A goose down custom clothing pattern processing system and method

Abstract

The present invention discloses a goose down custom clothing pattern processing system and method, the system includes a measuring unit, the measuring unit includes a processing module, and a BMI index acquisition module connected to the processing module, a 3D model generation module, a material selection unit, the material selection unit includes a pre-built pattern database, an RGB color selection module and an image synthesis module, a pattern sample production unit, the pattern sample production unit includes a mapping module and a proofing device, a storage unit, and is connected to the 3D model generation module, the image synthesis module and the mapping module. The present invention uses the actual human body shape measurement model and the cloth color and pattern selected by the consumer to generate a two-dimensional pattern, which can reduce the multiple modifications of the consumer to the custom clothing pattern, and the steps are simple.

Description

Technical Field

[0001] The present invention relates to the technical field of clothing customization, and in particular to a goose down customized clothing pattern processing system and method. Background Art

[0002] Goose down clothing is becoming more and more popular due to its excellent warmth retention, lightness, softness, water repellency, bright whiteness, and elegant appearance. Although goose down clothing is a high-end fabric, domestic manufacturers still use the traditional production model, introducing a few patterns and then mass production. This method not only cannot meet the needs of consumers, but also has the problem of product backlog and inventory that cannot be digested.

[0003] The clothing requirements of some high-end people have also changed from mass styles to customization, and at the same time, goose down clothing is required to have good comfort and uniqueness. It is necessary to meet consumers' needs for styles, fabrics, and patterns according to different consumer body shapes and aesthetic characteristics, and realize large-scale personalized customization of goose down clothing. The traditional method of manual private customization is to measure human body size parameters, make corresponding clothing patterns, and adjust the patterns according to experience. However, this operation process is not only cumbersome, but also cannot give high-end consumers an intuitive feeling of style, fabrics, and patterns, resulting in a long production cycle of clothing. If consumers are not satisfied with the finished product, they will return it to the factory. Therefore, a goose down customized clothing pattern processing system and method are proposed. Summary of the invention

[0004] The object of the present invention is to provide a goose down customized clothing pattern processing system and method, aiming to solve the problems mentioned in the background technology.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0006] The present invention provides a goose down custom clothing pattern processing system, the system comprising:

[0007] A measuring unit, the measuring unit comprising a processing module, and a BMI index acquisition module and a 3D model generation module connected to the processing module, wherein the BMI index acquisition module is used to collect the weight and height of the consumer to obtain the BMI index; the 3D model generation module is used to collect the first human body three-dimensional data of the consumer in a standard standing posture and the second human body three-dimensional data in a non-standard standing posture, the processing module is used to screen the first human body three-dimensional data and the second human body three-dimensional data to obtain reference human body three-dimensional data, and based on the pre-constructed measurement model, obtain the final human body three-dimensional data to generate a human body measurement model;

[0008] A material selection unit, the material selection unit comprising a pre-built pattern database, an RGB color selection module and an image synthesis module, wherein the RGB color selection module is used for consumers to select fabric colors according to an RGB color system, and the image synthesis module is used for consumers to generate pre-selected fabrics according to the selected fabric colors and the patterns selected from the pattern database, and store them in a receiving storage unit;

[0009] A pattern sample making unit, the pattern sample making unit comprising a mapping module and a proofing device, wherein the mapping module is used to map the human body measurement model to a two-dimensional plane to obtain a two-dimensional pattern, and input the two-dimensional pattern to the proofing device, and the proofing device is used to make and sew according to the two-dimensional pattern and pre-selected fabric to obtain a goose down customized clothing pattern;

[0010] The storage unit is connected to the 3D model generation module, the image synthesis module and the mapping module, and is used to receive and store the human body measurement model, the pre-selected fabric and the two-dimensional pattern.

[0011] A further improvement is that the standard standing posture is a standing state with the human body's arms naturally hanging down, and the non-standard standing posture includes a standing state with the human body's arms raised to shoulder level, a standing state with the human body's arms raised to perpendicular to the shoulders, a straddled state with the human body's hands behind the back and feet apart, a sitting state with the back straight and legs bent, and a squatting state with the hands on the knees.

[0012] A further improvement is that the three-dimensional human body data includes the following data items: chest circumference, waist circumference, hip circumference, neck circumference, back width, shoulder width, back length, upper arm width, wrist width, upper and lower arm length, waist length, chest depth, hip depth, inner and outer leg length, sitting height, thigh and calf width, crotch depth, and abdominal width.

