumference data, and upper chest circumference data, which reflect the chest state.

When the user customizes the underwear on line, the user can automatically perform qualitative judgment on the chest state of the user, and inputs the result obtained by the judgment into the underwear customizing system, namely the qualitative input aiming at the qualitative parameters of each chest. The user's input operation is performed for each chest qualitative parameter, and the qualitative input obtained by this operation corresponds to the chest qualitative parameter one by one.

In addition, the user can measure the chest measurement parameters according to the parameter description or measurement guide of the chest measurement parameters, and inputs the measured result into the underwear customization system, namely the measurement input aiming at the chest measurement parameters. The input operation by the user is performed for each chest measurement parameter, and the measurement input obtained by this operation corresponds to the chest measurement parameter one by one.

For example, aiming at the chest qualitative parameter of 'chest type', the underwear customization system can provide three options of 'Asia chest', 'Europe and America chest' and 'Lamei chest' for the current user, and the current user can click any one of the options according to the judgment of the current user to serve as the qualitative input aiming at the 'chest type'. For another example, for the chest measurement parameter "left breast inner radius vertical data", the underwear customization system may provide the current user with a text box for entering "left breast inner radius vertical data", into which the current user may enter the current user's measurement results directly as a measurement input of "left breast inner radius vertical data".

And step 120, based on the mutual error correction constraint relation between each chest qualitative parameter and each chest measurement parameter, performing error correction on the qualitative data of each chest qualitative parameter and the measurement input of each chest measurement parameter to obtain the chest data of the current user, wherein the chest data comprises the chest measurement data and the chest qualitative data, and the mutual error correction constraint relation is determined based on the chest data of the historical user.

Specifically, the chest measurement parameter and the chest qualitative parameter respectively reflect the chest state of the user from the quantitative and qualitative aspects, wherein the single parameter can reflect the condition of a certain aspect. For the overall chest state, mutual constraints and associations exist among various aspects, for example, the chest type "asian chest" is a chest type which is flat in visual inspection, relatively dispersed and similar to a poached egg in shape, and the chest type is generally not easy to generate cleavage, so when the chest type is selected as the "asian chest" and the chest softness is selected as the "compact" chest type, the selectable breast compression deformation generally comprises "cleavage can be generated by exerting force, no cleavage can be generated by exerting force" and "distance between breasts still exists after exerting force", and the breast compression deformation option "instant cleavage" can be directly excluded. For another example, the breast qualitative parameter "breast shape" is a ramen breast shape, which is more prominent and taller than asian breast shape and the euro-american breast shape, and has a shape similar to that of a pineapple breast shape, and usually the breast height of the ramen breast shape is not less than 10 cm, that is, when the breast qualitative parameter "breast shape" is the ramen breast shape, the breast measurement parameter "breast height" should be greater than or equal to 10 cm; for another example, when the breast qualitative parameter "breast drop condition" is a drop breast type, there should be a large difference between the breast superior radius and the breast inferior radius, and usually the breast superior radius/the breast inferior radius may be >3, i.e., when the breast qualitative parameter "breast drop condition" is a drop breast type, the ratio between the breast measurement parameters "breast superior radius" and "breast inferior radius" should be greater than 3.

Considering that the user may not know the parameters to be measured when measuring the body online, the self-qualitative judgment and measurement operation is not necessarily standard, and even the error data may be detected, the mutual error correction constraint relationship between the chest qualitative parameters and the chest measurement parameters may be applied, the qualitative input of the chest qualitative parameters judged by the current user and the measurement input of the chest measurement parameters measured by the current user may be checked for errors, at this time, the qualitative input and the measurement input which do not conform to the mutual error correction constraint relationship may be used as the error input caused by the user judgment or measurement error, and the error input may be corrected on the basis, for example, the error input may be directly deleted, or the input which should be corresponded to the error input actually may be deduced based on the mutual error correction constraint relationship and other correct inputs, so as to obtain the result after error correction, i.e. the chest data of the current user. The chest data referred to herein is data after error correction, and specifically includes chest qualitative data after error correction and chest measurement data, where the chest qualitative data includes data corresponding to each chest qualitative parameter, and the chest measurement data includes data corresponding to each chest measurement parameter.

Further, the mutual error correction constraint relationship between the chest qualitative parameters and the chest measurement parameters referred to herein is used to implement mutual error correction between the input of the chest qualitative parameters and the input of the chest measurement parameters, and may specifically include the constraint relationship between the chest qualitative parameters, the constraint relationship between the chest measurement parameters, and the constraint relationship between the chest qualitative parameters and the chest measurement parameters, which is not specifically limited in this embodiment of the present invention. The mutual error correction constraint relation between each chest qualitative parameter and each chest measurement parameter is obtained by performing correlation mining on the chest data of the historical user, wherein the chest data of the historical user comprises the chest qualitative data and the chest measurement data. The chest qualitative data of the historical user comprises data corresponding to chest qualitative parameters of the historical user and is different from qualitative input, the chest qualitative data can be reliable data judged by a typist, similarly, the chest measurement data of the historical user comprises data corresponding to chest measurement parameters of the historical user and is different from measurement input, and the sample chest measurement data can be reliable data measured by the typist.

Before step 120 is executed, the mutual error correction constraint relationship between each chest qualitative parameter and each chest measurement parameter may be obtained, and the specific obtaining manner includes: first, the chest data of the historical user is collected. On the basis, the constraint relation between the chest qualitative parameters and the data corresponding to the chest measurement parameters in the chest data of the historical user is mined, and the constraint relation is used as the mutual error correction constraint relation between the chest qualitative data and the chest measurement data.

At step 130, underwear customization is performed based on the chest data.

Specifically, after accurate and reliable chest measurement data and chest qualitative data are obtained, underwear type recommendation can be performed from the quantitative aspect and the qualitative aspect by combining the chest measurement data and the chest qualitative data, and therefore reliability and accuracy of recommended types are guaranteed.

Further, a recommended mold parameter corresponding to the chest data of the current user can be determined based on the mapping relation between the chest data and the underwear mold parameter, and underwear customization can be performed based on the recommended mold parameter. The mapping is determined based on historical user chest data and its adapted undergarment mold parameters.

Specifically, in consideration of the fact that different people have great differences in body types, compared with the scheme that the traditional underwear is made into a plate in a unified mode and only distinguished on the size, the underwear is split into the multiple components based on the concept of one person and one plate, and the corresponding die parameters are set for the components, so that the underwear fits the body types of the users on a finer granularity, and the reasonability and comfortableness of the recommended underwear type are guaranteed. The underwear components comprise cups, steel supports, hearts, rear ratios, earrings, oxter pieces, shoulder straps and the like, correspondingly, the underwear die parameters comprise cup parameters, steel support parameters, hearts parameters, rear ratios, earring parameters, oxter piece parameters, shoulder strap parameters and the like, wherein the cup parameters can comprise sizes, softnesses, thicknesses, materials, shapes and the like of the cups, the steel support parameters can comprise sizes, softnesses, materials and the like of the steel supports, the heart parameters can comprise sizes, widths and the like of the hearts, the rear ratios can comprise sizes, shapes and the like of the rear ratios, the earrings and the oxter piece parameters can be the presence or absence of the earrings and the oxter pieces and the like, and the shoulder strap parameters can be the widths, the lengths, the lifting and pulling point positions of the shoulder straps and the like.

After the chest data of the current user is obtained, the chest data of the current user can be mapped to the corresponding underwear die parameters based on the mapping relation between the chest data and the underwear die parameters which are obtained in advance, so that the underwear die parameters which can be recommended to the current user, namely the recommended die parameters, are obtained.

Compared with a mode of singly applying chest measurement data or chest qualitative data to recommend underwear formats, the underwear format recommendation method based on the chest measurement data and the underwear format recommendation system based on the chest measurement data and the chest qualitative data can comprehensively cover chest states and improve reliability and accuracy of underwear format recommendation.

The mapping relation between the chest data and the underwear die parameters is obtained by performing correlation mining on the chest data of the historical user and the underwear die parameters matched with the chest data. Before step 130 is executed, the mapping relationship between the chest data and the underwear mold parameters may be obtained, and the specific obtaining manner includes: the method comprises the steps of firstly collecting chest data of historical users and adaptive underwear die parameters thereof, wherein the chest data of the historical users can be obtained by store-to-store volume analysis or the historical users can be obtained by self-volume analysis, the adaptive underwear die parameters of the historical users can be determined by ordering records of the historical users, and likewise, the chest data of the historical users comprise chest measurement data and chest qualitative data of the historical users.

The information of each component of the underwear finally purchased by the historical user can be collated into the parameters of the underwear mould matched with the historical user. On the basis, the mapping relation between the chest data of the historical user and the underwear die parameters matched with the chest data can be mined, and the mapping relation is used as the mapping relation between the chest data and the underwear die parameters.

Furthermore, the mining of the mapping relation can be that the chest data of the historical user and the underwear mold parameters adapted to the chest data are used as training samples of the neural network model, the neural network model is trained, the trained neural network model is used as the embodiment of the mapping relation, the chest data of the current user are correspondingly input into the trained neural network model, and then the recommended mold parameters output by the neural network model can be obtained; the mining of the mapping relationship may also be to perform regression analysis based on the chest data of the historical user and the underwear mold parameters adapted thereto, and use a function obtained by the regression analysis as the mapping relationship, which is not specifically limited in the embodiment of the present invention.

After the recommended mold parameters are obtained, the underwear type obtained based on the recommended mold parameter combination can be recommended to the current user, so that the current user can determine whether to customize the underwear type based on the recommended underwear type, or the current user can provide modification suggestions on the basis of the recommended underwear type. And after the user confirms, customizing the underwear based on the recommended underwear type.

Further, the recommending of the underwear type to the current user may be to splice all underwear components corresponding to the recommended mold parameters into a set of solid underwear for the current user to try on, or to form an underwear three-dimensional model corresponding to the type based on the recommended mold parameters, and to match the underwear three-dimensional model with the human body three-dimensional model of the current user to simulate a wearing effect for reference by the current user, which is not specifically limited in the embodiment of the present invention.

According to the method provided by the embodiment of the invention, the qualitative input and the measurement input obtained by the user through measuring the body are corrected by excavating the mutual error correction constraint relation between the qualitative parameters of each chest and the measurement parameters of each chest, so that the reliability and the accuracy of the chest data for recommending the underwear form are ensured, the online measuring body customization can reduce the time cost of purchasing underwear by the user, meanwhile, the accurate and reliable underwear form recommendation service can be provided for the user, and the popularization of the online underwear customization is facilitated.

Based on the above embodiment, fig. 2 is a schematic flow chart of the input error correction method provided by the present invention, and as shown in fig. 2, step 120 includes:

step 121, correcting the qualitative input of each chest qualitative parameter based on the first constraint relation among the chest qualitative parameters to obtain chest qualitative data of the current user; the first constraint relationship is determined based on the chest qualitative data of the historical user.

Step 122, correcting the measurement input of each chest measurement parameter based on the second constraint relation between each chest qualitative parameter and each chest measurement parameter and the chest qualitative data of the current user to obtain the chest measurement data of the current user; the second constraint relationship is determined based on the historical user's qualitative data of the chest and measured data of the chest.

Specifically, the mutual error correction constraint relationship between each chest qualitative parameter and each chest measurement parameter can be divided into two parts, namely a first constraint relationship and a second constraint relationship, wherein the first constraint relationship is for qualitative analysis and reflects the constraint relationship between each chest qualitative parameter, and the second constraint relationship is for both qualitative analysis and quantitative measurement and reflects the constraint relationship between the chest measurement parameters corresponding to each chest qualitative parameter respectively.

When performing input error correction, error correction may be performed on each qualitative input based on the first constraint relationship to obtain corrected chest qualitative data. On the basis, the second constraint relation and the corrected chest qualitative data are combined to correct errors of all measurement inputs, and accordingly corrected chest measurement data are obtained.

Further, in step 121, the first constraint relationship between the qualitative chest parameters may be a data range corresponding to another qualitative chest parameter when determining data corresponding to the other qualitative chest parameter, or a data range not selectable by the other qualitative chest parameter. The first constraint relation among the chest qualitative parameters is obtained by performing association mining on the chest qualitative data of the historical user.

The first constraint relation among the chest qualitative parameters can be utilized to check the qualitative input of each chest qualitative parameter judged by the current user, at this time, the qualitative input which does not conform to the first constraint relation can be used as the error input caused by the judgment error of the user, and the error input is corrected on the basis, for example, the error input can be directly deleted, or the input which actually corresponds to the chest qualitative parameter corresponding to the error input is deduced based on the first constraint relation and other correct inputs, so that the result after error correction, namely the chest qualitative data of the current user, is obtained.

In step 122, the second constraint relationship between each chest qualitative parameter and each chest measurement parameter may be a data range in which one or more chest measurement parameters associated therewith may correspond or a data range in which one or more chest measurement parameters associated therewith may not be selectable when determining data corresponding to one chest qualitative parameter. And the second constraint relation between each chest qualitative parameter and each chest measurement parameter is obtained by performing correlation mining on the chest qualitative data and the chest measurement data of the historical user.

The second constraint relationship between each chest qualitative parameter and each chest measurement parameter can be applied, the measurement input of each chest measurement parameter obtained by the current user through self-measurement can be checked for errors by matching with the chest qualitative data of the current user obtained in step 121, at this time, the measurement input which does not conform to the second constraint relationship can be used as the error input caused by the user measurement error, and the error input can be corrected on the basis, for example, the error input can be directly deleted, or the input which actually should correspond to the chest measurement parameter corresponding to the error input is deduced on the basis of the second constraint relationship and the chest qualitative data, or the second constraint relationship and other correct inputs, so that the result after error correction, namely the chest measurement data of the current user, is obtained.