[0013] A further improvement is that the steps of constructing the measurement model are:

[0014] S1: Collecting BMI indexes and benchmark three-dimensional human body data of different consumers in historical data to obtain an initial data set, and normalizing the initial data set with the final three-dimensional human body data collected in advance to obtain a sample data set;

[0015] S2: Divide the sample data set into a training set and a test set, and import them into the RNN neural network for training verification, adjust the parameters in the RNN neural network, obtain a trained measurement model, and input the consumer's BMI index and benchmark human three-dimensional data collected in real time into the measurement model to output the predicted final human three-dimensional data.

[0016] A further improvement is that the method for obtaining the reference human body three-dimensional data is as follows:

[0017] In a standing state with both arms of a human body hanging naturally, collecting first three-dimensional data of a human body;

[0018] The second human body three-dimensional data is collected respectively in the following situations: a standing state with arms raised to shoulder level, a standing state with arms raised to shoulder vertical, a straddling state with hands behind the back and legs apart, a sitting state with back straight and legs bent, and a squatting state with hands on knees;

[0019] Subtract each item in the second human body three-dimensional data from each item in the first human body three-dimensional data to obtain Xi. When Xi is a positive number, the data item in the second human body three-dimensional data is used. When Xi is a negative number, the data item in the first human body three-dimensional data is used, and finally the screened benchmark human body three-dimensional data is obtained.

[0020] A further improvement is that the construction process of the pre-built database includes the following steps:

[0021] (1) Collect pattern images through different channels, build a pattern database, and unify the size and resolution of the images in the database;

[0022] (2) performing image recognition on each pattern image to obtain all characteristic keywords of the image;

[0023] (3) Name the image with all the characteristic keywords and store the image in the storage space of a single characteristic keyword.

[0024] The present invention also provides a goose down custom clothing pattern processing method, which is applied to the goose down custom clothing pattern processing system, and the steps include:

[0025] S1. Collecting the consumer's BMI index, first human body three-dimensional data in a standard standing posture, and second human body three-dimensional data in a non-standard standing posture through a measuring unit, screening the first human body three-dimensional data and the second human body three-dimensional data to obtain reference human body three-dimensional data, and inputting the reference human body three-dimensional data and the BMI index into a measurement model to obtain final human body three-dimensional data to generate a human body shape measurement model;

[0026] S2, the consumer selects a pattern and a fabric color respectively using the pattern database and the RGB color selection module, and the image synthesis module generates a pre-selected fabric according to the fabric color selected by the consumer and the pattern selected from the pattern database;

[0027] S3. Mapping the human body measurement model to a two-dimensional plane using a mapping module to obtain a two-dimensional pattern, and inputting the two-dimensional pattern into a proofing device, wherein the proofing device is used to make a pattern and sew according to the two-dimensional pattern and pre-selected fabrics to obtain a goose down customized clothing pattern.

[0028] The beneficial effects of the present invention are:

[0029] The two-dimensional pattern generated by using the actual human body measurement model and the fabric color and pattern selected by the consumer can reduce the multiple modifications of the customized clothing pattern by the consumer, and the steps are simple. Furthermore, using the BMI index and the three-dimensional data of the human body, based on the existing RNN neural network, the mathematical relationship between the BMI index and the final three-dimensional data of the human body is obtained, which improves the efficiency and accuracy of the customized clothing pattern customization work, makes the generated clothing more in line with the requirements of consumers, and is conducive to improving the user experience of personalized clothing customization services. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 System schematic diagram of the pattern processing system for custom goose down garments. DETAILED DESCRIPTION

[0031] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0032] like Figure 1 As shown, a goose down custom clothing pattern processing system, the system includes:

[0033] The measuring unit includes a processing module, a BMI index acquisition module connected to the processing module, and a 3D model generation module; wherein,

[0034] The BMI index collection module is used to collect the weight and height of the consumer to obtain the BMI index;

[0035] The 3D model generation module is used to collect the first human body three-dimensional data in the standard standing posture of the consumer and the second human body three-dimensional data in the non-standard standing posture, and the processing module is used to screen the first human body three-dimensional data and the second human body three-dimensional data to obtain the reference human body three-dimensional data, and obtain the final human body three-dimensional data based on the pre-constructed measurement model to generate the human body measurement model;

[0036] Preferably, the standard standing posture is a standing state with the human body's arms hanging naturally, and the non-standard standing posture includes a standing state with the human body's arms raised to shoulder level, a standing state with the human body's arms raised to perpendicular to the shoulders, a straddled state with the human body's hands behind the back and feet apart, a sitting state with the back straight and legs bent, and a squatting state with the hands on the knees.

[0037] Preferably, the three-dimensional human body data includes the following data: chest circumference, waist circumference, hip circumference, neck circumference, back width, shoulder width, back length, upper arm width, wrist width, upper and lower arm length, waist length, chest depth, hip depth, inner and outer leg length, sitting height, thigh and calf width, crotch depth, and abdominal width.