Based on any of the above embodiments, step 121 includes:

matching the first constraint relation among the chest qualitative parameters with the qualitative input of the chest qualitative parameters, and determining the matching times and failure times of the qualitative inputs;

determining correct input in each qualitative input based on the matching times and failure times of each qualitative input;

chest qualitative data for the current user is determined based on the correct one of the qualitative inputs.

In particular, the first constraint relation between the chest qualitative parameters may particularly be expressed as a first constraint relation between every two chest qualitative parameters. For a first constraint relationship between any two qualitative chest parameters, the first constraint relationship may be matched with qualitative inputs corresponding to the two, for example, the first constraint relationship between the "chest sag condition" and the "chest softness degree" indicates that "moderate sag" and "severe sag" are common to "very soft" and "soft" chests, and generally cannot occur to "tight" chests, while the qualitative input of the current user for the "chest sag condition" is "severe sag" and the qualitative input for the "chest softness degree" is "tight", i.e., the qualitative inputs of the two do not match with the first constraint relationship, and one matching failure may be counted for each of the qualitative inputs "severe sag" and "tight sag".

Considering that a first constraint relation may exist between one chest qualitative parameter and a plurality of other chest qualitative parameters, when matching, the first constraint relation matching needs to be performed on the qualitative input of the chest qualitative parameter in combination with the qualitative input of the other chest qualitative parameters one by one, and the number of times of participation in matching and the number of times of failure of matching are accumulated. For example, a first constraint relationship exists between the "chest softness degree" and the "chest sagging situation" and between the "breast compression deformation", if the qualitative input of the "chest softness degree" and the qualitative input of the "chest sagging situation" and the "breast compression deformation" are both matched with the corresponding first constraint relationship, the matching frequency of the "chest softness degree" is 2, and the failure frequency is 0; if the qualitative input of the chest softness degree and the qualitative input of the chest sagging situation are matched with the corresponding first constraint relation and the qualitative input of the breast compression deformation is not matched with the corresponding first constraint relation, the matching frequency of the chest softness degree is 2 and the failure frequency is 1; if the qualitative input of the chest softness degree, the qualitative input of the chest sagging condition and the qualitative input of the breast compression deformation are not matched with the corresponding first constraint relation, the matching frequency of the chest softness degree is 2, and the failure frequency is 2.

Considering that the reason for the failure of matching the qualitative input may be an error of the qualitative input itself, and may also be an error of other qualitative inputs associated with the qualitative input, the matching result alone cannot directly locate the wrong qualitative input. For this case, the matching times and the failure times of each qualitative input may be counted, where the failure times may also be understood as the number of other qualitative inputs that do not match the qualitative input, and in a case where the matching times are fixed, the higher the failure times is, the more other qualitative inputs that do not match the qualitative input is, and the higher the probability of the error of the qualitative input itself is. Conversely, if a qualitative input has a high number of matches and a low number of failures, the qualitative input may be considered to have a high probability of failing to match because of the matching qualitative input being incorrect, rather than the problem of the qualitative input itself.

After the matching times and failure times of each qualitative input are obtained, the reason of each matching failure can be analyzed to determine which qualitative inputs are possible to be wrong, so that wrong inputs in the qualitative inputs of the breast qualitative parameters are screened out, and correct inputs are reserved.

After the erroneous input is eliminated, the breast qualitative data of the current user can be determined based on the remaining correct input, for example, the breast qualitative parameters with the erroneous input to be eliminated can be directly left empty, only the correct input is used as the breast qualitative data, and the data actually corresponding to the breast qualitative parameters with the erroneous input can be deduced by combining the first constraint relation among the breast qualitative parameters on the basis of the correct input, so that the breast qualitative data is completed.

According to the method provided by the embodiment of the invention, the error input which does not conform to the first constraint relation is filtered out by counting the matching times and the failure times of each qualitative input, so that the reliability and the accuracy of the chest qualitative data are ensured.

Based on any of the above embodiments, in step 121, determining a correct input in each qualitative input based on the matching times and the failure times of each qualitative input includes:

determining the matching failure probability of the qualitative input to be distinguished based on the matching times and the failure times of the qualitative input to be distinguished, wherein the qualitative input to be distinguished is the undetermined qualitative input with the highest matching times;

if the matching failure probability of the qualitative input to be distinguished is higher than a preset probability threshold value, deleting the qualitative input to be distinguished, updating the matching times and the failure times of the qualitative input having a first constraint relation with the qualitative input to be distinguished, and otherwise, taking the qualitative input to be distinguished as correct input;

and updating the qualitative input to be judged until the failure times of all the remaining qualitative inputs are 0.

Specifically, the qualitative input of each chest qualitative parameter by the current user may be problematic, and each qualitative input needs to be distinguished one by one. When the order of the qualitative input is determined, the matching times can be taken as consideration conditions, and generally, the more the matching times of the qualitative input participating in the matching are, the higher the reliability of the corresponding matching failure probability is. In the embodiment of the invention, when input discrimination is carried out each time, one qualitative input with the highest matching times is selected from all the non-discriminated qualitative inputs as the qualitative input to be discriminated, and the quotient of the failure times and the matching times is used as the matching failure probability of the qualitative input to be discriminated.

After the matching failure probability of the qualitative input to be distinguished is obtained, the matching failure probability can be compared with a preset probability threshold, the preset probability threshold is the minimum matching failure probability of the error input, if the matching failure probability is higher than the preset probability threshold, the reason that the matching failure of the qualitative input to be distinguished is the input error of the user can be considered, at the moment, the qualitative input to be distinguished can be directly deleted, and the matching times and the failure times of the qualitative input having the first constraint relation with the qualitative input to be distinguished are updated.

If the matching failure probability is less than or equal to the preset probability threshold, the reason that the matching of the qualitative input to be distinguished is failed is considered to be that the rest of the qualitative inputs are wrong, and the qualitative input to be distinguished can be determined to be correct input at the moment.

After the qualitative input to be judged is judged, whether the qualitative input to be judged is correct or not, a qualitative input with the highest matching times is selected from all the remaining non-judged qualitative inputs to serve as a new round of qualitative input to be judged. At this time, whether the failure times of all the remaining qualitative inputs are 0 or not needs to be judged, and the condition that all the remaining qualitative inputs are 0 means that all the remaining qualitative inputs conform to the first constraint relation, namely all the remaining qualitative inputs are correct inputs, and the judging process is ended at this time; if not, the result shows that the error input which is not screened still exists in the residual qualitative input, and the next qualitative input to be distinguished is returned and updated.

Based on any of the above embodiments, in step 121, the updating the matching times and failure times of the qualitative input having the first constraint relationship with the qualitative input to be determined includes:

subtracting 1 from the failure times of the first fixed input, wherein the first fixed input is an unidentified fixed input which fails to be matched with the first constraint relation between the first fixed input and the fixed input to be identified;

and adding 1 to the failure times of the second qualitative input, wherein the second qualitative input is an unidentified qualitative input which is successfully matched with the first constraint relation between the second qualitative input and the qualitative input to be identified.

Specifically, for the case that the qualitative input to be determined is an erroneous input, the failure times of the qualitative input having the first constraint relationship with the qualitative input to be determined also need to be updated correspondingly. It can be understood that, under the condition that the qualitative input to be judged is known as the wrong input, if a qualitative input fails to be matched with the qualitative input to be judged, the large probability of the matching failure is caused by the qualitative input to be judged, the probability of correctness of the qualitative input is high, and the failure frequency of the qualitative input can be reduced by 1; if one qualitative input is successfully matched with the qualitative input to be judged, the high matching probability is caused by the fact that both the qualitative input to be judged and the qualitative input have errors, the probability that the qualitative input has errors is high, and the failure times of the qualitative input can be added by 1.

Based on any of the above embodiments, the method for determining the first constraint relationship includes:

based on the chest qualitative data of the historical user, performing correlation mining on each chest qualitative parameter to obtain a correlation qualitative parameter of each chest qualitative parameter;

selecting a first constraint data set of chest qualitative parameters and related qualitative parameters thereof from chest qualitative data of historical users;

a first constraint relationship is established based on the first constraint data set for each chest qualitative parameter and its associated qualitative parameter.

The association mining referred to herein is a data mining approach that can mine implicit relationships between objects from large-scale data. Specifically, in the embodiment of the present invention, the association mining is used to mine an implicit association relationship between chest qualitative parameters from the chest qualitative data of the historical user, so as to obtain the association qualitative parameters of the chest qualitative parameters. Here, for any chest qualitative parameter, its associated qualitative parameter, i.e. other chest qualitative parameters having an implicit association with the chest qualitative parameter, for example, the associated qualitative parameter of the chest qualitative parameter "chest type" includes "breast compression set". Further, the algorithm used for association mining herein may be Apriori algorithm, FP-growth algorithm, or the like.

After obtaining the associated qualitative parameter of each chest qualitative parameter, for any chest qualitative parameter and its corresponding associated qualitative parameter, a data pair of the two may be selected from a large number of chest qualitative data of historical users as a first constraint data set of the two. On the basis, a first constraint relation between the two can be established through modes of neural network training, regression analysis and the like, or the data of each chest qualitative parameter can be used as a participle, the co-occurrence relation between the participles is counted through a word co-occurrence algorithm, and then the co-occurrence relation between the participles is used as the first constraint relation between the chest qualitative parameters.

And finally, integrating the first constraint relations among all the chest qualitative parameters and the related qualitative parameters thereof to serve as the first constraint relation among all the chest qualitative parameters.

For example, for a chest qualitative parameter "chest shape" and its associated qualitative parameter "breast compression shape", the chest shape and breast compression shape of each historical user may be selected from the chest qualitative data of the historical users to construct a first constrained data set of "chest shape" and "breast compression shape". For the first constraint data set of the 'breast shape' and the 'breast compression deformation', a first constraint relation between the 'breast shape' and the 'breast elasticity measured by hands' can be obtained, for example, the 'Asian breast shape' usually corresponds to 'the breast sulcus can be generated only by exerting force', 'the breast sulcus can not be generated even by exerting force' and 'the distance between breasts still exists after exerting force', 'when the Asian breast shape' is matched with 'the instant breast shape' and 'the' very tight breast shape ', the' Lamei breast shape 'usually corresponds to' the instant breast sulcus ',' the Lamei breast shape 'does not generate the breast sulcus even by exerting force' and 'the distance between breasts still exists after exerting force'.

Based on any of the above embodiments, step 122 includes:

matching the chest qualitative data and the measurement input of each chest measurement parameter with a second constraint relation between each chest qualitative parameter and each chest measurement parameter, and determining the matching failure times of each measurement input;

determining correct input in each measurement input based on the matching failure times of each measurement input;

chest measurement data is determined based on the correct one of the measurement inputs.

In particular, the second constraining relationship between the respective breast qualitative parameter and the respective breast measurement parameter may particularly be expressed as a second constraining relationship between each breast qualitative parameter and one or more breast measurement parameters.

For the second constraint relationship between any one chest qualitative parameter and one or more chest measurement parameters, the second constraint relationship may be matched with data corresponding to each parameter, for example, when the second constraint relationship between "chest type" and "breast height" indicates that "chest type" is a ramet type, the chest measurement parameter "breast height" should be greater than or equal to 10 centimeters, and if the chest type of the current user is the ramet type, but the measurement input corresponding to the breast height is only 5 centimeters, the measurement input "breast height" is counted as one matching failure, and the corresponding number of matching failures is increased by 1.

The reason for the failure of matching the measurement input may be that the measurement input itself is wrong, or there may be a mistake in other measurement inputs associated with the measurement input, so that the wrong measurement input cannot be directly located by only one matching result. For this situation, the matching failure times of each measurement input may be counted, and when the matching times are fixed, the higher the matching failure times, the higher the probability that the measurement input itself is wrong is. Conversely, if the number of matches for a measurement input is high and the number of failures is low, it can be assumed that the reason for the failure of the measurement input match is due to other associated measurement input errors rather than the problem of the measurement input itself.

After the matching failure times of the measurement inputs are obtained, the reason of each matching failure can be analyzed, namely which measurement inputs are possible to be wrong, so that wrong inputs in the measurement inputs of the chest measurement parameters are screened out, and correct inputs are reserved.

After the erroneous input is eliminated, the chest measurement data of the current user can be determined based on the remaining correct input, for example, the chest measurement parameters with the erroneous input which are eliminated can be directly left empty, only the correct input is used as the chest measurement data, and the data actually corresponding to the chest measurement parameters with the erroneous input can be deduced by combining the second constraint relation between the qualitative parameters of each chest and the measurement parameters of each chest on the basis of the correct input, so that the chest measurement data can be completed.

According to the method provided by the embodiment of the invention, the error input which does not conform to the second constraint relation is filtered out by counting the matching failure times of each measurement input, so that the reliability and the accuracy of the chest measurement data are ensured.

Based on any of the above embodiments, in step 122, determining a correct input in each measurement input based on the number of times of failure in matching of each measurement input includes:

selecting the measurement input with the highest matching failure times as an error input;

and deleting the error input, and updating the matching failure times of the associated measurement input of the error input until the matching failure times of all the remaining measurement inputs are 0.

In particular, the measurement inputs of the current user for each chest measurement parameter may be problematic per se, and need to be distinguished one by one. When determining the order of the measurement input, the number of matching failures can be taken as a consideration condition, and generally, the more the number of matching failures obtained by statistics after the measurement input participates in the matching, the higher the corresponding probability of matching failures. In the embodiment of the invention, each time the input discrimination is carried out, the measurement input with the highest matching failure frequency in all the non-discriminated measurement inputs can be directly determined as the error input.