[0038] The 3D model generation module collects first human body three-dimensional data when the human body is in a standing state with arms naturally hanging down; collects second human body three-dimensional data when the human body is in a standing state with arms raised to shoulder level, a standing state with arms raised to shoulder vertical, a straddle state with hands behind the back and legs apart, a sitting state with back straight and legs bent, and a squatting state with hands on knees; subtracts each item in the second human body three-dimensional data from each item in the first human body three-dimensional data to obtain Xi, and when Xi is a positive number, the data item in the second human body three-dimensional data is used, and when Xi is a negative number, the data item in the first human body three-dimensional data is used, and finally the screened benchmark human body three-dimensional data is obtained.

[0039] Preferably, the steps of constructing the measurement model are:

[0040] S1: Collecting BMI indexes and benchmark three-dimensional human body data of different consumers in historical data to obtain an initial data set, and normalizing the initial data set with the final three-dimensional human body data collected in advance to obtain a sample data set;

[0041] S2: Divide the sample data set into a training set and a test set, and import them into an existing RNN neural network for training verification, adjust the parameters in the RNN neural network, and obtain a trained measurement model. The purpose is to obtain the relationship between the BMI index and the benchmark human three-dimensional data and the final human three-dimensional data. The BMI index and benchmark human three-dimensional data of the consumer collected in real time are input into the measurement model to output the predicted final human three-dimensional data.

[0042] A material selection unit, the material selection unit includes a pre-built pattern database, an RGB color selection module and an image synthesis module; wherein:

[0043] The RGB color selection module is used for consumers to select fabric colors according to the RGB color system. For example, consumers can select the color they like by sliding the R channel, the G channel and the B channel in the RGB color system at will. The image synthesis module is used to generate pre-selected fabrics according to the fabric color selected by the consumer and the pattern selected from the pattern database, and store them in the receiving storage unit;

[0044] The process of constructing the pattern database includes the following steps:

[0045] (1) Collect pattern images through different channels, build a pattern database, and unify the size and resolution of the images in the database;

[0046] (2) Perform image recognition on each pattern image to obtain all feature keywords of the image; for example, extract: geometric shapes, Chinese characters, English, flowers, birds, fish, insects, etc.;

[0047] (3) Name the image with all the characteristic keywords and store the image in the storage space of a single characteristic keyword. For example, store the image in the storage space of geometric shapes, Chinese characters, English, flowers, birds, fish, and insects respectively.

[0048] The pattern sample making unit includes a mapping module and a proofing device, wherein the mapping module is used to map the human body measurement model to a two-dimensional plane to obtain a two-dimensional pattern, and input the two-dimensional pattern into the proofing device, and the proofing device is used to make and sew according to the two-dimensional pattern and pre-selected fabric to obtain a goose down customized clothing pattern.

[0049] The system also includes a storage unit connected to the 3D model generation module, the image synthesis module and the mapping module, and is used to receive and store the human body measurement model, pre-selected fabrics and two-dimensional patterns to facilitate consumers to query records.

[0050] The present invention also provides a goose down custom clothing pattern processing method, which is applied to the goose down custom clothing pattern processing system, and the steps include:

[0051] S1. Collecting the consumer's BMI index, first human body three-dimensional data in a standard standing posture, and second human body three-dimensional data in a non-standard standing posture through a measuring unit, screening the first human body three-dimensional data and the second human body three-dimensional data to obtain reference human body three-dimensional data, and inputting the reference human body three-dimensional data and the BMI index into a measurement model to obtain final human body three-dimensional data to generate a human body shape measurement model;

[0052] S2, the consumer selects a pattern and a fabric color respectively using the pattern database and the RGB color selection module, and the image synthesis module generates a pre-selected fabric according to the fabric color selected by the consumer and the pattern selected from the pattern database;

[0053] S3. Mapping the human body measurement model to a two-dimensional plane using a mapping module to obtain a two-dimensional pattern, and inputting the two-dimensional pattern into a proofing device, wherein the proofing device is used to make a pattern and sew according to the two-dimensional pattern and pre-selected fabrics to obtain a goose down customized clothing pattern.