After determining the erroneous input, it is also possible to determine, based on the correlation between the chest measurement data indicated in the second constraint relationship, a measurement input associated with the erroneous input, that is, an associated measurement input of the erroneous input. For example, the second constraint relationship between the "breast drop condition" and the "radius above the breast" and the "radius below the breast" indicates that the ratio of the radius above the breast to the radius below the breast for the dropped breast shape should be greater than 3, i.e., there is a correlation between the "radius above the breast" and the "radius below the breast", and the "radius below the breast" associated with the "radius above the breast" can be located at the same time when the "radius above the breast" is determined to be an erroneous input.

After determining the erroneous input, the erroneous input may be directly deleted and the number of matching failures associated with the measurement input may be updated.

After that, whether the matching failure times of all the remaining measurement inputs are all 0 needs to be judged, and the condition that all the remaining measurement inputs are all 0 means that all the remaining measurement inputs conform to the second constraint relation, namely all the remaining measurement inputs are correct inputs, and the judging process can be ended at this moment; if not, the error inputs which are not screened still exist in the rest measurement inputs, and the error inputs are updated based on the matching failure times of the measurement inputs.

Based on any of the above embodiments, in step 122, the updating the number of times of matching failure of the associated measurement input of the erroneous input includes:

determining each match of an erroneous input to an associated measurement input;

if any matching condition is successful, adding one to the matching failure times input by the correlation measurement;

and if any matching condition is matching failure, subtracting one from the matching failure frequency input by the correlation measurement.

Specifically, there is a second constraint relationship between the erroneous input and its associated measurement input and several chest qualitative parameters, and correspondingly, the erroneous input and its associated measurement input may participate in several matches at the same time. The matching condition referred to herein may be a successful matching or a failed matching.

Aiming at any one time of matching with the two simultaneously participating, under the condition that the error input is determined to have problems, if the matching condition is successful, the associated measurement input associated with the error input is indicated to have problems, and the matching failure times of the associated measurement input is correspondingly added with 1; if the matching condition is matching failure, the reason for the matching failure is to input the error, and the number of matching failures of the associated measurement input is reduced by 1 regardless of the associated measurement input.

Based on any of the above embodiments, the method for determining the second constraint relationship includes:

based on chest qualitative data and chest measurement data of a historical user, performing correlation mining on each chest qualitative parameter and each chest measurement parameter to obtain a correlation measurement parameter of each chest qualitative parameter and a correlation measurement parameter of each chest measurement parameter;

selecting a second constraint data set of each chest qualitative parameter and related measurement parameters thereof and a third constraint data set of each chest measurement parameter and related measurement parameters thereof from the chest qualitative data and the chest measurement data of the historical user;

and establishing a second constraint relationship based on the second constraint data set and the third constraint data set.

Specifically, an implicit association relationship between each chest qualitative parameter and each chest measurement parameter and an implicit association relationship between each chest measurement parameter can be mined from the chest qualitative data and the chest measurement data of the historical user, so as to obtain an associated measurement parameter of each chest qualitative parameter and an associated measurement parameter of each chest measurement parameter. Here, for any chest qualitative parameter, its associated measured parameter, i.e. the chest measured parameter having an implicit association with the chest qualitative parameter, for example, the associated measured parameter of the chest qualitative parameter "chest type" includes "breast height"; for any chest measurement parameter, its associated measurement parameter is the other chest measurement parameters that have an implicit association with the chest measurement parameter, such as "radius above the breast" and "radius below the breast".

After obtaining the associated measurement parameters of each chest qualitative parameter, for any chest qualitative parameter and its corresponding associated measurement parameter, a data pair of the chest qualitative parameter and the chest measurement parameter may be selected from the chest qualitative data and the chest measurement data of a large number of historical users, and the selected data pair is used as a constraint data set of the chest qualitative parameter and the chest measurement data, that is, a second constraint set. Similarly, after obtaining the relevant measurement parameters of the chest measurement parameters, for any chest measurement parameter and its corresponding one, a data pair of the chest measurement parameter and its corresponding one may be selected from the chest measurement data of a large number of historical users, and the selected data pair may be used as a third constraint set, which is a constraint data set of the chest measurement parameter and its corresponding one.

On the basis, a second constraint relation between the two can be established through modes of neural network training, regression analysis and the like, or the co-occurrence relation between the participles can be calculated through a word co-occurrence algorithm by taking the data of each parameter as a participle, and then the co-occurrence relation between the participles is taken as the second constraint relation between the parameters.

And finally, integrating second constraint relations among all chest qualitative parameters and the related measured parameters thereof and between all chest measured parameters and the related measured parameters thereof as second constraint relations among all chest qualitative parameters.

Based on any of the above embodiments, the mapping relationship between the chest data and the underwear mold parameters comprises a first mapping relationship, and/or a second mapping relationship and a third mapping relationship;

the first mapping relation is a mapping relation between the combination of the chest measurement data and the chest qualitative data and the parameters of the underwear mold; the second mapping relation is the mapping relation between the chest measurement data and the underwear mould parameters; the third mapping relation is the mapping relation between the chest qualitative data and the underwear mould parameters.

Specifically, considering that the chest data comprises chest measurement data and chest qualitative data which describe chest states from different angles, when the mapping relation between the chest data and underwear mold parameters is mined, the chest measurement data and the chest qualitative data can be used as two independent individuals to respectively mine the mapping relation between the chest measurement data and the underwear mold parameters according to different reflected chest states, and also considering the implicit relation between quantification and the qualitative data, the chest measurement data and the chest qualitative data and the like can be considered together to mine the mapping relation between the chest measurement data and the underwear mold parameters as a whole.

Furthermore, the mapping relation between the chest measurement data and the chest qualitative data and the underwear mold parameters is mined as a whole, namely the first mapping relation is obtained. And respectively mining the mapping relation between the chest measurement data and the parameters of the underwear mould by taking the chest measurement data and the chest qualitative data as two independent individuals, namely respectively obtaining a second mapping relation and a third mapping relation.

Accordingly, the mapping relationship between the chest data and the underwear mold parameters can be divided into three cases, one is to only include the first mapping relationship, i.e. to analyze the mapping relationship between the chest data and the underwear mold parameters from the overall angle, the other is to only include the second mapping relationship and the third mapping relationship, i.e. to analyze the mapping relationship between the chest measurement data and the chest qualitative data and the underwear mold parameters from the independent angles, and the other is to include both the first mapping relationship and the second mapping relationship and the third mapping relationship, i.e. to analyze from the overall angle and analyze from the independent angles.

It should be noted that, when multiple mapping relationships exist simultaneously, the underwear mold parameters obtained by applying different mapping relationships may be different, and at this time, the underwear mold parameters obtained under various mapping relationships may be integrated to obtain the final recommended mold parameters, for example, the underwear mold parameters under various mapping relationships may be weighted based on weights preset for various mapping relationships.

Based on any of the above embodiments, the method for determining the first mapping relationship includes:

based on the chest data of the historical user and the underwear die parameters matched with the chest data, carrying out association mining on each chest measurement data, each chest qualitative data and each underwear die parameter to obtain the associated chest data of each underwear die parameter;

selecting mapping data sets of various underwear die parameters and relevant chest data thereof from the chest data of the historical user and the underwear die parameters matched with the underwear die parameters;

and establishing a first mapping relation based on the mapping data set of each underwear mould parameter and the associated chest data thereof.

Specifically, in the embodiment of the invention, the association mining is used for mining the implicit association relationship between each chest data and each underwear mold parameter from the chest data of the historical user and the underwear mold parameters adapted to the chest data, so as to obtain the associated chest data of each underwear mold parameter.

Here, for any underwear mold parameter, the associated chest data, that is, the chest data having an implicit association relationship with the underwear mold parameter, may be chest measurement data, chest qualitative data, or a combination of chest measurement data and chest qualitative data. For example, the associated chest data for the undergarment mold parameters "cup shape" includes chest measurement data "breast height," chest qualitative data "chest shape.

After obtaining the undergarment mold parameters and the associated chest data thereof, the associated chest data may be regarded as a whole for any undergarment mold parameter and the associated chest data corresponding thereto, and a data pair of the undergarment mold parameter and the chest data associated therewith may be selected from a large amount of historical data as a mapping data set of the two. On the basis, the mapping relation between the two can be established through the modes of neural network training, regression analysis and the like.

And finally, integrating the mapping relation between all the underwear mould parameters and the associated chest data thereof to serve as a first mapping relation.

For example, for the combination of the underwear mold parameter "cup shape" and its associated breast data "breast height, breast type", the breast height data, breast type data and its cup shape of the chosen underwear of each historical user can be selected from the breast data of the historical user and its adapted underwear mold parameter to construct a mapping data set of the underwear mold parameter "cup shape" and its associated breast data. Aiming at the mapping data set of the cup shape and the associated breast data, the mapping relation between the breast height, the breast shape and the cup shape can be obtained, namely the value range of the breast height and the selection range of the breast shape corresponding to each cup shape.

Based on any of the above embodiments, step 130 includes:

and determining recommended mold parameters corresponding to the chest data of the current user based on the mapping relation corresponding to the user type to which the current user belongs, and customizing underwear based on the recommended mold parameters.

The mapping relation corresponding to the user type to which the current user belongs is determined based on chest measurement data of historical users belonging to the same user type as the current user and underwear mold parameters matched with the chest measurement data, and the user type is determined based on at least one of user age, fertility condition, nursing condition and wearing habits of the corresponding user.

Specifically, different users may have different requirements for the function and wearing experience of underwear, for example, the preferences of the users themselves may be different, some users prefer the underwear of the body-shaping adjustment type, some users pursue the comfort of the underwear, prefer the underwear without steel rings, and for example, whether the users have a child or not and whether the users breast feed or not can affect the chest state of the users, thereby affecting the functional requirements of the users for the underwear.

In consideration of the situation, the users can be classified in advance, and the corresponding mapping relation is set for different user types, so that more suitable underwear styles can be recommended for different types of users in a more targeted manner. The user type referred to herein may be divided according to user preference, such as preference of comfortable users, preference of body-shaping users, preference of balance users, or may be divided according to user body status, such as lactation users, pregnancy users, and the like, which is not specifically limited in this embodiment of the present invention.

Correspondingly, the mapping relation of different user types is constructed only based on the chest data of each historical user under the corresponding user type and the underwear die parameters matched with the chest data, so that the obtained mapping relation can fully reflect the preference of the corresponding user type and can more easily meet the requirements of the corresponding user type.

Further, the method for determining the user type to which the current user belongs comprises the following steps:

acquiring user information of a current user, wherein the user information comprises at least one of user age, fertility status, lactation status and wearing habits;

and determining the user type of the current user based on the user information of the current user.

Specifically, the user information includes information related to the chest condition and/or underwear function requirement of the user, wherein the age of the user may be a specific age of the user or an age group to which the user belongs, the birth status may be whether the user has been born or not, or the number of times of the user has been born, the birth status may be whether the user has been born or not, whether the user is currently in the lactation period or not, or the time for finishing the lactation period, and the wearing habits may include the size of underwear which is frequently worn, whether the user often wears steel-ring-free underwear, whether the user often wears body-shaping underwear, like the width of shoulder straps, like the width of row buckles, like the degree of tightness, like the thickness of cups, and the like.

By comprehensively considering the user information, the chest condition of the user and the corresponding underwear function requirements can be comprehensively known, so that the user classification is realized. Here, the specific classification manner may be to collect user information of a large number of sample users in advance, perform user classification on the part of sample users manually to obtain a sample user type of each sample user, and then train a user classification model using the user information of the sample user and the sample user type as training samples to realize classification of each subsequent user, or may also set a classification rule for each item of user information, and perform user classification based on the classification rule, which is not specifically limited in the embodiment of the present invention.

Based on any of the above embodiments, the method for determining the user type to which the current user belongs includes:

encoding at least one of user age, fertility status, lactation status and wearing habits of the current user to obtain portrait characteristics of the current user;

calculating the similarity between the portrait characteristics of the current user and the type characteristics of each candidate user type, and taking the candidate user type corresponding to the type characteristic with the highest similarity as the user type of the current user;

the type feature of each candidate user type is determined based on each user cluster obtained by carrying out unsupervised clustering on the portrait features of each sample user.

Specifically, the user information includes information of each dimension, and user classification directly performed from the dimensions of age group, birth or lactation, and the like will generate a large number of user types, some of the user types may be relatively close to each other, and the large number of user types will cause a sudden increase in calculation amount when the mapping relationship is obtained.

Aiming at the problem, the portrait characteristics of a large number of sample users can be subjected to unsupervised clustering, so that a plurality of user types which are relatively reliable and have distinguished shapes are obtained, and the calculation amount is effectively reduced while the accurate classification of the users is ensured. The portrait features are obtained by coding user information of the application user, and the unsupervised clustering can be realized by a K-means algorithm, a DBSCAN algorithm and other clustering algorithms.

After the portrait features of the sample user are clustered, a plurality of clusters can be obtained, each cluster corresponds to one user type, and the mean value of all portrait features in a single cluster can be used as the clustering center of the cluster, namely the type feature corresponding to the user type.

On the basis, aiming at the current user, the user information of the current user can be coded to serve as the portrait characteristics of the current user, and the similarity between the portrait characteristics and the type characteristics of each candidate user type is calculated, so that the candidate user type with the highest similarity is selected to serve as the user type of the current user. Here, the similarity between the image feature and the type feature may be calculated by a common similarity calculation method such as euclidean distance and cosine similarity.

According to the method provided by the embodiment of the invention, each candidate user type is determined through unsupervised clustering, so that the reliability of user type division is ensured while massive calculation caused by fine-grained classification is avoided; on the basis, the type of the current user is divided through similarity calculation, so that the accuracy of user classification is guaranteed, and the pertinence and the reliability of underwear type recommendation are improved.