[0054] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A goose down custom clothing pattern processing system, characterized in that: The system includes: A measuring unit, the measuring unit comprising a processing module, and a BMI index acquisition module and a 3D model generation module connected to the processing module, wherein the BMI index acquisition module is used to collect the weight and height of the consumer to obtain the BMI index; the 3D model generation module is used to collect the first human body three-dimensional data of the consumer in a standard standing posture and the second human body three-dimensional data in a non-standard standing posture, the processing module is used to screen the first human body three-dimensional data and the second human body three-dimensional data to obtain reference human body three-dimensional data, and based on the pre-constructed measurement model, obtain the final human body three-dimensional data to generate a human body measurement model; The method of obtaining the final three-dimensional human body data based on the pre-built measurement model to generate a human body measurement model is specifically as follows: S1: Collecting BMI indexes and benchmark three-dimensional human body data of different consumers in historical data to obtain an initial data set, and normalizing the initial data set with the final three-dimensional human body data collected in advance to obtain a sample data set; S2: Divide the sample data set into a training set and a test set, and import them into the RNN neural network for training verification, adjust the parameters in the RNN neural network, obtain a trained measurement model, and input the BMI index and the reference human body three-dimensional data collected in real time into the measurement model to output the predicted final human body three-dimensional data; The method for obtaining the three-dimensional data of the reference human body is as follows: In a standing state with arms of a human body hanging naturally, the 3D model generation module collects first three-dimensional data of the human body; The 3D model generation module collects the three-dimensional data of the second human body in the following situations: a standing state with arms raised to the level of shoulders, a standing state with arms raised to the vertical position of shoulders, a straddling state with hands behind the back and feet apart, a sitting state with back straight and legs bent, and a squatting state with hands on knees; The processing module subtracts each item in the second human body three-dimensional data from each item in the first human body three-dimensional data to obtain Xi, and when Xi is a positive number, the data item in the second human body three-dimensional data is used, and when Xi is a negative number, the data item in the first human body three-dimensional data is used, and finally the filtered reference human body three-dimensional data is obtained; A material selection unit, the material selection unit comprising a pre-built pattern database, an RGB color selection module and an image synthesis module, wherein the RGB color selection module is used for consumers to select fabric colors according to the RGB color system, and the image synthesis module is used to generate pre-selected fabrics according to the fabric colors selected by consumers and the patterns selected from the pattern database, and store them in a receiving storage unit; The pattern sample making unit includes a mapping module and a proofing device, wherein the mapping module is used to map the human body measurement model to a two-dimensional plane to obtain a two-dimensional pattern, and input the two-dimensional pattern into the proofing device, and the proofing device is used to make and sew according to the two-dimensional pattern and pre-selected fabric to obtain a goose down customized clothing pattern.

2. A goose down custom clothing pattern processing system according to claim 1, characterized in that: The standard standing posture is a standing state with the human body's arms hanging naturally, and the non-standard standing posture includes a standing state with the human body's arms raised to shoulder level, a standing state with the human body's arms raised to perpendicular to the shoulders, a straddled state with the human body's hands behind the back and feet apart, a sitting state with the back straight and legs bent, and a squatting state with the human body's hands on the knees.

3. The goose down custom clothing pattern processing system according to claim 2, characterized in that: The three-dimensional human body data includes the following data: chest circumference, waist circumference, hip circumference, neck circumference, back width, shoulder width, back length, upper arm width, wrist width, upper and lower arm length, waist length, chest depth, hip depth, inner and outer leg length, sitting height, thigh and calf width, crotch depth, and abdominal width.

4. The goose down custom clothing pattern processing system according to claim 3, characterized in that: The construction process of the pre-built database consists of the following steps: (1) Collect pattern images through different channels, build a pattern database, and unify the size and resolution of the images in the database; (2) Perform image recognition on each pattern image to obtain all characteristic keywords of the image; (3) Name the image with all the characteristic keywords and store the image in the storage space of a single characteristic keyword.

5. A method for processing a goose down custom clothing pattern, applied to the goose down custom clothing pattern processing system according to claim 4, characterized in that the steps include: S1. Collecting the consumer's BMI index, first human body three-dimensional data in a standard standing posture, and second human body three-dimensional data in a non-standard standing posture through a measuring unit, screening the first human body three-dimensional data and the second human body three-dimensional data to obtain reference human body three-dimensional data, and inputting the reference human body three-dimensional data and the BMI index into a measurement model to obtain final human body three-dimensional data to generate a human body shape measurement model; S2, the consumer selects a pattern and a fabric color respectively using the pattern database and the RGB color selection module, and the image synthesis module generates a pre-selected fabric according to the fabric color selected by the consumer and the pattern selected from the pattern database; S3. Mapping the human body measurement model to a two-dimensional plane using a mapping module to obtain a two-dimensional pattern, and inputting the two-dimensional pattern into a proofing device, wherein the proofing device is used to make and sew patterns according to the two-dimensional pattern and pre-selected fabrics to obtain a goose down customized clothing pattern.

Images