Based on any of the above embodiments, step 130 further includes:

receiving feedback information of a current user;

based on the feedback information, adjusting the recommended mold parameters of the current user;

and updating the mapping relation based on the chest data of the current user and the adjusted recommended mold parameters.

Specifically, the feedback information is a feedback suggestion given by the current user for the recommended underwear form, the current user may return the feedback information after the fitting is completed, or the feedback information may return according to the simulated wearing effect of the recommended underwear form, which is not specifically limited in the embodiment of the present invention.

The feedback information carries the problems of the recommended underwear model indicated by the current user and/or adjustment suggestions aiming at the underwear model, such as 'chest pressing' and 'shoulder belt growing', and for example 'changing into a soft cup', and after the feedback information is received, the recommended mold parameters can be adjusted according to the problems or suggestions contained in the feedback information, so that the underwear model formed by the adjusted recommended mold parameters can better meet the requirements of the current user.

On the basis, the chest data of the current user and the adjusted recommended mold parameters can be used as a group of chest data of the historical user and the matched underwear mold parameters of the historical user, and the chest data and the adjusted recommended mold parameters are applied to updating of the mapping relation between the chest data and the underwear mold parameters, so that the mapping relation can be updated in real time in the underwear form recommending process, and the recommended underwear form can be more fit with the body form of the user.

Based on any of the above embodiments, fig. 3, 4, 5, and 6 are schematic diagrams of chest measurement parameters provided by the present invention, and the dashed lines in the diagrams represent the contours of the breasts. Fig. 3 and 4 reflect right and left milk inner radius vertical data, right and left milk inner radius fit data, right and left milk outer radius vertical data, right and left milk outer radius fit data, right and left milk upper radius vertical data, right and left milk upper radius fit data, right and left milk lower radius vertical data, right and left milk lower radius fit data, where a1 is milk inner radius vertical, a2 is milk outer radius vertical, a3 is milk upper radius vertical, a4 is milk lower radius vertical, b1 is milk inner radius fit, b2 is milk outer radius fit, b3 is milk upper radius fit, and b4 is milk lower radius fit. For example, the left and right milk inner radius vertical data may include a measurement of the left milk inner radius vertical a1 and a measurement of the right milk inner radius vertical a1, the left and right milk inner radius hugging data may include a measurement of the left milk inner radius hugging b1 and a measurement of the right milk inner radius hugging b1, the difference in the measurements of vertical and hugging may be seen in fig. 4. Fig. 5 shows left and right breast height data, fig. 6 shows data of the distance between the nipples of both breasts, the distance between the breast centers of the left and right breasts, and the distance between the breast shoulders of the left and right breasts, wherein the distance d1 between the breasts and the breast center is measured, the distance d2 between the breast centers is the linear distance from the clavicle groove to the breast point, the distance d3 between the breast point and the breast shoulder is vertical, the breast height h is the breast height, and the tape is placed horizontally and is vertical to the body. The upper chest circumference data is the measurement result of the fact that the body is inclined forward by 45 degrees, the circumference is measured horizontally for one circle, and the tape passes through the breast point; the lower chest circumference data is the measurement result of raising head, straightening chest, measuring horizontal circumference for a circle and passing through the lower breast root point by a tape.

Based on any of the above embodiments, the breast qualitative parameter includes at least one of breast shape, breast ptosis condition, breast flaring condition, axillary flesh, breast softness, fault condition, breast spacing condition, breast root dimension condition, breast axillary adhesion condition, and breast compression set.

Specifically, the chest shape can be an Asian chest shape, an European chest shape or a Laume chest shape, wherein the Asian chest shape is a chest shape which is flat and relatively scattered visually and is similar to a poached egg in shape, the European chest shape is relatively mellow and is similar to a steamed bun and approximately hemispherical, and the Laume chest shape is a chest shape which is more prominent and taller visually and is similar to a pineapple shape compared with the Asian chest shape and the European chest shape.

The chest sagging situation can be one of standard, slight sagging, moderate sagging and severe sagging, when the chest sagging situation is visually observed, a horizontal line 1 can be set by taking a nipple in the standard situation as a reference, a horizontal line 2 is set by taking a lowest point of a lower breast root in the standard situation as a reference, a specific chest sagging situation is determined according to a relative position relationship between a nipple of a user and two horizontal lines in an actual situation, the nipple is standard when being above the horizontal line 1, is slightly sagging when being between the horizontal lines 1 and 2, is moderate sagging when being below the horizontal line 2 but being closer to the horizontal line 2, and is severe sagging when being below the horizontal line 2 and being farther from the horizontal line 2.

The chest expansion condition can be one of normal chest type, slight expansion, whole expansion and severe expansion, when the chest expansion condition is visually observed, if the nipple is seen from right in front of a human body and is positioned at the left and right middle positions of the chest, the chest is normal, if two nipples are deviated to two sides of the body, the nipple is slightly expanded, if the breast is wholly deviated to two sides of the body, a triangular flat area is arranged between the breasts, the chest is wholly expanded, and if the breast is obviously deviated to two sides of the body, the interval between the breasts with more than two common fingers is determined as the severe expansion.

The axillary proud flesh may be one of skinned flesh, obvious flesh, diffuse flesh and non-skinned flesh, wherein skinned flesh refers to the condition of slight proud flesh, obvious flesh refers to the condition of visible excess proud flesh with naked eyes, and diffuse flesh refers to the condition of visible proud flesh, common dispersed flesh and relatively large area.

The chest softness degree can be one of very soft, firmer and firmer, wherein the very soft is the chest of a female who usually has a birth and a breast feeding, and skin and flesh are separated frequently; the breast mask is soft, is usually found in the breast of a female who does not nurse, and is easy to deform; the breast milk is compact and common to breast-feeding women, the breast milk has compact and elastic hand feeling, compact skin and flesh, no separation and strong integrity; the breast is compact, is usually in the female breast which is fond of body building and has low body fat content, and is not easy to deform.

The fault condition may be one of no fault, one finger fault, two finger fault, three finger fault and above fault, and is used for representing the fault condition of the upper part of the breast.

The milk-to-milk distance condition may be one of no distance, one-finger wide distance, two-finger wide distance, three-finger and above distances, for characterizing the distance between two breasts.

The breast root dimension condition can be one of full and clear, basically unclear, side lack and only lower periphery line, and can be obtained by judging the breast root contour of the chest side through visual inspection.

The axillary adhesion of the breast can be one of non-adhesion, slight adhesion and characteristic adhesion, and the axillary adhesion of the breast can be obtained by visually observing the adhesion of fat of the breast on the lateral side of the chest to the axilla.

The breast compression deformation reflects the breast elasticity, and can be one of the instant cleavage, cleavage only after applying force, cleavage not after applying force, and distance between breasts after applying force.

Based on any of the above embodiments, fig. 7 is a schematic structural diagram of an underwear customizing device in the mutual error correction mode provided by the present invention, as shown in fig. 7, the device includes:

an input acquisition unit 310 for acquiring qualitative input of a current user for each chest qualitative parameter and measurement input for each chest measurement parameter;

an input error correction unit 320, configured to perform error correction on the qualitative data of each chest qualitative parameter and the measurement input of each chest measurement parameter based on a mutual error correction constraint relationship between each chest qualitative parameter and each chest measurement parameter, to obtain the chest data of the current user, where the chest data includes chest measurement data and chest qualitative data, and the mutual error correction constraint relationship is determined based on the chest data of the historical user;

a recommending unit 330 for underwear customization based on the chest data.

According to the device provided by the embodiment of the invention, the qualitative input and the measurement input obtained by the user self-measuring body are corrected by excavating the mutual error correction constraint relation between the qualitative parameters of each chest and the measurement parameters of each chest, so that the reliability and the accuracy of the chest measurement data for recommending the underwear model are ensured, the online measuring body customization can reduce the time cost of purchasing underwear by the user, meanwhile, the accurate and reliable underwear model recommendation service can be provided for the user, and the popularization of the online underwear customization is facilitated.

Fig. 8 illustrates a physical structure diagram of an electronic device, and as shown in fig. 8, the electronic device may include: a processor (processor)410, a communication Interface 420, a memory (memory)430 and a communication bus 440, wherein the processor 410, the communication Interface 420 and the memory 430 are communicated with each other via the communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform an undergarment customization method in a mutual error correction mode, the method comprising: acquiring qualitative input of a current user for qualitative parameters of each chest and measurement input for measurement parameters of each chest; correcting errors of the qualitative data of the chest qualitative parameters and the measurement input of the chest measurement parameters based on the mutual error correction constraint relation between the chest qualitative parameters and the chest measurement parameters to obtain the chest data of the current user, wherein the chest data comprises the chest measurement data and the chest qualitative data, and the mutual error correction constraint relation is determined based on the chest data of the historical user; underwear customization is performed based on the chest data.

In addition, the logic instructions in the memory 430 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, which when executed by a computer, enable the computer to perform the underwear customization method in the mutual error correction mode provided by the above methods, the method comprising: acquiring qualitative input of a current user for qualitative parameters of each chest and measurement input for measurement parameters of each chest; correcting errors of the qualitative data of the chest qualitative parameters and the measurement input of the chest measurement parameters based on the mutual error correction constraint relation between the chest qualitative parameters and the chest measurement parameters to obtain the chest data of the current user, wherein the chest data comprises the chest measurement data and the chest qualitative data, and the mutual error correction constraint relation is determined based on the chest data of the historical user; underwear customization is performed based on the chest data.

In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor, implements a method for customizing an undergarment in the mutual error correction mode provided above, the method including: acquiring qualitative input of a current user for qualitative parameters of each chest and measurement input for measurement parameters of each chest; correcting errors of the qualitative data of the chest qualitative parameters and the measurement input of the chest measurement parameters based on the mutual error correction constraint relation between the chest qualitative parameters and the chest measurement parameters to obtain the chest data of the current user, wherein the chest data comprises the chest measurement data and the chest qualitative data, and the mutual error correction constraint relation is determined based on the chest data of the historical user; underwear customization is performed based on the chest data.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. An underwear customization method under a mutual error correction mode, characterized by comprising:

underwear customization is performed based on the chest data.

2. The method for customizing an undergarment in the mutual error correction mode according to claim 1, wherein the error correction of the qualitative data of each chest qualitative parameter and the measurement input of each chest measurement parameter based on the mutual error correction constraint relationship between each chest qualitative parameter and each chest measurement parameter to obtain the chest data of the current user comprises:

correcting the qualitative input of each chest qualitative parameter based on a first constraint relation among the chest qualitative parameters to obtain the chest qualitative data of the current user; the first constraining relationship is determined based on chest qualitative data of a historical user;

correcting errors of the measurement input of the chest measurement parameters based on second constraint relations between the chest qualitative parameters and the chest measurement parameters and the chest qualitative data of the current user to obtain the chest measurement data of the current user; the second constraint relationship is determined based on historical user's chest qualitative data and chest measurement data.

3. The method of claim 2, wherein the error correction of the qualitative input of each chest qualitative parameter based on the first constraint relationship between the chest qualitative parameters to obtain the chest qualitative data of the current user comprises:

determining chest qualitative data for the current user based on correct ones of the qualitative inputs.

4. The method for customizing an undergarment in the mutual error correction mode as claimed in claim 3, wherein the determining the correct input among the qualitative inputs based on the number of matching and the number of failure of the qualitative inputs comprises:

determining the matching failure probability of the qualitative input to be distinguished based on the matching times and failure times of the qualitative input to be distinguished, wherein the qualitative input to be distinguished is an unidentified qualitative input with the highest matching times;

if the matching failure probability of the qualitative input to be distinguished is higher than a preset probability threshold, deleting the qualitative input to be distinguished, updating the matching times and the failure times of the qualitative input having a first constraint relation with the qualitative input to be distinguished, and otherwise, taking the qualitative input to be distinguished as correct input;

and updating the qualitative input to be distinguished until the failure times of all the remaining qualitative inputs are 0.

5. The method for customizing an undergarment in the mutual error correction mode as claimed in claim 2, wherein the error correction of the measurement input of the chest measurement parameters based on the second constraint relationship between the chest qualitative parameters and the chest measurement parameters and the chest qualitative data of the current user to obtain the chest measurement data of the current user comprises:

determining the chest measurement data based on a correct one of the measurement inputs.

6. The method for customizing an undergarment in the mutual error correction mode according to claim 5, wherein the determining the correct input among the measurement inputs based on the number of matching failures of the measurement inputs comprises:

7. The undergarment customization method in the mutual error correction mode according to any of claims 1 to 6, wherein the underwear customization based on the chest data comprises:

determining recommended mold parameters corresponding to the chest data of the current user based on a mapping relation between the chest data and the underwear mold parameters, wherein the mapping relation is determined based on the chest data of historical users and the underwear mold parameters matched with the historical users;

customizing underwear based on the recommended mold parameters;

the mapping relation between the chest data and the underwear mold parameters comprises a first mapping relation and/or a second mapping relation and a third mapping relation;

the first mapping relationship is a mapping relationship between the combination of the chest measurement data and the chest qualitative data and the undergarment mold parameters;

the second mapping relation is a mapping relation between the chest measurement data and the underwear mold parameters;

the third mapping relationship is a mapping relationship between the breast qualitative data and the undergarment mold parameters.

8. An undergarment customization device in a mutual correction mode, comprising:

a recommending unit for customizing underwear based on the chest data.

9. An electronic device comprising a memory, a processor and a computer program stored on said memory and executable on said processor, characterized in that said processor, when executing said program, carries out the steps of the underwear customization method in mutual error correction mode according to any one of claims 1 to 7.

10. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the underwear customization method in the mutual error correction mode according to any one of claims 1 to 7.